Repair quest: who knows the HP 350D Attenuator??

[Othello]

I have three of these attenuators and they all display the same problems which I seem not to be able to fix, or even understand.

Its a simple T attenuator one would think.
Actually its 4 of them, which get linked depending on what attenuation you choose. Out of the 10, 20, 30 and 40dB you can create anything between 0 and 100 dB (I disregard the single number attenuator for the moment, but its the same principle).

So if one of those segments was not working, you would notice that in other settings where that segment was used as well.
Well, that's not the effect I am seeing.

So I wonder where I find anybody who worked on these things, maybe I can tell them all my problems.

Regards Uwe

 

[mneary]

It looks like there is a recent thread on this topic. Perhaps you should see how that one turned out.

 

[Othello]

Odd, I searched but didn't find such a topic.
What search term did you use??

Uwe

 

[JimB]

So I wonder where I find anybody who worked on these things, maybe I can tell them all my problems.

On the basis that there is probably nothing magical about the HP350, why not just tell us what the problem is?
Someone may have a similar experience with similar equipment, or be able to diagnose the problem from logical deduction.

JimB

 

[mneary]

What search term did you use??

I didn't search; I just recalled the topic and thought it unusual that two people had the same problem.

350D should find it. Yep, here it is.... oh, it's your thread from last week.

 

[kchriste]

I have three of these attenuators and they all display the same problems which I seem not to be able to fix, or even understand.

I sounds like they are probably all OK and you just don't know how to use them. Have you matched the source impedance and the load impedance to the attenuator? If not, you won't get the specified attenuation as set by the dials.

 

[Othello]

Well, JimB, here we go then.
Maybe Kchristie is right and I just don't know how to use them!
That would be the easiest repair...

The attenuator consists of two modules containing basically 4 sets of T attenuators each. One module covers 0 to 10 dB in 1dB steps and the other covers 0 to 100 dB in 10 dB steps.
I am concerning myself here with the 10 dB step unit.

This unit contains 10, 20, 30 and 40dB attenuators which are switched by a rotary switch and either are used individually or ganged up.
So the 100 dB attenuation, for example, uses all four sections, added up.

In all the units, I own three of them, I find the 10 dB attenuation section only providing about 5dB and this error is carried thru the next few steps, I do get 15 dB instead of 20 dB of attenuation and so on.

This is peculiar since up to 40 dB each position of the rotary switch uses a different one of the 4 sections, and yet the 5dB error continues, even when switching to the the other segments of the attenuato module.

Clearly one would think that this has to do with a bad contact in the rotary switch, an error independant of the 4 different attenuators of the unit.

I opened these units, they are beautifully made. And they looked pristine inside.

While I was in there I checked as many of the resistors as I could without having to unsolder them and they all seem to be right on, no signs of any abuse.
Then I used some contact spray on the rotary switches and worked them furiously.
I also measured each of the sliding contacts of the rotary switch, thinking (hoping!) that I would find a really bad one, but no, everything looks just fine, contact resistances of a fraction of 1 Ohm by my mutlilmeter and yet they have this 5 dB or so error, all of them.

This is why I thought an HP technician, someone who has seen many of these and maintained a few, would be my best bet to find out what systemic error I am running up against???

Uwe

 

[JimB]

How are you measuring the attenuation? Is that accurate.

What is the output impedance of your signal source?

What is the impedance of your load and level meter?

If the system is not matched, all bets are off, there WILL be errors.

JimB

 

[Othello]

Yep, you are right, I don't know what I am doing...

For the input I use an HP 200 series oscillator with 600 ohm, a perfect match for the attenuator.
The output from the attenuator goes into a Sennheiser UPM 550 AC voltmeter which as an input impedance of 1 M.
So I stuck a 680 ohm Resistor across the output jacks of the attenuator and apparently one end did not make such good contact.

I know, very funny.

When you wiggle that one the attenuation is very close to 10dB on the first step, as it should be.

I will stick my head in the snow for a few days before I say anything again.

Uwe, in snowy Maine