DM502 with problems


Colin Herbert
 

I have a DM502 which isn't working correctly. I have checked the power supply voltages at +5V, and +/- 12V and they are within spec. I have also adjusted the clock frequency to 20.48 kHz and that is fine. On initial switch-on and warming-up, the display "jumps about" rather than staying stable as one would expect, even with only a battery connected to the test-terminals and a suitable DC voltage range selected or with a resistor across the test terminals and a suitable ohms range selected. After being powered on for a while, the display just shows a constant sequence of digits, whatever the position of the range switch, only the position of the decimal point changing. I haven't got around to checking anything else, yet, but after the plug-in has been on for a while,the transformer T410 is rather hot to the touch, so much so that I can't keep a finger-tip in contact with it. Is this correct behaviour, or should I check other components (don't forget that the voltages out of the transformer are within spec.)?
TIA, Colin.


Roger Evans
 

Colin,

Some of the TM500 plugins had notoriously unreliable IC sockets. I would certainly remove the socketed ICs, clean pins and see if anything improves. Several people have commented that the long term solution is to replace with decent IC sockets, I expect someone will know how to identify the unreliable originals. Have you checked for ripple on the +/-12V and 5V lines? Also it is worth exercising the range switch a few dozen times! Is the DM502 a new acquisition or one that was working OK but has now failed?

Regards,

Roger


Roger Evans
 

Correction, I don't think the cam switches improve with use, if they are dirty they need cleaning with IPA and strips of paper.

Roger


Colin Herbert
 

I've had a look at that possibility, Roger. I have cleaned the attenuator cam-switch contacts on a couple of 400-series scopes, but I can't see how to clean them on the DM502. I have taken the cover off the cam-switch assembly, but I simply can't see the gold-plated (?) contact fingers and pcb contacts. It seems more like the time-base switch contacts on 400-series scopes than the attenuators on the same scopes.
Correction: I have taken another look after removing the cover and I can see the contacts underneath the cam-switch. I think I'll have a go with the IPA and paper, though whether IPA is as required as it is with attenuator contacts, I'm not sure. The pcb material looks as though it is fairly ordinary, rather than the polyphenylene oxide used in scope attenuators, which can be damaged by some solvents.
Another fault is that sometimes one or more sections of the seven-segment led displays tend to flick on or off - maybe this will improve with switch-contact cleaning?
Colin.

-----Original Message-----
From: TekScopes@groups.io [mailto:TekScopes@groups.io] On Behalf Of Roger Evans via groups.io
Sent: 05 June 2020 11:55
To: TekScopes@groups.io
Subject: Re: [TekScopes] DM502 with problems

Correction, I don't think the cam switches improve with use, if they are dirty they need cleaning with IPA and strips of paper.

Roger


Roger Evans
 

Colin,

If there are intermittent segments on the LEDs then the 74(LS)47 7 segment decoder is the likely culprit and is probably socketed (it is on my DM501).

Roger


Dave Daniel
 

I believe that the paper-soaked IPA cleaning method has benn discussed before and there probably messages in the archive.

I do recall that I had decided that when I needed to clean any TM-500 series plug-cam switches I would try that method.

DaveD

On Jun 5, 2020, at 07:59, Colin Herbert via groups.io <colingherbert=blueyonder.co.uk@groups.io> wrote:

I've had a look at that possibility, Roger. I have cleaned the attenuator cam-switch contacts on a couple of 400-series scopes, but I can't see how to clean them on the DM502. I have taken the cover off the cam-switch assembly, but I simply can't see the gold-plated (?) contact fingers and pcb contacts. It seems more like the time-base switch contacts on 400-series scopes than the attenuators on the same scopes.
Correction: I have taken another look after removing the cover and I can see the contacts underneath the cam-switch. I think I'll have a go with the IPA and paper, though whether IPA is as required as it is with attenuator contacts, I'm not sure. The pcb material looks as though it is fairly ordinary, rather than the polyphenylene oxide used in scope attenuators, which can be damaged by some solvents.
Another fault is that sometimes one or more sections of the seven-segment led displays tend to flick on or off - maybe this will improve with switch-contact cleaning?
Colin.

-----Original Message-----
From: TekScopes@groups.io [mailto:TekScopes@groups.io] On Behalf Of Roger Evans via groups.io
Sent: 05 June 2020 11:55
To: TekScopes@groups.io
Subject: Re: [TekScopes] DM502 with problems

Correction, I don't think the cam switches improve with use, if they are dirty they need cleaning with IPA and strips of paper.

