Author Topic: Test 3 failing on HP 3456A  (Read 3112 times)

0 Members and 1 Guest are viewing this topic.

Offline gadget73

  • Regular Contributor
  • *
  • Posts: 71
  • Country: us
Test 3 failing on HP 3456A
« on: April 28, 2017, 01:36:00 am »
As the title says, I've got a failed "test 3" on my HP 3456A meter.  I went through the manual to do the pinpoint tests, and it seems to be "T1" on the A3 board.  According to the manual, this is HP p/n 9100-2616, Transformer - Pulse bifilar wound, 18.0 mm, coce 28480.

link to manual:
https://www.sonoma.edu/esee/manuals/03456-90006.pdf

Troubleshooting steps I'm following start on pg 310, paragraph 8-A-37.  The pinpoint test for the transformer is 8-A-41 on the next page.

Some questions about this:

As a sanity check, is there any other means I can verify this?  The diagnostic process has me check for a particular waveform "at the core of A3T1".  I touched the probe to the ferrite core, and got no waveform.  I did get the waveform specified at the pins of U19.  Is this the proper means of testing the transformer?

Is this a part I can substitute, or am I looking for a very specific and probably grossly obsolete HP part or complete A3 board as a parts donor?  Any chance i could just re-wind this?  Looks like two pieces of magnet wire, nothing overly special.

Lastly, is it probably something totally unrelated to this part and I'm chasing my tail?  This is an all-original 1982 vintage unit.  All the caps look visually OK, no leakage or bulging but I have not tested or replaced any of them.

I appreciate any insight you fine folks might offer.
 

Offline Vgkid

  • Super Contributor
  • ***
  • Posts: 2598
  • Country: us
Re: Test 3 failing on HP 3456A
« Reply #1 on: April 28, 2017, 01:52:20 am »
Check the power supply. My capacitors looked good, yet were failling.
I started geting 1 error occasionally, then a whole bunch a few weeks later.
If you own any North Hills Electronics gear, message me. L&N Fan
 

Offline gadget73

  • Regular Contributor
  • *
  • Posts: 71
  • Country: us
Re: Test 3 failing on HP 3456A
« Reply #2 on: April 28, 2017, 03:26:09 am »
I found a list of part numbers for the PS board, maybe worth just swapping them all out.  Its near as old as I am and I'm not 100% anymore.

https://www.febo.com/pipermail/volt-nuts/2014-December/003958.html

I'll confirm those part numbers and such.

This had been giving me some odd floaty voltages at the input, but I thought that was leaky FETs.  Maybe power supply ripple?
 

Offline gadget73

  • Regular Contributor
  • *
  • Posts: 71
  • Country: us
Re: Test 3 failing on HP 3456A
« Reply #3 on: April 29, 2017, 09:32:21 pm »
part numbers look right, dimensions and values mesh with the parts in my unit.

Did some poking at the power supply.  Voltages are in spec, but ripple on the +33v line is pretty extreme.  I'm getting 1.7v p-p and its a nasty sawtooth wave.  The others are clean.  Hopefully replacing C8 will sort that out.  The post I linked previously has this in it

" C8, where going from 150uf (originally
>> selected part) to 330uf reduced p-p ripple from .45v to < .2v"

I take that to mean in properly working stock form it should be around 1/2 volt of ripple, not the 1.7 volts I'm getting.  Temporarily clipping a 220uf 50v cap across that line takes the ripple to 0.6 volts p-p.  Still doesn't pass the test though.

Anyone think this still sounds like power supply issues? 
 

Offline gadget73

  • Regular Contributor
  • *
  • Posts: 71
  • Country: us
Re: Test 3 failing on HP 3456A
« Reply #4 on: May 03, 2017, 03:16:03 am »
For better or worse, the caps are ordered.  Will update once they get in, hopefully this makes it go again.
 

Offline TimInCanada

  • Contributor
  • Posts: 48
  • Country: ca
Re: Test 3 failing on HP 3456A
« Reply #5 on: May 03, 2017, 09:36:08 pm »
Just to make sure I understand, did you the test in the previous step, i.e., 8-A-36e-i, and verify it is an Outguard problem?

Both boards A3 (Outguard) and A30 (Inguard) only use +5V power, so the ripple in the 33V line won't cause an Error 3.

To see if the transmitter section is working at A3T1 look at what is on the secondary winding at J7 pins 1 and 3.  It should match the manual figure.  You verified U19 is putting the right pulse signal into the primary, so if there is no signal on the secondary it means the problem is in between.

In that event either there is a bad connection to T1 or a winding on T1 is open.  In the latter case it's just a matter of wrapping on some turns of magnet wire equal to the original winding (the bad winding can be left on).

Another test is to turn on the meter and check for RESET pulses going to the Inguard processor U13 pin 4.  If the Inguard isn't communicating right the Outguard processor sends reset commands at about two per second if I recall right.  If these reset pulses are showing up it means the transmit path from A3 to A30 plus some A30 logic is working and the problem is between A30U13 and A3.

Good luck!

Tim
 

Offline Tektron

  • Contributor
  • Posts: 22
  • Country: ru
Re: Test 3 failing on HP 3456A
« Reply #6 on: May 07, 2017, 11:37:31 am »
All measurement are impossible?
Check is the U13 inguard controller installed in socket. If so, try accurately remove and install back the chip. HP recommended remove this socket and  soldering the chip directly to PCB, but I just changed it by another high quality socket with perfect result.
Good luck!
 

Offline enut11

  • Frequent Contributor
  • **
  • Posts: 510
  • Country: au
  • Love building/modifying/restoring test equipment
Re: Test 3 failing on HP 3456A
« Reply #7 on: May 08, 2017, 09:33:01 am »
I had a 3 error with my HP3456A. Someone suggested re-seating all PCBs. Worked for me. Also carefully check that all pins on the edge connectors are properly engaged and not bent.
an electronics nut from wayback...
 

Offline gadget73

  • Regular Contributor
  • *
  • Posts: 71
  • Country: us
Re: Test 3 failing on HP 3456A
« Reply #8 on: May 08, 2017, 11:55:32 pm »
Just to make sure I understand, did you the test in the previous step, i.e., 8-A-36e-i, and verify it is an Outguard problem?

Yes.  That was the first thing I did.  With the jumpers in place, it passes test 3.  From there I went to 8-A-38 and it failed at point b.  No waveform at the core of the transformer.

Quote
Both boards A3 (Outguard) and A30 (Inguard) only use +5V power, so the ripple in the 33V line won't cause an Error 3.

Thats what it looked like, but I needed a Mouser order anyway, so the few dollars worth of caps was no big deal to add.  I did just install them, it made no difference.

Quote
To see if the transmitter section is working at A3T1 look at what is on the secondary winding at J7 pins 1 and 3.  It should match the manual figure.  You verified U19 is putting the right pulse signal into the primary, so if there is no signal on the secondary it means the problem is in between.

In that event either there is a bad connection to T1 or a winding on T1 is open.  In the latter case it's just a matter of wrapping on some turns of magnet wire equal to the original winding (the bad winding can be left on).

Not sure precisely what config to use here, but I get no waveform at pin 1 or 3 with J7 connected, J7 disconnected, or with J7 unplugged and the header jumped out as per 8-A-37 step f (1 to 2, 3 to 4).

Quote
Another test is to turn on the meter and check for RESET pulses going to the Inguard processor U13 pin 4.  If the Inguard isn't communicating right the Outguard processor sends reset commands at about two per second if I recall right.  If these reset pulses are showing up it means the transmit path from A3 to A30 plus some A30 logic is working and the problem is between A30U13 and A3.

Good luck!

Tim

I do see a regular negative pulse appearing at that point.  This is the reset pulse I assume?


U13 is soldered, not socketed.  I can't measure anything, the display just shows -------. 

I have removed and re-installed the inguard board as well as the power supply board with no change.  I didn't mess with any of the other boards.


I'll pull A3T1 off and see if that tells me anything.  Is the wire gauge critical for this?  I probably have some magnet wire around, but it may not be the same gauge.
 

Offline gadget73

  • Regular Contributor
  • *
  • Posts: 71
  • Country: us
Re: Test 3 failing on HP 3456A
« Reply #9 on: May 09, 2017, 12:50:53 am »
I pulled T1 and measured it.  All seems fine.   Both windings have low resistance (<1 ohm) according to my trusty dusty Simpson 260.  I put it back in, and re-soldered U19 while I was in there.  Its actually working now.  I'll leave it on and see what it does.  When it first started to act up, it would read for a while, freeze, then fail a self-test on a power cycle. 

Thanks for the help folks.  Solid state isn't really my game, and gear as involved and modern as this meter is largely out of my league.  I know some probably scoff at the idea of a meter designed more than 30 years ago as being modern, but when most of your electronic work is done with vacuum tubes its a bit of a leap.  I feel like Doc Brown in 1955 whenever I open stuff with IC chips inside.
 

Offline gadget73

  • Regular Contributor
  • *
  • Posts: 71
  • Country: us
Re: Test 3 failing on HP 3456A
« Reply #10 on: May 09, 2017, 01:13:07 am »
nope, locked up and its failing test 3 again. 

Any ideas before I just re-flow a bunch of stuff on the A3 board?
 

Offline Tektron

  • Contributor
  • Posts: 22
  • Country: ru
Re: Test 3 failing on HP 3456A
« Reply #11 on: May 09, 2017, 03:51:17 am »
Try visually checking with a magnifying glass and, if necessary, resolder both sides of connectors P5 and P6. Sometimes in places of old solder joints contact disappears.
 

Offline gadget73

  • Regular Contributor
  • *
  • Posts: 71
  • Country: us
Re: Test 3 failing on HP 3456A
« Reply #12 on: May 10, 2017, 12:29:44 am »
I resoldered all of the connectors on A3, and re-seated all of the boards.  It ran for about an hour and stopped, test 3 fails again.  Something must be getting warm and breaking down in here. 

A close inspection of A3 under the magnifying glass didn't show me anything.  Starting to wonder if I should remove T1, connect it to an ohm meter, and hit it with a hair drier to warm it up.  Maybe one of the windings is bad and only opens when it warms up.
 

Offline gadget73

  • Regular Contributor
  • *
  • Posts: 71
  • Country: us
Re: Test 3 failing on HP 3456A
« Reply #13 on: May 14, 2017, 08:28:41 pm »
T1 ohms out fine.

Connecting the across pin 1 and 3 to directly read the secondary of T1 gives me the proper waveform.  Further on in the troubleshooting I guess.  Problem may be in the receiver portion?
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf