Author Topic: HP 3456A multimeter - ROM/RAM problem?  (Read 2632 times)

0 Members and 1 Guest are viewing this topic.

Offline rodpp

  • Regular Contributor
  • *
  • Posts: 245
HP 3456A multimeter - ROM/RAM problem?
« on: April 26, 2016, 05:39:05 pm »
Hi!

My HP 3456A multimeter is showing a intermittent problem.

When I turn it on it works without problems. After some minutes powered on, the front display freezes and the buttons don´t work. After that if I turn it off and on again, sometimes it shows "HP 3456" in the display and then freezes again; and more often it freezes just after the power on. Only after one day or more powered off that it works again, and only for some minutes.

I have the service manual but don´t have a HP signature analyser, so I can´t follow the manual´s throubleshooting instructions. Using the schematics I checked the basics: power supply, processor clock, logic gates and so on. Everything seens to be ok.

I read in this forum, and in other websites, reports of failures in ROM/RAM memories. The serial number of my unit starts with "2015" and there is reports of other users of RAM/ROM failures in that range. The problem is that I don´t have these ICs and it seems that they are almost unobtainium.

Are these intermittent symptons a hint for RAM/ROM problems?

Thanks in advanced.

Regards,
Rodrigo.










 

Offline pelule

  • Frequent Contributor
  • **
  • Posts: 509
  • Country: de
  • What is business? It’s other people’s money
Re: HP 3456A multimeter - ROM/RAM problem?
« Reply #1 on: April 26, 2016, 06:27:27 pm »
Out of my view this is not related to ROM/RAM.
A "cold" system runs well, the issue occures at a "warm" meter.
You may want to follow that primary cause first.
A good start is also to check the power supply (measure the vaules, check the caps).

BR
PeLuLe
You will learn something new every single day
 
The following users thanked this post: rodpp

Offline rodpp

  • Regular Contributor
  • *
  • Posts: 245
Re: HP 3456A multimeter - ROM/RAM problem?
« Reply #2 on: April 26, 2016, 06:47:29 pm »
Out of my view this is not related to ROM/RAM.
A "cold" system runs well, the issue occures at a "warm" meter.
You may want to follow that primary cause first.
A good start is also to check the power supply (measure the vaules, check the caps).

BR
PeLuLe


Thank you!

I´ll check again the power supply. I already replaced some capacitors and the different outputs seems ok.

Regarding the "cold" vs "warm", after 30 min powered off it should be cold, but it only works again after days! I resoldered almost all pins of the outguard logic board and no success...



 

Offline pelule

  • Frequent Contributor
  • **
  • Posts: 509
  • Country: de
  • What is business? It’s other people’s money
Re: HP 3456A multimeter - ROM/RAM problem?
« Reply #3 on: April 26, 2016, 07:55:59 pm »
That "cold" / "warm" is not so simple to judge in distance.
I once had a failure, which was gone, when the room temp was less than ~18°C (caused by bad fet).
When the meter was powered on, it need some minutes to heat up internally and to cause the error.
I finally found the problem part with cooling spray and hot air gun (hair dryer).
You should test with the open system (slower heating up, for faster cooling down, easier to heat/cool areas).

BR
PeLuLe
You will learn something new every single day
 
The following users thanked this post: rodpp

Offline rodpp

  • Regular Contributor
  • *
  • Posts: 245
Re: HP 3456A multimeter - ROM/RAM problem?
« Reply #4 on: April 26, 2016, 09:05:59 pm »
That "cold" / "warm" is not so simple to judge in distance.
I once had a failure, which was gone, when the room temp was less than ~18°C (caused by bad fet).
When the meter was powered on, it need some minutes to heat up internally and to cause the error.
I finally found the problem part with cooling spray and hot air gun (hair dryer).
You should test with the open system (slower heating up, for faster cooling down, easier to heat/cool areas).

BR
PeLuLe

Yes, I´ll investigate carefully the cold/warm problem. Here in Brazil the ambient temperature is almost always over 25ºC...





 

Online lowimpedance

  • Super Contributor
  • ***
  • Posts: 1088
  • Country: au
  • Watts in an ohm?
Re: HP 3456A multimeter - ROM/RAM problem?
« Reply #5 on: April 27, 2016, 12:16:33 am »
Another issue could be your IC sockets on the processor board (red color) as noted in these service notes from HP. see this old thread.
https://www.eevblog.com/forum/testgear/hp3456a-service-notes-for-owners-info!/msg144820/#msg144820
Also you really should replace ALL the power supply electrolytics as they do cause all manner of unusual problems.
These are great meters, good luck with the repair.

edit: typo!
« Last Edit: April 27, 2016, 12:56:25 am by lowimpedance »
The odd multimeter or 2 or 3 or 4...or........can't remember !.
 
The following users thanked this post: rodpp

Offline rodpp

  • Regular Contributor
  • *
  • Posts: 245
Re: HP 3456A multimeter - ROM/RAM problem?
« Reply #6 on: April 27, 2016, 12:34:50 am »
Another issue could be your IC sockets on the processor board (red color) as noted in these service notes from HP. see this old thread.
https://www.eevblog.com/forum/testgear/hp3456a-service-notes-for-owners-info!/msg144820/#msg144820
Also you really should replace ALL the power supply electrolytics as they do cause all manner of unusual problems.
These are great meters, good luck with the reapair.

Thank you!

I already eliminated the red socket under the processor of the outguard logic board, according with the service note linked in that thread. It says to eliminate all red sockets, but I only worked in the outgard logic board at the moment.
 

Offline JFJ

  • Regular Contributor
  • *
  • Posts: 231
  • Country: gb
Re: HP 3456A multimeter - ROM/RAM problem?
« Reply #7 on: April 27, 2016, 01:18:05 pm »
... After some minutes powered on, the front display freezes and the buttons don´t work...

While the meter is frozen, check the voltage level on the Outguard Controller RST-L line (pin 40 of the MC68A00 microprocessor). If it is low (logic 0), then suspect capacitor C9.

When the meter is switched on, the Power Supply Ready line is, initially, held low. This causes the microprocessor's reset (RST-L) to also be held low - via IC U14. Once the Power Supply Ready line is released, R13 should quickly charge C9 to logic 1, which causes U14 to take the RST-L line high. Thereby, starting the microprocessor.

If C9 (a tantalum capacitor) has become leaky, then the voltage level across it may not reach logic 1, and the microprocessor will be constantly held in reset. After a day, the capacitor could recover sufficiently to allow the meter to boot-up normally - only to fail again as the capacitor starts to break down.
 
The following users thanked this post: rodpp

Offline rodpp

  • Regular Contributor
  • *
  • Posts: 245
Re: HP 3456A multimeter - ROM/RAM problem?
« Reply #8 on: April 28, 2016, 07:00:44 pm »
... After some minutes powered on, the front display freezes and the buttons don´t work...

While the meter is frozen, check the voltage level on the Outguard Controller RST-L line (pin 40 of the MC68A00 microprocessor). If it is low (logic 0), then suspect capacitor C9.

When the meter is switched on, the Power Supply Ready line is, initially, held low. This causes the microprocessor's reset (RST-L) to also be held low - via IC U14. Once the Power Supply Ready line is released, R13 should quickly charge C9 to logic 1, which causes U14 to take the RST-L line high. Thereby, starting the microprocessor.

If C9 (a tantalum capacitor) has become leaky, then the voltage level across it may not reach logic 1, and the microprocessor will be constantly held in reset. After a day, the capacitor could recover sufficiently to allow the meter to boot-up normally - only to fail again as the capacitor starts to break down.

Thank you, its a good justification for the meter work again only after one day.

I remember that the service manual says to check the processor reset line, I checked it with the scope and it was ok, but maybe I should pay more attention to be sure that the logical 1 is really 5V. Normally I only look careless to the scope screen.

Next week I´ll be travelling and I can only work again in this repair by the end of May. If I find and fix the problem I post here.

Thanks for all!

 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf