Restoration : HP 4262A LCR meter

[TiN]

No, just soapy water Well, perhaps that would be a problem if we had test set, but since there is none here, not an issue.
I already have cheapo probe for 4263B which using BNC's so I'll just use same one here for testing and comparisons.

[TAMHAN]

Tin, sorry I do not want to sound disrespectful. But Fail 1 is "normal" in that it points to unhappyness with the DUT. For self test, some ranges need to be shorted dead.

Wait, ill look it up for you *stands up*

L and R need all four binding posts shorted! The resistance must be VERY low - I managed to get my 4262A to bicker if the cable resistance is not low enough.

[cncjerry]

tamhan,You beat me to it.  If you don't have the shorting bars on or good leads with low resistance from cur to pot you will get the fail1 error.

[TiN]

That doesn't fix fail 2 ones though. I'll have a check after, currently time not allowing more on this.

[TAMHAN]

Hello,
on mine, I always get some self test errors in the C range after start up. The self test of this unit is really complex. I am thinking about making a video.

Tam

[cncjerry]

I had a Fali2 as well, can't remember what it was, it was a simple fix though.  That unit is very, very picky about the either the shorting straps or test leads.  You must have a good 4-wire connection.  Actually your tweezer setup looks pretty cool.  Thinking back, as I'll never have a test set either, I should have put BNCs on it when I replaced the banana jacks as that would have provided a shield.

I have the comparator on mine, it works well, I've used it for sorting and it is pretty handy.  You can set it to 1/10th a picofarad.  There is a asian guy on ebay that is selling the GPIB cable and what looks to be the adapter card.  I was thinking of grabbing it to have a fully configured box but then again, I doubt I would ever use it.

Let me know if you need any voltages tested.  I don't use mine much so if you need to swap some cards for trouble shooting we might be able to work something out.

[TiN]

Worked a bit on this forgotted box. Got cables are replaced, so since -12V was faulty, time to get on power supply repair.
It's simple linear regulator, so after checking power transistors Q2 and Q3 (which were OK) only few parts remain to troubleshoot.

Replaced electrolytic capacitors C1,C2,C3,C4,C5 and U3 opamp (used far too good Linear LT1007 here, since usually don't have much stock for DIP8 opamps).
As result negative -12V rail powered up fine, just like new.

Schematics for those thinking at home:



Now front panel is responsive and LED displays are working. I think some buttons still flaky as it does not reliably react to key presses/sometimes lockes up.

[dietert1]

Today i fixed one of our HP 4262A LCR meters. About half a year ago it started crashing during measurements and some days ago it would no longer run for more than one minute. I observed voltage drops on the +5 V supply. Error vanished after resoldering the six parts on the power supply board used for +5 V regulation (3R, 2T, 1OpAmp). Before i had already reseated the two diodes and the power transistors in their sockets. The unit has been running now for more than 10 hours without any problem.
It was made in 1977 and needs attention every four or five years. In 2017 i replaced the old capacitor C1 0.22uF after it exploded in another unit.

Regards, Dieter

[dietert1]

Since someone mentioned bit rot above, in an act of preventive maintenance i also replaced the two 2316 ROMs with an AT29C256 flash part we had. First i made an adapter to read the original ROMs and to verify the hex files available at k04bb. Then i made another adapter as shown in the image. No mods on processor module.
Processor module runs much cooler with the flash part. The nanoprocessor also got a heatsink some years ago.

Regards, Dieter