RM503 ongoing repair saga T620 UNWOUND.


Daniel Koller
 

Hi All, and particularly Chuck Harris, John Snyder and Brad Thompson,...
   My RM503 still sits on my workbench, inoperable, but at present, I know it's T620 that is fried, and I have a working spare in hand (thanks Brad).
Before I resort to the spare, I'm going to test my coil winder and re-wind a transformer, but before I do that I UNWOUND my original T620 that had an open in the primary circuit.   I unwound it while taking notes in gory detail, but I have a few questions as a result. 
  First of all, here are the windings, counted one at a time as I unwound the transformer, one end to the other, from the outside in to the core.Coil                    TurnsHV Filament     1HV                     1161 
Primary             60-100V                34 12.6V                5
+100                  29 (on top of the +12.6V, total 34 from ground)+250                  51 (on top of the +100V, total 85 from ground)Primary             60

Now, I can get loopy headed while mind-numbingly counting hundreds of turns by hand, with the wire breaking a dozen or so times during the unwind, so I measured the lengths of each piece of wire I removed from the coil.  Using those lengths, I was able to confirm I got the counts right.  So the HV coil is 1161 turns, +/-1-2 turns maximum error.  
  The low voltage coils are very consistent.  If you assume 2.9412 V/turn, then the +250, +100 and -100 coils are dead on.  But then that means that 1161T * 2.9412V/T = ~-3400V.   This is 400V higher than the nominal -3000V filament supply.  
  Question #1)  Why is that so high?  Is there a huge drop in the 5642 rectifier tube?  Is there reduced flux at the outer ends of the HV coil so that it's not really 2.94V/turn?
  Another very interesting discovery, which I think may have been discussed before but I can't find the message, is that the primary is split.  First, 60 turns are wound around the core, then the +250 coil, then +100, +12.6 and finally -100 comes out on the top.  Then ANOTHER 60 turns of the primary coil are wound on top of the -100 between a couple layers of mylar tape and then the HV coil.  The -100V probably connected directly to the low end of the HV coil, but that connection was broken during removal of the core from the epoxy, so I can only assume it.
   Question #2)  Is this right?  1161 turns?  Again, is there less flux at the larger radii, and that's why it's more turns that expected?
  In the process of unwinding the trasnsormer, I discovered two things.  The HV coil was burnt about half way down, in the middle of the winding, where I found charred varnish.  And at the lowest end of the HV coil, at the bend in the HV wire where it is brought out to the side of the coil, it arced and burned through the mylar tape and varnish to the primary coil, and melted it open.  This is where my primary failed.In retrospect, it makes some sense.  The measured primary voltage is 880V peak-peak at the V620 anode, referred to ground!  +880 - (-100) = 990V difference, so no wonder the HV  punched through the tape.  
  Question #3)  Why is the primary split-wound?   Would I do better to wind all 120 turns on the bottom most layer?  Is there a good reason to split the coil?   The one obvious advantage of NOT splitting the primary is that the -100 leads directly in to the HV coil, and so there are not huge voltage gradients within the coil.
  Question #4)  Relates to #2 and #3 - I intend to pot the transformer in beeswax/parafin, so I expect reduced loss.  Will I then have *too good* a transformer, and thus too high a voltage if I split the primary and wind 1161 turns on the HV secondary?  
Finally, for Chuck,   I have mylar tape.  I have GE 7031 varnish and polystyrene Q-dope (GC electronics). 
Question #5)  Is one,  either or neither of these preferred?  My coil winder can't (yet) wind a universal wind.  I was thinking of a straight wind, but with a few extra layers of mylar tape between sections, perhaps 5.  The original coil was universal wound with one layer of fiberglass tape 1/2 way down.  Am I dead in the water without a universal wind?   When I am all done, I will vacuum impregnate the coil with parafin/beeswax 50/50 in my vacuum oven. 
  Finally,  I broke the ferrite core in removing it.  
Question #6) Can I glue it back together?  Will this core still work?  I don't see why not, especially considering that the core is gapped to begin which.  A crack is just a very small gap.
  Thanks all for your suggestions.  I'm looking for advice so I can learn to make this work.  The easy way is to just put the working spare transformer in and be done with it, but I'd like to be able to make another one when *that one* finally goes.
  Dan




[Non-text portions of this message have been removed]


Glenn Little WB4UIV <wb4uiv@...>
 

Could this be resent as true ASCII text, no tabs, just spaces, no HTML, no RTF?

Thanks
Glenn

At 06:28 PM 3/22/2015, you wrote:


Hi All, and particularly Chuck Harris, John Snyder and Brad Thompson,...
  My RM503 still sits on my workbench,
inoperable, but at present, I know it's T620
that is fried, and I have a working spare in hand (thanks Brad).
Before I resort to the spare, I'm going to test
my coil winder and re-wind a transformer, but
before I do that I UNWOUND my original T620 that
had an open in the primary circuit. Â I unwound
it while taking notes in gory detail, but I have a few questions as a result.Â
 First of all, here are the windings, counted
one at a time as I unwound the transformer, one
end to the other, from the outside in to the
core.Coil           TurnsHV
Filament   1HV           1161Â
Primary       60-100V
        34 12.6V         5
+100 Â Â Â Â Â Â Â Â Â 29 (on top of the
+12.6V, total 34 from ground)+250
         51 (on top of the
+100V, total 85 from ground)Primary       60

Now, I can get loopy headed while mind-numbingly
counting hundreds of turns by hand, with the
wire breaking a dozen or so times during the
unwind, so I measured the lengths of each piece
of wire I removed from the coil. Â Using those
lengths, I was able to confirm I got the counts
right. Â So the HV coil is 1161 turns, +/-1-2 turns maximum error. Â
 The low voltage coils are very consistent. Â
If you assume 2.9412 V/turn, then the +250, +100
and -100 coils are dead on. Â But then that
means that 1161T * 2.9412V/T = ~-3400V. Â This
is 400V higher than the nominal -3000V filament supply. Â
 Question #1)  Why is that so high?  Is
there a huge drop in the 5642 rectifier tube? Â
Is there reduced flux at the outer ends of the
HV coil so that it's not really 2.94V/turn?
 Another very interesting discovery, which I
think may have been discussed before but I can't
find the message, is that the primary is split.
 First, 60 turns are wound around the core,
then the +250 coil, then +100, +12.6 and finally
-100 comes out on the top. Â Then ANOTHER 60
turns of the primary coil are wound on top of
the -100 between a couple layers of mylar tape
and then the HV coil. Â The -100V probably
connected directly to the low end of the HV
coil, but that connection was broken during
removal of the core from the epoxy, so I can only assume it.
  Question #2)  Is this right?  1161 turns?
 Again, is there less flux at the larger radii,
and that's why it's more turns that expected?
 In the process of unwinding the trasnsormer,
I discovered two things. Â The HV coil was burnt
about half way down, in the middle of the
winding, where I found charred varnish. Â And at
the lowest end of the HV coil, at the bend in
the HV wire where it is brought out to the side
of the coil, it arced and burned through the
mylar tape and varnish to the primary coil, and
melted it open. Â This is where my primary
failed.In retrospect, it makes some sense. Â The
measured primary voltage is 880V peak-peak at
the V620 anode, referred to ground! Â +880 -
(-100) = 990V difference, so no wonder the HV Â punched through the tape. Â
 Question #3)  Why is the primary
split-wound? Â Would I do better to wind all
120 turns on the bottom most layer? Â Is there a
good reason to split the coil? Â The one
obvious advantage of NOT splitting the primary
is that the -100 leads directly in to the HV
coil, and so there are not huge voltage gradients within the coil.
 Question #4)  Relates to #2 and #3 - I
intend to pot the transformer in
beeswax/parafin, so I expect reduced loss. Â
Will I then have *too good* a transformer, and
thus too high a voltage if I split the primary
and wind 1161 turns on the HV secondary? Â
Finally, for Chuck, Â I have mylar tape. Â I
have GE 7031 varnish and polystyrene Q-dope (GC electronics).Â
Question #5) Â Is one, Â either or neither of
these preferred? Â My coil winder can't (yet)
wind a universal wind. Â I was thinking of a
straight wind, but with a few extra layers of
mylar tape between sections, perhaps 5. Â The
original coil was universal wound with one layer
of fiberglass tape 1/2 way down. Â Am I dead in
the water without a universal wind? Â When I am
all done, I will vacuum impregnate the coil with
parafin/beeswax 50/50 in my vacuum oven.Â
 Finally,  I broke the ferrite core in removing it. Â
Question #6) Can I glue it back together? Â Will
this core still work? Â I don't see why not, especially considering that the core is gapped
to begin which. Â A crack is just a very small gap.
 Thanks all for your suggestions.  I'm
looking for advice so I can learn to make this
work. Â The easy way is to just put the working
spare transformer in and be done with it, but
I'd like to be able to make another one when *that one* finally goes.
 Dan




Daniel Koller
 

Sorry, my post got mangled by Yahoo. Here it is again in PLAIN TEXT.
-------------------------------------------------------------------------------------------------------



Hi All, and particularly Chuck Harris, John Snyder and Brad Thompson,...

My RM503 still sits on my workbench, inoperable, but at present, I know it's T620 that is fried,
and I have a working spare in hand (thanks Brad).

Before I resort to the spare, I'm going to test my coil winder and re-wind a transformer, but before
I do that I UNWOUND my original T620 that had an open in the primary circuit. I unwound it while
taking notes in gory detail, but I have a few questions as a result.

First of all, here are the windings, counted one at a time as I unwound the transformer, one end
to the other, from the outside in to the core.
Coil Turns
HV Filament 1
HV 1161
Primary 60
-100V 34
12.6V 5
+100 29 (on top of the +12.6V, total 34 from ground)
+250 51 (on top of the +100V, total 85 from ground)
Primary 60

Now, I can get loopy headed while mind-numbingly counting hundreds of turns by hand, with the wire
breaking a dozen or so times during the unwind, so I measured the lengths of each piece of wire I
removed from the coil. Using those lengths, I was able to confirm I got the counts right. So the
HV coil is 1161 turns, +/-1 or 2 turns maximum error.

The low voltage coils are very consistent. If you assume 2.9412 V/turn, then the +250, +100 and
-100 coils are dead on. But then that means that 1161T * 2.9412V/T = ~-3400V. This is 400V
higher than the nominal -3000V filament supply.

Question #1) Why is that so high? Is there a huge drop in the 5642 rectifier tube? Is there
reduced flux at the outer ends of the HV coil so that it's not really 2.94V/turn?

Another very interesting discovery, which I think may have been discussed before but I can't find
the message, is that the primary is split. First, 60 turns are wound around the core, then the +250
coil, then +100, +12.6 and finally -100 comes out on the top. Then ANOTHER 60 turns of the primary
coil are wound on top of the -100 between a couple layers of mylar tape, and then the HV coil. The
-100V probably connected directly to the low end of the HV coil, but that connection was broken
during removal of the core from the epoxy, so I can only assume it.

Question #2) Is this right? 1161 turns? Again, is there less flux at the larger radii (outer coils), and
that's why it's more turns that expected?

In the process of unwinding the trasnsormer, I discovered two things. The HV coil was burnt about
half way down, in the middle of the winding, where I found charred varnish. And at the lowest end
of the HV coil, at the bend in the HV wire where it is brought out to the side of the coil, it arced
and burned through the mylar tape and varnish to the primary coil, and melted it open. This is where
my primary failed. In retrospect, it makes some sense. The measured primary voltage is 880V
peak-peak at the V620 anode, referred to ground! +880 - (-100) = 990V difference, so no wonder the
HV punched through the tape.

Question #3) Why is the primary split-wound? Would I do better to wind all 120 turns on the
bottom-most layer? Is there a good reason to split the coil? The one obvious advantage of NOT
splitting the primary is that the -100 leads directly in to the HV coil, and so there are not huge
voltage gradients within the coil.

Question #4) Relates to #2 and #3 - I intend to pot the transformer in beeswax/parafin, so I
expect reduced loss. Will I then have *too good* a transformer, and thus too high a voltage if
I split the primary and wind 1161 turns on the HV secondary?

Finally, for Chuck, I have mylar tape. I have GE 7031 varnish and polystyrene Q-dope
(GC electronics).

