HP 8903A Need help subbing an op amp

[gdnichols31]

I was using my 8903A the other day testing an amplifier and noticed the output wasn't right (amplitude and wave shape was wrong).

I found that the problem is on the attenuator card. I also have 2 of these cards and the other one works. I set the freq to 1kHz and the output at 1V for both cards and recorded the voltages at test points TP2 thru TP6 for each card. the voltages were the same up to TP4. the voltage is .88V. The floating output amplifiers have a gain of 1.12. So .88 times 1.12 = .99V. The voltages at TP5 and TP6 are different.

The problem is in the floating output amps somewhere.

I think I have tracked the problem down to U1 and U2 (see the attached partial schematic). I checked the transistors (Q1 thru Q4, 2n2222 & 2n2907A) and they appear good. I also checked diodes CR4 thru CR7 and they are good. All the 2.2uF caps check good. That only leaves U1 and U2.

In the BOM U1 and U2 are called out as NE5534AT and I can't find them any where. Data sheet attached

Can someone give me some guidance on a substitute? I'm not very good at this and I would hate to mess it up. I think there should be several op amps that would work. The bandwidth of the 8903a is only 100kHz and the gain is only 1.12.

Thanks
Gary

[Tony_G]

I think the biggest problem you're going to have it finding them in the T package.

Dave did a video on using parametric search to find an op amp recently - https://youtu.be/y5qjGPZ5VhQ?t=1064

Failing that you might want to ask over on the HP groups.io forum - https://groups.io/g/HP-Agilent-Keysight-equipment

Sorry I couldn't help more.

TonyG

[gdnichols31]

I had the same thought.

I didn't post a pic of the board, but the pcb holes are actually in an 8 pin DIP pattern. My other attenuator board actually has the 8 pin DIPS in it.

Which just made me think to go look at THAT part number......

Why didn't I think of that earlier?

Gary

[Wallace Gasiewicz]

Put in a plug in socket so you can try different DIP OP amps  if necessary
Usually you can find a DIP that works in place of the can type.
Saves a lot of aggravation if you have to switch. Also easier to install.
Sometimes a lot of different OP AMPS will work especially if it is just a DC volt comparison
OP AMPS are not perfect on their input, they need adjustment, sometimes even the same part number amp will vary, even those in the same batch.
Sometimes a cheaper amp will work just fine, HP has a habit of using the most expensive parts.

[Tony_G]

I had the same thought.

I didn't post a pic of the board, but the pcb holes are actually in an 8 pin DIP pattern. My other attenuator board actually has the 8 pin DIPS in it.

Which just made me think to go look at THAT part number......

Why didn't I think of that earlier?

Gary

DigiKey has the P part for $1.10 - https://www.digikey.com/products/en/integrated-circuits-ics/linear-amplifiers-instrumentation-op-amps-buffer-amps/687?k=NE5534A&k=&pkeyword=NE5534A&sv=0&pv69=411897&sf=0&FV=-8%7C687&quantity=&ColumnSort=0&page=1&stock=1&pageSize=25

TonyG

[Kleinstein]

The NE5534A is still a relatively common audio OP - a bit old but still produced. One could probably get away with the non A version and marginally more noise / THD. The 5534 has external compensation and could be used in a over-compensated way. So a substitute to a different type may not work in all circuits.

Chances are the pin-out would match the DIP version. There is not very much difference in performance, mainly cheaper and maybe slightly more long term offset drift - however if the offset would be a problem one would not use an NE5534 but more like an OP27.  A socket and DIP version is a pretty good bet.

[gdnichols31]

Well after my last post I did some more trouble shooting. While doing that I remembered I never checked the output fues on the board F1 and F2. They are 1/8A each. I found the low side fuse was blown. Replaced it and everything worked perfect.

While I was looking at the second board for the part numbers I did something to mess it up. It wouldn't work either. So now I'm out of a distortion analyzer to use.

After I found the fuse on the first card bad I checked the fuses on the second card. Yes the low side fuse was also blown!

Note to self: Be more careful with the output leads when using this analyzer!

The second board still has a problem though. The output voltage is low by a few mV. Working my way thru that now.

Gary

[Golds]

I can't provide technical assistance, but if you need parts of the equipment, it may be possible to help you.

[gdnichols31]

Update on the second board;

A resistor, can't remember which one, on the attenuator board had changed value. Replaced the resistor and all is good again. 

[gdnichols31]

I can't provide technical assistance, but if you need parts of the equipment, it may be possible to help you.

I have 2 HP 8903As.

One is fine and I'm using it.

The second one has a problem with the CPU card. Haven't gotten back to working on it yet. When I figure out what's up with it I'll keep you in mind.

Thanks
Gary

[David Hess]

That was boring.  You have it fixed now but my first thought to replace the NE5534 if a DIP package is acceptable is the LT1124, but the NE5534 should be available and less expensive.

[gdnichols31]

That was boring.  You have it fixed now but my first thought to replace the NE5534 if a DIP package is acceptable is the LT1124, but the NE5534 should be available and less expensive.

Sorry, it was kind of boring. Next time I'll remember to check the fuses first. 

I'll certainly consider the NE5534s if I need to replace any in the future.

Gary

[David Hess]

Note that I made a mistake in my post.  Somehow I got it in my mind that the NE5534 is a dual part so the LT1124 would be an improved replacement.  For a single, the LT1007 is what I would use.  But of course as pointed out earlier, the NE5534 is still produced.