Date   

Re: PG502 power lamp off

Christoph
 

Hi Dennis,

your objection may be justified. Only, in this case I had the strong suspicion that the 5V won't be there since the device doesn't show any sign of putting out any pulse. Right after posting my question here I was confirmed that I was right.

The +5V after the 1Ohm resistor were 0.7 V or so. I unsoldered the series regulator transistor Q650 also and measured open in every direction between b-c-e in either direction.


BTW, there was a mica isolation foil underneath the heat sink floating around and the hole for the fastening screw was empty.

Is there a screw normally in there? And is it isolated also from the backside if the PCB?
This is a plated through hole and I believe it carried GND, if I'm right. Since that kind of transistors normally have one electrode (c) connected to the heat sink I thinnk there should be a complete isolation kit in place.

--
Christoph


Re: 7904 - Very Dim Trace

 

On Mon, 15 Feb 2016 00:12:42 -0600, you wrote:

At 12 kilovolts, it was not as bright or sharp as when the 7904 is
using a 24 kilovolt PDA but it was close. Deflection was down about
10% versus being halved with no PDA at all.
Oh, and this means that it is possible that the high voltage
multiplier has been bad for some time and the calibration of your 7904
compensated for the slightly low deflection.


Re: 7904 - Very Dim Trace

 

One reason it will not work is that a standard high voltage multimeter
probe is not designed to operate much above 60 Hz and the heater
voltage is about 50 kHz.

A high voltage differential oscilloscope probe could do it.

On 14 Feb 2016 21:13:44 -0800, you wrote:

I don't see that working. The idea was to measure 6.3 VAC that's floating on -3kV DC.


Re: 7904 - Very Dim Trace

 

On 14 Feb 2016 20:04:24 -0800, you wrote:

No joy. The Service Manual Parts List calls out a 10 ohm series resistor in the heater, and that's what's in there. It measures 9.8 ohms. Re-checked deflection. Both X and Y are within 2%.
I'm still not convinced the 14kV is the only problem. This scope is new to me, and it could well have been 'mal-adjusted' to make up for the low PDA voltage. Still noodling on this. Continued comments welcomed...
I am not convinced either so I ran another test ...

I was considering how I could operate my 7904 with a 14 kilovolt PDA
to see how the CRT would behave but I lack a suitable power supply. I
do however have a 7603 sitting right next to the 7904 on the same desk
and it uses a 12 kilovolt PDA and the connectors are the same ...

So I reversed the 7603 to allow the anode lead to reach and connected
the PDA lead from the 7904 CRT to the 7603 power supply and powered
the 7904 up and then the 7603. It worked! I could even turn the 7603
off and watch the effect on the 7904 CRT as the PDA slowly discharged.

At 12 kilovolts, it was not as bright or sharp as when the 7904 is
using a 24 kilovolt PDA but it was close. Deflection was down about
10% versus being halved with no PDA at all.

So reducing the PDA to half of its nominal value had a minor but
noticeable effect on brightness and deflection. I think you have a
different problem causing a very dim CRT.

You have already verified that the output from the z-axis amplifier is
good and that the resistor in series with the heater is probably good
so I suspect the z-axis DC restorer. Did you make a measurement of
the actual grid to cathode voltage? This can also be done by floating
your Fluke...

The way the DC restorer works is that the difference between the
z-axis amplifier output and the grid bias voltage is subtracted from
the cathode voltage making the grid voltage more negative. Maybe
R1675 at the grid bias adjustment is open or the grid bias
potentiometer is bad.


Re: 7904 - Very Dim Trace

 

Nuke the Fluke, err, I mean float the Fluke.

It is the only way to be sure. :)

Carbon composition resistors have a high voltage coefficient of
resistance but that only applies at high voltages which this decidedly
is not.

If the thermal coefficient of resistance is a problem then the
resistor is bad and in that case, I would expect the nominal value to
be wrong as well.

On 14 Feb 2016 20:32:14 -0800, you wrote:

Sure - anything's possible I suppose. I now know that carbon comp resistors have a large voltage coefficient, but that's not really related to this. Anyway, changing out this resistor is pretty easy. I'll can give that a try...

It's easier and less risky than attempting to float my TRMS Fluke at -3 kV...


Re: 7904 - Very Dim Trace

Jim McIntyre
 

I don't see that working. The idea was to measure 6.3 VAC that's floating on -3kV DC.


Re: 7904 - Very Dim Trace

Brian Bloom
 

Apologies if my lack of experience + comments come as a possible irritation by some chance.

After all my time in construction and automotive, I have learned that it is better to state my ideas and thoughts regardless of their faults in the event they may trigger a new thought process in someone else and lead to the best outcome.

I wouldn't want to float a Fluke dmm at that voltage, either. What about making your own voltage divider x1000 probe for the application?


---In TekScopes@..., <jimmcintyre22@...> wrote :

Sure - anything's possible I suppose. I now know that carbon comp resistors have a large voltage coefficient, but that's not really related to this. Anyway, changing out this resistor is pretty easy. I'll can give that a try...

It's easier and less risky than attempting to float my TRMS Fluke at -3 kV...


Re: Introduction of myself!

 

On Sun, 14 Feb 2016 19:49:18 -0800, you wrote:

Hi David,
The delay is 75nSec as you said, but the risetime is 175pSec which is also a
consideration, although probably not for the 106 pulse generator which, I
suspect, has about a 1nSec risetime.
The service manual for the 106 recommends a Tektronix 4S3 which only
has a bandwidth of 1 GHz nicely matching the 7S14 or S1 so a 7M11
would be plenty fast.

I tried several ways to measure John Addis' little TD Pulse Generator
(067-0681-01) which had a 30pSec risetime and I could not use the 7M11. The
Random Sampling Mode of the 7T11 had so much time jitter on the fastest
sweep speeds that it didn't work either. In the end I had to give up.
I have had good results with random sampling mode on my 7T11A however
it is very finicky at best. Under ideal conditions, it easily met the
specified 30 picoseconds of jitter in random sampling mode.

The difficulty of knowing if the 7T11 is working correctly and
repairing it if not is why I would recommend a 7S14 instead despite
the battery issue.

Dennis Tillman W7PF


Re: Introduction of myself!

Brian Bloom
 

Certainly sounds like it would be easier to mod a 7s14 than deal with multiple units and their own independent costs and issues.

I managed to find Mercury cells of the correct voltage, though. Not to say that they are the correct physical size.. at $50 for 4 of them, I may as well just go thru with the mods so it'll just be done in the event those cells inevitably die. (hopefully I find a 7s14 in the near-er future!)


---In TekScopes@..., <davidwhess@...> wrote :

Each sampling input in the 7S14 takes 2 mercury cells so there are 4
cells total.

One option is to replace the original batteries with silver oxide or
alkaline cells. If the slightly higher voltage is a problem, then
either small signal schottky rectifiers can be added in series with
the batteries or the sampling drive circuit can be adjusted to produce
a slightly higher output by lowering the value of R15 and R17.

Another option is to use a photovoltaic optocoupler with a pair of
shunt diodes on the output to produce floating 1.35 volt supplies for
each sampling head.

I think doing the above would be easier than dealing with a
7S11/7T11/S1/S2.


Re: 7904 - Very Dim Trace

Jim McIntyre
 

Sure - anything's possible I suppose. I now know that carbon comp resistors have a large voltage coefficient, but that's not really related to this. Anyway, changing out this resistor is pretty easy. I'll can give that a try...

It's easier and less risky than attempting to float my TRMS Fluke at -3 kV...


Re: Introduction of myself!

 

Each sampling input in the 7S14 takes 2 mercury cells so there are 4
cells total.

One option is to replace the original batteries with silver oxide or
alkaline cells. If the slightly higher voltage is a problem, then
either small signal schottky rectifiers can be added in series with
the batteries or the sampling drive circuit can be adjusted to produce
a slightly higher output by lowering the value of R15 and R17.

Another option is to use a photovoltaic optocoupler with a pair of
shunt diodes on the output to produce floating 1.35 volt supplies for
each sampling head.

I think doing the above would be easier than dealing with a
7S11/7T11/S1/S2.

On 14 Feb 2016 19:39:13 -0800, you wrote:

I am definitely open to modifying a unit to meet modern requirements - or whatever must be done to get it working. With the price of a 7M11 at well over $200, if I could score a 7S14 for cheap it would be fantastic, regardless of the battery issues. I'll have to get my hands on one to see what you mean about the space being limited for modifications and alternatives, though. In that scenario, I suppose that you mean I would need to externally power it from a floating supply.... and I'd have no problem with doing so if it provides me a functional sampling unit.

On another note, I just smacked myself for buying a 7584 when I need sampling plugins - not another scope!! I suppose I'm not alone here in my addiction to impulsive purchases of old high-end Tek gear, though...


Re: Introduction of myself!

 

On 14 Feb 2016 19:25:05 -0800, you wrote:

I have a 067-0625-00 peak to peak detector that has a pair of good mercury batteries in it, but according to part numbers and specs, not only are they the wrong ones, they are much higher capacity. Same voltage, though. Regardless of whether that is a feasible option, I would definitely be more than willing/happy to perform modifications to the unit to accommodate modern batteries. There is a local machine shop that would be more than happy to design and make adapters if necessary. I may not have all that much experience, but I've got the determination and an open mind!

As for the 7M11 dual analog delay lines, they appear to be more expensive than I just paid for my 7854, a matching 7B87 timebase, AND a 7A24 amplifier..... so unless I have no choice, I prefer to avoid. Doesn't the 564 use the same type of custom line as the 7M11 for it's delay, though?
I do not know about the 564 but I was not recommending a 7M11 for the
reason you identify; it is both rare and usually expensive.

The 7S14 is more common and easier to use than a 7T11 in random
sampling mode. Just thinking about everything which can go wrong with
a 7T11 raises the hair on the back of my neck.


Re: 7904 - Very Dim Trace

Brian Bloom
 

From what I've been reading (books and forums), it doesn't have to be just the resistance that changes, does it? Or will that change first and indicate a change in other characteristics?

What I'm getting at, is from what I gather so far, a carbon comp resistor can act totally different when it's loaded and under high stress yet still measure relatively close to it's nominal value. Much in the same way that a transistor can check out fine from a dmm or even simple tester, but when hooked up to a curve tracer it will become apparent that the device has failed.

I could have misinterpreted what I've read so far, and I welcome corrections to my understanding & interpretation of info.

---In TekScopes@..., <jimmcintyre22@...> wrote :

No joy. The Service Manual Parts List calls out a 10 ohm series resistor in the heater, and that's what's in there. It measures 9.8 ohms. Re-checked deflection. Both X and Y are within 2%.

I'm still not convinced the 14kV is the only problem. This scope is new to me, and it could well have been 'mal-adjusted' to make up for the low PDA voltage. Still noodling on this. Continued comments welcomed...


Re: Introduction of myself!

Brian Bloom
 

Dennis,

Yes, the 106 is specified at 1nSec or less.
I would love to get my hands on a 284, which has a 250ps risetime, but I missed my last ebay opportunity for it. There's currently a 110 pulse gen up for grabs. It's missing half the case, though.

That little 30ps TD pulser is another piece I want/may need for the 485s in my stash. I suppose I could steal a TD from one and build my own if I must, though. The 113 delay cable is spec'd for less than 100psec risetime, which is what concerned me about using alternatives for the 106 cal. That's the best I've seen in my web/Tekwiki surfing regarding risetime of delay line. It doesn't sound like even that would be up to the task of measuring the '0681-01 though.

As for the 113, there's one up for grabs on ebay which the seller may be negotiable on if someone here is interested in one. I can't justify buying it right now as much as I want it.

-Brian


---In TekScopes@..., <dennis@...> wrote :

Hi David,
The delay is 75nSec as you said, but the risetime is 175pSec which is also a
consideration, although probably not for the 106 pulse generator which, I
suspect, has about a 1nSec risetime.

I tried several ways to measure John Addis' little TD Pulse Generator
(067-0681-01) which had a 30pSec risetime and I could not use the 7M11. The
Random Sampling Mode of the 7T11 had so much time jitter on the fastest
sweep speeds that it didn't work either. In the end I had to give up.

Dennis Tillman W7PF


Re: 7904 - Very Dim Trace

Jim McIntyre
 

No joy. The Service Manual Parts List calls out a 10 ohm series resistor in the heater, and that's what's in there. It measures 9.8 ohms. Re-checked deflection. Both X and Y are within 2%.

I'm still not convinced the 14kV is the only problem. This scope is new to me, and it could well have been 'mal-adjusted' to make up for the low PDA voltage. Still noodling on this. Continued comments welcomed...


Re: Introduction of myself!

 

Hi David,
The delay is 75nSec as you said, but the risetime is 175pSec which is also a
consideration, although probably not for the 106 pulse generator which, I
suspect, has about a 1nSec risetime.

I tried several ways to measure John Addis' little TD Pulse Generator
(067-0681-01) which had a 30pSec risetime and I could not use the 7M11. The
Random Sampling Mode of the 7T11 had so much time jitter on the fastest
sweep speeds that it didn't work either. In the end I had to give up.

Dennis Tillman W7PF

-----Original Message-----
Sent: Sunday, February 14, 2016 6:42 PM
Subject: Re: [TekScopes] Introduction of myself!

The fastest display window for the 7T11 extends from 50 nanoseconds to
100 nanoseconds so a 75 nanosecond delay is optimum and that is what the
7M11 has.

On Sun, 14 Feb 2016 18:26:52 -0800, you wrote:

Off the top of my head I don't remember the risetime of the 106 or the
setup time the 7T11 needs to trigger. One possible solution might be the
7M11 Dual Delay Line plugin. It can sometimes be found for a reasonable
price.

Dennis Tillman W7PF
------------------------------------
Posted by: David <@DWH>
------------------------------------


Re: Introduction of myself!

 

You need patience to snag bargains on Ebay. Some sellers want a lot for the
7M11 because they don't know what it is or because it may seem rare. But
others are happy to get rid of it since there is little demand for it. The
asking price will vary accordingly. I have seen them for a lot of money and
for very little money. Often, the price you are willing to pay will depend
on how fast you want it. Some of the things I have been looking for don't
show up more often than once every 5 years. I don't mind waiting for 5 years
if the price is right.
Dennis Tillman W7PF

-----Original Message-----
Sent: Sunday, February 14, 2016 7:25 PM
Subject: RE: [TekScopes] Introduction of myself!

I have a 067-0625-00 peak to peak detector that has a pair of good mercury
batteries in it, but according to part numbers and specs, not only are they
the wrong ones, they are much higher capacity. Same voltage, though.
Regardless of whether that is a feasible option, I would definitely be more
than willing/happy to perform modifications to the unit to accommodate
modern batteries. There is a local machine shop that would be more than
happy to design and make adapters if necessary. I may not have all that
much experience, but I've got the determination and an open mind!

As for the 7M11 dual analog delay lines, they appear to be more expensive
than I just paid for my 7854, a matching 7B87 timebase, AND a 7A24
amplifier..... so unless I have no choice, I prefer to avoid. Doesn't the
564 use the same type of custom line as the 7M11 for it's delay, though?

------------------------------------
Posted by: analogaddict013@...
------------------------------------


Re: Introduction of myself!

Brian Bloom
 

I am definitely open to modifying a unit to meet modern requirements - or whatever must be done to get it working. With the price of a 7M11 at well over $200, if I could score a 7S14 for cheap it would be fantastic, regardless of the battery issues. I'll have to get my hands on one to see what you mean about the space being limited for modifications and alternatives, though. In that scenario, I suppose that you mean I would need to externally power it from a floating supply.... and I'd have no problem with doing so if it provides me a functional sampling unit.

On another note, I just smacked myself for buying a 7584 when I need sampling plugins - not another scope!! I suppose I'm not alone here in my addiction to impulsive purchases of old high-end Tek gear, though...


---In TekScopes@..., <dennis@...> wrote :

Hi David and Brian,
Yes, the 7S14 is much easier to use, and it is dual channel with a dual
timebase which makes it an excellent sampling combination. Plus you can
usually find them for next to nothing. Bandwidth is limited to 1GHz which is
probably going to be fine.

This is the bad news... Mercury is banned in most countries.
Unfortunately the mercury cell in the sampling front end of each channel is
certainly dead by now. These plugins are all 40+ years old by now. The
mercury cells lasted 10 times longer than anything made today but 40 years
is impossible for any battery. There are no direct replacements for these
cells. Space inside the front ends is very tight which limits the possible
alternatives.

Alternative ways of powering the sampling front end have been discussed
(look in the archives) and if you can live with one of those solutions a
7S14 would be a excellent solution.

Dennis Tillman W7PF


Re: Introduction of myself!

 

The fastest display window for the 7T11 extends from 50 nanoseconds to
100 nanoseconds so a 75 nanosecond delay is optimum and that is what
the 7M11 has.

On Sun, 14 Feb 2016 18:26:52 -0800, you wrote:

Off the top of my head I don't remember the risetime of the 106 or the setup time the 7T11 needs to trigger. One possible solution might be the 7M11 Dual Delay Line plugin. It can sometimes be found for a reasonable price.

Dennis Tillman W7PF


Re: Introduction of myself!

 

Hi David and Brian,
Yes, the 7S14 is much easier to use, and it is dual channel with a dual
timebase which makes it an excellent sampling combination. Plus you can
usually find them for next to nothing. Bandwidth is limited to 1GHz which is
probably going to be fine.

This is the bad news... Mercury is banned in most countries.
Unfortunately the mercury cell in the sampling front end of each channel is
certainly dead by now. These plugins are all 40+ years old by now. The
mercury cells lasted 10 times longer than anything made today but 40 years
is impossible for any battery. There are no direct replacements for these
cells. Space inside the front ends is very tight which limits the possible
alternatives.

Alternative ways of powering the sampling front end have been discussed
(look in the archives) and if you can live with one of those solutions a
7S14 would be a excellent solution.

Dennis Tillman W7PF

-----Original Message-----
Sent: Sunday, February 14, 2016 5:56 PM
Subject: Re: [TekScopes] Introduction of myself!

I was going to suggest the 7S14 instead of the 7S11/7T11/S1/S2 because it is
easier to use and I believe easier to maintain.

On Sun, 14 Feb 2016 16:54:50 -0800, you wrote:

Hi,

Welcome to the group!

Buy the 7S11 and a 7T11 sampling plugins for your 7613. Then get an S1 or
S2 head for the 7S11. That will allow you to calibrate the 106. It will also
teach you about sampling measurements which are a fascinating aspect of
oscilloscope measurements.

Dennis Tillman W7PF
------------------------------------
Posted by: David <@DWH>
------------------------------------