Roger







 

Only one company is associates with bad IC sockets - that is Texas Instruments. Their sockets are easy to find by removing the IC because the TI logo is stamped underneath the IC in the middle of the socket where it can be easily seen.

TI saw an opportunity to make money selling sockets when IC sales were booming in the late 1960s and early 1970s. Unfortunately for them other companies already had all the patents locked up. Other manufacturers sockets used two flat spring contacts (one on each side of each pin of the IC) to grip the pin from both sides of the pin's very wide flat dimension. TI could not find a way around these ironclad patents. The only way left to grip an IC pin was by the very thin thickness of the metal the pins were stamped from once the IC was finished in manufacturing. A metal die stamps away all of the metal from this metal frame except for the pins themselves and the die leaves behind very rough edges where it stamps the metal frame. As a result the flat side of the IC pin is smooth and an excellent place for a socket to grab the pin and the edge of the pin is thin and very rough. The worst place to grab the pin is on the thin rough edges. These rough edges gouge into the contacts in the TI socket and the result is they will eventually develop intermittent problems.

There is not much you can do except replace them which is a tedious job that can cause other problems without the proper equipment to do that with.

Dennis Tillman W7pF

-----Original Message-----
From: TekScopes@groups.io [mailto:TekScopes@groups.io] On Behalf Of Colin Herbert via groups.io
Sent: Friday, June 05, 2020 5:00 AM
To: TekScopes@groups.io
Subject: Re: [TekScopes] DM502 with problems

I've had a look at that possibility, Roger. I have cleaned the attenuator cam-switch contacts on a couple of 400-series scopes, but I can't see how to clean them on the DM502. I have taken the cover off the cam-switch assembly, but I simply can't see the gold-plated (?) contact fingers and pcb contacts. It seems more like the time-base switch contacts on 400-series scopes than the attenuators on the same scopes.
Correction: I have taken another look after removing the cover and I can see the contacts underneath the cam-switch. I think I'll have a go with the IPA and paper, though whether IPA is as required as it is with attenuator contacts, I'm not sure. The pcb material looks as though it is fairly ordinary, rather than the polyphenylene oxide used in scope attenuators, which can be damaged by some solvents.
Another fault is that sometimes one or more sections of the seven-segment led displays tend to flick on or off - maybe this will improve with switch-contact cleaning?
Colin.

-----Original Message-----
From: TekScopes@groups.io [mailto:TekScopes@groups.io] On Behalf Of Roger Evans via groups.io
Sent: 05 June 2020 11:55
To: TekScopes@groups.io
Subject: Re: [TekScopes] DM502 with problems

Correction, I don't think the cam switches improve with use, if they are dirty they need cleaning with IPA and strips of paper.

Roger









--
Dennis Tillman W7pF
TekScopes Moderator


Colin Herbert
 

Hi Roger,
I haven't had a go at cleaning the cam-switches contacts, because I took your advice and re-seated and cleaned the pins of the ICs. The DM502 now seems good. There was a little hiccup when I replaced the temperature IC the wrong way around, but I have been soak-testing the device for some time now and it has been pretty much rock solid and no flicker/loss of display segments. I will probably leave well enough alone on the cam-switch contacts and have a bash at some kind of calibration, but it all looks fairly good so far.
Incidentally, the DM502 came to me with the initial condition - it was a cheap buy on eBay.
Many thanks to all who gave me advice.
Colin.

-----Original Message-----
From: TekScopes@groups.io [mailto:TekScopes@groups.io] On Behalf Of Roger Evans via groups.io
Sent: 05 June 2020 11:53
To: TekScopes@groups.io
Subject: Re: [TekScopes] DM502 with problems

Colin,

Some of the TM500 plugins had notoriously unreliable IC sockets. I would certainly remove the socketed ICs, clean pins and see if anything improves. Several people have commented that the long term solution is to replace with decent IC sockets, I expect someone will know how to identify the unreliable originals. Have you checked for ripple on the +/-12V and 5V lines? Also it is worth exercising the range switch a few dozen times! Is the DM502 a new acquisition or one that was working OK but has now failed?

Regards,

Roger


Roger Evans
 

Colin,

Glad to hear it is working OK, if there is no need to clean the cam switches then don't rush into it, there is advice around that says energetic cleaning with the wrong type of paper can remove the gold plating. There are 40 odd segments to the cam switch so it could be a tedious job anyway!

