Keithley 199 -- FIXED!!!!

[staze]

So, looking at buying a ~$100 bench multimeter (used).

At this point, it's pretty much 3 options I can see based on really 1 need: 5.5 digit (or better).

I was at first thinking an Fluke 8810a, but they only do volts and ohms, no current. =/ Guess that isn't THAT big a deal, but figured I'd try for the "big three" measurements.

So at this point, it's pretty much the:

Fluke 8840a/8842a
HP 3468a/3478a
Keithley 199

The Fluke and the Keithley are both Made in the US (a nicety, but not mandatory), all three do the measurements I'm after. And for this instance, let's assume all 3 do TRMS, as well as have GPIB (though I realize the 3468a does not).

Seems the biggest difference is the displays. The Fluke is a VFD, the HP is a non-backlit LCD, and the Keithley is a LED. At this point I'm leaning heavily toward the Keithley. The HP display seems like it just sucks unless you hack it for a backlight. The Fluke and the Keithley are both highly legible in bad/no lighting.

Can anyone enlighten as to the repairability of any of these? I think I've found Service Manuals for them all (the Keithley and Fluke both have their service manuals included in the instruction manual, I'm not sure about the HP). The Keithley is actually the most recent of the 3 as both the Fluke and HP seem to have been mid-eighties to early nineties models. And at least the Keithley provides front panel (or remote) calibration. I can't recall if this is the case with the Fluke and/or HP.

Are there others to consider? Are there any showstoppers? I've heard/read that some of the more modern Keithley's are a bit, dodgy. I also don't overly like the whole battery backed calibration data on the HPs. I get the impression the Fluke and Keithley both store this info in some type of non-volatile way?

Thoughts? Help?

[branadic]

Doh, ~$100 for one of these bench multimeters? Are you serious? In what condition are they or do you expect them to be for that price?

[bingo600]

OP might just squeezze this one home

*Bay 221237809510

And that seems to be an "only" in that pricerange , and has the "just sucks" display , "battery backed calibration" and no GPIB.

Or the fluke (maybe with all the blessings of a VFD)
330935612705

This is in the (or better) category and currently $100 , but i suppose you'd be lucky to get it within your ~100$ budget.
281117141789       

Edit: Watch out for the 281117141789 , pict looks fake , and seller has 0 in feedback ... PayPal though

/Bingo

[alm]

Two things to watch out for with the Fluke 8840a: AC and GPIB were optional (and options are sometimes removed without changing the label), and VFDs get dim after many years of 24/7 service. The advantage of the Keithley and HP is that they had all important features as standard (the 3468a obviously lacks GPIB).

Note that current ranges are somewhat limited: the Fluke is limited to one 2 A range, the HP 3468a only has one 3 A range, the HP 3478A has a 300 mA and a 3 A range, and the Keithley 199 has the oddball combination of 30 mA and 3 A (best current resolution of these models).

The HPs and Keithleys are limited to 300 V, the Fluke goes to 1000 V. The specs of the Keithley and Fluke are extremely close, I would probably rate a Fluke 8840a with AC/GPIB in good condition slightly higher than the Keithley 199, and the HP 3468a/3478a inferior due to the lousy LCD and lack of relative function. On the other hand, an 8840a in similar condition is usually a bit more expensive than the less popular Keithley 199. The huge LED display on the Keithley is very nice in my opinion.

The rebadged Keithley 2100 was quite terrible, but all designs up to the 20xx series (eg. the 2015 that Dave recently tore down) are well regarded. The Keithley 199 uses non-volatile storage for calibration constants (it's the newest of these models). Not sure about the Fluke, should be in the service manual. The Keithley 199 is late eighties (introduced in 1989), the 3468a in 1983 and the 3478a a few years later. Not sure about the Fluke.

I'm quite sure all meters in this list support front panel calibration. I know it for a fact on the Keithley 199 and HP 3468a, since I actually adjusted these.

