HP 3465A DMM Repair?

[inteljoe]

I acquired this HP 3465A digital bench multi-meter from a local guy on Craigslist for $20.
I knew it had some issues before purchasing, but I was not aware to what extent until I got it home.
I opened it up and proceeded to take it apart. 

That's when reality hit me like a brick to the face.
I did not take a picture of it at the time, but what I found inside was a battery block leaking and corroding everything it touched. The block was comprised of what appeared to be several AA style batteries.

I have no idea how many years those batteries have been inside the DMM leaking, but it was a mess. I managed to clean up the area where the batteries were installed. You can see that in the 2nd and 3rd pictures with the power switch and transformer. Unfortunately some of the traces where the battery block was installed are gone, but the majority of the corrosion has been cleaned off.

Now, in the rest of the pictures there are several components that are directly below where the battery block was installed, have been damaged. They all exhibit severe corrosion. I'm hoping to identify each component and its exact placement, then proceed to de-solder, clean the board and replace the damaged components with new components. I don't know if this will even be possible and if I will have to somehow rebuild any traces.

That said, once I removed the battery pack and cleaned up that PCB, I reassembled the DMM. But unfortunately I was unable to get it to power on. I'm not sure if there is something else damaged or just something I'm overlooking. I doubt those crusty batteries had any power left, and other than those batteries acting as a possible conductor, I have no idea how to supply power to the DMM. 

Also, there appears to be some damage to the front panel. I have no idea what happened, but it looks like some sort of residue or permanent discoloration to parts of the front panel, specifically around the display and the input jacks. This is merely cosmetic and otherwise, besides the corrosion damage, there does not seem to be any physical damage.

I have not tested it any further. In fact, I'm actually looking for advice on how best to proceed. Maybe someone who has more experience than I do, specifically someone who has preformed those type of repair before.
























.
.
.

[FlyingHacker]

Neat unit.

Did you get the service manuals? It is main powered, right? I might take a look at the power supply first. If it is modular disconnect it from the unit and see if you can get it to produce some output on its own.

You may be able to feed power from a bench supply into the battery inputs and power it that way. I think this would actually be a preferred initial power up, as you could limit the current with your power supply.

First I would take a good look over the board and look for damaged traces, perhaps measure ESR on caps, and maybe check resistors as well, at least look for shorts.

[Kleinstein]

The corrosion damage does not look that bad. It seems to be only on the relatively simple power supply board.  For the part a repair of the traces is no real problem. May not be nice looking but functional.

Its possible to test the supply board and the main part separately, to see if it is worth fixing the supply. The DMM is rather old and not that good by today's standards - so it's not so much about having a well working instrument, it's more the historic value.

For these old instruments chances are good to find service manuals.

[inteljoe]

Neat unit.

Did you get the service manuals? It is main powered, right? I might take a look at the power supply first. If it is modular disconnect it from the unit and see if you can get it to produce some output on its own.

You may be able to feed power from a bench supply into the battery inputs and power it that way. I think this would actually be a preferred initial power up, as you could limit the current with your power supply.

First I would take a good look over the board and look for damaged traces, perhaps measure ESR on caps, and maybe check resistors as well, at least look for shorts.

I don't have a paper copy of the service manual, but I was able to find a PDF online.
I assume it can be powered by batteries or A/C. What's interesting is the unit did not power on without being plugged in, but once the crust battery pack was removed, it won't power back on.

I did not think of powering just the main board itself, I will definitely try that. I've got 2 nice HP PSU's for the job.
As for the power board, I will have to check it again by itself. See if I can get it to do anything. I do remember probing it at one point, but I don't recall the results.

Thanks for the advice.

The corrosion damage does not look that bad. It seems to be only on the relatively simple power supply board.  For the part a repair of the traces is no real problem. May not be nice looking but functional.

Its possible to test the supply board and the main part separately, to see if it is worth fixing the supply. The DMM is rather old and not that good by today's standards - so it's not so much about having a well working instrument, it's more the historic value.

For these old instruments chances are good to find service manuals.

I will definitely test the main board by itself. As for repairing traces, any advice on that? I've never had to repair traces before. (I know, I know, I am an amateur).

As for the DMM not being "good by today's standards", that is not really why I bought it. I'm using it to learn with, and if I can get a functional bench meter for $20 + some components I likely already have and my time, that's not a bad deal.

I will take some time to study the service manual PDF, see if I can understand how this thing works.

Thanks for the advice.
.
.
.

[FlyingHacker]

There are lots of ways to repair traces. The simplest is just to bypass the trace with a small jumper wire, connecting the components directly.

Short breaks can usually be bridged with solder (use flux! I like the paste flux (rosin), as it stays where you put it).

Slightly longer ones can be fixed with a short piece of uninsulated wire, such as a lead from a resistor. You lay it across and solder it down. Use flux! Also, you may have to gently scrape the solder mask (usually the translucent green over the traces) to expose the actual metal of the traces.

They make thin copper tape that can be used, and lots of other board repair kits for things like through holes, etc.


Google around, as there are some good guides for repairing damaged boards. I have also seen some YouTube videos on the topic.

You will want some low temp solder (always WITH lead), and flux. Consult the guides for more info.

[inteljoe]

Thanks for the advice.

So, here is the current status...

#1. Connecting A/C to the power board, I can't see any voltage passed the transformer and what I think is a PTC fuse of some sort directly to the right of the transformer.
Now, the next device in line is the glass fuse, and I think the fuse may be bad and I will try replacing that next.

#2. I tried connecting my power supplies directly to the main board and I was unable to see anything happen. No signs of life. I will have to double check the service manual, but it's possible my power supplies may not be able to supply enough current (Amps) and/or voltage (volts) to run the DMM.

[e-doc]

A few jears ago, I had the same instrument with same problem.
If I remember correctly, it was an issue with the switch that switches between battery and mains supply.
Later I will open my 3465A an have a look at...

[inteljoe]

A few jears ago, I had the same instrument with same problem.
If I remember correctly, it was an issue with the switch that switches between battery and mains supply.
Later I will open my 3465A an have a look at...

Yes, there was a switch. With the DMM taken apart, I'm not sure how the switch functions.

[PaulAm]

With the nicad pack, plugging in the power cord disconnects the batteries and connects the internal power transformer.

There are four output voltages generated from the single battery/transformer supply.  Check those if it's not working.

The DMM is rather old and not that good by today's standards

+/- 0.03% of reading + 0.02% of range on the 10mv range and +/-0.02% +0.01% on ranges up to 100V
10**10 ohm input resistance on 10mv - 1V ranges.

That's still not too shabby.

I have one of these I use as my regular bench meter.

[e-doc]

Yes, there was a switch. With the DMM taken apart, I'm not sure how the switch functions.

The switch is mechanically connected to the mains socket by a small rocker arm. Maybe you have lost it or put it together in an incorrect way.
You should look at the switch when pushing the plug of the power chord into the backside socket. The slider of the switch have to move then. If it does not, the switch stays in "battery position", and if you don't have batteries in or if they are empty, your DMM does not work.

I just disassembled my 3465A. All I could see was a resoldered switch, 2 resoldered power transitors, glued plastic posts of the back cover, that hold the screws of the power supply board, and renewed battery cells. I am not 100% sure, but I think I also had to open the switch to clean the contacts...

BTW: Are the spring contacts in correct position to connect the power supply to the main PCB? I think I made a mistake at that point when I tried the first time to reassemble.

The battery packs seem to be the same HP pocket calculators used these days (HP35, 45, 55, 65, 67, 75C/75D & 80)...
http://www.ebay.com/itm/1300-mAH-Hewlett-Packard-Calculator-Battery-HP-35-45-55-65-67-75C-75D-80-/322038037393

[inteljoe]

Yes, I remember the switch. I believe it's all in tact, or at least I did not lose any parts when taking apart the meter.
I have all the small parts in a plastic bag with the meter.

Thanks for the battery pack link. I may have to pick up one of those and see if I can somehow reassemble the power board.
The more I think of it, I want to tackle this project in steps, and the first step is getting it to power on, so I can do further testing (probing) and see what is actually damaged and what can stay. I've already started to map the components, though I'm a bit concerned I may not be able to find an exact match for some components, I just have to hope to something similar will be suitable.

Thanks again for all the advice and the ideas. I hope to learn valuable skills from this experience for future endeavors.