Susy and test equipment.

[Netlist]

Being the proud owner of a pair of AlephX’s and recently an old but nice working distortion analyser, I started testing the amp with the analyser.
Since AlephX is a bridged amp, one cannot connect test equipment in a normal way.
It would be nice if people experienced in measuring such amps could explain a bit on how to hook up instruments to bridged and in particular X-amps.
Like Henrik said, measuring each channel apart will not give accurate figures either as the X-effect is not taken into account.
From a practical experience I notice that the X-circuit can be measured only with low output levels. (Max 1V RMS)
The results at higher levels, say 10V are not at all accurate. Something like 30% distortion which is definitely not correct.
Dave proposed the opamp and the battery here
Do we have to take another approach?

Nelson, Grey, others..., please come in.

/Hugo – with a nice smile

[Helix]

Temporarily removing the safety earth?

[uli]

Put either the amp under Test or better the analyzer on
a mains isolatin Xformer!
Then you can put the analyzer ground to e.g. -out and
the probe on +out or the way round.

Uli

[grataku]

Uli,
I don't think so. What is the isolation transformer going to do if its a earth issue?

Netlist,
I don't have a distortion meter...how does that work? It has se input and output right? I have a differential probe but I think one can use a high quality balanced-se converter with high impedance input and then calibrate the whole measuring circuit to account for the THD from the converter.
I'll check out the other link you provided.

[uli]

Usually you use this setup when you want to isolate the ground
of an oszilloscope or any analyzer to measure eg bridged
dut´s

Uli

[Pjotr]

Hi,

You definitely need a differential or “instrumentation” amp as a front end for your distortion analiser. Or better an isolation instrumentation amp. You also need this with single ended amps. Otherwise ground return currents will play havoc with your measurements. Pretty good instrumentation amps are the INA chips from BurrBrown, now TI.

Cheers

[grataku]

Originally posted by uli
Usually you use this setup when you want to isolate the ground
of an oszilloscope or any analyzer to measure eg bridged
dut´s

Uli

Do you mean you don't connect the ground to the plug ground?

[Netlist]

It is an earth problem. So, the floating ground has already been separated from the chassis ground.
First tests with the mains isolation trafo are pretty good so far.
Measuring results are now within acceptable ranges.
The output (from the analyser connected to the scope) however doesn’t show the distortion as in a normal setup.
But that’s not of great concern; the percentage is properly displayed on the tester’s meter.

Grataku, may I refer to http://www.slack.com/pdf/HP334A.pdf This is a pretty good manual, covering the principles of distortion meters.

Pjotr,
This instrumentation amp has been mentioned before, that’s why I’m happy to hear it again.
Only, you use the word ‘definitely’. Now when the chassis ground is lifted, there shouldn’t be any ground problem anymore?

The setup is now like this:
1) Signal generator to SE input of amp.
2) +Out of amp to +in of distortion analyser (DA)
3) –Out of amp to floating ground of DA
4) Chassis ground of DA lifted from floating ground.
5) DA powered via mains isolator trafo
6) Amp powered via mains.
7) No other ground connections.

/Hugo

[Pjotr]

Hi Hugo,

Problem with an isolation transformer is that it only suppresses low frequency signals. There is much HF noise around, e.g. radiated from mains cables, that still enters your measurement equipment. I am using a soundcard for distortion analysis and there the gnd from de signal generator (DAC) is connected to the input gnd (ADC). In many power amps input gnd is isolated fom LS gnd by a 100 ohm resistor and there can be sustantial signal across it.

I have used an INA103 as a front end with switcheable gain between 1 and 100 according to the app. note on the data sheet and a switcheable attenuator of 1, 10 and 100 in front of it for large signals. The INA103 is not that expensive, around 15 EU.

Cheers

[Netlist]

Originally posted by Pjotr
Hi Hugo,

Problem with an isolation transformer is that it only suppresses low frequency signals. There is much HF noise around, e.g. radiated from mains cables, that still enters your measurement equipment.

Hmm, nice, but isn't that type of circuitry already build in the DA?
Looking at the manual section IV, they talk about a rejection amplifier. Or is that something else?

Another thing: Is it essential to connect a dummy load, say 8ohm
to the amp? A few days ago I had to do that to get some decent measurements. With the new setup it looks like it's not necessary anymore. Besides that, connecting a load results in the DA starting to smell pretty bad at higher power levels.

/Hugo