HP34401A Ohms function problem

[WaveyDipole]

I recently acquired a HP34401A with a failed Ohms function. The DC and AC voltage and current functions work fine. The ohms function shows approximately 3 ohms when the probes are shorted, but resistance readings are all over the place. I researched the problem online and on this site and found that the usual culprits are either the COTO9947 relay contacts or the AD706 op amp in the Ohms current source circuit. The relay is good and I have just carefully replaced the AD706, but the problem persists.

The usual reports of an AD706 failure indicate that there is no current present on the terminals when Ohms mode is selected. In this case, there is approximately 10ma present rather than the expected 1ma, both before and after the replacement of the AD706 so clearly something is still wrong with the current source.

Voltages around the AD706 with reference to AGND (pin 4) terminals open / shorted are:

1.  4.09  / 4.07
2.  6.92  / 6.92
3.  6.92  / 6.92
4.  0 / 0
5.  8.28  / 8.76
6.  7.65  / 8.78
7.  13.25 / 12.44
8.  18.88 / 18.88

The 18V line looks a bit high but other than that I am not sure where to look next and any insights would be appreciated. There is some suggestion on one thread that the problem might be U101 (custom HP chip 15K6-0001), but how do I determine this?

[Dr. Frank]

Hello,
you certainly refer to e.g. this thread here: https://www.eevblog.com/forum/repair/hp-34401a-error-612-613-615-617-618-619-621/msg489687/#msg489687

I you don't mind, please get a copy of the schematic, measure all these voltages indicated in the diagram, write them down and copy this here.

Your description of the test current is not clear, i.e. is it faulty, or not? Do you use 2W or 4W Ohm, and what's the difference in behavior?

Please chose 4W Ohm mode and measure all different test currents in each range with another DMM, from 100 Ohm up to 10MOhm. The nominal values are listed in the service manual or in the specification.

In first place, I would not suggest that U101 is faulty, there are some other components around which might cause trouble, more probably.

Frank

[Kleinstein]

It looks like there is more current flown than supposed too. So the first suspect would be Q202, as this is the transistor (P-JFET) to control the current. With some 12-13 V from the AD706 the JFET should be essentially tured off all the way. So the voltages around Q202 would be interesting.

[WaveyDipole]

Hello,
you certainly refer to e.g. this thread here: https://www.eevblog.com/forum/repair/hp-34401a-error-612-613-615-617-618-619-621/msg489687/#msg489687

Thank you for that link. I hadn't come across that particular thread yet.

I you don't mind, please get a copy of the schematic, measure all these voltages indicated in the diagram, write them down and copy this here.

Ok, I will work on that and post later.

Your description of the test current is not clear, i.e. is it faulty, or not? Do you use 2W or 4W Ohm, and what's the difference in behavior?

Please chose 4W Ohm mode and measure all different test currents in each range with another DMM, from 100 Ohm up to 10MOhm. The nominal values are listed in the service manual or in the specification.

I have measured on both 2W and 4W settings and the results are approximately the same.

Auto   7.861  mA
100Ω   7.924 mA
1kΩ      8.516 mA
10kΩ   8.541 mA
100kΩ   8.494 mA
1MΩ      8.490 mA
10MΩ   8.509 mA
100MΩ   8.507 mA

The current readings are approximate as the current starts high and gradually sinks by a few 10ths of uA.
The following errors are reported:

+612,"Ohms 500 nA source failed"
+613,"Ohms 5 uA source failed"
+615,"Ohms 10 uA source failed"'
+617,"Ohms 100 uA source failed"
+618,"DC high voltage attenuator failed"
+619,"Ohms 1 mA source failed"

In first place, I would not suggest that U101 is faulty, there are some other components around which might cause trouble, more probably.

Frank

I am certainly hoping that the problem is elsewhere and not U101....

It looks like there is more current flown than supposed too. So the first suspect would be Q202, as this is the transistor (P-JFET) to control the current. With some 12-13 V from the AD706 the JFET should be essentially tured off all the way. So the voltages around Q202 would be interesting.

