Author Topic: Fluke PM3394 Microprocessor Fault  (Read 1485 times)

0 Members and 1 Guest are viewing this topic.

Offline EHT

  • Regular Contributor
  • *
  • Posts: 110
  • Country: gb
Fluke PM3394 Microprocessor Fault
« on: May 25, 2017, 06:06:05 pm »
.. and for my 2nd post I'm after some advice on fixing a Fluke PM3394 bench scope.

I believe the problem is in the logic board (A3 - "microprocessor") and annoyingly appears to be either the microcontroller itself or the PAL, both programmed parts.  :'( So i have two questions about this followed by my diagnosis. I'm referring to a scanned copy of the service manual which has been posted on the net (FlukePhilips-4778.pdf) and this copy which refers to the 'A' versions https://archive.org/stream/FLUKE_pm3390a/#page/n0/mode/2up.

1) Which other scopes from the range might the logic board be compatible with?
Assuming the way to proceed may be to find another dead scope and hope the A3 board is OK, it would be good to know if I could use one from any of those scopes and the 'A' or 'B' versions. The service manual covers PM3382 / PM3384 / PM3392 / PM3394. I see in the 'A' version manual that the board is somewhat different, so i'm thinking maybe i could use 3382/84/92/94 but not 'A' or 'B' ? No idea if the programmed devices are running the same code...

If anyone has a scrap one of these scopes which they would consider selling the whole thing or the logic board please let me know via PM.

2) Might I be able to get replacement programmed parts from somewhere, or the binary images to program the blanks?
Haven't asked Fluke Service yet. Google doesn't find any so not looking hopeful.

The copy of the non-'A' manual is missing the component list so i can't see the Fluke part #.

D1009 PLS153 (173 on 'A' version)
D1001 80C196KB (programmable version 87C196KB)

Diagnosis

I'm checking all these using the Fluke Scopemeter so i can see the waveforms as well as measure the voltages.

External symptoms: The scope appears 'dead'. Nothing on the CRT, no beeps, no relay clicks, the only sign of life is the fan. The CRT heater can be seen to be glowing as well.

PSU
I have checked all of the PSU outputs including the 2kV line and the input to the tripler. These are all fine and present as expected according to the manual. After finding the power supply is working fine i moved to the CPU.

A3 'microprocessor' board:
I have measured many of the signals around the CPU D1001 by attaching flying leads to D1007, D1014, D1010, D1016. (see pic)

CPCLOCK: Oscillating healthily.

Having found the power and clock OK I looked at the resest and chip-select circuits:
POWER-HT: low to start with then high after 250ms as described in the manual. I gather this is the signal from the PSU to the CPU and controls the reset sequence.
CPRESET-LT: low after 250ms as expected (shown in the attached picture. The leading trace is the 5v power coming on; 250ms/div.)
EA: stays high. FAULT, should go low some or all of the time to allow access to external ROM/RAM
this is generated by D1014. I then checked CPRSTCSLT which is the input that triggers EA:

CPRSTCSLT: stays high - FAULT. Should go low after 250ms
this is generated by D1016, a 74LS138. I looked up the truth table for this. It needs to have CPA04-06 all high, IOCS--LT and CPWR--LT low, CPA13 high.

Out of all those, only CPA13 is high.
CPA04-06 are high for the 1st 250ms boot period then all remain low.
IOCS--LT and CPWR--LT are high.

These signals all either come from the CPU or the PAL. So.. i could proceed if I could get hold of either of these parts but at present.. hit a brick wall. PAL is in a DIP socket, CPU is a QFP package, which would be tricky but not impossible.
 |O

On the positive side it seems these scopes are fairly common. Any advice much appreciated.

Thanks!
 

Offline grumpydoc

  • Super Contributor
  • ***
  • Posts: 2705
  • Country: gb
Re: Fluke PM3394 Microprocessor Fault
« Reply #1 on: May 25, 2017, 06:27:22 pm »
CPU boards are pretty interchangeable, in fact I think a board from any 33XX 'scope will at least fire up the 'scope

The big exception is that although a CPU card from one of the non-digital versions (30XX) is the same board it has the text generator on it and that wouldn't work in a 33XX 'scope while the digital board is plugged in (although I think it might even work like that with the A8 board removed).

Where are you in the UK? - I might be able to loan you a CPU card or take yours and confirm whether it is truly dead.
 

Offline EHT

  • Regular Contributor
  • *
  • Posts: 110
  • Country: gb
Re: Fluke PM3394 Microprocessor Fault
« Reply #2 on: May 25, 2017, 06:37:50 pm »
Thanks for the rapid response grumpydoc. That would be great. I'm in London.
 

Offline grumpydoc

  • Super Contributor
  • ***
  • Posts: 2705
  • Country: gb
Re: Fluke PM3394 Microprocessor Fault
« Reply #3 on: May 25, 2017, 06:40:45 pm »
Thanks for the rapid response grumpydoc. That would be great. I'm in London.
OK, we'll have to do it by post - PM me if you want to send up your CPU card
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf