Date   
Re: Type 503 Oscilloscope Issues

Evan
 

Going the tube removal route first, removing V484 and V474 resulted in the -100 V bus measuring -91 V.

With the tubes back in, removing V45 caused the voltage to be -87.0 V.


Removing V135 resulted in it being - 90.0 V.


Removing V145 resulted in it being -88.6 V.


Removing V160 resulted in it being -87.8 V.


I also checked V24 and V152, since both did involve the -100 V bus at some point in the circuit (just not directly). Removing V24 resulting in the -100 V bus measuring -87.8 V, while removing V152 resulted in it measuring at -85.2 V.


Again, with all of the tubes installed, the bus measures at -86.9 V. Overall, all of the tubes had marginal effects on the -100 V bus voltage when removed, so I will next take a look at the different adjustable components and their effect on the measurement.


- Evan

Re: Another scope 7854

Joe Laffey
 

I'd love a 7B87, but they seem tough to find. If anyone has one they want to get rid of give me a shout off list.


--
Joe Laffey

On May 6, 2017, at 10:08 PM, Dave Casey polara413@... [TekScopes] <TekScopes@...> wrote:

Correct, and I believe it is necessary if you want to capture single shot
events before the trigger (pre-trigger acquisition). When you cut your
teeth on modern DSOs, you take features like this for granted.

Dave Casey

On Sat, May 6, 2017 at 9:21 PM, 'bobh' bobh@... [TekScopes] <
TekScopes@...> wrote:





From: Joe Laffey joe@... [TekScopes]
Sent: Saturday, May 06, 2017 2:59 PM
To: TekScopes@...
Subject: Re: [TekScopes] Another scope 7854

I will seek out a better timebase.
Joe Laffey
I use a 7B92A and a 7B87 in mine. I think the 7B87 was basically built for
the 7854 scope.
Bob.

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


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


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

Re: Easy method for NVRAM battery replacement

Merchison Burke
 

All versions of the link worked for me with Win XP Pro SP3 and Thunderbird 52.1.0 (32 bit) and Firefox 52.1.0 (32 bit).

I download all posts to my computer and read them with Thunderbird. When I clicked on the link, I was able to see the photos with Firefox.

On 2017-May-06 2:55 PM, szigszabolcs@... [TekScopes] wrote:
Argh. Let's try the link again, and see if it works now

https://goo.gl/photos/6u7veVp1usNe4MmQA https://goo.gl/photos/6u7veVp1usNe4MmQA


Re: Another scope 7854

 

One of the more specialized 7000 series plug-in extenders could swap
the signal out and trigger out signals allowing the trigger output to
be viewed on the CRT.

Another test which might show something is the vertical signal output
on the back of the 7854. It is a copy of the currently selected A
trigger signal.

Re: Another scope 7854

Dave Casey
 

Correct, and I believe it is necessary if you want to capture single shot
events before the trigger (pre-trigger acquisition). When you cut your
teeth on modern DSOs, you take features like this for granted.

Dave Casey

On Sat, May 6, 2017 at 9:21 PM, 'bobh' bobh@... [TekScopes] <
TekScopes@...> wrote:





From: Joe Laffey joe@... [TekScopes]
Sent: Saturday, May 06, 2017 2:59 PM
To: TekScopes@...
Subject: Re: [TekScopes] Another scope 7854

I will seek out a better timebase.
Joe Laffey
I use a 7B92A and a 7B87 in mine. I think the 7B87 was basically built for
the 7854 scope.
Bob.

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



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

Re: Another scope 7854

bobh@joba.com
 

From: Joe Laffey joe@... [TekScopes]
Sent: Saturday, May 06, 2017 2:59 PM
To: TekScopes@...
Subject: Re: [TekScopes] Another scope 7854


I will seek out a better timebase.
Joe Laffey

I use a 7B92A and a 7B87 in mine. I think the 7B87 was basically built for the 7854 scope.
Bob.

Re: Type 503 Oscilloscope Issues

Dave Casey
 

The metal cover is a shield for electromagnetic interference (EMI). It may
be needed in order to meet a specification somewhere, but the scope should
work fine without it.

We're not necessarily looking for a bad tube, we're just disabling
different parts of the circuit by removing tubes because they are socketed
and thus easy to remove and reinstall. The actual cause of the problem may
be a resistor or a piece of loose metal that cause a short somwhere, etc.
Finding the tubes that prevent this fault from overloading the -100V bus
will help us determine where the fault is.

Just like V374 and V384 in the horizontal amplifier, there are V474 and
V484 in the vertical amplifier. You can try pulling those. You can also try
pulling V45, V135, V145, and V160 one at a time to see if any of them are
the smoking gun. If none of those point to the issue, then you'll have to
start un-soldering various connections or measuring a lot of different
voltages to suss out the issue.

Another thing to try with all the tubes installed and your meter hooked up
to monitor the -100V bus is to adjust all of the various controls and see
if any one control has a significant impact on the supply voltage. Again,
that doesn't mean that particular control is bad, but it might tell you
which part of the circuit is burning all that power.

It is also still possible that the problem is with the -100V supply itself.
A leaky filter capacitor such as C682 or C684 could be the problem. You
might see if either of these are particularly warm after the scope has been
on for a minute (but turn the scope off before you feel them up - not sure
if the whole can sees -100V or not).

Dave Casey


On Sat, May 6, 2017 at 7:29 PM, enchanter464@... [TekScopes] <
TekScopes@...> wrote:



Some updates on the scope:


First, to Kjo's concern, we did discuss using silver-bearing solder
previously. I have just been using the unused spool that came inside the
scope, which has worked out so far (although I may need to pick up more
later on).


Then to Dave's and John's points, with V334 and V434 removed, the -100 V
bus was at -90.3 V (better than before). In addition, I was able to better
adjust the horizontal on the trace (previously, twisting the knob for the
horizontal position to the left only made the trace move to about halfway
across the screen, but now it can span the entire width).

Of note, only V434 had the protective metal spring covering on top of the
tube (locks into the socket), while V334 lacked the metal covering. I am
not sure if this would affect anything.


With the tubes still removed, I adjusted R641 to see the effect.
Originally, it was turned all the way to the left (counter-clockwise), but
moving it to the exact middle of its range, the -100 V bus read -94.4 V.
Continuing to turn the resistor knob to its rightmost (clockwise) point,
the -100 V bus still only reached about -94.4 V (occasionally reaching -95
V), but mostly stable around -94.4).


Given the above, it was likely that another tube is also affecting the
-100 V bus, so I continued with Dave's advice on V374 and V384. With V334
and V434 plugged back in, and the other two removed, the -100 V bus
measured at -92 V, which is still better than when all tubes are plugged
in.


Are there any other tubes that should be checked, or any better way of
determining which tubes are bad? It seems that all of the tubes that were
removed helped the -100 V bus to some extent, but could this just be due to
simply removing the tube (whether the tube itself is good or bad)?


- Evan






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



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

Re: Type 503 Oscilloscope Issues

Evan
 

Some updates on the scope:


First, to Kjo's concern, we did discuss using silver-bearing solder previously. I have just been using the unused spool that came inside the scope, which has worked out so far (although I may need to pick up more later on).


Then to Dave's and John's points, with V334 and V434 removed, the -100 V bus was at -90.3 V (better than before). In addition, I was able to better adjust the horizontal on the trace (previously, twisting the knob for the horizontal position to the left only made the trace move to about halfway across the screen, but now it can span the entire width).

Of note, only V434 had the protective metal spring covering on top of the tube (locks into the socket), while V334 lacked the metal covering. I am not sure if this would affect anything.


With the tubes still removed, I adjusted R641 to see the effect. Originally, it was turned all the way to the left (counter-clockwise), but moving it to the exact middle of its range, the -100 V bus read -94.4 V. Continuing to turn the resistor knob to its rightmost (clockwise) point, the -100 V bus still only reached about -94.4 V (occasionally reaching -95 V), but mostly stable around -94.4).


Given the above, it was likely that another tube is also affecting the -100 V bus, so I continued with Dave's advice on V374 and V384. With V334 and V434 plugged back in, and the other two removed, the -100 V bus measured at -92 V, which is still better than when all tubes are plugged in.


Are there any other tubes that should be checked, or any better way of determining which tubes are bad? It seems that all of the tubes that were removed helped the -100 V bus to some extent, but could this just be due to simply removing the tube (whether the tube itself is good or bad)?


- Evan

Tek Blue paint source?

Reed Dickinson
 

The formula for the blue paint I put on the forum several days ago was
what was given to me. The base might be a proprietary product, I don't
know. The lack of a 'Y' in the yellow oxide entry is a mystery to me.
If in doubt, take the formula to a good paint company and ask the oldest
person there if it makes sense. I can't recall the name of the paint
company I used, it has been several years.

