HP 1741A Oscilloscope

[tryppyr]

Yes, that's how I felt too. Anyway, what's done is done.

I just ran two tests... first with LVPS with no load. Second with LVPS plugged onto A12 (but with HVPS removed. The no load test was almost identical to the previous no load test.

Across VR3 = 36.4
Across R18 = 0.0
Across R17 = 45.9
at +156 = 166.5
at +120 = 133.0
at -100 = -95.2
at -15 = -14.8
at +15 = +14.9
at +48 = 48.0
at +5 = 5.0
at +15 Unreg = 27.4

Then with A12 (only) plugged in everything started racing to zero. There was no way to get a real read. I tested two or three points, each after powering down and powering up again... the numbers just clicked downward non-stop.

After that I decided to unplug A12 and see if I got the right readings from the unloaded LVPS. They had moved a little, but not much... normal variance, probably. Still, I think A12 is now a prime suspect. Unfortunate, since it is going to be hard to test.

 

[KrisBlueNZ]

So is the +5V regulator working now?

Edit: Oops I didn't see your reply. This thread is up to five pages now!

So +5V is correct with everything plugged in except A12?

When you plug in A12 everything goes screwy?

Edit2: Where exactly are you connecting your multimeter earth lead?

 

[tryppyr]

So is the +5V regulator working now?

It seems to be under the no-load test

So +5V is correct with everything plugged in except A12?

I'll check that today. For the sake of clarity, "everything plugged in except A12" means everything except A12 and HVPS, because HVPS connects through A12.

When you plug in A12 everything goes screwy?

Yes

Where exactly are you connecting your multimeter earth lead?

In one of the screw holes of the chassis.

 

[KrisBlueNZ]

OK. I understand.

Can you post some photos of both sides of the A12 board please.

 

[tryppyr]

Yes, A12 will be the next challenge. I tried removing it last night but found I lack the tool needed to disconnect the dial extenders. Here's what I can see of A12 now. The tool, I believe, is a tiny thin hex key needed to loosen the set screw in those long white plastic rods.

 

[tryppyr]

I just ran three more sets of LVPS numbers. First was the test with all load applied except A12/A15.

At +156 rail it was 122.6
At +15 Unreg rail it was 27.6
At -15 rail it was 0.0
At +120 rail it was 96.6
At +48 rail it was 39.2
At -100 rail it was -94.3
At +5 rail it was 2.5

Next I removed the A6 connector, leaving only A17 attached.

At +156 rail it was 155.6
At +15 Unreg rail it was 26.0
At -15 rail it was -14.8
At +120 rail it was 120.1
At +48 rail it was 48.0
At -100 rail it was -94.9
At +5 rail it was 3.9

Finally, I did another baseline run with no load.

At +156 rail it was 167.5
At +15 Unreg rail it was 27.6
At -15 rail it was -14.8
At +120 rail it was 134.4
At +48 rail it was 48.0
At -100 rail it was -95.3
At +5 rail it was 5.0

Edit:

In the unloaded test I also checked the following:

Across VR2 (should be 10V) was 9.1
Across VR3 (should be 36V was 36.4
Across R17 was 46.7
Across R18 was 0.0

 

[tryppyr]

I finally liberated A12.

 

[tryppyr]

One of the first things I noticed is that one of the zeners has a discolored lead.

 

[tryppyr]

One of the first things I noticed is that one of the zeners has a discolored lead.

Edit: I found a second zener with similar lead discolorations.

Edit 2: Even closer inspection shows that all the zeners on this board show signs of carbon on the leads.

Edit 3: I'm really struggling to figure out which zener is which. I have most of the transistors mapped now, but the zeners elude me.... especially around U2 (the IC I can't locate on the Schematic)

 

[tryppyr]

Here's what I think so far.

 

[tryppyr]

Here's the one I labeled VR1, now replaced with a fres 10V Zener.

 

[tryppyr]

I did a little internet digging and found I am not the only one who hit the differences between the A12 units in the scope with the schematics available in the Agilent-provided manual. Apparently older versions of the scope didn't have the additional circuitry involving the CA3140T op amp, and fed the signal from the first anode of the CRT straight through. That's all I have been able to find so far. Certainly nothing that resembles an accurate schematic of this newer rev of the A12 board.

I downloaded the datasheet for the CA3140, and now that I better understand how it operates, I need to alter my assumptions about the circuit it's a part of, throwing away the assertions made by the manual about parts in that vicinity of the board, since they don't apply.

 

[tryppyr]

After studying the part of A12 that is different from the schematic and diagram provided in the manual, I've drawn this admittedly crude schematic of the new circuit. I also noticed what looks like another problem. Every other transistor mounted on the boards I have examined put the flat part of the transistor facing the connection for the base. However, with the 2N3906 in this circuit, it's mounted the other way around, so that the emitter is tied to the base of the adjoining 2N3904.

 

[tryppyr]

I did a little more cogitating tonight and deduced that the zeners labeled 02 00 41 are, in fact, 5.1V 1/2 watt. I base this opinion on the fact that VR2 (on the side of the board still covered by the schematic) was a 5.1V and it carried those numbers. This resolves one of the two mysteries, and means that one of the other two burned up zeners in the mystery circuit is also a 5.1V. I'll pick that one up tomorrow.

While I was working on the board, I pulled the transistor I labeled Q8 and installed a new 2N3906 the right way around (I hope).

 

[tryppyr]

The update for today (so far):

I have replaced A12Q6 (the transistor nearest the U1 chip) because it too was backwards. This could mean it was a properly oriented chip with a different pinout, but I wanted to be sure rather than sorry.

I also replaced one of the zeners in the mystery circuit that I identified as a 5.1V.

- Greg

 

[(*steve*)]

I have moved posts relating to another one of these scopes to here: https://www.electronicspoint.com/another-hp-1741a-oscilloscope-t266117.html

 

[tryppyr]

This morning I wrapped up what I had to do on A12, and decided it was time to put everything together and see what the current status is. Short version of the story is that there is still something somewhere pulling a load on LVPS. With everything connected the +156 line climbs to +131 and stays there. +120 gets to about +95. +48 is at about +36. It probably goes without saying that HVPS never lights up and there is no ozone smell, and nothing happens apart from lighting the LEDs on the front panel. In fact, the neon that was lighting on A12 now remains dark. It seems likely I did something to A12 that needs to be redone, but at this point I can't be sure what that is since I don't have an accurate schematic to reference.

So on that cheerful note I think I'll put this project aside for a bit, do some more reading, and see if any more bright ideas occur to me.

 

[tryppyr]

Well, I decided to get back at it. Having received the parts from Texas, I decided I might as well use what I paid for, even if there was no hard indication that the 48V rail was in need of a refresh. So I pulled and replaced the 3 transistors (Q5, Q6 and Q8) that I had not already replaced. That diminished the pile of uninstalled parts a bit.

Next up I intend to replace the zeners on the HVPS (A15), primarily because I saw the same sort of carbon on the leads that I saw on A12. Is it significant? I surely don't know, but it gives me a good reason to use the soldering iron, and I have grown quite fond of that. Once I have those five diodes replaced, the pile of uninstalled parts would be significantly diminished, and the world will surely be a better place. Something tells me the HVPS will be no better off... but we take our successes where we can find them. Who knows, I might actually get lucky.

Anyway, once all that is done I have a few more replacement parts for the good old A12 mystery board. I still have no idea what's going on there, or whether the changes I made last week were right or wrong. Alas, that's why it is a mystery. But these extra parts won't install themselves, so I must step up and solder like a mad man, for that is my new calling.

 

[tryppyr]

Maybe it's time for me to throw in the towel. I just went out to test LVPS after the transistor swaps. With absolutely NO load, all of the rails run to zero within seconds.

 

[(*steve*)]

It's worthwhile testing those zeners as you remove them.

Get a couple of 9V batteries in series (3 might be good) and connect a 4k7 resistor in series. Then place your multimeter across the ends and try the zener. It should measure 0.6V one way and something close to the zener voltage the other way.

It will let you know if the device is open or short as well, and you'll be able to tell if the zener voltage exceeds your battery voltage.

The limit of 5mA (for 3 x 9V batteries) shouldn't hurt any zener, and the relatively high voltage means that you get reasonable current for the low voltage zeners with mushy knees (it's less of an issue at higher voltages)