Author Topic: Keithley 197A owners' corner  (Read 7744 times)

0 Members and 1 Guest are viewing this topic.

Offline technogeeky

  • Frequent Contributor
  • **
  • Posts: 540
  • Country: us
  • Older New "New Player" Player Playa'
Re: Keithley 197A owners' corner
« Reply #25 on: October 11, 2017, 06:21:37 pm »
wow look at all that via stitching or ground plane action going on. very nice.
 

Offline rastro

  • Frequent Contributor
  • **
  • Posts: 353
  • Country: 00
Re: Keithley 197A owners' corner
« Reply #26 on: October 11, 2017, 08:22:31 pm »
Additional 197M Observations:
1. The display board has 197A written on it.  However it does not have the additional components for the back light. 
2. I doubt it would support the GPIB option due to physical constraints of the added shield.  However it still has the connector for the option.
3. TheMC68HC711E9CFN2 probably has the system firmware loaded into it.  These chips are available but I don't know if the firmware is readable.

rastro
 

Offline technogeeky

  • Frequent Contributor
  • **
  • Posts: 540
  • Country: us
  • Older New "New Player" Player Playa'
Re: Keithley 197A owners' corner
« Reply #27 on: October 12, 2017, 03:12:46 am »
Some differences from looking at pictures posted:

Y101 was 3.276800 MHz, is 2.62144 MHz (HEF4040 is still used to divide this down, and the ratio between the two crystals is 1.25. so... what does that mean w.r.t. the timing of the charge pump and integration cycle)

U109 was TI741, is AD711 ("enhanced replacement for LF411 and TL081")

U101 was TLC272lcp, is TLC27L2cp (not the same part at all: the 27L2 uses about 200 times less power (~ worst-case 100 µA versus ~ avg. 2000 µA typical))[1]



[1]: I'm confused by this substitution. This is a pretty important part in this multimeter, but these chips are very different. Both of these devices were first available in the same year. A lot of figures in the datasheet are incomparable because they have different reference points (TLC27L2 is specified into a 1M, TLC272 is specified into 10k?). CMRR is ~ 97 dB for the TLC27L2 chip, and ~ 88 dB for the TLC272.

TLC27L2 (new) versus TLC272 (old):
  • ~ 100 µA vs 2000 µA - supply power (worst case vs typical case) -  :-+
  • 97 dB vs 88 dB - CMRR -  8)
  • ~ 100 kHz vs ~ 2 MHz -  ???




Speculation about things I can't see:

It looks like all three zeners VR101, VR102, and VR103 are different from the unit I have. I'm curious if you can identify the part numbers on these.

I'm curious to know if you can identify which lines have inductors/ferrites on them (one long one above the CPU, two blocky ones near 74HC79 and CD4075) - and perhaps the inductors/ferrites themselves (probably too much trouble for the ferrite).

 

Offline technogeeky

  • Frequent Contributor
  • **
  • Posts: 540
  • Country: us
  • Older New "New Player" Player Playa'
Re: Keithley 197A owners' corner
« Reply #28 on: October 12, 2017, 03:40:30 am »
There are a number of JFETs that have Keithley only part numbers.  These are used for switching.  I ran across this thread which has some information on their characteristics.  Has anyone else looked into identifying substitutes?

https://groups.io/g/TekScopes/topic/jfet_needed_for_range/966846?p=Created,,,20,1,20,0

First, one handy thing that I've learned is that basically all of the ways people have produced JFETS (at least until this multimeter was made, and perhaps afterwards) are enumerated by their construction process. This PDF lists them all, and gives you an idea of why one process might be used as opposed to another. It also shows sample device numbers, and die construction.

Yep, I have seen it and I bought a few of a bunch of different nearby JFETs to test. I wanted to draw their characteristic curves and compare them for this purpose (or at least be able to understand how the existing ones were matched or paired (which in this case might not be the same thing)). However, the curve tracer I built only tests BJTs and only NPNs at that. I need to modify the circuit to test PNPs (which might be easy, maybe trivial) and then modify it to do JFETS (at the very least meaning going negative with voltage).

The two things to figure out is why they were chosen in the first place, and then why they were matched.

I think the current guesses for each are:

Why were Q104-Q113 chosen?

  • ultra-low leakage (process 50, process 51 ~ J113)
  • low on resistance (process 50, process 58 ~ J110)

I don't have any info to guess what's right or wrong here, yet.

Why are those particular ones paired (e.g. Q104 and Q105)?

  • They're not, but Q104-Q109 (and maybe the rest) are matched overall to fall on two pinch-off voltages, so they can control 6 devices with only 2 voltage levels.
  • They are matched in pairs, for _____?

At the end of the day, if someone has one of these parts which is dead for sure, perhaps they can dissolve and microscope it (or send it to me, or someone who can) so we can just look at the die and figure out which process they are.

I have already received the following parts for attempting to see if they could be matches (along with their process, if I know it):

  • J107 (p. 59)
  • J109 (p. 58)
  • J111 (p. 51) (from 3 different mfgs, one is a PF5102)
  • J112 (p. 51)
  • J113 (p. 51)
  • J175 (p. 88) (not sure why I got this?)
 

Offline technogeeky

  • Frequent Contributor
  • **
  • Posts: 540
  • Country: us
  • Older New "New Player" Player Playa'
Re: Keithley 197A owners' corner
« Reply #29 on: October 12, 2017, 08:34:47 am »
Well that's a little fortuitous. After reading my above comments about TLC272CP vs TLC27L2CP, I went and found the original part I removed from my 2nd unit (which was intermittent OL and a bad reading).

This unit has some other marking (SN something something). In any case, this unit now (mostly) works. It makes sense, too: putting a TLC272CP in there draws way too much current, which is far more than the 6.4V zener (which is the precision reference) can handle. Putting this lower power opamp in makes it work.

Now there are two remaining problems:

* everything is a little bit off in the first two or three ranges of each function
* in some of the upper ranges, things are way off

The worst problem:
* in ohms mode, after some unspecified amount of time, the multimeter puts more and more voltage (and current) out. When it's reading roughly correct (only roughly), it's putting about 0.019 volts and some small amount of current. However, after some time (or even right after turn-on sometimes), it's putting 4.030 volts and 0.1008 mA into the red and black terminals. At the same time, it oscillates between measuring ~ 173.xxx ohms and OL.

So I bet the current is correct, but the voltage is not. I'm not sure, though. I'm too tired to keep working on it.

Any hints on where to begin?

edit: perhaps I spoke too soon. Now my -V rail is at +1V. :/
« Last Edit: October 12, 2017, 12:55:33 pm by technogeeky »
 

Offline rastro

  • Frequent Contributor
  • **
  • Posts: 353
  • Country: 00
Re: Keithley 197A owners' corner
« Reply #30 on: October 12, 2017, 01:58:26 pm »
...
It looks like all three zeners VR101, VR102, and VR103 are different from the unit I have. I'm curious if you can identify the part numbers on these.

I'm curious to know if you can identify which lines have inductors/ferrites on them (one long one above the CPU, two blocky ones near 74HC79 and CD4075) - and perhaps the inductors/ferrites themselves (probably too much trouble for the ferrite).
I'll take a look.  I might take a few days since I'm in the middle of replacing my work bench.
-rastro
 

Offline rastro

  • Frequent Contributor
  • **
  • Posts: 353
  • Country: 00
Re: Keithley 197A owners' corner
« Reply #31 on: October 12, 2017, 02:10:34 pm »
... I went and found the original part I removed from my 2nd unit (which was intermittent OL and a bad reading)....
... In any case, this unit now (mostly) works. It makes sense, too: putting a TLC272CP in there draws way too much current, which is far more than the 6.4V zener (which is the precision reference) can handle. Putting this lower power opamp in makes it work.

Glad to hear you making some progress.  It sounds like you tried upgrading to a part with better specifications but it introduced other problems.  Reverting clears those issues.

One thing I'm a little confused on is that in reply # 27 you indicate that TLC27L2 (new) uses less power than the  TLC272 (old) original part.  Maybe I'm not reading something right.  :-//

-rasto
 

Offline technogeeky

  • Frequent Contributor
  • **
  • Posts: 540
  • Country: us
  • Older New "New Player" Player Playa'
Re: Keithley 197A owners' corner
« Reply #32 on: October 12, 2017, 03:18:17 pm »
... I went and found the original part I removed from my 2nd unit (which was intermittent OL and a bad reading)....
... In any case, this unit now (mostly) works. It makes sense, too: putting a TLC272CP in there draws way too much current, which is far more than the 6.4V zener (which is the precision reference) can handle. Putting this lower power opamp in makes it work.

Glad to hear you making some progress.  It sounds like you tried upgrading to a part with better specifications but it introduced other problems.  Reverting clears those issues.

One thing I'm a little confused on is that in reply # 27 you indicate that TLC27L2 (new) uses less power than the  TLC272 (old) original part.  Maybe I'm not reading something right.  :-//

-rasto

Yeah, basically: at some point in the last, I decided (through pictures or through reading on the forum) that U101 is TLC272CP. But it's possible it has always been TLC27L2CP, and I just identified it wrong and had ordered the wrong part, etc.

I did not consciously pick TLC272CP as an upgrade, I thought it was the intended part. But I'm not 100% sure that is the problem yet.

I really don't understand why they don't just use standard voltage regulators everywhere in this unit. Even the Keithley 175 uses 7805 and 7915 regulators. Perhaps they consume too much power for the battery option.  :-//

 

Offline timb

  • Super Contributor
  • ***
  • Posts: 2528
  • Country: us
  • Pretentiously Posting Polysyllabic Prose
    • timb.us
Keithley 197A owners' corner
« Reply #33 on: October 12, 2017, 04:58:26 pm »
197M tear down...additional pictures.

For what it’s worth, that battery board appears to be identical to the 197A’s battery option PCB. It’s essentially a discrete shunt regulator (using a TO220 transistor and thermistor) to trickle charge the NiCD pack, along with a charge pump inverter to generate the negative rail, plus a comparator to monitor the battery voltage, the output of which goes to the low battery symbol on the LCD.

Awhile back I sat down and tried to come up with a modern replacement using a single cell LiOn battery, charger IC and boost converter (the NiCD pack is 12V), however it’s surprisingly difficult!

Why? Well, the power switch on these meters is placed on the low voltage side of the main power transformer (after the bridge rectifier even). The battery connects between the rectifier and power switch. Since the transformer puts out between 13.5 and 16V and that voltage is higher than the battery pack’s nominal voltage (12V), the battery charges through that connection when the unit is plugged in.

When the unit is unplugged, the battery supplies 12V out through that same connection. The charging is completely dumb, relying on the way NiCD batteries charge. (Don’t try replacing the pack with NiMH cells, they’ll overcharge and eventually be ruined.)

So, you can see how it would be difficult to create a replacement that powers a lithium charger and feeds back power from a boost converter over the same connection. I couldn’t figure out a simple way of doing it, short of monitoring the voltage with a micro or comparator and using that to switch the charging circuit off and start the boost converter. (Actually, there is a connection to the “ON” side of the switch, so that could be used to start the inverting and boost converters somehow. That still doesn’t solve the problem of only running the charging circuit when plugged in.)

Anyway, it’s an interesting problem. One I a might take another crack at one day.

Oh yeah, there’s a post on the forums somewhere, where a member reverse engineered the display protocol used by 197 and 197A. If I recall, he used a micro to intercept the data and format it for another LCD. (The LCD in his unit had gone bad.)

I actually found an OLED display with nearly identical dimensions as the OEM LCD. That was another project I never quite got around to. (I ended up fixing my LCD by disassembling it and cleaning the zebra strips and PCB contacts with DeoxIT. I also applied a bit of tape around the LCD’s frame, so it would press harder against the front panel when screwed in, thus keeping more pressure on the zebra strips. Additionally  I found that applying a steel strip to the back of the top of the front panel, with epoxy, also helps to alleviate any flexing, which in turn keeps constant pressure on the LCD’s contacts.)
« Last Edit: October 12, 2017, 08:10:59 pm by timb »
Any sufficiently advanced technology is indistinguishable from magic; e.g., Cheez Whiz, Hot Dogs and RF.
 

Offline rastro

  • Frequent Contributor
  • **
  • Posts: 353
  • Country: 00
Re: Keithley 197A owners' corner
« Reply #34 on: October 12, 2017, 07:48:32 pm »
...
Oh yeah, there’s a post on the forums somewhere, where a member reverse engineered the display protocol used by 197 and 197A. If I recall, he used a micro to intercept the data and format it for another LCD. (The LCD in his unit had gone bad.)

I started a thread on the LCD replacement a few years back - but couldn't find a clean way to mod it into the DMM.  I think the nibbler also did some work on this.
https://www.eevblog.com/forum/projects/keithley-197-led-display-hack/msg493365/#msg493365

I also recall someone else staring another thread on this - I think he was from Europe.  He did a much better job of of analyzing the signals.  I just can't find his thread on the eevblog.  Maybe someone knows?

timb;
Please post the P/N & Mfg. of the display.  Who knows someone may want to carry this forward.

-rastro
 

Offline technogeeky

  • Frequent Contributor
  • **
  • Posts: 540
  • Country: us
  • Older New "New Player" Player Playa'
Re: Keithley 197A owners' corner
« Reply #35 on: October 12, 2017, 07:59:09 pm »
...
Oh yeah, there’s a post on the forums somewhere, where a member reverse engineered the display protocol used by 197 and 197A. If I recall, he used a micro to intercept the data and format it for another LCD. (The LCD in his unit had gone bad.)

I started a thread on the LCD replacement a few years back - but couldn't find a clean way to mod it into the DMM.  I think the nibbler also did some work on this.
https://www.eevblog.com/forum/projects/keithley-197-led-display-hack/msg493365/#msg493365

I also recall someone else staring another thread on this - I think he was from Europe.  He did a much better job of of analyzing the signals.  I just can't find his thread on the eevblog.  Maybe someone knows?

timb;
Please post the P/N & Mfg. of the display.  Who knows someone may want to carry this forward.

-rastro

Yeah, it was me. I have a fully working Arduino code which deciphers 100% of the data sent to the display controller (including display screens, including calibration, etc). It displays the data currently to a standard 16x2 display. I thought I had shared it already, but perhaps I did not.

Another advantage of the 16x2 display situation is that, with the aid of only two small square plastic sleeves (to cover the area outside of the 16x2 display but still visible through the original aperture), the original PCB can be re-used. This process puts a tiny amount of strain on the PCB, but it can easily take it. This way you can retain the original buttons easily and don't even need to manufacture a new PCB.

I promise I'll post my code soon (by the end of the weekend) because the decode table (which converts SPI bytes to ASCII bytes) will be useful regardless of all the other code.

This project is what killed my first unit: my arduino board was not optoisolated from the power supply of the K197, and when I measured an AC mains signal (I think polarity-reversed), the resulting explosion fried the regulator on my arduino mega, the LCD, and a few of the logic chips on the K197. (the arduino board was connected to my PC at the time, which brought a real ground reference back into the picture).

For what it's worth, I am also pretty certain that all of the same information (and perhaps a little bit more information) can be gotten from the 2 or 3 pins which go from the K197 to the GPIB board. I was in the process of trying to prove this/capture some correlated bits when I blew up my meter.