Reed

On 5/6/2017 3:16 PM, Chris Elmquist chrise@... [TekScopes] wrote:

Thanks for the additional follow-up guys. I will pursue several of
these paths. If I need to buy 25 cans, watch for my new eBay store :-)

Reed, can you comment on which reputable paint company since I suspect
the base "7-916" will be a number proprietary to them.

Also, is there possibly a typo in the Yellow Oxide code since it does
not have a "Y" prefix while all of the others do?

Chris

On Friday (05/05/2017 at 03:53PM -0700), Reed Dickinson
reed714@... [TekScopes] wrote:

Hi Gang:
I had a very reputable paint company do a spectrograph analysis of a
sample of Tektronix blue paint. The formula they came up with is as
follows:
Black B 4Y22
Yellow oxide C 33
Green D 1Y29
White W 1Y32
Blue E 1Y45
Base 7-916

I was told that Y = ounces. Other than that I have no information. The
paint they sold me using this formula was an exact match to the
original
sample.

Reed Dickinson
--
Chris Elmquist

Re: Easy method for NVRAM battery replacement

mosaicmerc
 

Wanted: Shelf for Tek Lab Cart model 3

Joe Laffey
 

If anyone has a spare shelf for a Tek Lab Cart Model 3 I would be interested. Please contact me off list.


--
73
Joe Laffey

Re: Tek Blue paint source?

Chris Elmquist
 

Thanks for the additional follow-up guys. I will pursue several of
these paths. If I need to buy 25 cans, watch for my new eBay store :-)

Reed, can you comment on which reputable paint company since I suspect
the base "7-916" will be a number proprietary to them.

Also, is there possibly a typo in the Yellow Oxide code since it does
not have a "Y" prefix while all of the others do?

Chris

On Friday (05/05/2017 at 03:53PM -0700), Reed Dickinson reed714@... [TekScopes] wrote:

Hi Gang:
I had a very reputable paint company do a spectrograph analysis of a
sample of Tektronix blue paint. The formula they came up with is as
follows:
Black B 4Y22
Yellow oxide C 33
Green D 1Y29
White W 1Y32
Blue E 1Y45
Base 7-916

I was told that Y = ounces. Other than that I have no information. The
paint they sold me using this formula was an exact match to the original
sample.

Reed Dickinson
--
Chris Elmquist

Re: Another scope 7854

Joe Laffey
 

On May 6, 2017, at 4:19 PM, David @DWH [TekScopes] <TekScopes@...> wrote:

On Sat, 6 May 2017 14:45:56 -0500, you wrote:

At the timebase plugging, measuring the rails on the trigger board I see

-15V rail. 0.32miliVAC RMS
+15V rail 0.73miliVAC RMS
+5V rail 2.07milVAC RMS

When the plugin in NOT triggering the ripple on the +5V goes to 0V

I will see about removing the power supply for the scope and testing it. This whole thing may go on hold for a week due to work.
The peak to peak high frequency noise matter more than the low
frequency RMS noise in this case.

Ah... well I did put a 1X probe on the 5V rail and Peak detect mode on the 2232 I saw edge peaks of around 75-90mV P-P. For grins and experimentation/learning tried clipping in various sized capacitors on the 5V rail and it did not alleviate the issue.

I do not think the common mode rejection of the trigger circuit path
in the 7B53A is all that good so it is possible that your 7A18 is
producing a lot of common mode noise which would not be immediately
apparent. Comparing the trigger signal waveforms at the output of the
7A18 to the 7A26 might reveal something.

I may compare these just to see what I see.
It may not be worth spending too much time on this. The 7B53A is not
a very good timebase to use with the 7854; the timebase itself
triggers too slowly to show the leading edge with the 400MHz 7854's
delay line which is only 65 nanoseconds versus 120 nanoseconds of the
100MHz 76xx mainframes. And like I said earlier, all of my 7B53As so
this to one extent or another; adjusting the trigger level and
coupling is usually enough to solve it.

Of course some of us have a lot of fun tracking down obscure problems
like this. :)
I am learning stuff in the process-- so that is good. In the real world I would just enable HF reject in the trigger filter and it is stable.

I will seek out a better timebase.

Thanks for all your help. I really appreciate it.

