HP 3440a digital voltmeter repair/troubleshooting: moving on to PSU ripple

[valley001]

New Q3 installed now looks like I have proper wave form per 5-11 B at A3 (19). 

Also:

Ramp looks good per 5-11 D

5-11E looks like the correct shape but it is -1v to -13.2v where if I am interpreting 5-11 correctly it should be 0v to -12ish volts.  EDIT: If I switch to AC coupling it suddenly looks correct?

EDIT again:  I have not been paying a lot of attention to the pulse timing and that it matches, is this important?  For example, I get a mostly accurate shape of 5-11E with the scope set to 200us, but if I am correct is should be showing at .1 msecs (100us right?)

[Armadillo]

Both the Volts/Div and Time/Div settings should be same as the 5-11.

When you set in DC mode, firstly set "oscilloscope" Channel input to GND and adjust the horizontal line to be same as the 5-11, then adjust the Channel input back to the Volts/Div in accordance to 5-11. Then do the measurement, so the shape and the time should be the same, almost as the figures in 5-11.

You should be alright there.

EDIT: But for functional troubleshooting now, as long as there are pulses, you should be good to proceed to other troubles. So what are the issues next at A2?

[valley001]

OK, A2 wave forms per 5-11:

5-11 F A2TP1- No wave form that I can find.

5-11 G A2TP2- Pulse present but does not match exactly. 

5-11 H A2Q7 Collector- Nothing, no wave form that I can see.

5-11 J A2Q9 Collector- Nothing, no wave form -15v or so.

5-11 K A2TP3- Wave present but distorted.

5-11 L A2TP4- Wave present looks good.

5-11 M A2Q15 Collector- Square wave -25 to +1v or so.

[Armadillo]

Let's approach it one by one.

Depress S3, and confirm that J7 [pin at R1], there is voltage coming in when you press S3.

Take out C4 and check it outside of circuit to make sure it is OK because C4 is important part.

then;

Install or Replace C4 back and measure BCE voltage at Q1 to Q5.

That will be quite a handful for now.

Edit: Also while you checking C4, also check C3 and C13.

Check cathode of CR7 that the Ramp Waveform is coming in.

[valley001]

Depressing S3 yields -7.9ish volts at J7 (referencing power supply ground).  Releasing S3 I get voltage readings -5ish volts fluctuating. 

That's all I could do this morning.  Will check the caps and CR7 tonight, should have C2 Friday will replace and test Q1-5.

[Armadillo]

Good, Sign of Life at least.   
No hurry, relax.   

[Edit: Releasing S3 and there is a readout does "suggest" that CR5 is not doing its comparison job and Q1-Q5 did fired over the single pulse created by the releasing action. Cross-Finger On CR5 and CR27, Dynasaur age device, I am sure its hard to find.]

Priority Changed:

Check cathode of CR7 that the Ramp Waveform is coming in.
Check cathode of CR5A anode that the Ramp Waveform is coming in.
Check C4.

[valley001]

OK, meeting was cancelled so I get to have some more fun   

Cathode CR7 I see the ramp waveform
Cathode CR5A I see the ramp waveform
Anode of CR5A I see the ramp waveform
For fun I checked: Anode of CR5B  I see a sawtooth ish or "pointed square" wave form.  -.8v to -4.4v.

EDIT: Also Checked CR27

Cathode- Rampish wave
Anode- Nothing .12 to -.040v
Anode- Ramp wave form

[Armadillo]

Please take out C4 and check carefully.

Because.;

If C4 don't pulse, TP1 see nothing.
C4 must pass the CR5B anode pointed square wave thru to the base of Q2 to be amplified by the trains of transistors.

[valley001]

OK, I lifted the leg opposite the diodes on C4, component tester results:


Vloss 0.0%
20.5nf  

It does not give an ESR reading.  Do I need to remove it completely to get proper readings?  Just worried about heating the diode leads. 

I also tested it with an analog ohm meter set to 1K, no deflection of the meter.  I don't have further testing abilities unfortunately.  I do have a supply of .022UF 630v 5%  metalized polypropylene film capacitors, perhaps I should sub one in? 

[Armadillo]

Let assume C4 is good, so you should get wave at collector of Q1, collector of Q3, Q4 and TP1.
If you can't get then all upstreams BCE, we need the voltage measurement.
Any wave get through any of them, we are happy, because then we eliminated the CR5 from the list.
If non get through to Q1, then we only hope C4 by replace it and try again until lastly - CR5.
I think that should be the safe procedure to approach.

[valley001]

Ill check those points.

Just now, for fun,  I took wave forms on either side of C4, on the diode side I get a wave, on the other side I get no pulse.  Does this fact tell me C4 is bad?

[valley001]

Q1 C - Nothing

Q3 C - I get a sharp pointed wave, about 5v pk-pk.

Q4 C - More or less the same as above, sharp pointed wave.

TP1- Nothing

[Armadillo]

Can you then measure BCE voltage of Q5, let hope it's defective there.

Press "Auto Setup" depending on your scope, in case you don't see anything cause the pulse may be sharp and short.

EDIT: USE AC MODE of the scope to see pulse.

[valley001]

Q5

C -.760 to -.700
B sharp point wave -.480 to 5.5v.
E -.760 to -740

Q5 scope on AC mode

C -.040 to -.080
B  nothing -.040 to .080v
E -.040 to -.080

[Armadillo]

My apology;

I mean use AC mode to see the pulse;

But I still need the DC voltage at BCE to know if the Transistor is Dead. You can measure by voltmeter hook to it instead of scope if that's easier for you.

Cause only with the DC voltage, we can determine if the Transistor OR external bias problem.

[valley001]

Q5- Multi meter on DC

C -.737
B -.326
E -.736

I definitely get a wave on the base with the scope set to DC mode, if I switch to AC it disappears. 

[6PTsocket]

Hi all,

I am new here, been working on repairing this old HP volt meter.  I also have a thread going on antique radio forum. 

Short story on this meter, it is a parts meter I bought to repair my once (and now again) working HP3440A, that I damaged while adjusting the -8.000v reference (screwdriver slipped and touched the back of A2  ).  The meter then stopped functioning.  I used A2 from this parts unit and installed it in my good meter and it is now functional, so I know the problem is in A2, but there are other problems with the parts unit.  I would like to get it going if possible. 

I have verified the voltages out of the PSU per 5-17 in the manual.

I have verified 400hz out of the oscillator per 5-18.

I do not get a reset pulse per 5-32.

I do not get any waveform out of A2TP1.

I do not get a "narrow positive pulse" out of A2CR5B on my scope (I am also not certain what a "narrow positive pulse" is but I get nothing on the scope.

Voltage readings per table 5-6:

A2Q1: -.672
A2Q2: -1.33
A2Q3: -.519
A2Q4: -.321
A2Q5: -.734

A2Q11: -.378
A2Q12: -.832
A2Q13: -20.5

This is the extent of my testing so far, and has revealed some problems.  A2Q1-5, all voltages are low.  I removed Q4 and 5 and tested them out of circuit with the diode check function on my DMM and they seem fine.  I wonder if some other issue might cause all the voltages to be low on these transistors? A2Q11-13 seem to read mostly in spec. 

I am interested in ideas on where to go next. Am I at the point of just subbing in transistors and see if things change?  Is there a more thoughtful way to approach these problems?

Just as a side note, a bipolar transitor "diode" test can read good b-c and b-e  and test shorted c-e which should read high both ways. This is probably a more common failure with power transistors. A cheap beta tester is probably more useful. Even the Harbor Freight freebee meter has this function or better yet one of those ATmega 328 testers that can be had for a few bucks.

Sent from my SM-G900V using Tapatalk

[Armadillo]

This warrant that Q5 be taken out to check it. Suspect defective. Expect Collector to be -31V [somewhere]

So, if you take out Q5, the R25 leg should read -31 ~ 35, right?

Component Test Q5.

Q5- Multi meter on DC

C -.737
B -.326
E -.736

[valley001]

No voltage at R25 testing with my multimeter on DC, with Q5 removed.

Component tester calls Q5 "two diodes .365v on both sides.

[Armadillo]

Replace Q5.

But you need to check R25 is 6800 ohms and why -35 power rail is or not available to it. Confirm power rail problem or R25 problem.

You can then proceed to Ground comparator and Ground Amplifier.

Same deal as before.

I will be out for a while. Let's communicate 2 hours later. It's getting interesting and close.

 

[valley001]

R25 should be 220K no?  Did you mean R26? I do get -36v/-38v there.

Installing replacement Q5

[Armadillo]

Oh my mistake.... can't see clearly if the writing is R25 or R26, is connected to the collector of Q5 to -35V rail.

R25 is 220 ohms not 220K ohms.

OK great to hear that R26 is not fried.

After Q5 installed, please measure TP2 as well.

[valley001]

UPDATE:    Replaced Q5 and now displays are counting normally, settling at .006v reading.  Will check test points now and see what things look like. 

[Armadillo]

WOW! man that's GREAT news. 

I thought it's going to drag with further multiple faults.... to TEST OUR resolves at troubleshooting..... heheheheehe and actually a bit voodoo feeling with my big mouth with that "YOO, WE ARE ABLE HERE" [see beginning post]... LOL...

And to think that the dinasaur diodes actually "SURVIVE"!.

 

[valley001]

I could not have done it without your help and another on the forum who helped me procure parts.  This place is super cool.

The funny thing is I DID find those two matched pair diodes, im going to buy them anyway.

A2TP2 shows a pulse but it does not match 5-11 G, could still be me being green at using the scope, or I need more than 70mhz to see it. 

I still need to replace C2, arriving today. 

Then I can attempt calibration.