Author Topic: FLUKE 7261a repair advice requested  (Read 382 times)

0 Members and 1 Guest are viewing this topic.

Offline ejd.polTopic starter

  • Contributor
  • Posts: 18
  • Country: nl
FLUKE 7261a repair advice requested
« on: June 10, 2023, 02:13:36 pm »

I have a Fluke 7261a on my workbench that needs a repair. It would not turn on. I replaced the fuse, and turned it on. The fuse blew. Hmm...
A first step, I searched for the service manual, and found it easily, only to discover that it does not have the schematics.  :(
Unfortunately, I have not been able to locate the schematics for this meter anywhere, so any pointers are really welcome.

Second step was to open it up, and do a visual inspection. I could find nothing obviously wrong (bulging caps, black components, strange smells).

Time for step three: a bit of reverse engineering of the power supply.
I measured as well as I could all the components of the supply, and specifically, how they are connected. The on/stby and batt/line switches were helpful!
This resulted in the hand-drawn schematics in the first attachment. The design of the supply seems relatively simple.
Of course, completeness can not be guaranteed, but at least I could understand how it works.


None of the measurements showed any sign of something wrong, so I decided to do an experiment with replacing the fuse, and turning it on while open.
Again, the fuse blew, but this time I could smell something. So I repeated the experiment, and this time I could see black smoke escaping from under the transformer.
There was no chance of inspecting the underside, so I desoldered the thing from the PCB. (Not so easily done, it being so large and with many pins!)

The transformer has separate winding sections for the primary and secondary windings. Once I got it out from the PCB, a black spot could be seen on the primary winding.
Figuring I had nothing to lose, I dismantled the core (that was quite some work, as the core was heavily coated with lacquer), and unwound the first primary winding.
The black spot clearly can be seen.


So, the diagnosis is that the primary winding has developed a short in it, and some further unwinding needs to be done to reach it.
However, unwinding the first winding was already difficult, as the wire was frequently sticking strongly to the remaining wound part, especially near the short where it got hot.
The insulation of the wire spooled off has been damaged in many places, so it needs to be replaced with new wire.
So the repair is not so easy, and I don't know what to expect when unwinding the second primary winding.

This is when I started to consider alternatives. The counter seems to run on three voltages: 5V, 12V, and -12 V.
By coincidence, I do happen to have a switched mode power supply that produces these three voltages: an Astec LPT42.

[ Specified attachment is not available ]

Knowing the disadvantages of switching supplies, I was wondering if it would be possible to leave in and use the linear regulation section,
and feed that from the switch mode supply. That means the output voltages need to be a bit higher.
It so happens that this particular model does have a sense input on the 5V line, so that gives hope it is possible to persuade it to generate, say, 6 V.
The +-12 volt rails are generated by 7812 and 7912 regulators in the LPT42, so those are linearly regulated anyway.

I do not consider building the Astec into the Fluke, I fear that would give too much trouble with disturbances from the switching.
But perhaps it could be built into an external box, and feed the supply into the Fluke through a new hole in the rear panel.
Perhaps it would then even be a possibility to attach the external supply directly to the transformer holes in the PCB (optionally shorting parts of the rectifier bridges).

At this point I am scratching my head how best to proceed. Continuing on the transformer is a lot of work, with no guarantee that I can fully restore it.
The switch mode power supply is a second option, but also not without risk. A third option would be to build a custom-design power supply.
That power supply would then have to be a linearly regulated one (otherwise I better take the second option), and that brings me right back to the transformer.
I have been looking around for one, but my guess at this point is that it would have to be built by hand...
A fourth option is to shop around for a broken 7261a, and hope the transformer is alive and well and can be transplanted.
The fifth option is to abandon my counter, and offer it for parts.

[ Specified attachment is not available ]

Soooo.... This is where I am right now. I am open to hear your thoughts, comments, suggestions. Any feedback is welcome!

Many thanks in advance, Evert-Jan




Offline ejd.polTopic starter

  • Contributor
  • Posts: 18
  • Country: nl
Re: FLUKE 7261a repair advice requested
« Reply #1 on: June 10, 2023, 02:30:59 pm »
Funny, the third and fourth attachments did not make it out to the post....
Second try!


Offline MathWizard

  • Super Contributor
  • ***
  • Posts: 1147
  • Country: ca
Re: FLUKE 7261a repair advice requested
« Reply #2 on: June 15, 2023, 12:45:40 am »
Well there's nothing to the original PS, much like my 5.5D Keithley DMM. I would have expected more than 78/79 linear regulators for a DMM. One of these days I should power it from a SMPS and see if it reads anything different.

Seeing as how there's so much digital stuff in your Counter, I'd try the SMPS. I'm guessing it doesn't use much current. You could try bigger output caps. I have no idea what the front end on a counter looks like, really fast peak detector circuits?

I wouldn't think it's be that hard or expensive just to get another similar transformer. I'd look on ebay, how many things with about the same power, run off +/-12 and 5. But yeah most of them are still in stuff.

Offline ejd.polTopic starter

  • Contributor
  • Posts: 18
  • Country: nl
Re: FLUKE 7261a repair advice requested
« Reply #3 on: June 29, 2023, 08:16:23 am »
Hey, many thanks for the response! Apologies for the delay in answering.

For a DMM, the voltage source that defines its accuracy is the reference, but that does not deliver any power.
The measurement circuits are designed such that supply voltage drops out of the equations, so the supply itself does not need to be accurate at all.
(Dual or multiple ramp integration is a good example.)

Let me know how your DMM does with an SMPS! If there is a difference,
it might be that the switching activity radiates magnetically into the measurement unit (an SMPS transforms the voltages by switching coils after all),
and causes disturbances in the sensitive part. That would also be the first thing that I am concerned about in my case.

I have looked around for replacement transformers, but so far have not found any that comes even close to what the counter needs.
So I think I will do the following. First, I'll try to get the SMPS working, using the sense input to persuade it to deliver a bit higher voltage for the digital supply.
So 6 V or so as input for the 5 V regulator circuit in the counter. I inspected how the SMPS produces the +-12 Volt rails,
and guess what: it has 7812 and 7912 regulators on board! So I can hook those outputs directly into my counter.

That setup hopefully will allow me to see if the counter is still operational, even though maybe its accuracy is impaired by the SMPS somehow.
If the counter is still alive and kicking, then I will continue working on the transformer. That would entail also unwinding the second primary.
And there are some more smaller primary windings. I need to take good note of how the windings are arranged! (Direction and such.)
And then replace the windings with new wire. (I'll try to measure the length and the resistance to get an estimate of the cross section of the copper....)
That process will take some time, but at least then I know I have a fighting chance to get the counter back into working condition.

I will post progress every now and then!

Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo