Keithley 197A owners' corner

[Vtile]

Hello.
 
There only seems to be one or two threads on this old meter here at forums and pretty scarsely on the net. I thought some dedicated general thread on it could be handy.

The reason is ofcourse I became an owner of one of these as I just noticed a reasonable good (physically almost unused condition, even the range selectors are still matte finish) Keithley 197A 'Autoranging microvolt DMM' in a local pawnshop and couldn't walk by (after revisiting home to pick a few "precision" =0.1% 25ppm resistors and a 1.5V AA battery and test leads). The sales persons were showing symptoms of worry while I spend 15 minutes testing the scope with these "precision" references. I only kick myself that I could have twisted the price down more, he were too happy my "reasonable" discaunt of shelf-price.  The volt-nuttery is a few digits closer (3½ to 5½).

My unit seems to load some capacitor on the frontend while unit is on DC: 200 mV range and test leads are disconnected. This seems to be the only "oddity", but since the 200mV and 2V ranges have >1Gohm input impedance I assume this is should be normal phenomenom. My other high impedance meter (100meg) throws needle to FSD in similar test so I weren't overly conserned with this.

[Smokey]

I've got two 197A on the bench (along with two Keithley 175, two Keithley 199, and one Keithley 196).  I like 'em brown!

[technogeeky]

I have two Keithley 197s that are still in need of repair. I also have manually edited, corrected, and cleaned-up (remove artifacts from drafting and scanning) the schematics.

I have been building the equipment I need to test and repair them recently, so I should be fixing them soon.

[Vtile]

Reserved for the future.

[Vtile]

I have two Keithley 197s that are still in need of repair. I also have manually edited, corrected, and cleaned-up (remove artifacts from drafting and scanning) the schematics.

I have been building the equipment I need to test and repair them recently, so I should be fixing them soon.

Oh, nice.

Are the 197 and 197A internals (except the display) the same or is there a lot difference between the models?

@Smokey, how your two 197A meters react in the two lowest DC voltage ranges without probes?

[technogeeky]

I have two Keithley 197s that are still in need of repair. I also have manually edited, corrected, and cleaned-up (remove artifacts from drafting and scanning) the schematics.

I have been building the equipment I need to test and repair them recently, so I should be fixing them soon.

Oh, nice.

Are the 197 and 197A internals (except the display) the same or is there a lot difference between the models?

@Smokey, how your two 197A meters react in the two lowest DC voltage ranges without probes?

They are essentially identical. The only differences involved are the plastics and the support for the backlight.

[tooki]

I have two Keithley 197s that are still in need of repair. I also have manually edited, corrected, and cleaned-up (remove artifacts from drafting and scanning) the schematics.

I have been building the equipment I need to test and repair them recently, so I should be fixing them soon.

Oh, nice.

Are the 197 and 197A internals (except the display) the same or is there a lot difference between the models?

@Smokey, how your two 197A meters react in the two lowest DC voltage ranges without probes?

I've got one of each.

The 197A is fundamentally identical, other than adding an extra power rail on the display connector for the backlight, and the PCB layout is also different near the power transformer, because the 197A uses an IEC power cord, whereas the 197 has a fixed cord.

On the low ranges without probes, the numbers should be jumping all over the place.

[Vtile]

On the low ranges without probes, the numbers should be jumping all over the place.

Hmm... My unit is going up starting from 0 mV adding about 1 mV/s. Shorting the sense ports will zero it (and keep it on zero) as do shorting the input jacks.

[technogeeky]

On the low ranges without probes, the numbers should be jumping all over the place.

Hmm... My unit is going up starting from 0 mV adding about 1 mV/s. Shorting the sense ports will zero it (and keep it on zero) as do shorting the input jacks.

Perfectly normal.

[tooki]

On the low ranges without probes, the numbers should be jumping all over the place.

Hmm... My unit is going up starting from 0 mV adding about 1 mV/s. Shorting the sense ports will zero it (and keep it on zero) as do shorting the input jacks.

Yes, that's what meters with high input impedance do. If it didn't it would be indicative of a fault.

[Vtile]

Good to hear. What a lovely meter, how I have managed without, especially the 4-wire resistance and the PN-junction test (only 1.5% typical accuracy unfortunately) are handy. Hmm... I finally have tool to make a voltage and current reference, need to do one now. 