I will certainly do so soon. I am determined to get my two units working properly. Even if they are out of calibration, I can figure that out later.
« Last Edit: October 12, 2017, 08:05:56 pm by technogeeky »
 

Offline tooki

  • Super Contributor
  • ***
  • Posts: 3882
  • Country: ch
Re: Keithley 197A owners' corner
« Reply #36 on: October 16, 2017, 12:55:40 pm »
I’d love to make a custom replacement display board. Just too elegant!

technogeeky, I sent you a PM, we should hop on Skype again and work on our 197/197A issues some more.
 

Offline rastro

  • Frequent Contributor
  • **
  • Posts: 353
  • Country: 00
Re: Keithley 197A owners' corner
« Reply #37 on: October 16, 2017, 05:00:06 pm »
Finally found the other contributor "Maxior" who was also decoding the LCD display communications.  Very nice summary.  This was a couple of years back as well.
https://www.eevblog.com/forum/projects/modding-a-keithley-197how-to-print-on-front-panels/msg790844/#msg790844

Also here's some work on installing a back light @ reply#11:
https://www.eevblog.com/forum/testgear/modding-a-keithley-197/msg725392/#msg725392

-rastro
 

Offline rastro

  • Frequent Contributor
  • **
  • Posts: 353
  • Country: 00
Re: Keithley 197A owners' corner
« Reply #38 on: October 19, 2017, 03:56:20 pm »
...
It looks like all three zeners VR101, VR102, and VR103 are different from the unit I have. I'm curious if you can identify the part numbers on these.

I'm curious to know if you can identify which lines have inductors/ferrites on them (one long one above the CPU, two blocky ones near 74HC79 and CD4075) - and perhaps the inductors/ferrites themselves (probably too much trouble for the ferrite).
I'll take a look.  I might take a few days since I'm in the middle of replacing my work bench.
-rastro

Back with some information on 197M parts:

1. L101 & L102 - I don't have much information.  You can see some of the connections on the back picture of the board.  The front side is difficult to trace out because of vias and passing under components.  On the top of these inductors are 3 wire loops that measure less than 1-ohm between them.
2. E101 - Inductor near CPU.  Connected to  U120/PIN-04 and U124/See Pic                           
3. VR's : I did not want to unsolder components.  Could identify marking on 2 components.
    - 101: Gray; marked IN75/6A304(Possible 1N756A)
    - VR 102: Glass; marked 357                       
    - VR103 : conveniently soldered face down
-rastro
 

Offline rastro

  • Frequent Contributor
  • **
  • Posts: 353
  • Country: 00
Re: Keithley 197A owners' corner
« Reply #39 on: October 25, 2017, 08:56:16 pm »
Looking for some suggestions or ideas on resurfacing the 4 soft mode keys next to the LCD display.  These are rubber with the buttons with some kind of conductor (carbon??) impregnated.  I have a set that is very difficult to make good contacts.  It seems like the conductive treatment is worn-off or gone so resistive it can't activate the mode button. 

I've done all the cleaning to the PCB and rubber button to no avail.  I want to keep these buttons original.  Has anyone tried to resurface this type of button?

-rastro
 

Offline tooki

  • Super Contributor
  • ***
  • Posts: 3882
  • Country: ch
Re: Keithley 197A owners' corner
« Reply #40 on: October 26, 2017, 05:04:32 am »
Looking for some suggestions or ideas on resurfacing the 4 soft mode keys next to the LCD display.  These are rubber with the buttons with some kind of conductor (carbon??) impregnated.  I have a set that is very difficult to make good contacts.  It seems like the conductive treatment is worn-off or gone so resistive it can't activate the mode button. 

I've done all the cleaning to the PCB and rubber button to no avail.  I want to keep these buttons original.  Has anyone tried to resurface this type of button?

-rastro
Though those conductive pads are apparently available online, I think it’s extremely unlikely that they’ve failed. What kind of resistance do you get if you probe a conductive pad with a DMM? Depending on where and how hard I probe, I get between about 1-25Kohms. If I measure the resistance of a pressed button on the connector, it’s up to about 40K for a light press, under 1K for pressing down really hard. (Follow the PCB traces, you’ll see that the buttons have one common, plus one line for each button, leading straight to the ribbon cable.) Test the resistance at a few different points to see if the signal is getting lost somewhere.
 