Regards,

Roger


Richard Peterson
 

Sorry for hijacking this discussion, but i'm working on a DM502 as well and wonder if you all could help. The unit powers on and has +5V and +/- 12V (step 1 of the check/adjustment procedure) but I don't have a clock signal - step 2. I'm guessing the NE555 timer IC is fried, but before i start randomly replacing parts, I checked out as much as i could. Voltages at various other points i tested seem OK and the resistors, capacitors and transistors within the display drive circuit all test OK as well. When the unit is powered up, i get a '0' on the display no mater what setting the switch is on with the decimal point showing up on a couple of the switch settings. Sometimes the display will go dark after being on for a while but not always. The sn7447 and LD110 ICs both feel very warm but not too hot to touch - maybe OK?

I tried Roger's suggestion to pull, clean and reseat the ICs but that didn't change anything. So . . . should i try just changing the NE555 to see if it comes back on line. Replace all the ICs? Are there other tests I can try to isolate the root cause of the problem? possibly a dirty switch?

Mouser has the NE555, the SN7447 and the SN7416N but the LD110 and LD111 appear to be approaching extinction although there is one seller on the fleabag selling sets.

Thanks for your help.

Rich


Harvey White
 

This is a multiplexed display.  There's a switch/transistor applying power to each of the displays in turn.  That same switch changes the input of a single decoder to the various counters, likely through a multiplexer.

There's a counter that selects a digit and the appropriate input.  If that counter doesn't run, you get whatever input there is to the decoder and the single digit much brighter than normal. Scanning averages out the peak current (selected) to the average current desired, hence the digits would be dimmer.

FIrst there needs to be a scanning clock.  Secondly, the counter and decoding circuit that selects the digits needs to work.

If it's a 555, they're easy to come by.  RadioShack/Tandy sold them, as did/does Jameco, etc.

If you had a signal generator that put out TTL pulses, I'd suggest pulling the 555 and patching a 1 Khz (roughly) output into the former output of the 555 timer and see how it goes.  What their exact frequency is will be in the manual.

Depending on what state the decoding circuitry comes up in, you could be looking at any one of the counter outputs.

Analyze the circuit starting at the oscillator and go from there.

Harvey

On 7/16/2021 7:36 PM, Richard Peterson wrote:
Sorry for hijacking this discussion, but i'm working on a DM502 as well and wonder if you all could help. The unit powers on and has +5V and +/- 12V (step 1 of the check/adjustment procedure) but I don't have a clock signal - step 2. I'm guessing the NE555 timer IC is fried, but before i start randomly replacing parts, I checked out as much as i could. Voltages at various other points i tested seem OK and the resistors, capacitors and transistors within the display drive circuit all test OK as well. When the unit is powered up, i get a '0' on the display no mater what setting the switch is on with the decimal point showing up on a couple of the switch settings. Sometimes the display will go dark after being on for a while but not always. The sn7447 and LD110 ICs both feel very warm but not too hot to touch - maybe OK?

I tried Roger's suggestion to pull, clean and reseat the ICs but that didn't change anything. So . . . should i try just changing the NE555 to see if it comes back on line. Replace all the ICs? Are there other tests I can try to isolate the root cause of the problem? possibly a dirty switch?

Mouser has the NE555, the SN7447 and the SN7416N but the LD110 and LD111 appear to be approaching extinction although there is one seller on the fleabag selling sets.

Thanks for your help.

Rich





Tom Lee
 

If the 555 is being fed with the right Vcc and the other components are ok (did you verify that the pot was working?), then you should see a clock output.

Finding a 555 is trivial, so you should have no problem finding a sub.

Good luck!

Cheers
Tom

Sent from my iThing, so please forgive brevity and typos

On Jul 16, 2021, at 17:52, Harvey White <madyn@dragonworks.info> wrote:

This is a multiplexed display. There's a switch/transistor applying power to each of the displays in turn. That same switch changes the input of a single decoder to the various counters, likely through a multiplexer.

There's a counter that selects a digit and the appropriate input. If that counter doesn't run, you get whatever input there is to the decoder and the single digit much brighter than normal. Scanning averages out the peak current (selected) to the average current desired, hence the digits would be dimmer.

FIrst there needs to be a scanning clock. Secondly, the counter and decoding circuit that selects the digits needs to work.

If it's a 555, they're easy to come by. RadioShack/Tandy sold them, as did/does Jameco, etc.

If you had a signal generator that put out TTL pulses, I'd suggest pulling the 555 and patching a 1 Khz (roughly) output into the former output of the 555 timer and see how it goes. What their exact frequency is will be in the manual.