--
Joe Laffey

Re: Another scope 7854

 

On Sat, 6 May 2017 14:45:56 -0500, you wrote:

At the timebase plugging, measuring the rails on the trigger board I see

-15V rail. 0.32miliVAC RMS
+15V rail 0.73miliVAC RMS
+5V rail 2.07milVAC RMS

When the plugin in NOT triggering the ripple on the +5V goes to 0V

I will see about removing the power supply for the scope and testing it. This whole thing may go on hold for a week due to work.
The peak to peak high frequency noise matter more than the low
frequency RMS noise in this case.

On Sat, 6 May 2017 15:45:48 -0500, you wrote:

Well, it was a good thing that my schematic no vertical amplifier came today. Because the same timebase works just fine with the second amplifier 7A26

So now I am really confused as to why the signals coming out of the two switching ICs on the 7A18 looked fine, but still had the issue.

Kinda confused at this point. Good thing I am taking a break this week for work. Perhaps things will be more clear when I get back into it.
I do not think the common mode rejection of the trigger circuit path
in the 7B53A is all that good so it is possible that your 7A18 is
producing a lot of common mode noise which would not be immediately
apparent. Comparing the trigger signal waveforms at the output of the
7A18 to the 7A26 might reveal something.

It may not be worth spending too much time on this. The 7B53A is not
a very good timebase to use with the 7854; the timebase itself
triggers too slowly to show the leading edge with the 400MHz 7854's
delay line which is only 65 nanoseconds versus 120 nanoseconds of the
100MHz 76xx mainframes. And like I said earlier, all of my 7B53As so
this to one extent or another; adjusting the trigger level and
coupling is usually enough to solve it.

Of course some of us have a lot of fun tracking down obscure problems
like this. :)

Re: Another scope 7854

 

That sounds like an eraser shield used for drafting.

<https://www.amazon.com/Alvin-Stainless-Steel-Erasing-Shield/dp/B000HF6VK6>

I always thought one of those rotary electric erasers used for
drafting with one of the more abrasive eraser elements for ink would
work for this.

On Sat, 6 May 2017 19:08:31 +0200, you wrote:

One standard tool used in NASA Deep Space Network soldering courses in
the 70's was a kind of stainless steel stencil which has holes for the
common solder pads (round, square, droplet, etc.) that was used to
remove the gold from the PCBs before soldering, using a common ink
eraser. It was a standard item but unfortunately I lost mine.

Regards,
Ignacio, B4APL

Re: Another scope 7854

Joe Laffey
 

Well, it was a good thing that my schematic no vertical amplifier came today. Because the same timebase works just fine with the second amplifier 7A26

So now I am really confused as to why the signals coming out of the two switching ICs on the 7A18 looked fine, but still had the issue.

Kinda confused at this point. Good thing I am taking a break this week for work. Perhaps things will be more clear when I get back into it.
--
Joe Laffey
The Stable
Visual Effects

On May 6, 2017, at 2:45 PM, Joe Laffey <joe@...> wrote:

At the timebase plugging, measuring the rails on the trigger board I see

-15V rail. 0.32miliVAC RMS
+15V rail 0.73miliVAC RMS
+5V rail 2.07milVAC RMS

When the plugin in NOT triggering the ripple on the +5V goes to 0V

I will see about removing the power supply for the scope and testing it. This whole thing may go on hold for a week due to work.


--
Joe Laffey
The Stable
Visual Effects

On May 6, 2017, at 12:40 PM, Joe Laffey joe@... [TekScopes] <TekScopes@...> wrote:



On May 6, 2017, at 10:21 AM, David @DWH [TekScopes] <TekScopes@...> wrote:
Still digging through the docs.
I have never studied the timebase logic in enough detail to make
directed suggestions.

I would take a really close look however at noise on the power supply
outputs at the plug-in compartment. The internal trigger path inside
the mainframe is differential however it is single ended inside the
7B53A making it susceptible to power supply noise.
Sounds reasonable. I haven't examined the power supply at all yet, though I didn't see any ripple on the power rails of the vertical amplifier.

Will measure power supply itself. It may have original caps.

--
Joe Laffey




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

Re: Another scope 7854

Joe Laffey
 

At the timebase plugging, measuring the rails on the trigger board I see

-15V rail. 0.32miliVAC RMS
+15V rail 0.73miliVAC RMS
+5V rail 2.07milVAC RMS

When the plugin in NOT triggering the ripple on the +5V goes to 0V

I will see about removing the power supply for the scope and testing it. This whole thing may go on hold for a week due to work.


--
Joe Laffey
The Stable
Visual Effects

On May 6, 2017, at 12:40 PM, Joe Laffey joe@... [TekScopes] <TekScopes@...> wrote:



On May 6, 2017, at 10:21 AM, David @DWH [TekScopes] <TekScopes@...> wrote:
Still digging through the docs.
I have never studied the timebase logic in enough detail to make
directed suggestions.

I would take a really close look however at noise on the power supply
outputs at the plug-in compartment. The internal trigger path inside
the mainframe is differential however it is single ended inside the
7B53A making it susceptible to power supply noise.
Sounds reasonable. I haven't examined the power supply at all yet, though I didn't see any ripple on the power rails of the vertical amplifier.

Will measure power supply itself. It may have original caps.

--
Joe Laffey




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

Re: Easy method for NVRAM battery replacement

ArtekManuals
 

Never mind I forgot that yahole doubles up on the link when using the
WEB interface ( but not an email client

https://goo.gl/photos/6u7veVp1usNe4MmQA


On 5/6/2017 2:33 PM, szigszabolcs@... [TekScopes] wrote:

Hi,


While fixing a 2430A, it turned out that it has two DS1235 NVRAM, with
built in battery. I could have replaced it with DS1230, as that is
easily available, but this would not have been economical at all. So I
looked into how to add an external battery. As these chips are potted,
it's pretty hard and messy to get to the original battery. All methods
that I have seen on the web involved grinding the potting, and
potentially risk damaging the part.


So after some trials, I have found a method that enabled me to quickly
and easily add a battery.
All you need is a heat gun with a thin nozzle. Heating the potting
will make it soft and more importantly, make it release from the
battery. Fortunately the battery is at the bottom of the part (not all
such similar devices are like this). First I heated it at about where
the battery is and when it became a bit soft, it was easy to scrape it
off with a small screwdriver. Then the the negative tab could be pried
off the battery. Next comes the hard part, remove the battery. Make
sure that the potting is removed all from the area of the battery, and
at the edge of the device, remove it at such a depth that you see the
whole thickness of the battery. It's not very thick, and there is a
PCB underneath, so be careful. I then drilled a small hole into the
battery, so it would not explode, when heated and used the heat gun,
to heat the battery. When sufficiently heated, you can simply pry/flip
it out from its place from the side, by putting a small screwdriver
under it. Be careful, as the positive tab is spot-welded to it under,
wiggle a bit and then it's possible to snip off the tab.
There are two batteries, but it is enough to do this with one, as the
circuitry in the NVRAM will use the battery with the higher voltage.
Then I simply soldered some wire and a battery holder for a 2032
lithium battery to it. I used some epoxy to fix the wires in the
cavity, where the old battery was. Originally, I planned to place the
new batteries somewhere where there is place inside the scope, but it
turned out that actually there is just about enough height, so i
epoxied the battery holder to the top of the NVRAM, put some quite
ugly looking kapton tape on the top, just in case it would touch the
PSU cover. I could probably shorten the wires, but that is ok. It now
works fine. Hope this can help someone with similar needs.


Here are some pictures:


https://goo.gl/photos/6u7veVp1usNe4MmQA
https://goo.gl/photos/6u7veVp1usNe4MmQA



Szabolcs





--
Dave
Manuals@...
www.ArtekManuals.com

Re: Easy method for NVRAM battery replacement

Szabolcs Szigeti
 

Argh. Let's try the link again, and see if it works now

https://goo.gl/photos/6u7veVp1usNe4MmQA https://goo.gl/photos/6u7veVp1usNe4MmQA

Re: Easy method for NVRAM battery replacement

ArtekManuals
 

The problem is Yahole doubles the link

Try this

https://goo.gl/photos/6u7veVp1usNe4MmQA


On 5/6/2017 2:51 PM, szigszabolcs@... [TekScopes] wrote:

Ok, an other try, for some reason my message disappeared.

So her is the link, this is to a google photos shared album, you
should be able to see it without any login etc.


https://goo.gl/photos/6u7veVp1usNe4MmQA
https://goo.gl/photos/6u7veVp1usNe4MmQA



Let's see if ti works now. I was unable to upload it to the yahoo groups.


Szabolcs





--
Dave
Manuals@...
www.ArtekManuals.com