Question #5) Is one, either or neither of these preferred? My coil winder can't (yet) wind a
universal wind. I was thinking of a straight wind, but with a few extra layers of mylar tape
between sections, perhaps 5 sections. The original coil was universal-wound with one layer of fiberglass
tape 1/2 way down. Am I dead in the water without a universal wind? When I am all done, I will
vacuum impregnate the coil with parafin/beeswax 50/50 in my vacuum oven.

Finally, I broke the ferrite core in removing it.

Question #6) Can I glue it back together? Will this core still work? I don't see why not,
especially considering that the core is gapped to begin which. A crack is just a very small gap.

Thanks all for your suggestions. I'm looking for advice so I can learn to make this work. The
easy way is to just put the working spare transformer in and be done with it, but I'd like to be
able to make another one when *that one* finally goes.

Dan


Chuck Harris <cfharris@...>
 

Hi Dan,

A couple of notes:

First, HV windings in the multiple KV range scare the poogies
out of transformer designers. A stray arc over into the low
level circuitry, and you can wipe out a whole bunch of preamps,
and stuff... So, when you design a transformer that is going
to handle all of the voltages in a hybrid system, like the 503,
you very much need to have an electrostatic shield between the
HV winding and the windings driving the solid state electronics...

The Tek engineer that designed this transformer obviously thought
that an easy way to do that would be to simply use half of the
very low impedance, and very much immune to stray arc over,
primary winding as a protective shield between the solid state
amplifiers and the EHT windings.

Next, the flux difference at the center of the core and the
outer most reaches of the last winding on the coil are the
same.

The reason that the HV winding has more volts than you might
expect is many-fold. First, the resistance of all of that
fine wire and the leakage between turns does add up, second,
the 5642 rectifier drops about 10V just for being a rectifier,
and third, you also get a significant drop because of ripple
on the 1/2 wave configuration.

And, finally, there is always the fact that it just doesn't
matter much what the HV is on a scope like the 502. They don't
even give you an adjustment for it. Any small variation would
easily be calibrated out by adjusting the vertical and horizontal
gain.

Ok, insulation materials. Tek seemed to use a yellow adhesive
clear plastic tape... It looks a lot like the tape that CDE
used to wind the old yellow mylar capacitors... probably mylar.

I use Kapton tape, because I always have, and I understand it.
Mylar would be just as good in my opinion.

The biggest reason for using a universal wind is to reduce
the voltage difference between wires and layers. If you use
a plain wind, you will need to put a layer of tape between
each layer of wire. If you don't, you will have enough voltage
between the adjacent turns at the ends of the windings to arc
over the wire's insulation.

You can calculate it:

(volts/turn) * 2 * (width of winding) / (diameter of wire)

These fine wires can only handle a couple of hundred volts
before they arc over... and that assumes no flaws in the
wire's insulation, and no corona pecking away at the insulation.

There is already a gap in the core that would completely
overshadow any glue used to fix the core. Although I have
not broken any of these cores... yet... I have broken others,
and I just glue them back together with a little crazy glue,
and I have never had a problem. I have even glued toroids
where there is no gap without a problem.

-Chuck Harris



Daniel Koller kaboomdk@... [TekScopes] wrote:

Sorry, my post got mangled by Yahoo. Here it is again in PLAIN TEXT.
-------------------------------------------------------------------------------------------------------



Hi All, and particularly Chuck Harris, John Snyder and Brad Thompson,...

My RM503 still sits on my workbench, inoperable, but at present, I know it's T620 that is fried,
and I have a working spare in hand (thanks Brad).

Before I resort to the spare, I'm going to test my coil winder and re-wind a transformer, but before
I do that I UNWOUND my original T620 that had an open in the primary circuit. I unwound it while
taking notes in gory detail, but I have a few questions as a result.

First of all, here are the windings, counted one at a time as I unwound the transformer, one end
to the other, from the outside in to the core.
Coil Turns
HV Filament 1
HV 1161
Primary 60
-100V 34
12.6V 5
+100 29 (on top of the +12.6V, total 34 from ground)
+250 51 (on top of the +100V, total 85 from ground)
Primary 60

Now, I can get loopy headed while mind-numbingly counting hundreds of turns by hand, with the wire
breaking a dozen or so times during the unwind, so I measured the lengths of each piece of wire I
removed from the coil. Using those lengths, I was able to confirm I got the counts right. So the
HV coil is 1161 turns, +/-1 or 2 turns maximum error.

The low voltage coils are very consistent. If you assume 2.9412 V/turn, then the +250, +100 and
-100 coils are dead on. But then that means that 1161T * 2.9412V/T = ~-3400V. This is 400V
higher than the nominal -3000V filament supply.

Question #1) Why is that so high? Is there a huge drop in the 5642 rectifier tube? Is there
reduced flux at the outer ends of the HV coil so that it's not really 2.94V/turn?

Another very interesting discovery, which I think may have been discussed before but I can't find
the message, is that the primary is split. First, 60 turns are wound around the core, then the +250
coil, then +100, +12.6 and finally -100 comes out on the top. Then ANOTHER 60 turns of the primary
coil are wound on top of the -100 between a couple layers of mylar tape, and then the HV coil. The
-100V probably connected directly to the low end of the HV coil, but that connection was broken
during removal of the core from the epoxy, so I can only assume it.

Question #2) Is this right? 1161 turns? Again, is there less flux at the larger radii (outer coils), and
that's why it's more turns that expected?

In the process of unwinding the trasnsormer, I discovered two things. The HV coil was burnt about
half way down, in the middle of the winding, where I found charred varnish. And at the lowest end
of the HV coil, at the bend in the HV wire where it is brought out to the side of the coil, it arced
and burned through the mylar tape and varnish to the primary coil, and melted it open. This is where
my primary failed. In retrospect, it makes some sense. The measured primary voltage is 880V
peak-peak at the V620 anode, referred to ground! +880 - (-100) = 990V difference, so no wonder the
HV punched through the tape.

Question #3) Why is the primary split-wound? Would I do better to wind all 120 turns on the
bottom-most layer? Is there a good reason to split the coil? The one obvious advantage of NOT
splitting the primary is that the -100 leads directly in to the HV coil, and so there are not huge
voltage gradients within the coil.

Question #4) Relates to #2 and #3 - I intend to pot the transformer in beeswax/parafin, so I
expect reduced loss. Will I then have *too good* a transformer, and thus too high a voltage if
I split the primary and wind 1161 turns on the HV secondary?

Finally, for Chuck, I have mylar tape. I have GE 7031 varnish and polystyrene Q-dope
(GC electronics).

Question #5) Is one, either or neither of these preferred? My coil winder can't (yet) wind a
universal wind. I was thinking of a straight wind, but with a few extra layers of mylar tape
between sections, perhaps 5 sections. The original coil was universal-wound with one layer of fiberglass
tape 1/2 way down. Am I dead in the water without a universal wind? When I am all done, I will
vacuum impregnate the coil with parafin/beeswax 50/50 in my vacuum oven.

Finally, I broke the ferrite core in removing it.

Question #6) Can I glue it back together? Will this core still work? I don't see why not,
especially considering that the core is gapped to begin which. A crack is just a very small gap.

Thanks all for your suggestions. I'm looking for advice so I can learn to make this work. The
easy way is to just put the working spare transformer in and be done with it, but I'd like to be
able to make another one when *that one* finally goes.

Dan


------------------------------------
Posted by: Daniel Koller <kaboomdk@...>
------------------------------------


------------------------------------

Yahoo Groups Links




ykochcal
 

Dan

Good to hear you are making progress.

I had done some work and was actually going to write it up today but did not
get to it.

Anyway some thoughts

I went a bit sideways and have been looking at the cores to see if I could
find similar cores and do some test windings. I also have been taking apart
HV transformers. TV switching supply/PC power supply/microwave with
switching supply/TV EHT/Laser printer HV.

So I would say that doing the primary in two sections is often done in TV/PC
switching supplies. I did find some pdf files that I did not fully read on
the subject. If I understand correctly It distributes the magnetic field
better in the secondary with less magnetic loss and can reduce effective
capacitance.

If you calculated the one turn voltage for the filament of 2.9412 V/turn

That matches what I calculated
5643 Tube filament
1.25V RMS
200ma RMS
Series 4.3 ohm resistor
.86V RMS drop
2.11V RMS out of coil into resistor and filament in parallel.
*1.414
2.9835 V P (probably a design number of 3V/Turn)

So I figured (if wound from ground with out any diode etc. losses)
-3000 Vp = 1000 T
-100 Vp = 33T
+12.6Vp = 4 or 5 T
+100Vp = 33 T
+250Vp = 84 T
Primary
880Vp-p (your number)
=440Vp = 146 T

1161 is 16% more so I am not sure what that means it's not that much higher
when look at from a % point of view.

I did not remove the sides yet but It looks to me as if the -3000V goes to
ground and not the -100V. I don't see any other connection to the -100 rivet


I did glue the core I broke back together. I broke both cores on one side by
the center leg so I had an L and a U. After I glued them back I turned one
side so I have a broken leg in each outside core path.
I figure the gap is .020 so a .002-4 more would work. (not based on any
reality, fact or figures)

At the same time I have tried to understand the nature of the cores. I built
a saturation tester to measure the basic core with a 16T coil. For the glued
core with a .020 gap, "loosely" taped together, I measure 47uH on the GR
1657 and 50uH on the cheep transistor indicator. At first I was not that
close with the core saturation tester. So I reconfigure the tester in a
second version with an op-amp to keep the voltage constant. That got closer
but still I was out from the GR 1657 by quite a bit. Then I looked at the
math for resistance of the coil and if my math is correct that brought the
uH must closer.

I would like it if someone could check my math after I get that up on the
web.

Anyway with a gap of .020 and 16 Turns I saw a saturation around 22 amps and
about 48uH (depending a bit on the place in the curve)

With no gap and 16 turns I saw a saturation around 9.5 amps, 110uH with no r
Comp.

The 16 turns went to 45 to 60 Amps or so on my tester so with the tape I had
holding it together it was more like a speaker. Epoxy still holding.

In taking apart transformers I note that TV flybacks are potted but the
compound does not penetrate the actual HV windings. I also note that based
on two switching supplies for microwaves (which I think are in the multi KV
output range) do not have any potting but I don't know what wire was used
yet.

So I have been trying to acquire wire with better insulation, why not
Belden 8085 38 gage 155C with two coat
200C rated enameled 38 gage with insulation that can not be soldered off
And some Quad coated 39 gage, in shipping now.

Along with mylar and kapton tapes

And have put 5 similar sized cores on the Saturation tester but have not
done any calc's to see if the nature is similar. They are all wide and short
vs the Tek skinny and tall. But for example a larger core could fit in the
305 can.

And was going to just start winding some coils that are similar to a 647's
turns and crank them up on the PA amp, make some measurements and get some
practice.


I acquired a few TV vector monitors for a few bucks that included a few 528A
and 1420 Vector scopes. The 528 has what looks like the same core but with a
clip holding dip in each outside corner. I have not taken it apart to
measure it.

Does anyone know if that is the same core? If so there are a lot of those
being dumped on the market now and it would be a good source.

I was think at some point to take the sides of the 305 coil you gave me and
then place it on the winder in the reverse unwind mode to count the turns.
But I was not mentally there yet until I wind a test of some kind.

I tried Q-dope on my first try and it did not last, I think the Q-dope might
not have much flex in it with it's too thick and that may result in cracking
later, but I had other issues and have not total given up on it's use.

John

If anyone has access to the complete article, I would be very interested in
seeing all the text and charts.

<http://www.appliancedesign.com/articles/print/90001-fine-magnet-wire-for-hi
gh-temperature-and-high-voltage-applications>


Daniel Koller
 

Thanks John,

I guess the issue of where the HV coil connects is a key one. If you can determine that, it will be helpful. My coil was too damaged during removal to really see where the wire came out. However, it is certainly closer to the -100V end than the grounded end. But if the wire ran along the side of the coil, I would not have seen it during dis assembly.

Thanks.

Dan


--------------------------------------------

On Mon, 3/23/15, 'John Snyder' Kochcal@... [TekScopes] <TekScopes@...> wrote:

Subject: RE: [TekScopes] RM503 ongoing repair saga T620 UNWOUND.
To: TekScopes@...
Date: Monday, March 23, 2015, 9:29 AM


 









Dan



Good to hear you are making progress.



I had done some work and was actually going to write it up
today but did not

get to it.



Anyway some thoughts



I went a bit sideways and have been looking at the cores to
see if I could

find similar cores and do some test windings. I also have
been taking apart

HV transformers. TV switching supply/PC power
supply/microwave with

switching supply/TV EHT/Laser printer HV.



So I would say that doing the primary in two sections is
often done in TV/PC

switching supplies. I did find some pdf files that I did not
fully read on

the subject. If I understand correctly It distributes the
magnetic field

better in the secondary with less magnetic loss and can
reduce effective

capacitance.



If you calculated the one turn voltage for the filament of
2.9412 V/turn



That matches what I calculated

5643 Tube filament

1.25V RMS

200ma RMS

Series 4.3 ohm resistor

.86V RMS drop

2.11V RMS out of coil into resistor and filament in
parallel.

*1.414

2.9835 V P (probably a design number of 3V/Turn)



So I figured (if wound from ground with out any diode etc.
losses)

-3000 Vp = 1000 T

-100 Vp = 33T

+12.6Vp = 4 or 5 T

+100Vp = 33 T

+250Vp = 84 T

Primary

880Vp-p (your number)

=440Vp = 146 T



1161 is 16% more so I am not sure what that means it's
not that much higher

when look at from a % point of view.



I did not remove the sides yet but It looks to me as if the
-3000V goes to

ground and not the -100V. I don't see any other
connection to the -100 rivet



I did glue the core I broke back together. I broke both
cores on one side by

the center leg so I had an L and a U. After I glued them
back I turned one

side so I have a broken leg in each outside core path.

I figure the gap is .020 so a .002-4 more would work. (not
based on any

reality, fact or figures)



At the same time I have tried to understand the nature of
the cores. I built

a saturation tester to measure the basic core with a 16T
coil. For the glued

core with a .020 gap, "loosely" taped together, I
measure 47uH on the GR

1657 and 50uH on the cheep transistor indicator. At first I
was not that

close with the core saturation tester. So I reconfigure the
tester in a

second version with an op-amp to keep the voltage constant.
That got closer

but still I was out from the GR 1657 by quite a bit. Then I
looked at the

math for resistance of the coil and if my math is correct
that brought the

uH must closer.



I would like it if someone could check my math after I get
that up on the

web.



Anyway with a gap of .020 and 16 Turns I saw a saturation
around 22 amps and

about 48uH (depending a bit on the place in the curve)



With no gap and 16 turns I saw a saturation around 9.5 amps,
110uH with no r

Comp.



The 16 turns went to 45 to 60 Amps or so on my tester so
with the tape I had

holding it together it was more like a speaker. Epoxy still
holding.



In taking apart transformers I note that TV flybacks are
potted but the

compound does not penetrate the actual HV windings. I also
note that based

on two switching supplies for microwaves (which I think are
in the multi KV

output range) do not have any potting but I don't know
what wire was used

yet.



So I have been trying to acquire wire with better
insulation, why not

Belden 8085 38 gage 155C with two coat

200C rated enameled 38 gage with insulation that can not be
soldered off

And some Quad coated 39 gage, in shipping now.



Along with mylar and kapton tapes



And have put 5 similar sized cores on the Saturation tester
but have not

done any calc's to see if the nature is similar. They
are all wide and short

vs the Tek skinny and tall. But for example a larger core
could fit in the

305 can.



And was going to just start winding some coils that are
similar to a 647's

turns and crank them up on the PA amp, make some
measurements and get some

practice.



I acquired a few TV vector monitors for a few bucks that
included a few 528A

and 1420 Vector scopes. The 528 has what looks like the same
core but with a

clip holding dip in each outside corner. I have not taken
it apart to

measure it.



Does anyone know if that is the same core? If so there are a
lot of those

being dumped on the market now and it would be a good
source.



I was think at some point to take the sides of the 305 coil
you gave me and

then place it on the winder in the reverse unwind mode to
count the turns.

But I was not mentally there yet until I wind a test of some
kind.



I tried Q-dope on my first try and it did not last, I think
the Q-dope might

not have much flex in it with it's too thick and that
may result in cracking

later, but I had other issues and have not total given up on
it's use.



John



If anyone has access to the complete article, I would be
very interested in

seeing all the text and charts.



<http://www.appliancedesign.com/articles/print/90001-fine-magnet-wire-for-hi

gh-temperature-and-high-voltage-applications>













#yiv3311445403 #yiv3311445403 --
#yiv3311445403ygrp-mkp {
border:1px solid #d8d8d8;font-family:Arial;margin:10px
0;padding:0 10px;}

#yiv3311445403 #yiv3311445403ygrp-mkp hr {
border:1px solid #d8d8d8;}

#yiv3311445403 #yiv3311445403ygrp-mkp #yiv3311445403hd {
color:#628c2a;font-size:85%;font-weight:700;line-height:122%;margin:10px
0;}

#yiv3311445403 #yiv3311445403ygrp-mkp #yiv3311445403ads {
margin-bottom:10px;}

#yiv3311445403 #yiv3311445403ygrp-mkp .yiv3311445403ad {
padding:0 0;}

#yiv3311445403 #yiv3311445403ygrp-mkp .yiv3311445403ad p {
margin:0;}

#yiv3311445403 #yiv3311445403ygrp-mkp .yiv3311445403ad a {
color:#0000ff;text-decoration:none;}
#yiv3311445403 #yiv3311445403ygrp-sponsor
#yiv3311445403ygrp-lc {
font-family:Arial;}

#yiv3311445403 #yiv3311445403ygrp-sponsor
#yiv3311445403ygrp-lc #yiv3311445403hd {
margin:10px
0px;font-weight:700;font-size:78%;line-height:122%;}

#yiv3311445403 #yiv3311445403ygrp-sponsor
#yiv3311445403ygrp-lc .yiv3311445403ad {
margin-bottom:10px;padding:0 0;}

#yiv3311445403 #yiv3311445403actions {
font-family:Verdana;font-size:11px;padding:10px 0;}

#yiv3311445403 #yiv3311445403activity {
background-color:#e0ecee;float:left;font-family:Verdana;font-size:10px;padding:10px;}

#yiv3311445403 #yiv3311445403activity span {
font-weight:700;}

#yiv3311445403 #yiv3311445403activity span:first-child {
text-transform:uppercase;}

#yiv3311445403 #yiv3311445403activity span a {
color:#5085b6;text-decoration:none;}

#yiv3311445403 #yiv3311445403activity span span {
color:#ff7900;}

#yiv3311445403 #yiv3311445403activity span
.yiv3311445403underline {
text-decoration:underline;}

#yiv3311445403 .yiv3311445403attach {
clear:both;display:table;font-family:Arial;font-size:12px;padding:10px
0;width:400px;}

#yiv3311445403 .yiv3311445403attach div a {
text-decoration:none;}

#yiv3311445403 .yiv3311445403attach img {
border:none;padding-right:5px;}

#yiv3311445403 .yiv3311445403attach label {
display:block;margin-bottom:5px;}

#yiv3311445403 .yiv3311445403attach label a {
text-decoration:none;}

#yiv3311445403 blockquote {
margin:0 0 0 4px;}

#yiv3311445403 .yiv3311445403bold {
font-family:Arial;font-size:13px;font-weight:700;}

#yiv3311445403 .yiv3311445403bold a {
text-decoration:none;}

#yiv3311445403 dd.yiv3311445403last p a {
font-family:Verdana;font-weight:700;}

#yiv3311445403 dd.yiv3311445403last p span {
margin-right:10px;font-family:Verdana;font-weight:700;}

#yiv3311445403 dd.yiv3311445403last p
span.yiv3311445403yshortcuts {
margin-right:0;}

#yiv3311445403 div.yiv3311445403attach-table div div a {
text-decoration:none;}

#yiv3311445403 div.yiv3311445403attach-table {
width:400px;}

#yiv3311445403 div.yiv3311445403file-title a, #yiv3311445403
div.yiv3311445403file-title a:active, #yiv3311445403
div.yiv3311445403file-title a:hover, #yiv3311445403
div.yiv3311445403file-title a:visited {
text-decoration:none;}

#yiv3311445403 div.yiv3311445403photo-title a,
#yiv3311445403 div.yiv3311445403photo-title a:active,
#yiv3311445403 div.yiv3311445403photo-title a:hover,
#yiv3311445403 div.yiv3311445403photo-title a:visited {
text-decoration:none;}

#yiv3311445403 div#yiv3311445403ygrp-mlmsg
#yiv3311445403ygrp-msg p a span.yiv3311445403yshortcuts {
font-family:Verdana;font-size:10px;font-weight:normal;}

#yiv3311445403 .yiv3311445403green {
color:#628c2a;}

#yiv3311445403 .yiv3311445403MsoNormal {
margin:0 0 0 0;}

#yiv3311445403 o {
font-size:0;}

#yiv3311445403 #yiv3311445403photos div {
float:left;width:72px;}

#yiv3311445403 #yiv3311445403photos div div {
border:1px solid
#666666;height:62px;overflow:hidden;width:62px;}

#yiv3311445403 #yiv3311445403photos div label {
color:#666666;font-size:10px;overflow:hidden;text-align:center;white-space:nowrap;width:64px;}

#yiv3311445403 #yiv3311445403reco-category {
font-size:77%;}

#yiv3311445403 #yiv3311445403reco-desc {
font-size:77%;}

#yiv3311445403 .yiv3311445403replbq {
margin:4px;}

#yiv3311445403 #yiv3311445403ygrp-actbar div a:first-child {
margin-right:2px;padding-right:5px;}

#yiv3311445403 #yiv3311445403ygrp-mlmsg {
font-size:13px;font-family:Arial, helvetica, clean,
sans-serif;}

#yiv3311445403 #yiv3311445403ygrp-mlmsg table {
font-size:inherit;font:100%;}

#yiv3311445403 #yiv3311445403ygrp-mlmsg select,
#yiv3311445403 input, #yiv3311445403 textarea {
font:99% Arial, Helvetica, clean, sans-serif;}

#yiv3311445403 #yiv3311445403ygrp-mlmsg pre, #yiv3311445403
code {
font:115% monospace;}

#yiv3311445403 #yiv3311445403ygrp-mlmsg * {
line-height:1.22em;}

#yiv3311445403 #yiv3311445403ygrp-mlmsg #yiv3311445403logo {
padding-bottom:10px;}


#yiv3311445403 #yiv3311445403ygrp-msg p a {
font-family:Verdana;}

#yiv3311445403 #yiv3311445403ygrp-msg
p#yiv3311445403attach-count span {
color:#1E66AE;font-weight:700;}

#yiv3311445403 #yiv3311445403ygrp-reco
#yiv3311445403reco-head {
color:#ff7900;font-weight:700;}

#yiv3311445403 #yiv3311445403ygrp-reco {
margin-bottom:20px;padding:0px;}

#yiv3311445403 #yiv3311445403ygrp-sponsor #yiv3311445403ov
li a {
font-size:130%;text-decoration:none;}

#yiv3311445403 #yiv3311445403ygrp-sponsor #yiv3311445403ov
li {
font-size:77%;list-style-type:square;padding:6px 0;}

#yiv3311445403 #yiv3311445403ygrp-sponsor #yiv3311445403ov
ul {
margin:0;padding:0 0 0 8px;}

#yiv3311445403 #yiv3311445403ygrp-text {
font-family:Georgia;}

#yiv3311445403 #yiv3311445403ygrp-text p {
margin:0 0 1em 0;}

#yiv3311445403 #yiv3311445403ygrp-text tt {
font-size:120%;}

#yiv3311445403 #yiv3311445403ygrp-vital ul li:last-child {
border-right:none !important;
}
#yiv3311445403


Daniel Koller
 

Thanks Chuck!

Ok, I am going to go ahead with an attempt at winding my own, starting possibly as early as this week. Now I just need to get some toluene to thing out my GE varnish a little bit (the recommended solvent was 50/50 Acetone/Toluene).

I think I am going to "skimp" and wind the primary as one turn at the bottom of the coil, rather than break it in two halves. I understand the intended point about shielding, but A) my scope is all tubes (no solid state at the inputs in the older models) and B) I have a documented failure due to that "shield winding" which would probably not have occurred had it been connected to the -100V directly, thus reducing gradients. Also, for this first one, I need to keep the winding a little simpler.

For a given cold width, how much larger in diameter is a universal coil versus a simple layered stack? I need to make sure I have enough room for layers of mylar tape.

Dan

--------------------------------------------

On Mon, 3/23/15, Chuck Harris cfharris@... [TekScopes] <TekScopes@...> wrote:

Subject: Re: [TekScopes] RM503 ongoing repair saga T620 UNWOUND.
To: TekScopes@...
Date: Monday, March 23, 2015, 5:32 AM


 









Hi Dan,



A couple of notes:



First, HV windings in the multiple KV range scare the
poogies

out of transformer designers. A stray arc over into the
low

level circuitry, and you can wipe out a whole bunch of
preamps,

and stuff... So, when you design a transformer that is
going

to handle all of the voltages in a hybrid system, like the
503,

you very much need to have an electrostatic shield between
the

HV winding and the windings driving the solid state
electronics...



The Tek engineer that designed this transformer obviously
thought

that an easy way to do that would be to simply use half of
the

very low impedance, and very much immune to stray arc
over,

primary winding as a protective shield between the solid
state

amplifiers and the EHT windings.



Next, the flux difference at the center of the core and
the

outer most reaches of the last winding on the coil are
the

same.



The reason that the HV winding has more volts than you
might

expect is many-fold. First, the resistance of all of
that

fine wire and the leakage between turns does add up,
second,

the 5642 rectifier drops about 10V just for being a
rectifier,

and third, you also get a significant drop because of
ripple

on the 1/2 wave configuration.



And, finally, there is always the fact that it just
doesn't

matter much what the HV is on a scope like the 502. They
don't

even give you an adjustment for it. Any small variation
would

easily be calibrated out by adjusting the vertical and
horizontal

gain.



Ok, insulation materials. Tek seemed to use a yellow
adhesive

clear plastic tape... It looks a lot like the tape that
CDE

used to wind the old yellow mylar capacitors... probably
mylar.



I use Kapton tape, because I always have, and I understand
it.

Mylar would be just as good in my opinion.



The biggest reason for using a universal wind is to
reduce

the voltage difference between wires and layers. If you
use

a plain wind, you will need to put a layer of tape
between

each layer of wire. If you don't, you will have enough
voltage

between the adjacent turns at the ends of the windings to
arc

over the wire's insulation.



You can calculate it:



(volts/turn) * 2 * (width of winding) / (diameter of
wire)



These fine wires can only handle a couple of hundred
volts

before they arc over... and that assumes no flaws in the

wire's insulation, and no corona pecking away at the
insulation.



There is already a gap in the core that would completely

overshadow any glue used to fix the core. Although I
have

not broken any of these cores... yet... I have broken
others,

and I just glue them back together with a little crazy
glue,

and I have never had a problem. I have even glued
toroids

where there is no gap without a problem.



-Chuck Harris



Daniel Koller kaboomdk@... [TekScopes] wrote:

> Sorry, my post got mangled by Yahoo. Here it is again
in PLAIN TEXT.

>
----------------------------------------------------------

>

>

>

> Hi All, and particularly Chuck Harris, John Snyder and
Brad Thompson,...

>

> My RM503 still sits on my workbench, inoperable, but at
present, I know it's T620 that is fried,

> and I have a working spare in hand (thanks Brad).

>

> Before I resort to the spare, I'm going to test my
coil winder and re-wind a transformer, but before

> I do that I UNWOUND my original T620 that had an open
in the primary circuit. I unwound it while

> taking notes in gory detail, but I have a few questions
as a result.

>

> First of all, here are the windings, counted one at a
time as I unwound the transformer, one end

> to the other, from the outside in to the core.

> Coil Turns

> HV Filament 1

> HV 1161

> Primary 60

> -100V 34

> 12.6V 5

> +100 29 (on top of the +12.6V, total 34 from
ground)

> +250 51 (on top of the +100V, total 85 from
ground)

> Primary 60

>

> Now, I can get loopy headed while mind-numbingly
counting hundreds of turns by hand, with the wire

> breaking a dozen or so times during the unwind, so I
measured the lengths of each piece of wire I

> removed from the coil. Using those lengths, I was able
to confirm I got the counts right. So the

> HV coil is 1161 turns, +/-1 or 2 turns maximum
error.

>

> The low voltage coils are very consistent. If you
assume 2.9412 V/turn, then the +250, +100 and

> -100 coils are dead on. But then that means that 1161T
* 2.9412V/T = ~-3400V. This is 400V

> higher than the nominal -3000V filament supply.

>

> Question #1) Why is that so high? Is there a huge
drop in the 5642 rectifier tube? Is there

> reduced flux at the outer ends of the HV coil so that
it's not really 2.94V/turn?

>

> Another very interesting discovery, which I think may
have been discussed before but I can't find

> the message, is that the primary is split. First, 60
turns are wound around the core, then the +250

> coil, then +100, +12.6 and finally -100 comes out on
the top. Then ANOTHER 60 turns of the primary

> coil are wound on top of the -100 between a couple
layers of mylar tape, and then the HV coil. The

> -100V probably connected directly to the low end of the
HV coil, but that connection was broken

> during removal of the core from the epoxy, so I can
only assume it.

>

> Question #2) Is this right? 1161 turns? Again, is
there less flux at the larger radii (outer coils), and

> that's why it's more turns that expected?

>

> In the process of unwinding the trasnsormer, I
discovered two things. The HV coil was burnt about

> half way down, in the middle of the winding, where I
found charred varnish. And at the lowest end

> of the HV coil, at the bend in the HV wire where it is
brought out to the side of the coil, it arced

> and burned through the mylar tape and varnish to the
primary coil, and melted it open. This is where

> my primary failed. In retrospect, it makes some sense.
The measured primary voltage is 880V

> peak-peak at the V620 anode, referred to ground! +880
- (-100) = 990V difference, so no wonder the

> HV punched through the tape.

>

> Question #3) Why is the primary split-wound? Would I
do better to wind all 120 turns on the

> bottom-most layer? Is there a good reason to split the
coil? The one obvious advantage of NOT

> splitting the primary is that the -100 leads directly
in to the HV coil, and so there are not huge

> voltage gradients within the coil.

>

> Question #4) Relates to #2 and #3 - I intend to pot
the transformer in beeswax/parafin, so I

> expect reduced loss. Will I then have *too good* a
transformer, and thus too high a voltage if

> I split the primary and wind 1161 turns on the HV
secondary?

>

> Finally, for Chuck, I have mylar tape. I have GE
7031 varnish and polystyrene Q-dope

> (GC electronics).

>

> Question #5) Is one, either or neither of these
preferred? My coil winder can't (yet) wind a

> universal wind. I was thinking of a straight wind, but
with a few extra layers of mylar tape

> between sections, perhaps 5 sections. The original
coil was universal-wound with one layer of fiberglass

> tape 1/2 way down. Am I dead in the water without a
universal wind? When I am all done, I will

> vacuum impregnate the coil with parafin/beeswax 50/50
in my vacuum oven.

>

> Finally, I broke the ferrite core in removing it.

>

> Question #6) Can I glue it back together? Will this
core still work? I don't see why not,

> especially considering that the core is gapped to begin
which. A crack is just a very small gap.

>

> Thanks all for your suggestions. I'm looking for
advice so I can learn to make this work. The

> easy way is to just put the working spare transformer
in and be done with it, but I'd like to be

> able to make another one when *that one* finally
goes.

>

> Dan

>

>

> ------------------------------------

> Posted by: Daniel Koller <kaboomdk@...>

> ------------------------------------

>

>

> ------------------------------------

>

> Yahoo Groups Links

>

>

>

>











#yiv5142895427 #yiv5142895427 --
#yiv5142895427ygrp-mkp {
border:1px solid #d8d8d8;font-family:Arial;margin:10px
0;padding:0 10px;}

#yiv5142895427 #yiv5142895427ygrp-mkp hr {
border:1px solid #d8d8d8;}

#yiv5142895427 #yiv5142895427ygrp-mkp #yiv5142895427hd {
color:#628c2a;font-size:85%;font-weight:700;line-height:122%;margin:10px
0;}

#yiv5142895427 #yiv5142895427ygrp-mkp #yiv5142895427ads {
margin-bottom:10px;}

#yiv5142895427 #yiv5142895427ygrp-mkp .yiv5142895427ad {
padding:0 0;}

#yiv5142895427 #yiv5142895427ygrp-mkp .yiv5142895427ad p {
margin:0;}

#yiv5142895427 #yiv5142895427ygrp-mkp .yiv5142895427ad a {
color:#0000ff;text-decoration:none;}
#yiv5142895427 #yiv5142895427ygrp-sponsor
#yiv5142895427ygrp-lc {
font-family:Arial;}

#yiv5142895427 #yiv5142895427ygrp-sponsor
#yiv5142895427ygrp-lc #yiv5142895427hd {
margin:10px
0px;font-weight:700;font-size:78%;line-height:122%;}

#yiv5142895427 #yiv5142895427ygrp-sponsor
#yiv5142895427ygrp-lc .yiv5142895427ad {
margin-bottom:10px;padding:0 0;}

#yiv5142895427 #yiv5142895427actions {
font-family:Verdana;font-size:11px;padding:10px 0;}

#yiv5142895427 #yiv5142895427activity {
background-color:#e0ecee;float:left;font-family:Verdana;font-size:10px;padding:10px;}

#yiv5142895427 #yiv5142895427activity span {
font-weight:700;}

#yiv5142895427 #yiv5142895427activity span:first-child {
text-transform:uppercase;}

#yiv5142895427 #yiv5142895427activity span a {
color:#5085b6;text-decoration:none;}

#yiv5142895427 #yiv5142895427activity span span {
color:#ff7900;}

#yiv5142895427 #yiv5142895427activity span
.yiv5142895427underline {
text-decoration:underline;}

#yiv5142895427 .yiv5142895427attach {
clear:both;display:table;font-family:Arial;font-size:12px;padding:10px
0;width:400px;}

#yiv5142895427 .yiv5142895427attach div a {
text-decoration:none;}

#yiv5142895427 .yiv5142895427attach img {
border:none;padding-right:5px;}

#yiv5142895427 .yiv5142895427attach label {
display:block;margin-bottom:5px;}

#yiv5142895427 .yiv5142895427attach label a {
text-decoration:none;}

#yiv5142895427 blockquote {
margin:0 0 0 4px;}

#yiv5142895427 .yiv5142895427bold {
font-family:Arial;font-size:13px;font-weight:700;}

#yiv5142895427 .yiv5142895427bold a {
text-decoration:none;}

#yiv5142895427 dd.yiv5142895427last p a {
font-family:Verdana;font-weight:700;}

#yiv5142895427 dd.yiv5142895427last p span {
margin-right:10px;font-family:Verdana;font-weight:700;}

#yiv5142895427 dd.yiv5142895427last p
span.yiv5142895427yshortcuts {
margin-right:0;}

#yiv5142895427 div.yiv5142895427attach-table div div a {
text-decoration:none;}

#yiv5142895427 div.yiv5142895427attach-table {
width:400px;}

#yiv5142895427 div.yiv5142895427file-title a, #yiv5142895427
div.yiv5142895427file-title a:active, #yiv5142895427
div.yiv5142895427file-title a:hover, #yiv5142895427
div.yiv5142895427file-title a:visited {
text-decoration:none;}

#yiv5142895427 div.yiv5142895427photo-title a,
#yiv5142895427 div.yiv5142895427photo-title a:active,
#yiv5142895427 div.yiv5142895427photo-title a:hover,
#yiv5142895427 div.yiv5142895427photo-title a:visited {
text-decoration:none;}

#yiv5142895427 div#yiv5142895427ygrp-mlmsg
#yiv5142895427ygrp-msg p a span.yiv5142895427yshortcuts {
font-family:Verdana;font-size:10px;font-weight:normal;}

#yiv5142895427 .yiv5142895427green {
color:#628c2a;}

#yiv5142895427 .yiv5142895427MsoNormal {
margin:0 0 0 0;}

#yiv5142895427 o {
font-size:0;}

#yiv5142895427 #yiv5142895427photos div {
float:left;width:72px;}

#yiv5142895427 #yiv5142895427photos div div {
border:1px solid
#666666;height:62px;overflow:hidden;width:62px;}

#yiv5142895427 #yiv5142895427photos div label {
color:#666666;font-size:10px;overflow:hidden;text-align:center;white-space:nowrap;width:64px;}

#yiv5142895427 #yiv5142895427reco-category {
font-size:77%;}

#yiv5142895427 #yiv5142895427reco-desc {
font-size:77%;}

#yiv5142895427 .yiv5142895427replbq {
margin:4px;}

#yiv5142895427 #yiv5142895427ygrp-actbar div a:first-child {
margin-right:2px;padding-right:5px;}

#yiv5142895427 #yiv5142895427ygrp-mlmsg {
font-size:13px;font-family:Arial, helvetica, clean,
sans-serif;}

#yiv5142895427 #yiv5142895427ygrp-mlmsg table {
font-size:inherit;font:100%;}

#yiv5142895427 #yiv5142895427ygrp-mlmsg select,
#yiv5142895427 input, #yiv5142895427 textarea {
font:99% Arial, Helvetica, clean, sans-serif;}

#yiv5142895427 #yiv5142895427ygrp-mlmsg pre, #yiv5142895427
code {
font:115% monospace;}

#yiv5142895427 #yiv5142895427ygrp-mlmsg * {
line-height:1.22em;}

#yiv5142895427 #yiv5142895427ygrp-mlmsg #yiv5142895427logo {
padding-bottom:10px;}


#yiv5142895427 #yiv5142895427ygrp-msg p a {
font-family:Verdana;}

#yiv5142895427 #yiv5142895427ygrp-msg
p#yiv5142895427attach-count span {
color:#1E66AE;font-weight:700;}

#yiv5142895427 #yiv5142895427ygrp-reco
#yiv5142895427reco-head {
color:#ff7900;font-weight:700;}

#yiv5142895427 #yiv5142895427ygrp-reco {
margin-bottom:20px;padding:0px;}

#yiv5142895427 #yiv5142895427ygrp-sponsor #yiv5142895427ov
li a {
font-size:130%;text-decoration:none;}

#yiv5142895427 #yiv5142895427ygrp-sponsor #yiv5142895427ov
li {
font-size:77%;list-style-type:square;padding:6px 0;}

#yiv5142895427 #yiv5142895427ygrp-sponsor #yiv5142895427ov
ul {
margin:0;padding:0 0 0 8px;}

#yiv5142895427 #yiv5142895427ygrp-text {
font-family:Georgia;}

#yiv5142895427 #yiv5142895427ygrp-text p {
margin:0 0 1em 0;}

#yiv5142895427 #yiv5142895427ygrp-text tt {
font-size:120%;}

#yiv5142895427 #yiv5142895427ygrp-vital ul li:last-child {
border-right:none !important;
}
#yiv5142895427


ykochcal
 

Let me poke at the wires going to the ground rivet and take some more
pictures. The answer should be there.



John

-----Original Message-----
From: TekScopes@... [mailto:TekScopes@...]
Sent: Monday, March 23, 2015 6:56 AM
To: TekScopes@...
Subject: RE: [TekScopes] RM503 ongoing repair saga T620 UNWOUND.

Thanks John,

I guess the issue of where the HV coil connects is a key one. If you can
determine that, it will be helpful. My coil was too damaged during removal
to really see where the wire came out. However, it is certainly closer to
the -100V end than the grounded end. But if the wire ran along the side of
the coil, I would not have seen it during dis assembly.

Thanks.

Dan


Daniel Koller
 

John,

It occurs to me a DC voltage measurement should indicate if the -100V is attached to the low end of the HV coil or not. Measure V from ground to the -100V terminal. If you apply a small current to the HV coil (5V / 200 Ohm = 25 mA), and the -100V end is not connected, you will see no voltage rise. On the other hand, if the two coils are in series, the ~~1 Ohm of the -100V coil will show a voltage drop of about 25 millivolts. All assuming of course that the coils are still intact. I don't remember if the HV coil was open on that transformer. I guess if it's open, it's probably fried where mine was, near the bottom. So then a visual check is the only way to see where it is connected at the ends.

Dan

--------------------------------------------

On Mon, 3/23/15, 'John Snyder' Kochcal@... [TekScopes] <TekScopes@...> wrote:

Subject: RE: [TekScopes] RM503 ongoing repair saga T620 UNWOUND.
To: TekScopes@...
Date: Monday, March 23, 2015, 4:52 PM


 









Let me poke at the wires going to the ground rivet
and take some more

pictures. The answer should be there.



John



-----Original Message-----

From: TekScopes@...
[mailto:TekScopes@...]

Sent: Monday, March 23, 2015 6:56 AM

To: TekScopes@...

Subject: RE: [TekScopes] RM503 ongoing repair saga T620
UNWOUND.



Thanks John,



I guess the issue of where the HV coil connects is a key
one. If you can

determine that, it will be helpful. My coil was too damaged
during removal

to really see where the wire came out. However, it is
certainly closer to

the -100V end than the grounded end. But if the wire ran
along the side of

the coil, I would not have seen it during dis assembly.



Thanks.



Dan













#yiv4474950023 #yiv4474950023 --
#yiv4474950023ygrp-mkp {
border:1px solid #d8d8d8;font-family:Arial;margin:10px
0;padding:0 10px;}

#yiv4474950023 #yiv4474950023ygrp-mkp hr {
border:1px solid #d8d8d8;}

#yiv4474950023 #yiv4474950023ygrp-mkp #yiv4474950023hd {
color:#628c2a;font-size:85%;font-weight:700;line-height:122%;margin:10px
0;}

#yiv4474950023 #yiv4474950023ygrp-mkp #yiv4474950023ads {
margin-bottom:10px;}

#yiv4474950023 #yiv4474950023ygrp-mkp .yiv4474950023ad {
padding:0 0;}

#yiv4474950023 #yiv4474950023ygrp-mkp .yiv4474950023ad p {
margin:0;}

#yiv4474950023 #yiv4474950023ygrp-mkp .yiv4474950023ad a {
color:#0000ff;text-decoration:none;}
#yiv4474950023 #yiv4474950023ygrp-sponsor
#yiv4474950023ygrp-lc {
font-family:Arial;}

#yiv4474950023 #yiv4474950023ygrp-sponsor
#yiv4474950023ygrp-lc #yiv4474950023hd {
margin:10px
0px;font-weight:700;font-size:78%;line-height:122%;}

#yiv4474950023 #yiv4474950023ygrp-sponsor
#yiv4474950023ygrp-lc .yiv4474950023ad {
margin-bottom:10px;padding:0 0;}

#yiv4474950023 #yiv4474950023actions {
font-family:Verdana;font-size:11px;padding:10px 0;}

#yiv4474950023 #yiv4474950023activity {
background-color:#e0ecee;float:left;font-family:Verdana;font-size:10px;padding:10px;}

#yiv4474950023 #yiv4474950023activity span {
font-weight:700;}

#yiv4474950023 #yiv4474950023activity span:first-child {
text-transform:uppercase;}

#yiv4474950023 #yiv4474950023activity span a {
color:#5085b6;text-decoration:none;}

#yiv4474950023 #yiv4474950023activity span span {
color:#ff7900;}

#yiv4474950023 #yiv4474950023activity span
.yiv4474950023underline {
text-decoration:underline;}

#yiv4474950023 .yiv4474950023attach {
clear:both;display:table;font-family:Arial;font-size:12px;padding:10px
0;width:400px;}

#yiv4474950023 .yiv4474950023attach div a {
text-decoration:none;}

#yiv4474950023 .yiv4474950023attach img {
border:none;padding-right:5px;}

#yiv4474950023 .yiv4474950023attach label {
display:block;margin-bottom:5px;}

#yiv4474950023 .yiv4474950023attach label a {
text-decoration:none;}

#yiv4474950023 blockquote {
margin:0 0 0 4px;}

#yiv4474950023 .yiv4474950023bold {
font-family:Arial;font-size:13px;font-weight:700;}

#yiv4474950023 .yiv4474950023bold a {
text-decoration:none;}

#yiv4474950023 dd.yiv4474950023last p a {
font-family:Verdana;font-weight:700;}

#yiv4474950023 dd.yiv4474950023last p span {
margin-right:10px;font-family:Verdana;font-weight:700;}

#yiv4474950023 dd.yiv4474950023last p
span.yiv4474950023yshortcuts {
margin-right:0;}

#yiv4474950023 div.yiv4474950023attach-table div div a {
text-decoration:none;}

#yiv4474950023 div.yiv4474950023attach-table {
width:400px;}

#yiv4474950023 div.yiv4474950023file-title a, #yiv4474950023
div.yiv4474950023file-title a:active, #yiv4474950023
div.yiv4474950023file-title a:hover, #yiv4474950023
div.yiv4474950023file-title a:visited {
text-decoration:none;}

#yiv4474950023 div.yiv4474950023photo-title a,
#yiv4474950023 div.yiv4474950023photo-title a:active,
#yiv4474950023 div.yiv4474950023photo-title a:hover,
#yiv4474950023 div.yiv4474950023photo-title a:visited {
text-decoration:none;}

#yiv4474950023 div#yiv4474950023ygrp-mlmsg
#yiv4474950023ygrp-msg p a span.yiv4474950023yshortcuts {
font-family:Verdana;font-size:10px;font-weight:normal;}

#yiv4474950023 .yiv4474950023green {
color:#628c2a;}

#yiv4474950023 .yiv4474950023MsoNormal {
margin:0 0 0 0;}

#yiv4474950023 o {
font-size:0;}

#yiv4474950023 #yiv4474950023photos div {
float:left;width:72px;}

#yiv4474950023 #yiv4474950023photos div div {
border:1px solid
#666666;height:62px;overflow:hidden;width:62px;}

#yiv4474950023 #yiv4474950023photos div label {
color:#666666;font-size:10px;overflow:hidden;text-align:center;white-space:nowrap;width:64px;}

#yiv4474950023 #yiv4474950023reco-category {
font-size:77%;}

#yiv4474950023 #yiv4474950023reco-desc {
font-size:77%;}

#yiv4474950023 .yiv4474950023replbq {
margin:4px;}

#yiv4474950023 #yiv4474950023ygrp-actbar div a:first-child {
margin-right:2px;padding-right:5px;}

#yiv4474950023 #yiv4474950023ygrp-mlmsg {
font-size:13px;font-family:Arial, helvetica, clean,
sans-serif;}

#yiv4474950023 #yiv4474950023ygrp-mlmsg table {
font-size:inherit;font:100%;}

#yiv4474950023 #yiv4474950023ygrp-mlmsg select,
#yiv4474950023 input, #yiv4474950023 textarea {
font:99% Arial, Helvetica, clean, sans-serif;}

#yiv4474950023 #yiv4474950023ygrp-mlmsg pre, #yiv4474950023
code {
font:115% monospace;}

#yiv4474950023 #yiv4474950023ygrp-mlmsg * {
line-height:1.22em;}

#yiv4474950023 #yiv4474950023ygrp-mlmsg #yiv4474950023logo {
padding-bottom:10px;}


#yiv4474950023 #yiv4474950023ygrp-msg p a {
font-family:Verdana;}

#yiv4474950023 #yiv4474950023ygrp-msg
p#yiv4474950023attach-count span {
color:#1E66AE;font-weight:700;}

#yiv4474950023 #yiv4474950023ygrp-reco
#yiv4474950023reco-head {
color:#ff7900;font-weight:700;}

#yiv4474950023 #yiv4474950023ygrp-reco {
margin-bottom:20px;padding:0px;}

#yiv4474950023 #yiv4474950023ygrp-sponsor #yiv4474950023ov
li a {
font-size:130%;text-decoration:none;}

#yiv4474950023 #yiv4474950023ygrp-sponsor #yiv4474950023ov
li {
font-size:77%;list-style-type:square;padding:6px 0;}

#yiv4474950023 #yiv4474950023ygrp-sponsor #yiv4474950023ov
ul {
margin:0;padding:0 0 0 8px;}

#yiv4474950023 #yiv4474950023ygrp-text {
font-family:Georgia;}

#yiv4474950023 #yiv4474950023ygrp-text p {
margin:0 0 1em 0;}

#yiv4474950023 #yiv4474950023ygrp-text tt {
font-size:120%;}

#yiv4474950023 #yiv4474950023ygrp-vital ul li:last-child {
border-right:none !important;
}
#yiv4474950023


ykochcal
 

Dan and Chuck

As Chuck points out for what I would call the back and forth winding,

9
8765
1234

the voltage difference is:

You can calculate it:

(volts/turn) * 2 * (width of winding) / (diameter of wire)

Which is the same as

(volts/turn) * 2 * turns per layer

Because 8 is on top of 1

But there is winding where you come straight back

9
5678
1234

That has 1/2 the voltage per layer and lower capacitance

(volts/turn) * turns per layer

And that, in my mind with a single wire wind might work well as the "back"
could be in the overlap of the kapton tape. If that could be placed on the
sides of the coil that is not in the core window, the build would not over
fill the core.

I considered that on my first wind, but had side walls on my bobbin and two
wires so it did not work well on the test runs.

With out side walls on the bobbin when winding I think it could work well

John


ykochcal
 

Chuck

Did you use Kapton with the "regular" glue? Or something special?

John

A couple of notes:

I use Kapton tape, because I always have, and I understand it.
Mylar would be just as good in my opinion.


-Chuck Harris


Daniel Koller
 

Hi John,

The flaw I see in the straight back scheme is that the wire needs to be brought back, and yes, you have half the voltage between the return wire and the starting end, but then you are winding subsequent layers over top of the cross-wire. That may put undue pressure on it so it will need to have a layer of tape over AND under it. So, possibly it's more of a hassle than the back and forth wind with 1 layer of tape over each layer of wire.

With a back and forth wind, I am saving a little bit in the diameter over a universal wind, so I think I can add a few layers of tape and still be ok. Also, I can cheat a little. The original coil was about 0.2" wide and tall. I can go a little wider I think, and still be very far away from the PCBs, and thus have a few fewer layers, over which I will put more tape. At least for my first trial run, this is what I intend to try. I'm going to try to wind without the bobbin, and varnish each layer and let it dry, so I don't have to worry about the sides falling off.

Dan

--------------------------------------------

On Mon, 3/23/15, 'John Snyder' Kochcal@... [TekScopes] <TekScopes@...> wrote:

Subject: RE: [TekScopes] RM503 ongoing repair saga T620 UNWOUND.
To: TekScopes@...
Date: Monday, March 23, 2015, 5:09 PM


 









Dan and Chuck



As Chuck points out for what I would call the back and forth
winding,



9

8765

1234



the voltage difference is:



You can calculate it:



(volts/turn) * 2 * (width of winding) / (diameter of
wire)



Which is the same as



(volts/turn) * 2 * turns per layer



Because 8 is on top of 1



But there is winding where you come straight back



9

5678

1234



That has 1/2 the voltage per layer and lower capacitance



(volts/turn) * turns per layer



And that, in my mind with a single wire wind might work well
as the "back"

could be in the overlap of the kapton tape. If that could be
placed on the

sides of the coil that is not in the core window, the build
would not over

fill the core.



I considered that on my first wind, but had side walls on my
bobbin and two

wires so it did not work well on the test runs.



With out side walls on the bobbin when winding I think it
could work well



John













#yiv3485506816 #yiv3485506816 --
#yiv3485506816ygrp-mkp {
border:1px solid #d8d8d8;font-family:Arial;margin:10px
0;padding:0 10px;}

#yiv3485506816 #yiv3485506816ygrp-mkp hr {
border:1px solid #d8d8d8;}

#yiv3485506816 #yiv3485506816ygrp-mkp #yiv3485506816hd {
color:#628c2a;font-size:85%;font-weight:700;line-height:122%;margin:10px
0;}

#yiv3485506816 #yiv3485506816ygrp-mkp #yiv3485506816ads {
margin-bottom:10px;}

#yiv3485506816 #yiv3485506816ygrp-mkp .yiv3485506816ad {
padding:0 0;}

#yiv3485506816 #yiv3485506816ygrp-mkp .yiv3485506816ad p {
margin:0;}

#yiv3485506816 #yiv3485506816ygrp-mkp .yiv3485506816ad a {
color:#0000ff;text-decoration:none;}
#yiv3485506816 #yiv3485506816ygrp-sponsor
#yiv3485506816ygrp-lc {
font-family:Arial;}

#yiv3485506816 #yiv3485506816ygrp-sponsor
#yiv3485506816ygrp-lc #yiv3485506816hd {
margin:10px
0px;font-weight:700;font-size:78%;line-height:122%;}

#yiv3485506816 #yiv3485506816ygrp-sponsor
#yiv3485506816ygrp-lc .yiv3485506816ad {
margin-bottom:10px;padding:0 0;}

#yiv3485506816 #yiv3485506816actions {
font-family:Verdana;font-size:11px;padding:10px 0;}

#yiv3485506816 #yiv3485506816activity {
background-color:#e0ecee;float:left;font-family:Verdana;font-size:10px;padding:10px;}

#yiv3485506816 #yiv3485506816activity span {
font-weight:700;}

#yiv3485506816 #yiv3485506816activity span:first-child {
text-transform:uppercase;}

#yiv3485506816 #yiv3485506816activity span a {
color:#5085b6;text-decoration:none;}

#yiv3485506816 #yiv3485506816activity span span {
color:#ff7900;}

#yiv3485506816 #yiv3485506816activity span
.yiv3485506816underline {
text-decoration:underline;}

#yiv3485506816 .yiv3485506816attach {
clear:both;display:table;font-family:Arial;font-size:12px;padding:10px
0;width:400px;}

#yiv3485506816 .yiv3485506816attach div a {
text-decoration:none;}

#yiv3485506816 .yiv3485506816attach img {
border:none;padding-right:5px;}

#yiv3485506816 .yiv3485506816attach label {
display:block;margin-bottom:5px;}

#yiv3485506816 .yiv3485506816attach label a {
text-decoration:none;}

#yiv3485506816 blockquote {
margin:0 0 0 4px;}

#yiv3485506816 .yiv3485506816bold {
font-family:Arial;font-size:13px;font-weight:700;}

#yiv3485506816 .yiv3485506816bold a {
text-decoration:none;}

#yiv3485506816 dd.yiv3485506816last p a {
font-family:Verdana;font-weight:700;}

#yiv3485506816 dd.yiv3485506816last p span {
margin-right:10px;font-family:Verdana;font-weight:700;}

#yiv3485506816 dd.yiv3485506816last p
span.yiv3485506816yshortcuts {
margin-right:0;}

#yiv3485506816 div.yiv3485506816attach-table div div a {
text-decoration:none;}

#yiv3485506816 div.yiv3485506816attach-table {
width:400px;}

#yiv3485506816 div.yiv3485506816file-title a, #yiv3485506816
div.yiv3485506816file-title a:active, #yiv3485506816
div.yiv3485506816file-title a:hover, #yiv3485506816
div.yiv3485506816file-title a:visited {
text-decoration:none;}

#yiv3485506816 div.yiv3485506816photo-title a,
#yiv3485506816 div.yiv3485506816photo-title a:active,
#yiv3485506816 div.yiv3485506816photo-title a:hover,
#yiv3485506816 div.yiv3485506816photo-title a:visited {
text-decoration:none;}

#yiv3485506816 div#yiv3485506816ygrp-mlmsg
#yiv3485506816ygrp-msg p a span.yiv3485506816yshortcuts {
font-family:Verdana;font-size:10px;font-weight:normal;}

#yiv3485506816 .yiv3485506816green {
color:#628c2a;}

#yiv3485506816 .yiv3485506816MsoNormal {
margin:0 0 0 0;}

#yiv3485506816 o {
font-size:0;}

#yiv3485506816 #yiv3485506816photos div {
float:left;width:72px;}

#yiv3485506816 #yiv3485506816photos div div {
border:1px solid
#666666;height:62px;overflow:hidden;width:62px;}

#yiv3485506816 #yiv3485506816photos div label {
color:#666666;font-size:10px;overflow:hidden;text-align:center;white-space:nowrap;width:64px;}

#yiv3485506816 #yiv3485506816reco-category {
font-size:77%;}

#yiv3485506816 #yiv3485506816reco-desc {
font-size:77%;}

#yiv3485506816 .yiv3485506816replbq {
margin:4px;}

#yiv3485506816 #yiv3485506816ygrp-actbar div a:first-child {
margin-right:2px;padding-right:5px;}

#yiv3485506816 #yiv3485506816ygrp-mlmsg {
font-size:13px;font-family:Arial, helvetica, clean,
sans-serif;}

#yiv3485506816 #yiv3485506816ygrp-mlmsg table {
font-size:inherit;font:100%;}

#yiv3485506816 #yiv3485506816ygrp-mlmsg select,
#yiv3485506816 input, #yiv3485506816 textarea {
font:99% Arial, Helvetica, clean, sans-serif;}

#yiv3485506816 #yiv3485506816ygrp-mlmsg pre, #yiv3485506816
code {
font:115% monospace;}

#yiv3485506816 #yiv3485506816ygrp-mlmsg * {
line-height:1.22em;}

#yiv3485506816 #yiv3485506816ygrp-mlmsg #yiv3485506816logo {
padding-bottom:10px;}


#yiv3485506816 #yiv3485506816ygrp-msg p a {
font-family:Verdana;}

#yiv3485506816 #yiv3485506816ygrp-msg
p#yiv3485506816attach-count span {
color:#1E66AE;font-weight:700;}

#yiv3485506816 #yiv3485506816ygrp-reco
#yiv3485506816reco-head {
color:#ff7900;font-weight:700;}

#yiv3485506816 #yiv3485506816ygrp-reco {
margin-bottom:20px;padding:0px;}

#yiv3485506816 #yiv3485506816ygrp-sponsor #yiv3485506816ov
li a {
font-size:130%;text-decoration:none;}

#yiv3485506816 #yiv3485506816ygrp-sponsor #yiv3485506816ov
li {
font-size:77%;list-style-type:square;padding:6px 0;}

#yiv3485506816 #yiv3485506816ygrp-sponsor #yiv3485506816ov
ul {
margin:0;padding:0 0 0 8px;}

#yiv3485506816 #yiv3485506816ygrp-text {
font-family:Georgia;}

#yiv3485506816 #yiv3485506816ygrp-text p {
margin:0 0 1em 0;}

#yiv3485506816 #yiv3485506816ygrp-text tt {
font-size:120%;}

#yiv3485506816 #yiv3485506816ygrp-vital ul li:last-child {
border-right:none !important;
}
#yiv3485506816


ykochcal
 

Dan

Ok, use the good ohm meter you say, with 4 wires.

I get 204.1 ohms for Grn to Wht (ground) and 204.6 ohms for Grn to Brn

As a second check while making the measurement I put a second voltmeter on
the wht to Brn and I get -0.001V when ohms are measured Grn to Wht
And 0.052V when measured Grn to Brn.

And Wht to Brn is .53 ohms from an earlier measure.

So I would say the HV starts at ground.

It comes out a hole in the side plate and connects to the rivet

John

-----Original Message-----
From: TekScopes@... [mailto:TekScopes@...]
Sent: Monday, March 23, 2015 10:01 AM
To: TekScopes@...
Subject: RE: [TekScopes] RM503 ongoing repair saga T620 UNWOUND.

John,

It occurs to me a DC voltage measurement should indicate if the -100V is
attached to the low end of the HV coil or not. Measure V from ground to
the -100V terminal. If you apply a small current to the HV coil (5V / 200
Ohm = 25 mA), and the -100V end is not connected, you will see no voltage
rise. On the other hand, if the two coils are in series, the ~~1 Ohm of the
-100V coil will show a voltage drop of about 25 millivolts. All assuming of
course that the coils are still intact. I don't remember if the HV coil was
open on that transformer. I guess if it's open, it's probably fried where
mine was, near the bottom. So then a visual check is the only way to see
where it is connected at the ends.

