Agilent 34401A Bubble on PCB ... can it be saved ? (SOLVED)

[chilternview]

Lets recap, you say DCV is OK, and DCI appears to be so, but continuity is flaky and ohms not at all?
And self test gives errors 621 , 619 , 619 , 617, 615, 613 ,612 (you've got two 619s, I guess a typo) - which are AC RMS failing full scale, and all ohms current sources failing? Any more error numbers?

That tells you there are two problem areas:
- ohms after U101/2. So check Q202-211 and U201,CR201/202 (the +7V ref must be OK if DCV reads all OK).
- ACV. Could be anything on the AC schematic page 9-11, but you can apply an AC signal and check with a scope its following the circuit path.

It is possible someone's been in there and tried to repair it and not succeeded/given up, so the visual damage may have been caused by the initial fault and has been fixed, or by someone trying to fix the problem.

[The_Spectrum.A_idiot]

if you have them   that's ok

the worst from the bubble would be vias, they could have snapped open, and be very careful not to overheat parts before removal


i've seen some,  drilling an hole where there is no parts and filling the space with superglue  while giving some pressure on both sides, to bring back layers together and slimming down the bubble expansion  ... but not sure if i would try this,  it does seem recoverable, and it could continue to work fine

patience and be meticulous ... you'll get it repaired

i hope the special asic has not been damaged  ....

I've checked all of them with my Multimeter they seem ok.

if I missed any we will find out soon enough.

Now I need to find out a way to test the rest

I have this but I have no idea under what conditions the system is being tested. 10Mohm in the input ? 

[The_Spectrum.A_idiot]

Lets recap, you say DCV is OK, and DCI appears to be so, but continuity is flaky and ohms not at all?
And self test gives errors 621 , 619 , 619 , 617, 615, 613 ,612 (you've got two 619s, I guess a typo) - which are AC RMS failing full scale, and all ohms current sources failing? Any more error numbers?

That tells you there are two problem areas:
- ohms after U101/2. So check Q202-211 and U201,CR201/202 (the +7V ref must be OK if DCV reads all OK).
- ACV. Could be anything on the AC schematic page 9-11, but you can apply an AC signal and check with a scope its following the circuit path.

It is possible someone's been in there and tried to repair it and not succeeded/given up, so the visual damage may have been caused by the initial fault and has been fixed, or by someone trying to fix the problem.

Thank you very much for your time and support.
Yes you are right there is a typo.
errors 621 , 619 , **618** , 617, 615, 613 ,612

next how should I test said components properly ?
Check Q202-211 and U201,CR201/202 ?

Are there some expected values ? Do I have to desolder them ?


Thanks

[Kleinstein]

One can do a first test in circuit. A common failure mode for diodes and transistors is a short. The DMM diode test usually works OK for this and should also work in circuit here.

Form the meaured voltages things are not totally off. There may still be an issue with the protection, especially the diode used JFET.
It would make sense to measure the test currents to see which ranges are effected. Ideally one would measure the current directly with just a 2nd DMM and also with something like a LED or 2 or 3 diodes in series to see if it changes with external voltage.

[Dr. Frank]

Hello Unknown,
it's a pity that some people don't even reveal their first name. I don't like to cite SUCH alias names, in first place 

From a closeup of the damaged PCB, I strongly assume, that somebody tried to repair the instrument, using a hot air gun around U110, or the Ohm protection transistors Q203 - Q210, either to replace or re-insert them.
That guy used too much heat, also too long, and did not cover the other components, so creating this PCB delamination.
It's not clearly visible, but all components inside  the area of delamination might have been damaged by the heat, at least all the solder joints are under suspect.
Please make a macro photo exactly vertical above this area, so that every component and solder joint is sharp, and better illuminated, maybe from different sides.

The transistors  have been soldered very badly, and I see solder balls. The transistors might be defective, or there might be a short to GND as well , via Q211 which should be tested in situ.
If you select 4W Ohm, do you measure the appropriate test currents out of + Input and - Input for the different (manually selected) ranges?

You have to de-solder these transistors anyhow, so you can test them individually, later on.

W/o the transistors back in place, you might then solder a short from Q202 to CR202 temporarily, to make further tests on the Ohm mode.
Descriptions how to do that, has been written down elsewhere .. I guess you found that thread already, as you copied the OHM schematic here.

I guess, that U110, that is used for current mode protection, has been replaced, but that mode should work properly, and has nothing to do with your OHM problem. Anyhow, many of the solder joints look very bad. U304 also looks bad.

We will definitely continue analysis, when you have presented further results.

Frank
 

[The_Spectrum.A_idiot]

Hello Unknown,
it's a pity that some people don't even reveal their first name. I don't like to cite SUCH alias names, in first place 

From a closeup of the damaged PCB, I strongly assume, that somebody tried to repair the instrument, using a hot air gun around U110, or the Ohm protection transistors Q203 - Q210, either to replace or re-insert them.
That guy used too much heat, also too long, and did not cover the other components, so creating this PCB delamination.
It's not clearly visible, but all components inside  the area of delamination might have been damaged by the heat, at least all the solder joints are under suspect.
Please make a macro photo exactly vertical above this area, so that every component and solder joint is sharp, and better illuminated, maybe from different sides.

The transistors  have been soldered very badly, and I see solder balls. The transistors might be defective, or there might be a short to GND as well , via Q211 which should be tested in situ.
If you select 4W Ohm, do you measure the appropriate test currents out of + Input and - Input for the different (manually selected) ranges?

You have to de-solder these transistors anyhow, so you can test them individually, later on.

W/o the transistors back in place, you might then solder a short from Q202 to CR202 temporarily, to make further tests on the Ohm mode.
Descriptions how to do that, has been written down elsewhere .. I guess you found that thread already, as you copied the OHM schematic here.

I guess, that U110, that is used for current mode protection, has been replaced, but that mode should work properly, and has nothing to do with your OHM problem. Anyhow, many of the solder joints look very bad. U304 also looks bad.

We will definitely continue analysis, when you have presented further results.

Frank
 

Dear Dr. Frank ,

Regarding the screen name is a long story but it showed my superficial and lack of deep knowledge of the subject.

I took my time with the pictures and its WAY WORSE than I ever imagined.

please have a look and we will iron out the details later.

[The_Spectrum.A_idiot]

More pics

[The_Spectrum.A_idiot]

more

[The_Spectrum.A_idiot]

more

[The_Spectrum.A_idiot]

last batch I promise

[The_Spectrum.A_idiot]

Hello Unknown,
it's a pity that some people don't even reveal their first name. I don't like to cite SUCH alias names, in first place 

From a closeup of the damaged PCB, I strongly assume, that somebody tried to repair the instrument, using a hot air gun around U110, or the Ohm protection transistors Q203 - Q210, either to replace or re-insert them.
That guy used too much heat, also too long, and did not cover the other components, so creating this PCB delamination.
It's not clearly visible, but all components inside  the area of delamination might have been damaged by the heat, at least all the solder joints are under suspect.
Please make a macro photo exactly vertical above this area, so that every component and solder joint is sharp, and better illuminated, maybe from different sides.

The transistors  have been soldered very badly, and I see solder balls. The transistors might be defective, or there might be a short to GND as well , via Q211 which should be tested in situ.
If you select 4W Ohm, do you measure the appropriate test currents out of + Input and - Input for the different (manually selected) ranges?

You have to de-solder these transistors anyhow, so you can test them individually, later on.

W/o the transistors back in place, you might then solder a short from Q202 to CR202 temporarily, to make further tests on the Ohm mode.
Descriptions how to do that, has been written down elsewhere .. I guess you found that thread already, as you copied the OHM schematic here.

I guess, that U110, that is used for current mode protection, has been replaced, but that mode should work properly, and has nothing to do with your OHM problem. Anyhow, many of the solder joints look very bad. U304 also looks bad.

We will definitely continue analysis, when you have presented further results.

Frank
 

If you select 4W Ohm, do you measure the appropriate test currents out of + Input and - Input for the different (manually selected) ranges?
-- What do you mean exactly ? IF the instrument measures the proper value of a plugged in resistor ? no. not at all. you want me to plug a multimeter in mA mode and measure the output current between the LOW + and - ?


W/o the transistors back in place, you might then solder a short from Q202 to CR202 temporarily, to make further tests on the Ohm mode.
Descriptions how to do that, has been written down elsewhere .. I guess you found that thread already, as you copied the OHM schematic here.

I was also a bit in the dark as I didn't really grasp what that schematic meant , and under what conditions the instrument was under .. so I would like some further insights from you if you have the chance. Thanks

[inse]

That is a somewhat botched repair (attempt).
Someone with too much hot air and little intuition had been at work.
But don’t worry, with love, patience and enamel wire it can all be fixed.
First I would remove the SO8 chip to check where the traces went, then restore missing ones with enamel wire.
Use flux to reflow the other components.

[Sorama]

Why not replace all those visibly “impacted” devices  by default ?
Takes 10’ of soldering.

That would also allow one to clean and inspect the pcb.

[The_Spectrum.A_idiot]

Why not replace all those visibly “impacted” devices  by default ?
Takes 10’ of soldering.

That would also allow one to clean and inspect the pcb.

Due to the fact the PCB is so badly delaminated , I want to do the bare minimum to reincarnate the device. as far as I know if I whisper a bit louder it would break everything else that might be kind of working.

So I am trying to do as much testing and diag BEFORE loading the parts canon.

Thanks

[wraper]

I'm 99.9% sure someone fried it with hot air so much that melted front/rear switch and water vapor popped the bubble in PCB. As already mentioned the main concern is vias. As there does not seem to be any traces on internal layers, worst case broken vias can be fixed by running thin wires through them and soldering on both sides.

[m k]

Some joints in between bad ones seem to be fine.

[Kleinstein]

If you select 4W Ohm, do you measure the appropriate test currents out of + Input and - Input for the different (manually selected) ranges?
-- What do you mean exactly ? IF the instrument measures the proper value of a plugged in resistor ? no. not at all. you want me to plug a multimeter in mA mode and measure the output current between the LOW + and - ?

Yes use a different DMM (could be even handheld, no high accuracy needed) and measure the current between the COM and +input, when in the different ohms ranges. Ideally this could be 4 wire mode, but 2 wire mode would likely give the same result.

If there is a defect, the current is often way off, like essentially not current. The results can give more of a hint on where the problem is.

Though likely not a problem, another simple test is to check for the input bias current in voltage mode. To high a current there can also cause trouble in the resistance measurements. For a final check of the meter one should do that test anyway. For this test there are 2 ways: the simple is uning the 10 M input resistance mode and something like manual 1 V range to compare the reading with a short and open input. The difference should be well less than some 50 pA*10 M = 500 µV.

[The_Spectrum.A_idiot]

If you select 4W Ohm, do you measure the appropriate test currents out of + Input and - Input for the different (manually selected) ranges?
-- What do you mean exactly ? IF the instrument measures the proper value of a plugged in resistor ? no. not at all. you want me to plug a multimeter in mA mode and measure the output current between the LOW + and - ?

Yes use a different DMM (could be even handheld, no high accuracy needed) and measure the current between the COM and +input, when in the different ohms ranges. Ideally this could be 4 wire mode, but 2 wire mode would likely give the same result.

If there is a defect, the current is often way off, like essentially not current. The results can give more of a hint on where the problem is.

Though likely not a problem, another simple test is to check for the input bias current in voltage mode. To high a current there can also cause trouble in the resistance measurements. For a final check of the meter one should do that test anyway. For this test there are 2 ways: the simple is uning the 10 M input resistance mode and something like manual 1 V range to compare the reading with a short and open input. The difference should be well less than some 50 pA*10 M = 500 µV.

With a Fluke 117
At A mode DC and AC I get nothing on the 2 jacks HI-LO going though all ranges of 2W AND 4W MODE
AT mV DC I get nothing on 2 jacks high and low both on 2W and 4W mV AC I get a fixed 150-80mV AC I will get my better meter 287 and I will retry.

[The_Spectrum.A_idiot]

Okay good news .

After re-touching the CR203 Q201 and Q202 Continuity is perfect know , Diode mode has issues show 0.4 instead of 0.6V F-v of a diode

So I think it's safe to retouch everything with A LOT of flux and see how it goes.

Thanks

[Kleinstein]

The current to look at from the ohms source should be DC from 0.5 µA to 1 mA, depending on the range. Beside the current source part, that is also a chance that a realy is at fault. One could check this by taking the current off CR202 instead of the + input.

No current may also be faulty fuse at the external DMM - so maybe check the continuity.

The already "repaired" area needs to be cleaned anyway - so a touch-up with new flux can be worth a try.

If the resoldering changes things it may be woth to re-measure the voltages (to GND) at Q202, e.g. in the 1 K range, both with the terminals open and shorted. This should give a good idea if the issue is more with the actual current source of the protection.

[The_Spectrum.A_idiot]

The current to look at from the ohms source should be DC from 0.5 µA to 1 mA, depending on the range. Beside the current source part, that is also a chance that a realy is at fault. One could check this by taking the current off CR202 instead of the + input.

No current may also be faulty fuse at the external DMM - so maybe check the continuity.

The already "repaired" area needs to be cleaned anyway - so a touch-up with new flux can be worth a try.

If the resoldering changes things it may be woth to re-measure the voltages (to GND) at Q202, e.g. in the 1 K range, both with the terminals open and shorted. This should give a good idea if the issue is more with the actual current source of the protection.

I think my 117 has no ability to measure so small currents.
I need to test it out with 287.

Thanks

[The_Spectrum.A_idiot]

The current to look at from the ohms source should be DC from 0.5 µA to 1 mA, depending on the range. Beside the current source part, that is also a chance that a realy is at fault. One could check this by taking the current off CR202 instead of the + input.

No current may also be faulty fuse at the external DMM - so maybe check the continuity.

The already "repaired" area needs to be cleaned anyway - so a touch-up with new flux can be worth a try.

If the resoldering changes things it may be woth to re-measure the voltages (to GND) at Q202, e.g. in the 1 K range, both with the terminals open and shorted. This should give a good idea if the issue is more with the actual current source of the protection.

okay tested with the 289.
0.57-0.5μADC on any scale ohms.

the areas in question are cleaned and seem ok.

**If the resoldering changes things it may be woth to re-measure the voltages (to GND) at Q202, e.g. in the 1 K range, both with the terminals open and shorted. This should give a good idea if the issue is more with the actual current source of the protection.**

Are there any expected values ? so set manual omhs , at kohms range , short low and hi ( red and black connectors )

I will post my findings

[The_Spectrum.A_idiot]

The current to look at from the ohms source should be DC from 0.5 µA to 1 mA, depending on the range. Beside the current source part, that is also a chance that a realy is at fault. One could check this by taking the current off CR202 instead of the + input.

No current may also be faulty fuse at the external DMM - so maybe check the continuity.

The already "repaired" area needs to be cleaned anyway - so a touch-up with new flux can be worth a try.

If the resoldering changes things it may be woth to re-measure the voltages (to GND) at Q202, e.g. in the 1 K range, both with the terminals open and shorted. This should give a good idea if the issue is more with the actual current source of the protection.

based on this -see attached
-----------////////////---------------
kohms - short
R201 4.993v
R202 6.972v
U201-A 3 - 6.972v
U201-A 2 - 6.972V
+7REF - 6.972V

Q201-1 - 11.099
Q201-2 - 7.110
Q201-3 - 4.4636

CR203-A 1 - 13.594
CR203-A 3 - 18.811

1SRC Q202 - 1 12.415
Q202 - 2 - 1.2771
R290 - Q202 - 3- 18.008
 
CR201-A -3 - 18.008
CR201-A -1 14.006
 
U201-B 6 11.925
U201-B 5 11.098

[The_Spectrum.A_idiot]

The current to look at from the ohms source should be DC from 0.5 µA to 1 mA, depending on the range. Beside the current source part, that is also a chance that a realy is at fault. One could check this by taking the current off CR202 instead of the + input.

No current may also be faulty fuse at the external DMM - so maybe check the continuity.

The already "repaired" area needs to be cleaned anyway - so a touch-up with new flux can be worth a try.

If the resoldering changes things it may be woth to re-measure the voltages (to GND) at Q202, e.g. in the 1 K range, both with the terminals open and shorted. This should give a good idea if the issue is more with the actual current source of the protection.

**If the resoldering changes things it may be woth to re-measure the voltages (to GND) at Q202, e.g. in the 1 K range, both with the terminals open and shorted. This should give a good idea if the issue is more with the actual current source of the protection.
**
open
Q202-1 - 12.416
Q202-2 - 3.4-3.45
Q202-3 - 18.006

short
Q202-1 - 12.416
Q202-2 - 1.2460
Q202-3 - 18.006

[Kleinstein]

The full 18 V at the Q202 gate suggsts that the regulation tries to turn the fet off (it is a P channel FET and thus off with a positive voltage).

The voltages at the OP-amp inputs (pins 5 and 6) suggest that the output should go negative and not positive as observed. So chances are high that U201 B is bad.

Still having some voltage at the output side (Q202 pin 2)  of the curren source somewhat points to leakage at Q202 - so this part may also be damaged ( I would at least order or look for a replacement, if ordering an AD706 (or substitute).
This leaves the question if the protection is working OK. This could be  a shorted transistors, or maybe one with too low a break down votlage and also an open Q211. Off cause the reason of the damage could also be from just too much voltage for the transistors.