Yes, the current is much higher than it should be on all ranges. The voltages around Q202 measured with reference to AGND are:

Q202 (01H - MMBF5460 - P-Channel general purpose amplifier)
G   12.80
D   8.15
S   4.38

So yes, there is some 12-13V from the AD706 but evidently the FET is not turned off. The 18V feed is present at R290. There is a 0.78V drop across the diode and the R290 tests good. Its a P-Channel FET so as you point out, a positive voltage at the gate should turn it off. Looks like Q202 ought to be the next candidate for replacement.

[Kleinstein]

The gate voltage at Q202 still looks quite low. It looks like the current from R290 frows towards the gate. The zener diode should not conduct much with a low voltage, at least not to lower the voltage.

Before odering new part, I would also check Q203-Q210 in circuit. There is a chance they can fail from too high external voltage.  This should not cause the extra current from the ohms source, but it could be a path on how to damage Q202.

[Dr. Frank]

Hello,

these 8mA can only be delivered by R290 and U201-B and only if Q202 is damaged, i.e. the Gate is conducting towards its Drain.
Maximum current delivered by U101-E, from the 5k resistor would be 3.6mA.

Therefore Q202 is damaged probably by overvoltage, so you should check the HV protection transistor array Q203 .. Q210, Q211 and CR202 as well.

Frank

[Keith956]

I would check the +7 REF is OK, if that's out then the whole ohms current source will be wrong.

[WaveyDipole]

The reference voltage is 6.927V so pretty close to 7V. Not sure how far out it would need to be before obvious problem show. Voltage and current readings are spot on.

I checked the HV protection transistor array. Q203, Q204 and Q210 have at least on junction with resistance across it in revered bias mode. The remaining ones test good. Resistors R203 to R206 appear to be fine. Diode CR202 tests good. I am uncertain about the Zenner DR201A. Q211 is impossible to test in-circuit. I have been trying to work out the purpose of that arrangement?

A short, high voltage event would seem to be a reasonable explanation.

I will be ordering some replacement parts in due course.

[Kleinstein]

CR202 acts as a protection against a high positive voltage when in Ohms mode. That is the easy part.
Q203-Q210 arct a bit like a high voltage darlington. R203-R206 divide the votlage approximately equal over the transistors (actually not very accurate as the base current is neglected). Under normal conditions (no negative external votage) Q211 does no conduct and the votlage drop over the transistor string is relatively small (~1-2 V).
With a larger neg votlage at the terminal Q211 will start conducting and will limit the votlage for the emitter of Q203 to some +0.5 V or so. So the actual current source aroudn Q202 will not see the high votlage.

[WaveyDipole]

Q202 (MMBF5460) is proving rather difficult to source and showing as no longer manufactured at main suppliers such as Farnell. The only UK source, Littlediode, charges an extortionate £5 + postage for a single unit - and I rather suspect they are just re-packaging from China (generic graphic, no brand etc)! I am not sure whether to trust the cheap ones from China on eBay. I saw some branded NOS from Israel which would cost £13 to obtain with postage and + unknown tax, so still not cheap, but at least it is 10x rather than just one piece.

Has anyone successfully tried any alternatives? The IDSS current of 1mA min and 5mA max is proving impossible to match. How important is this? I have seen situations e.g. in oscilloscope inputs where the IDSS can be quite critical.

[Dr. Frank]

Sorry, no, haven't tried an alternative.. maybe I need to do that in a few days. 
 
The 34401A and the identical 34970A schematics both indicate a 2N5461, sel., i.e. I_DSS  = -2 .. -9mA.
I assume, that only the minimum current parameter is crucial, as at least 1mA is required for the test current. I would estimate, that hp selected them for 3mA at least.

Maybe it can be replaced by a  p-channel MOSFET, as this is drawn alternatively in the 34970A schematic.

Frank

PS: Maybe the J175 / J176 could be a direct replacement, as these are in series production  and cheap.