Dan


Dave Wise
 

I wound a 1000V 20kHz flyback transformer (3000V peaks) using the "Z" scheme. 1600 turns in nine layers. I put tape under and over the return wire, taking care to stagger the returns so they didn't build up a hump. I left some margin at each end, which I sealed with corona dope. I used acrylic packing tape for insulation, but I think it would be annoying to slit if your bobbin isn't exactly the normal width of the tape.

Z-winding reduces interlayer capacitance as well as voltage. It worked out fine for me.

FWIW,
Dave Wise

________________________________________
From: TekScopes@... [TekScopes@...]
Sent: Monday, March 23, 2015 10:28 AM
To: TekScopes@...
Subject: RE: [TekScopes] RM503 ongoing repair saga T620 UNWOUND.

Hi John,

The flaw I see in the straight back scheme is that the wire needs to be brought back, and yes, you have half the voltage between the return wire and the starting end, but then you are winding subsequent layers over top of the cross-wire. That may put undue pressure on it so it will need to have a layer of tape over AND under it. So, possibly it's more of a hassle than the back and forth wind with 1 layer of tape over each layer of wire.

With a back and forth wind, I am saving a little bit in the diameter over a universal wind, so I think I can add a few layers of tape and still be ok. Also, I can cheat a little. The original coil was about 0.2" wide and tall. I can go a little wider I think, and still be very far away from the PCBs, and thus have a few fewer layers, over which I will put more tape. At least for my first trial run, this is what I intend to try. I'm going to try to wind without the bobbin, and varnish each layer and let it dry, so I don't have to worry about the sides falling off.

Dan


Daniel Koller
 

Hi John,

Interesting! I agree with your measurments. Yes, 0.54Ohms is what I got for the +/-100V coils, NOT 1 Ohm as mentioned in my last message. 1.19Ohms for the 250V coil (to white = ground).

It's pretty clear then that the HV is a separate coil. Why did they do this instead of piggybacking it on to the -100V? I guess that also explains the split primary a little better too. The differential between the primary and the HV coil is not as high as I thought.

What a weird beast this little transformer is.

Dan

--------------------------------------------

On Mon, 3/23/15, 'John Snyder' Kochcal@... [TekScopes] <TekScopes@...> wrote:

Subject: RE: [TekScopes] RM503 ongoing repair saga T620 UNWOUND.
To: TekScopes@...
Date: Monday, March 23, 2015, 5:52 PM


 









Dan



Ok, use the good ohm meter you say, with 4 wires.



I get 204.1 ohms for Grn to Wht (ground) and 204.6 ohms for
Grn to Brn



As a second check while making the measurement I put a
second voltmeter on

the wht to Brn and I get -0.001V when ohms are measured Grn
to Wht

And 0.052V when measured Grn to Brn.



And Wht to Brn is .53 ohms from an earlier measure.



So I would say the HV starts at ground.



It comes out a hole in the side plate and connects to the
rivet



John



-----Original Message-----

From: TekScopes@...
[mailto:TekScopes@...]

Sent: Monday, March 23, 2015 10:01 AM

To: TekScopes@...

Subject: RE: [TekScopes] RM503 ongoing repair saga T620
UNWOUND.



John,



It occurs to me a DC voltage measurement should indicate if
the -100V is

attached to the low end of the HV coil or not. Measure V
from ground to

the -100V terminal. If you apply a small current to the HV
coil (5V / 200

Ohm = 25 mA), and the -100V end is not connected, you will
see no voltage

rise. On the other hand, if the two coils are in series,
the ~~1 Ohm of the

-100V coil will show a voltage drop of about 25 millivolts.
All assuming of

course that the coils are still intact. I don't
remember if the HV coil was

open on that transformer. I guess if it's open,
it's probably fried where

mine was, near the bottom. So then a visual check is the
only way to see

where it is connected at the ends.



Dan


Chuck Harris <cfharris@...>
 

What are you planning on doing with the varnish? I can't
recommend taking a random varnish material and impregnating
the coil with it... some varnish is good, some isn't.

A universal wind done correctly is just as compact as
when the wire is spooled neatly in layers without any
separators.

Calculate how many layers you are going to need by figuring
out the number of turns per layer, and dividing that into
the number of turns. Figure 3 mil per layer for the tape,
and a wire diameter per layer for the copper, and you
should be able to come up with a pretty good estimate.

-Chuck Harris

Daniel Koller kaboomdk@... [TekScopes] wrote:

Thanks Chuck!

Ok, I am going to go ahead with an attempt at winding my own, starting possibly as
early as this week. Now I just need to get some toluene to thing out my GE
varnish a little bit (the recommended solvent was 50/50 Acetone/Toluene).

I think I am going to "skimp" and wind the primary as one turn at the bottom of
the coil, rather than break it in two halves. I understand the intended point
about shielding, but A) my scope is all tubes (no solid state at the inputs in the
older models) and B) I have a documented failure due to that "shield winding"
which would probably not have occurred had it been connected to the -100V
directly, thus reducing gradients. Also, for this first one, I need to keep the
winding a little simpler.

For a given cold width, how much larger in diameter is a universal coil versus a
simple layered stack? I need to make sure I have enough room for layers of mylar
tape.

Dan


Chuck Harris <cfharris@...>
 

I use Permacel P221, which is a 1/2 inch wide,
amber Kapton 1 mil tape, with a silicone high
temperature adhesive. It is 180C temperature
class, and has 7KV dielectric strength.

I bought a lot of it back when I was making a
lot of transformers for switching supplies. As
I said, I understand it.

-Chuck Harris

'John Snyder' Kochcal@... [TekScopes] wrote:


Chuck

Did you use Kapton with the "regular" glue? Or something special?

John


Chuck Harris <cfharris@...>
 

The 503 schematic shows the HV coil is attached at the -100V tap,
not ground.

-Chuck Harris



Daniel Koller kaboomdk@... [TekScopes] wrote:

Hi John,

Interesting! I agree with your measurments. Yes, 0.54Ohms is what I got for the
+/-100V coils, NOT 1 Ohm as mentioned in my last message. 1.19Ohms for the 250V
coil (to white = ground).

It's pretty clear then that the HV is a separate coil. Why did they do this
instead of piggybacking it on to the -100V? I guess that also explains the split
primary a little better too. The differential between the primary and the HV coil
is not as high as I thought.

What a weird beast this little transformer is.

Dan


Daniel Koller
 

Yes, agreed, but it definitely does not agree with John's measurements.

Dan

--------------------------------------------

On Mon, 3/23/15, Chuck Harris cfharris@... [TekScopes] <TekScopes@...> wrote:

Subject: Re: [TekScopes] RM503 ongoing repair saga T620 UNWOUND.
To: TekScopes@...
Date: Monday, March 23, 2015, 8:26 PM


 









The 503 schematic shows the HV coil is attached at
the -100V tap,

not ground.



-Chuck Harris



Daniel Koller kaboomdk@... [TekScopes] wrote:

> Hi John,

>

> Interesting! I agree with your measurments. Yes,
0.54Ohms is what I got for the

> +/-100V coils, NOT 1 Ohm as mentioned in my last
message. 1.19Ohms for the 250V

> coil (to white = ground).

>

> It's pretty clear then that the HV is a separate
coil. Why did they do this

> instead of piggybacking it on to the -100V? I guess
that also explains the split

> primary a little better too. The differential between
the primary and the HV coil

> is not as high as I thought.

>

> What a weird beast this little transformer is.

>

> Dan











#yiv1776970434 #yiv1776970434 --
#yiv1776970434ygrp-mkp {
border:1px solid #d8d8d8;font-family:Arial;margin:10px
0;padding:0 10px;}

#yiv1776970434 #yiv1776970434ygrp-mkp hr {
border:1px solid #d8d8d8;}

#yiv1776970434 #yiv1776970434ygrp-mkp #yiv1776970434hd {
color:#628c2a;font-size:85%;font-weight:700;line-height:122%;margin:10px
0;}

#yiv1776970434 #yiv1776970434ygrp-mkp #yiv1776970434ads {
margin-bottom:10px;}

#yiv1776970434 #yiv1776970434ygrp-mkp .yiv1776970434ad {
padding:0 0;}

#yiv1776970434 #yiv1776970434ygrp-mkp .yiv1776970434ad p {
margin:0;}

#yiv1776970434 #yiv1776970434ygrp-mkp .yiv1776970434ad a {
color:#0000ff;text-decoration:none;}
#yiv1776970434 #yiv1776970434ygrp-sponsor
#yiv1776970434ygrp-lc {
font-family:Arial;}

#yiv1776970434 #yiv1776970434ygrp-sponsor
#yiv1776970434ygrp-lc #yiv1776970434hd {
margin:10px
0px;font-weight:700;font-size:78%;line-height:122%;}

#yiv1776970434 #yiv1776970434ygrp-sponsor
#yiv1776970434ygrp-lc .yiv1776970434ad {
margin-bottom:10px;padding:0 0;}

#yiv1776970434 #yiv1776970434actions {
font-family:Verdana;font-size:11px;padding:10px 0;}

#yiv1776970434 #yiv1776970434activity {
background-color:#e0ecee;float:left;font-family:Verdana;font-size:10px;padding:10px;}

#yiv1776970434 #yiv1776970434activity span {
font-weight:700;}

#yiv1776970434 #yiv1776970434activity span:first-child {
text-transform:uppercase;}

#yiv1776970434 #yiv1776970434activity span a {
color:#5085b6;text-decoration:none;}

#yiv1776970434 #yiv1776970434activity span span {
color:#ff7900;}

#yiv1776970434 #yiv1776970434activity span
.yiv1776970434underline {
text-decoration:underline;}

#yiv1776970434 .yiv1776970434attach {
clear:both;display:table;font-family:Arial;font-size:12px;padding:10px
0;width:400px;}

#yiv1776970434 .yiv1776970434attach div a {
text-decoration:none;}

#yiv1776970434 .yiv1776970434attach img {
border:none;padding-right:5px;}

#yiv1776970434 .yiv1776970434attach label {
display:block;margin-bottom:5px;}

#yiv1776970434 .yiv1776970434attach label a {
text-decoration:none;}

#yiv1776970434 blockquote {
margin:0 0 0 4px;}

#yiv1776970434 .yiv1776970434bold {
font-family:Arial;font-size:13px;font-weight:700;}

#yiv1776970434 .yiv1776970434bold a {
text-decoration:none;}

#yiv1776970434 dd.yiv1776970434last p a {
font-family:Verdana;font-weight:700;}

#yiv1776970434 dd.yiv1776970434last p span {
margin-right:10px;font-family:Verdana;font-weight:700;}

#yiv1776970434 dd.yiv1776970434last p
span.yiv1776970434yshortcuts {
margin-right:0;}

#yiv1776970434 div.yiv1776970434attach-table div div a {
text-decoration:none;}

#yiv1776970434 div.yiv1776970434attach-table {
width:400px;}

#yiv1776970434 div.yiv1776970434file-title a, #yiv1776970434
div.yiv1776970434file-title a:active, #yiv1776970434
div.yiv1776970434file-title a:hover, #yiv1776970434
div.yiv1776970434file-title a:visited {
text-decoration:none;}

#yiv1776970434 div.yiv1776970434photo-title a,
#yiv1776970434 div.yiv1776970434photo-title a:active,
#yiv1776970434 div.yiv1776970434photo-title a:hover,
#yiv1776970434 div.yiv1776970434photo-title a:visited {
text-decoration:none;}

#yiv1776970434 div#yiv1776970434ygrp-mlmsg
#yiv1776970434ygrp-msg p a span.yiv1776970434yshortcuts {
font-family:Verdana;font-size:10px;font-weight:normal;}

#yiv1776970434 .yiv1776970434green {
color:#628c2a;}

#yiv1776970434 .yiv1776970434MsoNormal {
margin:0 0 0 0;}

#yiv1776970434 o {
font-size:0;}

#yiv1776970434 #yiv1776970434photos div {
float:left;width:72px;}

#yiv1776970434 #yiv1776970434photos div div {
border:1px solid
#666666;height:62px;overflow:hidden;width:62px;}

#yiv1776970434 #yiv1776970434photos div label {
color:#666666;font-size:10px;overflow:hidden;text-align:center;white-space:nowrap;width:64px;}

#yiv1776970434 #yiv1776970434reco-category {
font-size:77%;}

#yiv1776970434 #yiv1776970434reco-desc {
font-size:77%;}

#yiv1776970434 .yiv1776970434replbq {
margin:4px;}

#yiv1776970434 #yiv1776970434ygrp-actbar div a:first-child {
margin-right:2px;padding-right:5px;}

#yiv1776970434 #yiv1776970434ygrp-mlmsg {
font-size:13px;font-family:Arial, helvetica, clean,
sans-serif;}

#yiv1776970434 #yiv1776970434ygrp-mlmsg table {
font-size:inherit;font:100%;}

#yiv1776970434 #yiv1776970434ygrp-mlmsg select,
#yiv1776970434 input, #yiv1776970434 textarea {
font:99% Arial, Helvetica, clean, sans-serif;}

#yiv1776970434 #yiv1776970434ygrp-mlmsg pre, #yiv1776970434
code {
font:115% monospace;}

#yiv1776970434 #yiv1776970434ygrp-mlmsg * {
line-height:1.22em;}

#yiv1776970434 #yiv1776970434ygrp-mlmsg #yiv1776970434logo {
padding-bottom:10px;}


#yiv1776970434 #yiv1776970434ygrp-msg p a {
font-family:Verdana;}

#yiv1776970434 #yiv1776970434ygrp-msg
p#yiv1776970434attach-count span {
color:#1E66AE;font-weight:700;}

#yiv1776970434 #yiv1776970434ygrp-reco
#yiv1776970434reco-head {
color:#ff7900;font-weight:700;}

#yiv1776970434 #yiv1776970434ygrp-reco {
margin-bottom:20px;padding:0px;}

#yiv1776970434 #yiv1776970434ygrp-sponsor #yiv1776970434ov
li a {
font-size:130%;text-decoration:none;}

#yiv1776970434 #yiv1776970434ygrp-sponsor #yiv1776970434ov
li {
font-size:77%;list-style-type:square;padding:6px 0;}

#yiv1776970434 #yiv1776970434ygrp-sponsor #yiv1776970434ov
ul {
margin:0;padding:0 0 0 8px;}

#yiv1776970434 #yiv1776970434ygrp-text {
font-family:Georgia;}

#yiv1776970434 #yiv1776970434ygrp-text p {
margin:0 0 1em 0;}

#yiv1776970434 #yiv1776970434ygrp-text tt {
font-size:120%;}

#yiv1776970434 #yiv1776970434ygrp-vital ul li:last-child {
border-right:none !important;
}
#yiv1776970434


Daniel Koller
 

Chuck,

I was planning on tacking the layers with the GE varnish. I'll impregnate with parafin/wax when all is done.

GE varnish is a very well known, stable, vacuum compatible varnish used in many low-temperature applications. I am not sure what it's dielectric strength is, but it's been studied extensively, so I am sure I will turn up something on it. There was some sort of varnish-like fill material used between layers of the lower coils, sometimes 3-6 mils or so thick. I flaked it off in chunks as I got down to certain layers.

I may have some Kapton tape. I will look for it. But the mylar tape I have has the advantage that it's exactly the right width for the LV coils.

Do you see any reason NOT to wind a single primary coil at the bottom of the stack?

Dan



--------------------------------------------

On Mon, 3/23/15, Chuck Harris cfharris@... [TekScopes] <TekScopes@...> wrote:

Subject: Re: [TekScopes] RM503 ongoing repair saga T620 UNWOUND.
To: TekScopes@...
Date: Monday, March 23, 2015, 8:07 PM


 









What are you planning on doing with the varnish? I
can't

recommend taking a random varnish material and
impregnating

the coil with it... some varnish is good, some isn't.



A universal wind done correctly is just as compact as

when the wire is spooled neatly in layers without any

separators.



Calculate how many layers you are going to need by
figuring

out the number of turns per layer, and dividing that into

the number of turns. Figure 3 mil per layer for the
tape,

and a wire diameter per layer for the copper, and you

should be able to come up with a pretty good estimate.



-Chuck Harris



Daniel Koller kaboomdk@... [TekScopes] wrote:

> Thanks Chuck!

>

> Ok, I am going to go ahead with an attempt at winding
my own, starting possibly as

> early as this week. Now I just need to get some
toluene to thing out my GE

> varnish a little bit (the recommended solvent was 50/50
Acetone/Toluene).

>

> I think I am going to "skimp" and wind the
primary as one turn at the bottom of

> the coil, rather than break it in two halves. I
understand the intended point

> about shielding, but A) my scope is all tubes (no solid
state at the inputs in the

> older models) and B) I have a documented failure due to
that "shield winding"

> which would probably not have occurred had it been
connected to the -100V

> directly, thus reducing gradients. Also, for this
first one, I need to keep the

> winding a little simpler.

>

> For a given cold width, how much larger in diameter is
a universal coil versus a

> simple layered stack? I need to make sure I have
enough room for layers of mylar

> tape.

>

> Dan

>











#yiv3350127267 #yiv3350127267 --
#yiv3350127267ygrp-mkp {
border:1px solid #d8d8d8;font-family:Arial;margin:10px
0;padding:0 10px;}

#yiv3350127267 #yiv3350127267ygrp-mkp hr {
border:1px solid #d8d8d8;}

#yiv3350127267 #yiv3350127267ygrp-mkp #yiv3350127267hd {
color:#628c2a;font-size:85%;font-weight:700;line-height:122%;margin:10px
0;}

#yiv3350127267 #yiv3350127267ygrp-mkp #yiv3350127267ads {
margin-bottom:10px;}

#yiv3350127267 #yiv3350127267ygrp-mkp .yiv3350127267ad {
padding:0 0;}

#yiv3350127267 #yiv3350127267ygrp-mkp .yiv3350127267ad p {
margin:0;}

#yiv3350127267 #yiv3350127267ygrp-mkp .yiv3350127267ad a {
color:#0000ff;text-decoration:none;}
#yiv3350127267 #yiv3350127267ygrp-sponsor
#yiv3350127267ygrp-lc {
font-family:Arial;}

#yiv3350127267 #yiv3350127267ygrp-sponsor
#yiv3350127267ygrp-lc #yiv3350127267hd {
margin:10px
0px;font-weight:700;font-size:78%;line-height:122%;}

#yiv3350127267 #yiv3350127267ygrp-sponsor
#yiv3350127267ygrp-lc .yiv3350127267ad {
margin-bottom:10px;padding:0 0;}

#yiv3350127267 #yiv3350127267actions {
font-family:Verdana;font-size:11px;padding:10px 0;}

#yiv3350127267 #yiv3350127267activity {
background-color:#e0ecee;float:left;font-family:Verdana;font-size:10px;padding:10px;}

#yiv3350127267 #yiv3350127267activity span {
font-weight:700;}

#yiv3350127267 #yiv3350127267activity span:first-child {
text-transform:uppercase;}

#yiv3350127267 #yiv3350127267activity span a {
color:#5085b6;text-decoration:none;}

#yiv3350127267 #yiv3350127267activity span span {
color:#ff7900;}

#yiv3350127267 #yiv3350127267activity span
.yiv3350127267underline {
text-decoration:underline;}

#yiv3350127267 .yiv3350127267attach {
clear:both;display:table;font-family:Arial;font-size:12px;padding:10px
0;width:400px;}

#yiv3350127267 .yiv3350127267attach div a {
text-decoration:none;}

#yiv3350127267 .yiv3350127267attach img {
border:none;padding-right:5px;}

#yiv3350127267 .yiv3350127267attach label {
display:block;margin-bottom:5px;}

#yiv3350127267 .yiv3350127267attach label a {
text-decoration:none;}

#yiv3350127267 blockquote {
margin:0 0 0 4px;}

#yiv3350127267 .yiv3350127267bold {
font-family:Arial;font-size:13px;font-weight:700;}

#yiv3350127267 .yiv3350127267bold a {
text-decoration:none;}

#yiv3350127267 dd.yiv3350127267last p a {
font-family:Verdana;font-weight:700;}

#yiv3350127267 dd.yiv3350127267last p span {
margin-right:10px;font-family:Verdana;font-weight:700;}

#yiv3350127267 dd.yiv3350127267last p
span.yiv3350127267yshortcuts {
margin-right:0;}

#yiv3350127267 div.yiv3350127267attach-table div div a {
text-decoration:none;}

#yiv3350127267 div.yiv3350127267attach-table {
width:400px;}

#yiv3350127267 div.yiv3350127267file-title a, #yiv3350127267
div.yiv3350127267file-title a:active, #yiv3350127267
div.yiv3350127267file-title a:hover, #yiv3350127267
div.yiv3350127267file-title a:visited {
text-decoration:none;}

#yiv3350127267 div.yiv3350127267photo-title a,
#yiv3350127267 div.yiv3350127267photo-title a:active,
#yiv3350127267 div.yiv3350127267photo-title a:hover,
#yiv3350127267 div.yiv3350127267photo-title a:visited {
text-decoration:none;}

#yiv3350127267 div#yiv3350127267ygrp-mlmsg
#yiv3350127267ygrp-msg p a span.yiv3350127267yshortcuts {
font-family:Verdana;font-size:10px;font-weight:normal;}

#yiv3350127267 .yiv3350127267green {
color:#628c2a;}

#yiv3350127267 .yiv3350127267MsoNormal {
margin:0 0 0 0;}

#yiv3350127267 o {
font-size:0;}

#yiv3350127267 #yiv3350127267photos div {
float:left;width:72px;}

#yiv3350127267 #yiv3350127267photos div div {
border:1px solid
#666666;height:62px;overflow:hidden;width:62px;}

#yiv3350127267 #yiv3350127267photos div label {
color:#666666;font-size:10px;overflow:hidden;text-align:center;white-space:nowrap;width:64px;}

#yiv3350127267 #yiv3350127267reco-category {
font-size:77%;}

#yiv3350127267 #yiv3350127267reco-desc {
font-size:77%;}

#yiv3350127267 .yiv3350127267replbq {
margin:4px;}

#yiv3350127267 #yiv3350127267ygrp-actbar div a:first-child {
margin-right:2px;padding-right:5px;}

#yiv3350127267 #yiv3350127267ygrp-mlmsg {
font-size:13px;font-family:Arial, helvetica, clean,
sans-serif;}

#yiv3350127267 #yiv3350127267ygrp-mlmsg table {
font-size:inherit;font:100%;}

#yiv3350127267 #yiv3350127267ygrp-mlmsg select,
#yiv3350127267 input, #yiv3350127267 textarea {
font:99% Arial, Helvetica, clean, sans-serif;}

#yiv3350127267 #yiv3350127267ygrp-mlmsg pre, #yiv3350127267
code {
font:115% monospace;}

#yiv3350127267 #yiv3350127267ygrp-mlmsg * {
line-height:1.22em;}

#yiv3350127267 #yiv3350127267ygrp-mlmsg #yiv3350127267logo {
padding-bottom:10px;}


#yiv3350127267 #yiv3350127267ygrp-msg p a {
font-family:Verdana;}

#yiv3350127267 #yiv3350127267ygrp-msg
p#yiv3350127267attach-count span {
color:#1E66AE;font-weight:700;}

#yiv3350127267 #yiv3350127267ygrp-reco
#yiv3350127267reco-head {
color:#ff7900;font-weight:700;}

#yiv3350127267 #yiv3350127267ygrp-reco {
margin-bottom:20px;padding:0px;}

#yiv3350127267 #yiv3350127267ygrp-sponsor #yiv3350127267ov
li a {
font-size:130%;text-decoration:none;}

#yiv3350127267 #yiv3350127267ygrp-sponsor #yiv3350127267ov
li {
font-size:77%;list-style-type:square;padding:6px 0;}

#yiv3350127267 #yiv3350127267ygrp-sponsor #yiv3350127267ov
ul {
margin:0;padding:0 0 0 8px;}

#yiv3350127267 #yiv3350127267ygrp-text {
font-family:Georgia;}

#yiv3350127267 #yiv3350127267ygrp-text p {
margin:0 0 1em 0;}

#yiv3350127267 #yiv3350127267ygrp-text tt {
font-size:120%;}

#yiv3350127267 #yiv3350127267ygrp-vital ul li:last-child {
border-right:none !important;
}
#yiv3350127267