[staze]

Two things to watch out for with the Fluke 8840a: AC and GPIB were optional (and options are sometimes removed without changing the label), and VFDs get dim after many years of 24/7 service. The advantage of the Keithley and HP is that they had all important features as standard (the 3468a obviously lacks GPIB).

Note that current ranges are somewhat limited: the Fluke is limited to one 2 A range, the HP 3468a only has one 3 A range, the HP 3478A has a 300 mA and a 3 A range, and the Keithley 199 has the oddball combination of 30 mA and 3 A (best current resolution of these models).

The HPs and Keithleys are limited to 300 V, the Fluke goes to 1000 V. The specs of the Keithley and Fluke are extremely close, I would probably rate a Fluke 8840a with AC/GPIB in good condition slightly higher than the Keithley 199, and the HP 3468a/3478a inferior due to the lousy LCD and lack of relative function. On the other hand, an 8840a in similar condition is usually a bit more expensive than the less popular Keithley 199. The huge LED display on the Keithley is very nice in my opinion.

Hey, alm, I was hoping you'd be the one that would respond. =) since it was largely your reviews that pointed me at the Keithley in the first place.

The 300V limit isn't that big a deal since I have a Fluke 189 handheld meter (which will do 1000V), so no worries there. And yeah, the VFD does seem a bit weird, but they are highly visible when new(ish).

I do like that Keithley is owned by Tek (a local (for me) company)... but then, Wikipedia makes that seem pretty confusing. Fluke is pretty local as well (Pacific Northwest). This actually reminds me a bit of the Nikon/Canon debate, and I'm over in the corner yelling "Pentax!" =)

But anyway, thanks! I'm assuming the Keithley 199's you have are still quite stable/good/etc? I still can't say I understand the "Scanner" capability.... in my mind, "Scanner" means either a Receiving only radio, or something for digitizing images/paper, etc.

[alm]

I have no complaints about the Keithleys. One of them has a scanner option, I've never used it. It's a relay card that allows the meter to measure 10 (?) channels in quick succession. Something often used for system applications. I imagine you could use it for some monitoring applications, but I just tend to throw more meters at the problem .

One minor annoyance is that AC/DC and auto ranging are per function, and default to manual ranging. I think the other meter defaults to auto ranging, so it may be something you can change.

[tinhead]

you can buy as well Yokogawa 7552, there are currently two on ebay (with RS232 instead of GPIB, which is why i didn't bought them). They area as well 5.5digit meters with 0.003% accuracy, 20A / 1000V / 4wire.

[staze]

makes sense... just weird they call it a "scanner". so you'd have a bunch of alligators at a test station that you'd hook to test points, hit "go", and it would run through all the test points. makes sense.

[staze]

Huh, the Yokogawa's seem interesting. They take a memory card!

The RS232 would be preferable... but they're a bit higher than my price point unless I could get them to come down a LOT on a "best offer" =P

[bingo600]

Someone either "got taken" , or made a super buy
http://www.ebay.com/itm/281117141789

/Bingo

[edavid]

Doh, ~$100 for one of these bench multimeters? Are you serious? In what condition are they or do you expect them to be for that price?

Yes, he's serious.  That's what they go for in the US, in good working condition (although the 8840A/8842A has been going up lately).

How about the Keithley 196?

There's also the Fluke 8502A/8505A/8506A if you have the space, have to watch out for the options though.

[staze]

Someone either "got taken" , or made a super buy
http://www.ebay.com/itm/281117141789

/Bingo

Wow, that's a hell of a deal. And yeah, either a "fluke" (not to be confused with Fluke), or it's counterfeit.

[edavid]

Someone either "got taken" , or made a super buy
http://www.ebay.com/itm/281117141789

/Bingo

Wow, that's a hell of a deal. And yeah, either a "fluke" (not to be confused with Fluke), or it's counterfeit.

Looks more like a simple scam.

[staze]

Sad if that's the case... at least eBay will help out in that case.

[Smokey]

I do like that Keithley is owned by Tek (a local (for me) company)...

Careful.  Keithley is owned by Tek, but they are both owned by the evil Danaher Corp: Destroyer of engineering departments!