Does anyone got further with the IEEE-488 module reverse engineering or the internal serial bus decoding? There were some discussion on this two or so years ago.

[psient]

Hello: I am learning. I have a Mastech handheld (Ms8228). I was hoping to get something better for the bench. I haven't much money. I was looking at the Flukes until I read their service is not very helpful. If I found a inexpensively priced 197a I thought I'd buy it. I'm retired and pensioned at the low end having only been a community college adjunct with a PhD from a highly regarded university. I'd have a little bit more to spend but I lost mysocSec because of my windfall in earning that pension from teaching lower income students . 

Is this  good decision in my seeking greater precision for the bench at a limited budget? I know this is the place owners comment but I figure this thread wouldn't be here if the DiMM wasn't worth about 75 bucks?

Thank you

Jon

[tooki]

It’s a great meter. The only thing that newer/more expensive bench meters have is a) much faster readings (the 197A is about as fast as a good handheld meter, around 5 readings per sec), and b) VFD displays that are a bit easier to read than LCDs. For $75 it’s definitely worth picking up. 

One thing I love about the 197/197A is automatic 2 and 4 wire resistance. My K2015 requires you to choose manually.

[Vtile]

197 also supports the battery operation so it can serve also as field instrument or true floating meter.

The real qurstion is though. What is sufficient resolution for your use.

So far I  have been pleased with the functionality.

[Cerebus]

Hadn't spotted this thread before. I picked up a 197 (non-A model) a few months back. It's become my daily driver bench meter. As LCDs go it's one of the easiest to read that I've encountered.

Mine needs a new switch for the AC/DC selection and it's a little troublesome tracking down one that's 'just right'. That style of switch used to be uniquitous but they're quite hard to find now. If anyone has any idea where to source one (short of cannibalising a dead meter) I'd be pleased to hear where.

[rastro]

There is also a 197 Military version.  Which is an uplift on the standard 197 model.

Added design/features:
- Additional shielding over the processor
- Different (shielded?) cable going to the display/panel.
- Torroids on the power input and on the meter inputs.  Probably to reduce noise.
- Detachable AC power cord.  This makes more sense for battery operation.
- Typically has battery/charger.
- It also has a 10A fuse on the high amp input.  The standard 197/197A aren’t fused for the 10A input!!!
- Like the 197A it has 4 references in dB mode including 600 and 50 ohm.  The 197 only has 600 ohm.

rastro

Edit: Just remembered the model: AN/USM-486A; http://www.liberatedmanuals.com/TM-11-6625-3055-24P.pdf (Correction http://radionerds.com/images/7/72/TM_11-6625-3277-14.PDF)

[tooki]

There is also a 197 Military version.  Which is an uplift on the standard 197 model.

Added design/features:
- Additional shielding over the processor
- Different (shielded?) cable going to the display/panel.
- Torroids on the power input and on the meter inputs.  Probably to reduce noise.
- Detachable AC power cord.  This makes more sense for battery operation.
- Typically has battery/charger.
- It also has a 10A fuse on the high amp input.  The standard 197/197A aren’t fused for the 10A input!!!
- Like the 197A it has 4 references in dB mode including 600 and 50 ohm.  The 197 only has 600 ohm.

rastro

Edit: Just remembered the model: AN/USM-486A; http://www.liberatedmanuals.com/TM-11-6625-3055-24P.pdf

So basically, the military 197 is halfway between a 197 and 197A, plus the added fuse and CPU shield, toroid on the inputs, and with silkscreen on the analog section shielding, it looks like. (The standard 197A has a toroid on the AC jack, and the removable cord.)


Did you look at that manual? No way is it the right one; nothing about its construction matches a 197/197A, nor does it match the pictures you posted. It looks like it’s probably a Fluke design. (Looks like the Keithley 197-M is AN/USM-486A; that manual is for AN/USM-486U

[tooki]

Wow, there’s a beauty of a military 197 on eBay right now: supposedly unused (and it looks it!), fairly recent calibration

http://www.ebay.com/itm/KEITHLEY-197-M-2-BENCHTOP-MULTIMETER-NOS-SURPLUS-TESTED-GOOD-COMPLETE-SET-/182518230394?_ul=CH

[technogeeky]

There is also a 197 Military version.  Which is an uplift on the standard 197 model.

Added design/features:
- Additional shielding over the processor
- Different (shielded?) cable going to the display/panel.
- Torroids on the power input and on the meter inputs.  Probably to reduce noise.
- Detachable AC power cord.  This makes more sense for battery operation.
- Typically has battery/charger.
- It also has a 10A fuse on the high amp input.  The standard 197/197A aren’t fused for the 10A input!!!
- Like the 197A it has 4 references in dB mode including 600 and 50 ohm.  The 197 only has 600 ohm.

rastro

Edit: Just remembered the model: AN/USM-486A; http://www.liberatedmanuals.com/TM-11-6625-3055-24P.pdf

So basically, the military 197 is halfway between a 197 and 197A, plus the added fuse and CPU shield, toroid on the inputs, and with silkscreen on the analog section shielding, it looks like. (The standard 197A has a toroid on the AC jack, and the removable cord.)


Did you look at that manual? No way is it the right one; nothing about its construction matches a 197/197A, nor does it match the pictures you posted. It looks like it’s probably a Fluke design. (Looks like the Keithley 197-M is AN/USM-486A; that manual is for AN/USM-486U

Agreed, that's just not a 197. Looks like a Fluke to me, too.

Also, it looks like the transformer has a shield around it too (neither of my 197s have that - I think the 197A does).

I am still having a ton of trouble repairing my 197s. Some of the circuits seem very complicated to me. Right now one of my units is reporting OL in every range, on every function. The other intermittently reports OL in between incorrect measurements. This 2nd one (OL then measure then measure then OL) seems to to have problems with the -9V rail intermittently. It's like the negative supply isn't sufficiently loaded or something. I really don't understand how the negative supply is created/regulated.

For noobs like myself, though, it turns out that two other models rely on essentially exactly the same signal path (modulo the details of the range switching): the Keithley 175, the 192, and the 197 all have basically the same process after signal conditioning, all the way to the microprocessor.

Now if only I could get an idea of what exactly is going wrong with my units. I might have to ask someone to take oscilloscope screenshots of certain measurements. I have so many measurements I doubt are correct, though, I almost don't know where to begin.

I actually have a battery board from the 2nd unit, but I never tried it until yesterday and it turns out it arrived dead. The CMOS inverter which provides the negative rail doesn't produce any negative voltage. Eventually, I'll have to order one of those ICs.

Unfortunately I don't yet have any low noise power power supplies (especially no split-rail or bipolar ones), so I can't bypass the AC power supply yet.

[technogeeky]

Here's the real military manual:

AN/USM-486A.

edit:

This is not just a 197 with shielding.  There are no schematics or part numbers in the manual, but there is a layout diagram. Page 129. At first glance (through eyes who have spent a lot of time staring at these schematics), the entire digital section is totally different. The entire analog front-end looks exactly identical though. Other than the aforementioned 10A fuse, I don't see a single part out of place.

This board also has (gasp!) power rail test points!


edit 2: It looks like this manual has some nice reference material (namely waveform examples captured with an oscilloscope) that might prove very handy indeed!


edit 3: I did discover kind of a neat trick when trying to isolate noise and power rail problems that are influenced by transients and noise generated by the normal multimeter operation: You can select the Ohms reading and also (intentionally) select the AC button ON. This will cause all of the FET switching to stop, which will calm a lot of sources of noise down to their minimums.

[rastro]

Ooops.

I grabbed the wrong URL in my browser.  I had a couple of tabs open trying to find different versions of the manual - hoping to find better quality schematics for Nibblers documentation work.  I've pasted a correction in my initial post.

Thanks
rastro

[technogeeky]

Not that it's strictly necessary, but here is a calibration instruction sheet for the 197 multimeter:

[rastro]

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

[rastro]

I did a further tear down of a 197M.  As pointed out, there are some significant differences compared to the 197/197A.  I'll attach a few pictures.

I'm going to leave the instrument apart for this next week.  So if some one has any questions or picture requests I can follow up.

-rastro

[rastro]

197M tear down...additional pictures.