hp 34401a, what to worry about when buying used?

[linux-works]

if buying a used 34401, what questions should be asked to the seller to see if there are expensive things that need to be fixed?  are there any common problems that can be show-stoppers on a 'used, good deal' item?

I often see these meters in the $500+ range but sometimes I see low prices and wonder if the seller knows about some issue that will cause a lot of expense to have it fixed, yet the unit might seem like it works ok enough to casual buyers and sellers.

any advice on how to buy one of these used and not get taken for a ride?

[Lightages]

The one thing I have learned to look for is the brightness of the display. I have not purchased one yet but am still looking for a good 6.5 meter. The displays fade over time of use and some of them get pretty dim. If you see a an image in the ebay advert and it looks grainy, or blurred, or the display seems to be showing digits changing during the image, then the seller probably took the picture with a long shutter speed to try and make the display look brighter.

This is not too big a deal as the displays can be purchased for replacement for around $70, but it is an extra cost and work. Ask the seller about the brightness of the display and if you get anything other than a straight answer then you might want to look for another seller.

[KJDS]

I've nothing to add, but want to know more about these as I've just bought 20 of them.

[tszaboo]

Test the current range, and bring a spare fuse just in case. Sometimes the ohms range can be dead, no excitation current. If the DMM is old the display could be faded. Handle might be broken or the rubber, not a big deal. The other parts are hard to destroy, but not impossible.
Good luck with it!

[linux-works]

good point about the photo trick.  I've seen some bright display photos but it could very well be doctored-up in post.

other than the display, I wonder if there are ranges that might be bad even if some are testing ok?  how about whole subsystems?  current, voltage, ohms and so on.  I'll try to get sellers to verify that at least one testable thing shows a value in each general mode, but I suspect that with complex things like this, there are many details that can be easily overlooked and cost too much to fix to be worth it.  even a calibration on the unit is often as much as the unit costs (from what I understand about proper calibration costs on units like this).

on older hp gear, if the internal battery goes dead, you lose all cal data.  so I'm worried about stuff like that, on used gear.  I just don't know this model at all and when I see such huge ranges in price for them, it makes me wonder.  lots of sellers who 'play dumb' are only acting and they know their gear is bad, but they claim 'unable to test' or 'tested for power-on only'.  I know those are almost always warning words.

[ManateeMafia]

A few people have also reported a bad front/rear switch.

https://www.eevblog.com/forum/testgear/hpagilentkeysight-34401a-2w-problem/

https://www.eevblog.com/forum/testgear/34401a-cal-error-707/

[eas]

Worry that you should be looking for a Keithley 2000 instead?

I considered both and decided to go the Keithley route. Now I have 3 K2000s and 2 of the related K2700. I am sure there are important differences between them, but some of my experience with the Keithley's probably translate:

• Screen brightness has been the most variable factor
• Rubber button reliability has generally been quite good, but one unit that seemed to keep company with a pipe smoker for a couple decades is a little reluctant despite my effort to clean the contacts.
• 100 mV AC range seems to be the most variable in my limited testing. I'm in the process doing a more detailed comparison between units now).
• Missing trim that lets gunk/air currents into the device. Not a huge problem with my keithley's but seems to be an issue with some of the HPs I've seen on eBay.
• Firmware version. Not sure how this works with the HPs, but with the K2000s, its a eprom chip swap and older version aren't supported by current VISA drivers.  Also, upgrading some versions require recalibration.
• Be sure to look at the history of past sales, not just the current listings, to figure out what a reasonable price is

[free_electron]

Worry that you should be looking for a Keithley 2000 instead?

bwahahaaaaaa . best joke i've heard in a long time. if you would have said a 2001 or 2002 i'd have replied : no contest.  the 2000 is not really that great of a machine..

run the extended selftest ( in the menu ). if it tells you no errors : good to go

[KJDS]

run the extended selftest ( in the menu ). if it tells you no errors : good to go

That's very useful to know.

[VK5RC]

I believe a relatively 'common' fault is with one of the Tant caps on an internal power rail, gives a false fault code/s and takes out one of the zeners with it,(mines has done it) it is well described on here I recall and also the HP/Agilent/KS yahoo group. It comes up in the self test routine .
https://www.eevblog.com/forum/testgear/hpagilent-34401a-ac-v-fail-error-621-due-to-short-in-cr305-zener/

The front back switch I think can go high contact resistance, it can be disassembled, cleaned and re-assembled with great care, lots of little plates  and springs of different sizes, I think one of the volt nut guys here describes it. ?GermanEE.

I have used mine a lot (it had a hard life before I got it as well), general comments about them are good.

[linux-works]

if you don't care about having front/rear access, is it a horrible crime to 'edit' the switch by soldering short jumpers where things should be connected?  or maybe even just taking it out entirely and wiring in your own shunts at the right holes on that switch pcb area?  is it bad enough to justify putting in small relays and controlling them, thus working around the bad switch problem?  I guess that actual switch is no longer buyable as a part?  if signal does go thru that, it seems like it could be a weak link and if you can live with not having rear connections, why not just bypass the switch and probably even get better performance than even a brand new switch would give (no?)

[dom0]

is it bad enough to justify putting in small relays and controlling them, thus working around the bad switch problem?

You would need high-quality low thermal EMF relays for that. Soldering needs to be done correctly, then it shouldn't be an issue.

[eas]

Worry that you should be looking for a Keithley 2000 instead?

bwahahaaaaaa . best joke i've heard in a long time. if you would have said a 2001 or 2002 i'd have replied : no contest.  the 2000 is not really that great of a machine..

Hmm, both the HP and K2K seem like fine machines with similar new and use prices, similar stability and specs. I picked the K2k because of the scanner card option, but hey, if the the Playskool bumpers tip the scales the other way for you, that's cool -- I wouldn't mind some bumpers to protect the instrument, my furniture and the top of my foot from damage.

[Deathwish]

I've nothing to add, but want to know more about these as I've just bought 20 of them.

Price ?.

[free_electron]

Worry that you should be looking for a Keithley 2000 instead?

bwahahaaaaaa . best joke i've heard in a long time. if you would have said a 2001 or 2002 i'd have replied : no contest.  the 2000 is not really that great of a machine..

Hmm, both the HP and K2K seem like fine machines with similar new and use prices, similar stability and specs. I picked the K2k because of the scanner card option, but hey, if the the Playskool bumpers tip the scales the other way for you, that's cool -- I wouldn't mind some bumpers to protect the instrument, my furniture and the top of my foot from damage.

That's not it. The Keithleys are plagued by bad caps in the power supplies. These things are timebombs. the electrolytics leak taking out the entire supply , eating half the board , shorting out the bridges which blow up , taking out the transformer after it has sent 25 volts AC up the ass of every single chip on the motherboard. MY count stands at 7 of machines that were in that state ...

There's no schematics for the 2001 or 2002. They are a contraption from old obsolete technology that is overly complicated.

Open a 2000 and a 34401 side by side. the 34401 is a schoolbook example of proper engineering. The layout is clean , simple. The Keithley is a hairball.

Don't get me wrong , 34401's have their share of problems.
- The tantalum cap is well known ( any engineer still using those should be barred of ever designing anything ever again and branded on his forehead with the text 'Tantalum User'. Tantalums are a PEST)  I lost the count of how many times i had to fix a machine or instrument and the cause was a fried tantalum. The only good Tantalum is one that is not fitted into anything i own.
- The swithc is also known
- the 2512 resistor ladder next to the front or rear jacks can crack loose in a range of serial numbers
- the cpu on the front panel can explode if tone of the VFD drivers calls it quits and commits harakiri in a fiery fireball.
There is another tantalum in the True RMS section that can kill that one off as well.

I can fix 34401's almost blind. The faults are simple and easily repairable ( if you are into instrument repair that is... ) The keithley on the otherhand ... documentation is inexistent and its full of handselected matched crap that is unobtainium. I bought all my 34401's ( i got 8 ) as parts units. They have all been repaired , have all gone through calibration are all in factory spec.

This handsesecting stuff may have been the bees knees in the 50's and 60's but in this day and age there is no need for that. The software and algorithms in these machines are so powerful they can compensate.

THe keitlheys are simply made with old technology and they need a massive load of parts to get it to behave. Agilent on the other hand simply fabs their own custom chip, laser trims it in the factory and , since all is integrated, the drift is much more under control than in the massive circuit board with 1000 parts..

Nope. Keithley was great , but now it runs on outdated technology. They havent released anything new in years... their last multimeter update predates the 34401....

[dadler]

That may very well be true--are you saying the 7510 is an old design with a spiffy touchscreen tacked on?

http://www.keithley.com/products/dcac/dmm/graphsampling/?mn=DMM7510

[free_electron]

Hold it. that's a 7 1/2 digit machine ! we're talking apples vs pears here. That thing is a totally different price bracket.

Take a look here:

http://www.keithley.com/products/dcac/dmm/broadpurpose

the 2000 is outdated.
the 2100 ands 2110 are rebadged machines ... not keithley designs.... Array / Picotest / one-hung-low

So sad how the brand has gone to snot.

[deadlylover]

The currently available Keysight VFD/front panel assembly drops right in only after a certain serial number IIRC, you can always just lift the VFD off of it, but it's something to consider. The specific number is mentioned somewhere in the service notes, it has something to do with the display firmware.

On that note, make sure when looking at a listing that the picture shows it in full 6.5digit mode, because it defaults to 5.5 digit. Sometimes it lives its life in 5.5digit mode with all of the digits relatively dim...and the last one in 6.5 digit mode is as bright as the sun. 

The factory cal count is around 35, so if it's higher than that then you know it's been adjusted (or tinkered with =P).

[dadler]

Hold it. that's a 7 1/2 digit machine ! we're talking apples vs pears here. That thing is a totally different price bracket.

Take a look here:

http://www.keithley.com/products/dcac/dmm/broadpurpose

the 2000 is outdated.
the 2100 ands 2110 are rebadged machines ... not keithley designs.... Array / Picotest / one-hung-low

So sad how the brand has gone to snot.

True. The 2100 series is an abomination. I'm not sure which market segment the 2010 is targeted at, either.

[HighVoltage]

I have 10 of the 34401A's in use and only two of them I bought new, a long long time ago in the early 90s.
The oldest one was used a lot over the years and still has a bright display and is calibrated specs.

All others I have bought used and most of them in "Broken" condition. Years ago I had a friend helping me with the repair, these days I do it myself. The beauty is that these instruments are relatively easy to repair and all schematics are available.

So, I would even recommend to buy them, if they are broken, because they can be repaired easily, most of the time, with a few exceptions.

I have a friend in town that repairs equipment professionally for a living (Calibration Lab) and he loves these 34401A, 34410A, 34411A repairs a lot, because it is always a success and his customers are always happy. On the other side, he hates to repair the Keithley 2000, 2001, 2002 because it is always a nightmare for him and usually the customers are not happy with either the repair or the high invoice. He told me once, to stay far away from the older Keithley DMMs.

[TiN]

There's no schematics for the 2001 or 2002.

Minor correction, we do have full official schematics for 2001.
And lot of stuff in 2002 is exactly same (heck, they share same digital PCB even ), so that's something as well.
As positive side, these old Keithley's have old parts, and only one custom chip - ADC CPLD, but rest is market-grade parts, so even if half of board fried - you still have chance of fixing it with buying replacement from digikey.

Not really something you do in 5 minutes during lunchtime, but doable.
But yeah, this is offtopic in 6.5 digit grade instruments thread.

[KJDS]

I've just had a quick test of the first six that I brought into the lab. The only issue I have is that one has a blown current fuse, and someone has epoxied the fuse cover on so I can't get at it.

They are in various states, some with the remains of rack mount, some with bumpers and handles and some with the added pouch with manuals and leads.

[HighVoltage]

I've nothing to add, but want to know more about these as I've just bought 20 of them.

Wow, 20 at once? That is a lot.
What are you planning to do with them?

It seems that they get more stable with age.
My oldest one from the early 90's is extremely stable from cal cycle to cal cycle
The latest one I have is built in 2014 and has the Keysight logo and needed re-calibration in the 10V DC range
 

[rsjsouza]

There's no schematics for the 2001 or 2002.

Minor correction, we do have full official schematics for 2001.

And non-official Keithley 2000 schematics are here, if anyone is interested.

[free_electron]

There's no schematics for the 2001 or 2002.

Minor correction

oops. i stand corrected. gonna siphon that off immediately and add to my collection. i only had the 2000