Depending on what state the decoding circuitry comes up in, you could be looking at any one of the counter outputs.

Analyze the circuit starting at the oscillator and go from there.

Harvey



On 7/16/2021 7:36 PM, Richard Peterson wrote:
Sorry for hijacking this discussion, but i'm working on a DM502 as well and wonder if you all could help. The unit powers on and has +5V and +/- 12V (step 1 of the check/adjustment procedure) but I don't have a clock signal - step 2. I'm guessing the NE555 timer IC is fried, but before i start randomly replacing parts, I checked out as much as i could. Voltages at various other points i tested seem OK and the resistors, capacitors and transistors within the display drive circuit all test OK as well. When the unit is powered up, i get a '0' on the display no mater what setting the switch is on with the decimal point showing up on a couple of the switch settings. Sometimes the display will go dark after being on for a while but not always. The sn7447 and LD110 ICs both feel very warm but not too hot to touch - maybe OK?

I tried Roger's suggestion to pull, clean and reseat the ICs but that didn't change anything. So . . . should i try just changing the NE555 to see if it comes back on line. Replace all the ICs? Are there other tests I can try to isolate the root cause of the problem? possibly a dirty switch?

Mouser has the NE555, the SN7447 and the SN7416N but the LD110 and LD111 appear to be approaching extinction although there is one seller on the fleabag selling sets.

Thanks for your help.

Rich








Jared Cabot
 

I repaired one of these not too long ago, the video I made might help.

https://www.youtube.com/watch?v=Tx6DtLooCOs

My unit had drifting carbon composite resistors that were replaced, the usual sticky switches that were disassembled, cleaned, DeOxit-ised and relubricated (the gold wafer cam switches were only cleaned with alcohol and paper as mentioned earlier in this thread), and a faulty LD110 (faulty as it warmed up) that was successfully replaced with an eBay sourced part.


Richard Peterson
 

Great feedback Harvey. I should have thought to inject the timer signal . . . I'll give that a try.

Jared - i found your video a few days ago. Thanks for taking the time to create/upload it.


 

On Sat, Jun 6, 2020 at 09:40 AM, Dennis Tillman W7pF wrote:


Only one company is associates with bad IC sockets - that is Texas Instruments
Not all TI sockets were bad even if of course some could fail just like any other brands.
The one that stands out of the crowd is called C-95. In my opinion there is not a question
if it would fail but when. They were not only in TM500's but all over the place at that time.
Typically faults are intermittent and cleaning and reseating may help for some time but it
will eventually come back. If you see those in your instrument count on problems.
They are easily identified by the very unique profile on the short end.

Check on TekWiki here: https://w140.com/tekwiki/wiki/Bad_TI_IC_sockets

My three pictures at the bottom show it while the larger picture shows what I would not generally consider bad.

/Håkan


Richard Peterson
 

Quick update to close the loop. I went through the components related to the NE555 timer and concluded that it was, indeed, a dead IC. I injected a 20.48 kHz signal at the output of the NE555 and the display came to life. I replaced the IC and its working great. Calibration was pretty much spot on as far as i can tell with my limited testing. Thanks again for all the feedback. Another TM504 lives along with the PS503A, FG502 and of course the DM502 that came with it. The fourth slot has a storage box that had an unused temperature probe & manual and a variety of test leads, banana jumpers and some odd push-in BNC jumpers with molded plastic ends that fit into a female BNC and will accept a male BNC on the back side of the plug. Looks great along side the TM503 . . .

Rich


Tom Lee
 

Glad it was something so simple and cheap to fix -- just a 555 instead of some left-handed unobtainium hydroxide.

-- Cheers,
Tom

--
Prof. Thomas H. Lee
Allen Ctr., Rm. 205
350 Jane Stanford Way
Stanford University
Stanford, CA 94305-4070
http://www-smirc.stanford.edu

On 7/26/2021 19:36, Richard Peterson wrote:
Quick update to close the loop. I went through the components related to the NE555 timer and concluded that it was, indeed, a dead IC. I injected a 20.48 kHz signal at the output of the NE555 and the display came to life. I replaced the IC and its working great. Calibration was pretty much spot on as far as i can tell with my limited testing. Thanks again for all the feedback. Another TM504 lives along with the PS503A, FG502 and of course the DM502 that came with it. The fourth slot has a storage box that had an unused temperature probe & manual and a variety of test leads, banana jumpers and some odd push-in BNC jumpers with molded plastic ends that fit into a female BNC and will accept a male BNC on the back side of the plug. Looks great along side the TM503 . . .

Rich