restoration of a ENI 325LA power RF amplifier

[JacquesBBB]

In my favorite dumpster, I got yesterday a ENI 325LA RF power amplifier.

Characteristics :

- 50 Db
- 250khz - 150 Mhz
- 25 w

I powered it and it seems to work, but if I input a square wave, it get pretty much distorted :
In the attached pics,
the ch1 ( yellow) is the input signal, and the ch2 (blue) is the output signal from the ENI 325LA.
 
Note that I had  only a usual 50 ohm terminal plug, so I did not push the power very far, and stayed below 1 w.

For Sine wave, there is no problem, and the  amplification seems quite OK.

I checked the power supply, but it seems to be  well in par with the  manual
(25.62 V 26.52V)  with not much noise (see the pics).

I have thus several questions :

- Is the behavior of this RF amplifier normal ?  They claim a flat response in the
 250khz - 150 Mhz range. I would expecte that the  (nearly)  square wave should be conserved.

- How to fix it ?
Besides the power supply, I  do not know much on these RF amplifiers.

Thanks for any hint.

[JacquesBBB]

I realise I put a pic in the previous post that was out of the  range of the amplifier.
Here are two pics  that are with 975khz frequency, so they should be well in the range,
including their harmonics.

Can someone tell  me  if  such distortion is expected, or is the amplifier faulty ?

[jfphp]

Funny to test a linear power amplifier with a square wave !

[JacquesBBB]

I   am far from a specialist, but I dont see the problem.

If you use a sine wave, you test only one harmonic.

If you use a square wave, you test at the same time all the harmonics. Unless I am wrong,  If an amplifier
is even on all the spectrum range it advertises, it should give an output with the same shape as the input.

[edpalmer42]

I'm not an RF expert either, but to get a square wave to come through an amplifier, you have to have both flat frequency response and flat phase response.  Look at all the inductors in that unit!  Do you really think that the harmonics are going to line up right?

AFAIK, the job of an RF amp is usually to amplify at most a few discrete frequencies plus closely-spaced modulation products, not harmonically-related signals.  Normally, harmonics are a bad thing!

Ed

[m3vuv]

is this class a, ab, c etc realy need to know the class of amplification to comment. 73 m3vuv.

[JacquesBBB]

This is a class A amplifier. Attached is the doc.

Indeed, it states Table 1.1
"All harmonics more than 23dB below main signal at 20 Watts output. Lower at reduced power output."

I was at about 1 Watt output or less.

But I dont understand then the relation with
"The ENI Model 325LA is an solid state amplifier which has a flat response from 250khz to 150 Mhz.
It provides up to 25 watts of linear power with low harmonic and intermodulation distortion".

Can someone explain ?
( I know nothing about RF amplifiers ).

[dave_k]

"All harmonics more than 23dB below main signal at 20 Watts output. Lower at reduced power output."

This will be any harmonics created by the amplifier itself, when fed with a pure sine wave at one frequency.

( I know nothing about RF amplifiers ).

What are you doing with one, then?

[floobydust]

The whole thing is Class C?
I keep looking for a bias voltage (adjustment) for the first stage but it looks like somebody messed around with that, changing parts.
I would draw a schematic of the input stage, well the whole thing, but I can't see the input stage (first transistor) biased into linear mode.

I'd replace the antique Sprague electrolytic cap, it's surely open.

[JacquesBBB]

"All harmonics more than 23dB below main signal at 20 Watts output. Lower at reduced power output."

This will be any harmonics created by the amplifier itself, when fed with a pure sine wave at one frequency.

OK thanks, now it make sense.

( I know nothing about RF amplifiers ).

What are you doing with one, then?

Curiosity. I know nothing, but I am willing to know. I know already more than two days ago. I like to do things that I have never done before. Opening this piece of equipment is like reading a new book. I had the chance to get it for free in the dumpster. It is beautiful construction, beautiful high end electronics, and I want to see it working properly.

Maybe it is already working well, and this is what I want to know. I want also to learn how to use it.

[JacquesBBB]

The whole thing is Class C?

It is written class A in the doc.

I keep looking for a bias voltage (adjustment) for the first stage but it looks like somebody messed around with that, changing parts.

I dont agree. It looks to me that nothing has been changed in this device which seems to date from 1985.
 

I would draw a schematic of the input stage, well the whole thing, but I can't see the input stage (first transistor) biased into linear mode.

The doc is in my previous post with a full schematics.

I'd replace the antique Sprague electrolytic cap, it's surely open.

I checked the value and ESR of the caps ( 5OuF in the power supply and 12000 uF from Sprague).
They are  perfect (0.35 ohm ESR) . I dont  see why I would change them.
 
I checked the whole power supply.  Which looks fine. The only thing I found is that the  power transistors have very different gain values, but this may not be a problem as they are supposed to be used at saturation.

The output of the PSU was 26.52 V, not far from the 26.4 V of the manual. I   tuned it back to 26.4V, but I doubt its an issue.

[jfphp]

On the manual there is a performance test chapter. Follow the insructions and eventually replace the 8601A with another sweeper with similar harmonics, subharmonics and spurious . Not evident as most of sweepers don't go down to 1 Mhz and have very bad data between 10 mhz and 2 Ghz. Either take a 8640A/b and turn the button and change the ranges while observing the power meter, or take a 8753A/B/C/D. The manual does'nt need a square wave generator ! Last but not the least : meeting in Seigy France in april...

[floobydust]

OK, I missed the posted schematic, it's Class A and linear. I would not "saturate" it or inject square-waves, that seems mean.

I would measure quiescent DC voltages on each transistor, that would tell how healthy things are.
The Sprague 50uF cap looks 40 years old, so I replace those on sight, the rubber bung can dry out like those Philips blue ones.

Service manual says the bias resistors are *300R "* indicates variable value" and I see an extra in parallel for the input stage. Most are 360R.
The factory may match transistors and select bias by hand, although each stage does have local feedback.

The RF transistors look like Electronic Navigation Industries house-numbered. I think that would be difficult to procure if you have a failed transistor.
Motorola ENI-16 I could not find.

[capt bullshot]

You don't expect such an amplifier to reproduce a square wave. IMO this thing is OK, nice catch BTW.

Typically they're used to amplify the sine wave output from a signal generator to higher power levels.

One of their main usage is EMC testing for conducted imission - the amplifiers output is applied to a DUT (which way is specified in the standards), and the signal generator sweeps a range of frequencies while the DUT is observed for malfunctions ...
You could set up a nice AM/FM radio transmitter with it.

To roughly test it (if you don't have a suitable dummy load), connect it's output to a ordinary scope probe and feed a 1MHz signal with increasing amplitude and watch for waveform distortion at max. specified output amplitude - the manual says, this is allowed (don't take this for granted with an RF amplifier, usually they _need_ to be terminated properly).