[Kleinstein]

The OP can provide quite a large voltage range. So this should also work with a MOSFETs.
A much larger JFET may compromise the protection when the supply is off. However normally then the relay should also be diconnected.
I also don't see an obvious reason why a different p-JFET like J176 should not work. It is larger current, and thus a little more leakage, but the 34401 is not using such a small Ohms test current.

[WaveyDipole]

I am going to try a J177 that I have been able to source (pmbfj177, 1.5mA to 20mA). The MOSFET is an interesting observation as well which I will keep in mind should the jfet not work out.

Does it make sense to replace all eight transistors in the Q203-Q210 array, or just the faulty ones? My thinking is that although the remaining ones have not failed, they will have been stressed. I have had to order 10-off anyway, so it makes little difference how many I use.

[Kleinstein]

Chances are the others are still OK. With these transistors it is not about leakage or noise. Failure was likely from exceeding the SOA from high votlage and than more thermal.
So chances are the other are OK and tested.

[Dr. Frank]

Hello,
today I got my 'new' 34970A on the desk, similar Ohm mode fault.
The AD706 very probably is broken, as the voltages on the 2nd OpAmp pin 5, 6, 7 are similar to the defective 34401A here: https://www.eevblog.com/forum/repair/hp-34401a-error-612-613-615-617-618-619-621/msg489687/#msg489687. The transistors of the +/- inputs are obviously blown. Characteristically, the +input is on a lower potential than the -input, but the output on pin 7 is positive, instead of being negative.
As this is the first instrument ('401A or '970A) which I have to repair myself, I found out that the defective inputs obviously drive or draw excessive currents, because I measured an excessive current out of pins 33, 35, 36, 39 of U101-E, which could only be induced via pin 40.
This aspect was new to me.

Your replaced AD706 seems to be fine, as the voltages on pins 5, 6 are nearly identical, and the output seems also to be reasonable. Therefore, the j-FET seems most probable to be the root cause, as a few of the Q203 .. Q210 were also damaged.
But anyhow, according to Murphys law, I propose that you should de-solder the j-Fet quite carefully so to be able to reuse it, in case your already replaced AD706 is also defective, and might show a different defect characteristics at the moment, compared to the usual ones, as described above.

Frank

[WaveyDipole]

The FET has been replaced, as have the three faulty transistors in the array.
There is now no current whatsoever between LO and the output from the array.

The array tests good except that Q209 seems to be showing about 2.7 volts across B-C in the reverse direction on diode test when all other transistors show infinity on both junctions. Transistors Q203, Q204 and Q209 were replaced as they showed anomalous readings. Q210 has also now been replaced for good measure so I am puzzled by this reading.

Voltage readings are as follows:

AD706:-
1.  4.08
2.  6.92
3.  6.92
4.  0
5.  8.774
6.  8.800
7.  2.862
8.  18.88

Q202 (01H - MMBF5460 - P-Channel general purpose amplifier)
G   7.568
D   8.284 0
S   0 8.284

Voltage across the Zenner CR201-A is exactly 4.70V.

Full voltage reading details are contained in the attached image.

Since there is now no output from the FET it would whereas there was an excessive output previously would seem to indicate that the original FET was faulty, but wouldn't there have to be a negative output from U101 to turn it on and produce current.

The ohms reading now fluctuates randomly unless I short the leads when it shows something like 0.2ohm.

This now appears to be the classic AD706 failure mode, although that IC has already been replaced. It is likely to have failed again?

[Dr. Frank]

Hi,
it may be really helpful, if you would measure the voltages and write them into the schematic , like in the other thread. It's quite confusing to analyze this way.

Also, you obviously swap the measurements of Drain and Source voltages of this FET, as Source goes to the ASIC, at about 8..9V  and Drain is the output and should have a low voltage.
pin5 and 6 should be identical, if the circuit works correctly, but pin7 seems to be about the correct output voltage to drive the FET open, as its Gate is at about 8V.