Offline rastro

  • Frequent Contributor
  • **
  • Posts: 353
  • Country: 00
Re: Keithley 197A owners' corner
« Reply #41 on: October 26, 2017, 01:17:26 pm »
tooki,

Thanks for the recommendations.  I replaced the 4 keys using one from a K175 parts unit.  It fixed the problem. So it's not a connection/signal issue elsewhere. 
I didn't realize they sold sheets and buttons of 'conductive pads' but apparently they do.  Thanks for the suggestion and search description.

I agree these usually don't go bad.  I've always been able to clean them to remedy poor contact issues.  I'll have to look at the resistance on the failed pads.  Visually it looked like they had the imprint of the PCB contact side imprinted in it - suggesting mechanical wear or hardening of the rubber.

-rasto
 

Offline tooki

  • Super Contributor
  • ***
  • Posts: 3882
  • Country: ch
Re: Keithley 197A owners' corner
« Reply #42 on: October 26, 2017, 05:39:35 pm »
Well, substitution is absolute proof of the culprit! Weird, in my 197, they're still soft and work fine, and your 197M should be younger, if anything. I wonder if maybe storage conditions affect the rubber aging.

In any case, you should be able to get the little pads and fix it! :)
 

Offline technogeeky

  • Frequent Contributor
  • **
  • Posts: 540
  • Country: us
  • Older New "New Player" Player Playa'
Re: Keithley 197A owners' corner
« Reply #43 on: October 28, 2017, 12:49:12 am »
Some good and OK news on my repair processes:

My old unit (killed by measuring AC and some ground connection via USB to earth mains -- from my LCD replacement solution) I finally got some life back into it. The ADC and integrator basically had no activity. It turned out to be one of the traces on the board for U105.

I had already replaced U105 (it was one of the chips which died), and I added a socket. But either the trace was vaporized by the original problem, or bad soldering work. In any case, no matter how many times I tried to reflow the joint to include the trace (the cut in the trace was only a millimeter or two away from the U105 pin that it connected to), it wouldn't connect. So I gave up and just used a short bodge wire.

After that, now the ADC is running again!

That older meter is not out of the woods yet, though. The only range that is even close to accurate is the two lowest voltage ranges. All resistance ranges and all but the lowest current ranges show basically nonsense values.

I'm also suspicious of the stability of the voltage measurements. I used my Keithley power supply (2304), which isn't the cleanest supply (although the noise is very ... random? gaussian?). And basically no matter what setting I pick for the Keithley, the meter instantly locks on to some (reasonable, if a little bit off) voltage value -- all 5.5 digits, no drift at all.

This makes me very suspicious, but it is certainly progress compared to before.


As for the newer of my two meters, I have been testing by completely disconnecting the negative voltage rail entirely (by removing the resistor and capacitor), and powering it using the aforementioned Kiethley power supply. One of the cool features of the power supply is that it reads back (and sinks) the voltage and current present on the connected nodes, no matter if the supply is ON or OFF. Furthermore, you can "force" the node into compliance.

This came in handy when I noticed that, when the power supply was OFF, the negative rail came in at -0.640V with some small (sub 1mA) current. In other words, the negative rail was going positive somehow. I suspected this was because of some short to a positive rail, but I was unable to find it. In any case, I tried correcting it by forcing the rail to zero (by setting the compliance voltage to 0.001 volts). Because the current measurement on the supply is so good (it has a 5mA range), I could easily see leakage currents change as certain multiplexing JFETs were turned on or off! We're talking values like 0.0145 mA to 0.0165 mA!

In any case, when I powered the negative rail with 9V, things lit up. Again, pretty much all of the measured values are pretty far off; but at least it's alive. And there isn't the kind of stability the old meter has. I also am pretty sure that two of the multiplexing JFETs are bad: Q110 and Q112 both exhibit a pretty huge drift (like, 100s of mV) in the gate voltage after being turned on (that is, when the voltage is near or above 0V). I can't imagine that's correct, and it's slow enough (and bidirectional) that it seems thermal.