Scanners are awesome!  It's pretty much just a relay card which allows you to hook up multiple probes and have the meter switch between them.  It's really only useful for automated measurements but when you need that it's SUPER useful.  If you can, get the scanner card.

I've been rocking the Keithley 199 for a while now.  I got mine for about $120 from the ebay in really good condition. 

[staze]

That pretty much explains Tek recently. At least they're trying to refocus on what they're good at... and sold off all the stuff they were not so good at (interested in), like the Phaser printers.

My only problem with the older Keithley's is the color scheme. Brown? Really? The 70's called, and they want their color schemes back.... not that beige is much better, but at least it's "common".

Anyway, I purchased a meter. Once I have it, and know it works, I'll post back with details. =)

[codeboy2k]

Since the Keithley 199 was mentioned in this thread.. I have a question.. 

I have a chance to get a broken 199 shipped to me for $100.
The seller says that the Ohms and Volts display is always 0.0000 but the current works.

Seems like an easy fix ?? I had a look at the schematic, there is a whole lot of FETS being used on the input circuit, some as low leakage protection diodes, others as switches... I just wonder if these FETS are (a) selected for their characteristics and (b) unobtanium now

Thanks for any guidance.

Cheers!

[staze]

Since the Keithley 199 was mentioned in this thread.. I have a question.. 

I have a chance to get a broken 199 shipped to me for $100.
The seller says that the Ohms and Volts display is always 0.0000 but the current works.

Seems like an easy fix ?? I had a look at the schematic, there is a whole lot of FETS being used on the input circuit, some as low leakage protection diodes, others as switches... I just wonder if these FETS are (a) selected for their characteristics and (b) unobtanium now

Thanks for any guidance.

Cheers!

So this is actually one of the ones I bought... got it for less than $100 (the seller has "make offer"). I'm actually guessing the issue is a configuration one (either it's set to single shot, or control/output over GPIB, etc). Once I get one, I'll post back and let ya'll know. =)

[alm]

Could also be a calibration issue. I had one meter that always read 0 V on one ACV range. Turned out the cal constant was just zero for that particular range, probably due to someone using the same voltage for the first and second point. A calibration with a known AC voltage fixed it. This would explain the hard 0.0000 V without any noise. My other suspicion would be a short in the input circuit, since an open would result in noisy readings on the lower ranges.

[codeboy2k]

It just seems like it could be an easy fix no matter what..

Yes, the seller has more than 1 available, hard to believe they are all in the same condition though, i.e. all not working in the same way??

I  think the seller has tested a few and doesn't have a whole lot of time to test them all, so maybe it's just a hit and miss...

however, the input circuit seems rather straightforward and easy to follow the signal path, the schematic is readily available, everything is soldered through hole and DIP packaging... could be an inexpensive Keithley 199 if I could get it working again. 

@staze... yes, thanks... I'd like to hear your results when you get it.

Thanks!

[staze]

It just seems like it could be an easy fix no matter what..

Yes, the seller has more than 1 available, hard to believe they are all in the same condition though, i.e. all not working in the same way??

I  think the seller has tested a few and doesn't have a whole lot of time to test them all, so maybe it's just a hit and miss...

however, the input circuit seems rather straightforward and easy to follow the signal path, the schematic is readily available, everything is soldered through hole and DIP packaging... could be an inexpensive Keithley 199 if I could get it working again. 

@staze... yes, thanks... I'd like to hear your results when you get it.

Thanks!

No problem. I should get mine Friday.

I'm actually guessing he/she didn't have time to really test them at all. If you look at their seller page, they have 9500 listings! So they get stuff, probably briefly test it, then throw it up either as is, or working. First thing I'm going to do is reset to factory defaults.

One thing I did notice is that the one pictured didn't have a power button. The seller said he/she would make sure to send me one that did. So, we'll see.

I will definitely report back.

[codeboy2k]

Great thanks!  If it's just a calibration issue as alm says, or possibly just some settings...then it's a bargain for sure.

Since the seller says volt/ohms shows 0.000 and current works I assumed it was something shorted or open in the volt/ohms path (which is clearly seen on the schematic as distinct from the current path ) .. so I was thinking maybe one of the many DG211 analog switches was fried and stuck open, or one of those FETS that are used directly in the signal path (but probably not the FETs used as protection diodes, since those go to + and - supplies, so the display wouldn't show 0.000 in that case, it would have a voltage offset)

I'm still thinking to take a risk myself and buy one anyways... really can't go wrong for less than $100 for this decent 5-1/2 digit meter which seems highly repairable

[staze]

okay, got the meter. good and bad.

Good: powers up fine, and displays. Board/Unit revision is A4.
Bad: none of the measurements work. Volts (DC or AC), Ohms, or Current. Did a reset, no change. Occasionally the volts/amps show "overflow" and the ohms fluctuates a bit, but that's with nothing hooked up. So I'm guessing it's something with references, transformer output, or something along those lines. I did try a calibration for DC voltage using my PSU, but no luck. no change in readings.

Sadly, none of the components are labeled on the board... so finding test points is a bit of a pain.

The manual: http://www.ko4bb.com/Manuals/09_Misc_Test_Equipment/Keithley/Keithley_199man.pdf

Any comments, guidance would be greatly appreciated. I'm guessing once we get one of these squared away, they'll all be the same issue.

Thanks!

P.S. Question for Alm: Do you ever get any relays that audibly click... when switching between AC and DC, for example? I don't get anything. Not with the scanner either.

[alm]

The relays in the 199 click, but they are very quiet: a quite high pitched click which is barely audible over the sound of the button click. Nothing like the chatter of some other meters. You only hear them when opening, not closing. I hear them: at power on/off, when switching from AC to DC (not the other way around), when switching from ohms to volts/amps, when switching from the 3 V DC/AC range to 30 V, when switching from the 3 Mohm range to 30 Mohm, and when pressing the scanner button after selecting one of the channels other than 0.

Yes, often no silkscreen on equipment from this era. You get used to identifying parts from the parts location diagram. So no current measurement either. Not what the seller described. Then my suspicion it would be somewhere in the multiplexer/ADC circuit. I would first go through a visual inspection and the power supply checks from the manual (obviously), and after that use the troubleshooting mode on page 6-18/6-19 (just pick the 3 VDC range) to test the ADC, mux and reference/zero circuits.

[staze]

The relays in the 199 click, but they are very quiet: a quite high pitched click which is barely audible over the sound of the button click. Nothing like the chatter of some other meters. You only hear them when opening, not closing. I hear them: at power on/off, when switching from AC to DC (not the other way around), when switching from ohms to volts/amps, when switching from the 3 V DC/AC range to 30 V, when switching from the 3 Mohm range to 30 Mohm, and when pressing the scanner button after selecting one of the channels other than 0.

Yes, often no silkscreen on equipment from this era. You get used to identifying parts from the parts location diagram. So no current measurement either. Not what the seller described. Then my suspicion it would be somewhere in the multiplexer/ADC circuit. I would first go through a visual inspection and the power supply checks from the manual (obviously), and after that use the troubleshooting mode on page 6-18/6-19 (just pick the 3 VDC range) to test the ADC, mux and reference/zero circuits.

Hmmm... I'm not hearing ANY relays at all. Which obviously makes me suspicious. Visual inspection looks good so far, but I need to pull the digital card, and get under the shield over the bulk of the analog board. With the power supply checks, I'm curious, they reference checking certain points (which is fine) but they say the voltages are referenced to digital or analog common. Where are these points?

I'm guessing the +5V digital is fine (since the display works, etc), but I wouldn't be surprised if the +5V or +15V analog are not working (which would explain the relays not being heard), which they say to check pin 3 on U51 or U52/U50 (respectively), and that these are referenced to analog common (which I'm unsure of where that is.

I will also say I hooked up the "meter complete" BNC output to my scope, and the output looks pretty darn noisy. It's not just a nice TTL output.

Thanks!