Maybe the transistor array is still not working correctly.

You can check the current source by an external DMM, current range, if you connect minus to the low input of your 34401A (4W Ohm mode), and the current plus in mA range directly to the Drain of the FET, i.e. the output of the FET, pin2 and Q210, pin2.

Frank

[WaveyDipole]

Thanks. I was working on it after the last post. I have attached the result.

I checked the output current of the FET by directly measuring it from the drain as suggested but am still seeing no output.

[Dr. Frank]

Have you tried another range?
the new FET seems not to work properly, as its Gate voltage is too low, in conducting mode.
have you selected the FET somehow, for minimum 2..3mA for example?
the current seems to be sunk by the opamp, instead of output to the jacks

[Kleinstein]

The JFETs seems to be not working correnct - maybe broken or wrong pin connection (as D and S are essentially symmetric it would be that gate that would be wrong).

The voltages for the OP make sense, so likely no porblem there.

[WaveyDipole]

Thanks.

Have you tried another range?
the new FET seems not to work properly, as its Gate voltage is too low, in conducting mode.
have you selected the FET somehow, for minimum 2..3mA for example?
the current seems to be sunk by the opamp, instead of output to the jacks

If you mean the different resistance ranges on the meter, then yes, the measurements are with the DMM set to Auto, but I did try all of the manual ranges with the same result.

The J177 is in the same line as J175 and J176, just has a minimum IDSS down to -1.5mA whereas the minimum for J176 is -2mA and J175 is -7mA. I have made no selection other than to choose the J177 because it goes to -1.5mA. I had to purchase a minimum quantity of 10 so I have spares available and could try another one from the pack?

BTW, Dr. Frank, does the 34970A manual list the alternative p-channel MOSFET part number please?

The JFETs seems to be not working correnct - maybe broken or wrong pin connection (as D and S are essentially symmetric it would be that gate that would be wrong).

I did check to make sure that the pinout of the J177 is the same as that if the original part.

The voltages for the OP make sense, so likely no porblem there.

That is good to note.

[Dr. Frank]

unfortunately,  no alternative part mentioned.
this CC circuit might have a parasitic modus operandi, which will depend on the jFET characteristics. maybe you select your FETs for different pinch off and zero GS current
Frank

[Kleinstein]

The source voltage should be at some 9 or 12 V depending on the range for the current source.
So a depeletion JFET would need a gate votlage from some 15 to about 9 V.
An emhancement MOSFET should need some 6-11 V.

With the current zener diode the OP should be able to work from about 7 V to 17 V.  With MOSFET one may have to change the zener to a slightly lower voltage one.

There are not that many small THT p-MOSFETs  to choose from. I don't think the parameters should be that critical. Maybe avoid a modern type with low limit on the gate voltage and possibly a little more gate leakage. A rather high high threshold could be a problem to.

[WaveyDipole]

I think I rather overlooked Q211. I figured since source and drain where shorted it was impossible to test in circuit and didn't check between gate and source/drain. This measures 80Ω in both directions on resistance and 0.08V on diode test which presumably cannot be correct. A leak/short would pull Q204 base to ground which would presumably be enough to shut down the array? Initially I will just remove Q211 to see whether that makes a difference but I don't think this would explain the absence of current from Q202? Still, its another item to sort out. I do have some J390's to hand but the operating voltage is 25V rather than 40V for the MMBF4117.

I will have to give the Q202 FET issue some thought and maybe order those NOS MMBT5460s from Israel. At least then we know its the correct part.

[Kleinstein]

Q211 is uses as a low leakage diode.  It does not absultely have to be a JFET.  A low leakage diode would work as well. They just selected the 2N4117 because it is sopposed to be tested for low leakage. A different FET may not be tested as well.  For a substitute one could consider the collector - base diode of the jmall BJT like 2N3904 or BC548.

For a curde test to see if the rest is OK one could also consider a PNP transistor instead of Q202: it should also operate the current soruce, just not very accurate.