• The move to the new server is done. There are some software and database maintenance updates in process. This has us passing the hat around to help out. We appreciate any donations. Seriously, even a dollar helps. The payment page may be found here - https://www.audiokarma.org/support.html

pac1085's rambling test equipment thread

pac1085

pac1085

AK Subscriber
Subscriber
I figured I'd start my own thread to ramble on about my never ending test equipment repairs. I post some of these as vids on my YouTube channel: Phil's Analog Basement

This weekend I'm working on calibrating my HP 428B clip on DC milliammeter. It's a really neat tool and as far as I know, HP's last vacuum tube device as it was in the catalog up until around 1986. I bought it last year, and it didn't work - I found a tube had gone to air. I replaced that tube and it now works. However, it reads low on all ranges. The tube I replaced was V8, a 12AU7. I replaced it with a Hewlett Packard labelled GE tube that has perfectly matched triodes for both plate current and transconductance. I actually tried a few others in there and it didn't behave any better, so I don't think that's it.

This is a later production unit, around 1980 which used a PCB. The earlier versions were point to point wired.

The calibration process doesn't seem too bad, but once I really started digging into it, I found that I ended up needing quite a bit of equipment. Luckily I have everything I need to make with happen.
IMG_7779.jpeg
Analog scope for XY mode to perform the final adjustment.
IMG_7830.jpeg

One of the steps calls for an oscilloscope with 10 vertical divisions. They suggest a HP 130C scope in the manual. I remembered the 11403A displays 10 vertical divisions so I pulled that off the shelf for that step. The rest of my scopes have 8 or less. I could have made it work by scaling it down to 5, but this scope needed some runtime anyways.
IMG_7840.png

Fluke 5101B calibrator as the current source. Most of the calibration requires DC current however there is one where it needs to see AC current, which this does. The manual calls for a HP 6920B which can apparently supply up to 5 amps. The fluke only goes up to 1.99999 so I'm just using some 26awg wire and looping it through the probe 5 times for the 10A range (they have you do this on other ranges since the 6920B has limited selections - will be easier with the Fluke since I can key in whatever I want)

One of the tests required a very low frequency square wave (5 uHz), but at a higher amplitude than my Siglent SDG1032X or HP 3325B could provide, so the 467A amplifier to the rescue.

IMG_7846.jpeg
 
Last edited:
Register to hide this ad
arts said:
Is it just me, or does it seem like every time there's a job to do you end up having to repair the test equipment first? I really think I repair/rebuild more test equipment than anything else.:dunno:
Click to expand...
I haven't really had that experience yet, but I tend to spend way too much time/money overhauling gear for the lab when I first receive it, maybe thats why
 
Well, things aren’t looking too good for the 428B. I think it’s actually a bad probe. Page 19 of the manual has some basic resistance checks of the probe. They are supposed to read 5 ohms. One reads 15. The manual goes on to say if any pair of pins is significantly higher than between the other pairs (by about 3 times), then an open probe coil is indicated.

I see someone has one for sale, I messaged them and asked if they would check the resistance for me before buying.

Edit: I disassembled the probe and cut back the wires just to make sure there wasn’t a break or anything. Nope. It’s definitely bad. Bummer IMG_7848.jpeg
 
Last edited:
With all of the modern SOTA gear available out there these days, it's nice to see that there are still some people out there ready and willing to give those beautiful old pieces of gear the TLC they need and deserve!:thumbsup::trebon:
 
New probe ordered. Guy on eBay had two for sale. One of them failed the ohm test but the other passed. I think it’s from an earlier 428a but as far as I know it’s the same probe. I got him to throw in the broken one too, maybe I can use half of it to fix mine and have a spare. I think each half of the transformer might be matched to each other though
 
Vint Age said:
With all of the modern SOTA gear available out there these days, it's nice to see that there are still some people out there ready and willing to give those beautiful old pieces of gear the TLC they need and deserve!:thumbsup::trebon:
Click to expand...
For this tool at least there is only one modern alternative that I know of. Model 428E - Applied Physics Systems

It’s basically the same thing, they took over after HP stopped making these in 86.
 
pac1085 said:
I got him to throw in the broken one too, maybe I can use half of it to fix mine and have a spare. I think each half of the transformer might be matched to each other though
Click to expand...
Nice. At the very least you have a no cost dissection candidate ...

Probably not your situation but what I recall from the Tek AC current scope probes we used years ago was if they were dropped or took a reasonable hit, the core would crack and or chip and then the probe was toast.
 
Looks like it has similar specs to the Tek AM503/AM503B but i like the meter rather than having to use a scope. I have a couple of the Tek AM503s and an assortment of A6302 and A6303 probes waiting for some love on the shelf . . . You've motivated me to dig them out in the near future.
 
Vint Age said:
Nice. At the very least you have a no cost dissection candidate ...

Probably not your situation but what I recall from the Tek AC current scope probes we used years ago was if they were dropped or took a reasonable hit, the core would crack and or chip and then the probe was toast.
Click to expand...
Yea that’s definitely the case with them still, I’m terrified of dropping my TCP202! I was afraid of buying it used for that reason too, but luckily it worked out.

For this one though, I’m guessing someone overloaded it, shrug.

Edit: ok maybe it was dropped. I just found this in an old tech note about it

1774798894642.png
 
I like that 11403 :-)my secondary workbench is also a home for the DSA602 here... they are heavy metal and they can provide a great job, still fast enough for something.
 
Very cool scope. One of the reasons I keep the 11403A around is that it has the 3GHz 11A81 module, which is handy for checking out my RF signal generators
 
pac1085 said:
Edit: ok maybe it was dropped. I just found this in an old tech note about it
Click to expand...
What would happen with the AC current scope probe for us is when it was overloaded the core would saturate giving us false reading but the winding never opened or changed.
 
a sampling unit in a 1GHz scope that is awesome :-)
I have here the classic setup: 11A72 and a 1,5GHz Fet-Probe.
I need a 11A34 for something more then 2CH.
 
Gathering some parts for another project. I have a HP 3325B function generator, but it does not have option 002 - High voltage output. As you could see in the first post, I needed to use a 467A to amplify the signal from my function generator for that step of the 428B calibration.

Option 002 is somewhat rare, enough to not be able to find any pulls for sale etc. Luckily someone cloned the board and released the Gerber files and BOM for free.

HP3325A Opt 002 - Google Drive

drive.google.com drive.google.com

I ordered the PCB's from JLCPCB. There is a metal shield that needs to be fabricated as well. I uploaded the dxf file to send cut send, they can make it cheap but their shipping price is pretty high at $21. However if I buy $30 or more worth of stuff, the shipping is free. Hmm, wonder what other sheet metal parts I could get laser cut ...
 
Pulled the broken Keithley 6517A Electrometer out today since someone replied to an old post elsewhere, with a suggestion to try. Sadly it didn't get me anywhere. I am still getting the error codes 421 and 414 when attempting the Offset-Adj (which is also the first step in the calibration process).

A few examples of what it's doing ... if you give it a 5v signal, it shows 5.2v. If you give it a 1v signal it shows 1.2v. If you give it a 1mv signal, it shows 1.2mv. It does the same thing on all amperage ranges.

If you run the offset adj, they both fail, but it does correct it a bit and get it closer to 5.000v for example. But it goes away next time you turn it on, even if you run it with the cal unlocked. It also strangely drifted down to like 4.78v and stayed there after a while. Shrug.

This would be so cool to get working for the lab.

IMG_7999.jpeg
IMG_8004.jpeg

I also notice that in current mode on the 200nA and 20nA range, it shows 18nA even with the triax cap over the triax port. I can not go down range any further (2nA, 200pA, 20pA) since they all see 18nA and throw an overflow error. Perhaps that is the key to the whole issue? Leakage somewhere?
tempImagehH4Cwr.jpg


I noticed some flux residue around Q414 (top right) - it's an LS313 dual NPN transistor. The collector legs are cut off and only the B & E terminals are used. The manual suggested a IT121 should be in that spot but with this being a 2012 production board, I suppose it's possible those were already out of production (4 leg TO-52 package) and the LS313 was subbed in.
IMG_8043.jpeg
 
Last edited:
The 428B works with the new probe. The seller sent me the second broken probe, and I was able to fix that one, too. It must have had a broken wire somewhere inside the cable within a few inches of the connector. I cut it back a bit and put the end back on. It was pretty rusty, must have been in a wet location, so I swapped the good parts from my bad probe onto it. Now I have two working probes. The calibration is almost dead on with one of the probes. The other one needs some tweaking. I need to decide which one I'm keeping with the instrument and calibrate it against that one.


IMG_8063.jpeg

IMG_8062.jpeg


For the 6517A electrometer, I posted on eevblog hoping for help. Not too much more insight but they think the LMC6001 or the LS313 are the most likely candidates. I ended up ordering an LMC6001AIN, LS313 and two LT1013ACN8 (used in the offset circuit). Those parts weren't cheap but I'll risk a bit of $$ on this meter... I also ordered some methanol and better cleaning swabs so I can get it really clean...
 
Richard P said:
Are you a member of the HP/Agilent/Keysight Groups.io group? If not, you should join . . . they probably have the best test equipment experience out there and may be able to help.

HP-Agilent-Keysight-equipment groups.io Group
Click to expand...
I am, but this is a Keithley not a keysight. Keithley is owned by Tek, I posted on the Tek groups.io page but no replies. I don’t think there are too many out there who have repaired these, the most comprehensive info seems to be at eevblog
 
You must log in or register to reply here.
Back
Top Bottom