I'm sorry for the rambling, I'm just putting a log of what's going on in case it randomly helps someone else. I think I would have made more progress quickly if I had just went ahead and removed the multiplexing JFETs, but I really don't want to unnecessarily contaminate the sensitive/guarded sections of the meter, because if I screw up calibration I really have no way to get back to calibration.


For what it's worth,  the military manual that was posted above is a great resource. None of the Keithley 197 manuals so far (there are three or four editions out there) have an actual picture of any waveforms (be they oscilloscope traces or just theoretical diagrams). I knew what the waveform of the 197's integrator/ADC step should look like because of the Keithley 175, but you never know.

In any case, despite the fact that the military version manual uses a different CPU (and a different crystal frequency), all of the troubleshooting values in the military manual are identical to the steps listed in the normal Keithley 197 manuals, except oscilloscope
 traces are pictured. I don't know what the "alternate" trace is, but I don't see it on either of my two units.




 

Offline technogeeky

  • Frequent Contributor
  • **
  • Posts: 540
  • Country: us
  • Older New "New Player" Player Playa'
Re: Keithley 197A owners' corner
« Reply #44 on: October 31, 2017, 09:40:34 am »
I finally have success on my older meter. The final thing that needed repair was the PIA chip. Pin 8 was dead - I discovered this because it was effectively floating, which I could notice when I probed things with a 1x scope probe. The output (pin 8) which goes to U103 pin 11, which turns Q112 on/off.

As is the downside with JFETs, if you aren't actively turning them off, then they are on. So Q112 wasn't getting turned off.

So as of now, all voltage (both AC and DC) and current (AC and DC) ranges work!

Resistance measurement is still problematic, and I'm not sure why yet. I thought it might be Q110 but since that's used in the higher DC voltage ranges, I guess it's fine.

Anyway, good progress!
 

Offline rastro

  • Frequent Contributor
  • **
  • Posts: 353
  • Country: 00
Re: Keithley 197A owners' corner
« Reply #45 on: October 31, 2017, 06:03:33 pm »
Great News.  Nice Tshooting;
 

Offline technogeeky

  • Frequent Contributor
  • **
  • Posts: 540
  • Country: us
  • Older New "New Player" Player Playa'
Re: Keithley 197A owners' corner
« Reply #46 on: November 01, 2017, 05:33:51 pm »
So I worked on the resistance stuff last night. Normally, I get completely nonsense readings. For a long time last night, I was getting sensible readings on all ranges except the lowest range. But there are two bizzare things:

1) The way I was able to get sensible readings was by connecting an oscilloscope probe to the forward side of R111 (though the backward side of R111 worked as well) -- along with connecting it to ground. Both x1 and x10 settings worked. If i disconnected this oscilloscope probe, then the readings once again cycled around and were nonsensical - often times just OL, but sometimes actual values. But again, when the probe was connected here, I would get readings that are within 1-2% of the correct value on several ranges (and again, except the 200 Ohm range which reads OL).

2) However, after some time later, and after I added a ferrite to the power input of the meter overall (to try and filter some of the noise that my UPS is putting back on the power/ground lines), this trick no longer worked. I am currently unable to get any sensible resistance readings.

I really don't understand what is going on here. I can't imagine that stray capacitance or inductance is making a huge difference. I can only imagine that somehow the 1M or 10M connection to ground is what made a difference, and this somehow reset or overcame some connection that is otherwise floating (like a JFET which is totally open but should itself have a 1M or so resistance).

But if it's Q112 that is broken (or for that matter, any of Q110-Q113) then why are the voltage and current ranges working (granted: they are actually somewhat off, but only a few percent).

Any advice at all would be very welcome.
 

Offline technogeeky

  • Frequent Contributor
  • **
  • Posts: 540
  • Country: us
  • Older New "New Player" Player Playa'
Re: Keithley 197A owners' corner
« Reply #47 on: November 03, 2017, 05:49:20 am »
 :scared:

My original Keithley 197 is fixed! And even better news, it's essentially exactly back in the calibration state it was before!

The last remaining problem was basically that the LO input was floating. This meant that any measurements I was taking (voltage, current, or resistance) were being taken through the 220k resistor to the LO SENSE input. This explained the slightly wrong readings for voltage, and why resistance measurement didn't work. As for why the above happened (why a 1M/10M resistor to ground from the scope probe caused resistance measurements to run, even if incorrect), I can't fully explain that (other than connecting something to ground, which was floating).

The trace from the LO input which snakes off under the range switches and ultimately to the Voltage switch on pin 4. This was completely blown, lifted off the board. I didn't see it because it's hidden underneath a current shunt and the range switches. I cut this trace off from both ends (to avoid coupling any noise into it), and ran a bodge wire in its place.

Now all the measurement functions work fine. My 1M 0.5% resistor reads 0.99998 MOhm. My voltage calibrator readings are basically the same as before the explosion. Bottom line, everything's back to working!

Now I have to find out what's wrong with my 2nd unit, re-create the Arduino (or probably ESP32-wireless so I can use these meters for datalogging without connecting anything to a PC directly. USB isolators are expensive.

I still need to find a replacement for the protection transistors GES5818. Or I need to find out why they picked those specific transistors. I'm guessing it's for the low (< 5pF max) capacitance. One of them was dead, and I broke the leads off of another one.

I'm so happy!

Thanks also to joe_z on ##electronics on FreeNode who helped a great deal.
 

Offline Cerebus

  • Super Contributor
  • ***
  • Posts: 3396
  • Country: gb
Re: Keithley 197A owners' corner
« Reply #48 on: November 03, 2017, 01:52:10 pm »
I still need to find a replacement for the protection transistors GES5818. Or I need to find out why they picked those specific transistors. I'm guessing it's for the low (< 5pF max) capacitance. One of them was dead, and I broke the leads off of another one.

This thread: Using Transistors as Diodes, is on the go, and there have been mentions of some obsolete transistors that were much loved in these rôles. Might be worth dipping in and seeing if anyone recognises that part number.
Anybody got a syringe I can use to squeeze the magic smoke back into this?
 

Offline technogeeky

  • Frequent Contributor
  • **
  • Posts: 540
  • Country: us
  • Older New "New Player" Player Playa'
Re: Keithley 197A owners' corner
« Reply #49 on: May 08, 2018, 04:49:13 am »
I wanted to chime in here in case anyone is watching. I had promised several times to release some code for doing 16x2 LCD replacement of the broken LCDs for these instruments.

As far as I know, the mapping from native LCD to a 16x2 LCD is complete and total. All displays for all modes translate to something on the 16x2 display. This includes status information, measurements, store/recall modes, dB mode, and troubleshooting modes. I even found a mode that I couldn't find in the manual: with GPIB card installed it sometimes says "OUT" using the alphanumerics.

I am in the process of extending the functionality shown below further, including:

* porting to Arduino Uno (trivial, just test which pins work)
* porting to ESP32 (wifi and/or bluetooth)
* release better documentation
* add back plotting support (native to Arduino)
* adding exporting for something to do plotting (like for sigrok)
* adding the ability for two of these devices to combine, displaying a combined stat on screen (power, for instance)

But since I just got back to a workable point, I figured I'd release the code so you guys can try to use it and/or comment or ask questions.

The Arduino project is located on GitHub. It is designed for a Arduino 2650.

Once I make a little more progress, I'll make an new thread about this with more documentation, pictures. I am essentially assuming you can do the 16x2 display wiring yourself (there are plenty of guides), and then you only need to pick off the GND, SCK, (MO)SI, and SS pins. If you want, you can skip the 16x2 display altogether and use serial as a display.
« Last Edit: May 08, 2018, 04:52:00 am by technogeeky »
 
The following users thanked this post: tooki, Vtile


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf