HP 3465A DMM repair

[bd139]

The voltage is clamped at +/- 5V by the opposing circuit so it will never hit the 6v breakdown voltage, thus the B-E junction is fine.

(checked with LTspice)


Edit: this is actually wrong. I am clearly drunk driving LTspice at the moment. There is 9v across the junction 

JFET it is!

[Zoli]

The voltage is clamped at +/- 5V by the opposing circuit so it will never hit the 6v breakdown voltage, thus the B-E junction is fine.

(checked with LTspice)


Edit: this is actually wrong. I am clearly drunk driving LTspice at the moment. There is 9v across the junction 

JFET it is!

Sorry to ask, but CR10/11 isn't 3.9V Zener? Better check in reality, because I wouldn't trust the schematics...

[bd139]

CR10/11 are 3.92v zeners. CR8/9 are low leakage diodes. So the zeners set the clamp constraint plus one diode drop across the low leakage diodes.

[Zoli]

CR10/11 are 3.92v zeners. CR8/9 are low leakage diodes. So the zeners set the clamp constraint plus one diode drop across the low leakage diodes.

OK so (3.9+0.6)*2=9V; now I understand.

[Caliaxy]

CR10/11 are 3.92v zeners. CR8/9 are low leakage diodes. So the zeners set the clamp constraint plus one diode drop across the low leakage diodes.

Dat scan... The one in MrYakimovYA's link is much better. What's the expected voltage range in the node pointed by the arrow "See Note 5" during normal operation? What happens if you apply there, say, +3V?

[bd139]

It’s also a different revision of the DMM. I think that part of the front end is the same though.

Range from my LTspice sim applying +/-1000V to the left of the 100k resistor is +/-5v at the clamp point.

[Caliaxy]

Range from my LTspice sim applying +/-1000V to the left of the 100k resistor is +/-5v at the clamp point.

Right, but for applied voltages that result in more than 2.1V (6-3.9V) and less than (3.9V + 2N3904 B-E drop) at the clamping point (which might or not might not be within the intended operating range, I don't know), the other B-E junction will clamp (or try to...) by breaking down due to reverse bias >6V. This might work fine for a while, but it's not ideal (and would make the B-C junction a safer bet). I believe this is the main point bdunham7 was making above (Reply #20).

[MrYakimovYA]

There are also afforable low leakage diodes BAV199 available in a SMD case. Not tested for super low current, but typical leakage current < 10 pA.

I looked over a couple of datasheets. It's strange for me but there aren't any detailed information of the leakage current. Either a typical value (3 pA) or max. value (80 nA at reverse voltage 80 V). It's very strange inspite of BAV199 is specified as low-leakage diode set... So I guess they must be sorted and tested before installation in equipment...

[MrYakimovYA]

It’s also a different revision of the DMM.

I just forget about a lot of revisions, sorry

[Robert763]

If only it was that easy. The revision of the DMM I have is not covered by that manual and the component placement and front end are slightly different.

The only copy is the Keysight one and that's awful quality scan:

Wow crap scan indeed, can't even see which one is CR9 ! 

Looks like the one on the up side of the input path, going to the negative 7V rail.... OK... so you were wrong, your offset repair DOES pertain to my old Rochar Nixie DVM, as this is general purpose stuff and my meter has something similar a s well

To be pedantic these HP (and a lot of US military document digital archive documents) are not bad scans, they are scans of bad microfilm / microfiche "originals". These were typically done to comply with archive legal requirements by third party companies who didn't really care about quality.

[xrunner]

Looks like the adjustment procedure (Cal) is pretty easy. Power supply uses jumpers to adjust the 10 V reference. Says you need 1k & 10 M 1% resistors, and an AC low voltage source "standard" up to 20 kHz.

Of course you can compare its readings to another reliable DMM too if you have one - I think you do ... 

[bd139]

Yes it's fairly easy. Actually I checked the 10V cal against the 34401A and it's showing 10.00012 which is surprisingly good. Either that or both are surprisingly wrong in the same direction. Probably should get the 34401A calibrated as last cal was with HP in 1999.

AC voltage standard is going to be an AWG with a comparative measurement. The resistors I have and didn't manage to throw away in the grand cock up. A few years back I bought a full set of Dale 0.1% resistors from 10 ohms to 10M in magnitude steps just for this scenario

[BU508A]

Yes it's fairly easy. Actually I checked the 10V cal against the 34401A and it's showing 10.00012 which is surprisingly good. Either that or both are surprisingly wrong in the same direction. Probably should get the 34401A calibrated as last cal was with HP in 1999.

AC voltage standard is going to be an AWG with a comparative measurement. The resistors I have and didn't manage to throw away in the grand cock up. A few years back I bought a full set of Dale 0.1% resistors from 10 ohms to 10M in magnitude steps just for this scenario

And you didn't throw them away by accident? 

[bd139]

No they were in the "really important stuff" box 

I needed more of those boxes 

[xrunner]

No they were in the "really important stuff" box 

I needed more of those boxes 

Yea that's the key - label all important boxes with "really important stuff" so you don't lose them. Trouble is I forgot where my labeller is! 

[AVGresponding]

No they were in the "really important stuff" box 

I needed more of those boxes 

Yea that's the key - label all important boxes with "really important stuff" so you don't lose them. Trouble is I forgot where my labeller is! 

Obviously you should have put it in a box labelled "really important stuff"...

[bd139]

Ordered a 2N4117A JFET for £3.50 from Langrex. Will install and validate again.

[tggzzz]

AC voltage standard is going to be an AWG with a comparative measurement.

Cheat: with a 0.1Hz signal, use the 34401 to measure min and max DC voltages. Easier with a 34410, though.

Then presume the AWG will have the same peak output voltage at 400Hz.

The resistors I have and didn't manage to throw away in the grand cock up. A few years back I bought a full set of Dale 0.1% resistors from 10 ohms to 10M in magnitude steps just for this scenario

I'm just about to buy some 100/1k/10k 0.1% resistors, to replace the o/c wirewound (silk covered enamelled wire on a PCB bobbin!) resistors in a Muirhead decade resistance box.

[bd139]

Ok 3465A is now fixed. 2N4117A subbed in as a diode connected FET as suggested by both HPAK and EEVblog forum members. Works perfectly, clamp works too.





Did DC and ohms cal. I don't have any test cables (  ) to do the AC cal yet.

Thanks all for suggestions and support 

[MrYakimovYA]

Ok 3465A is now fixed.

Oh, my congratulations! You've done a great work! I'm glad you have this nice DMM in working state now!