Behringer DCX2496 digital X-over

[mige0]

Lets not loose track on the interesting small signal transformer topic, guys .

Here is my today's measurement of one I have available.
Its a 1:1 purchased in Germany way back and though I do not really know its origin, I have a strong feeling it possibly could be a re-labelled Amplimo TM8 – dunno.

Measurement was done for convinience directly with my soundcard which specs as

source impedance 100 Ohm balanced
load impedance 20kOhm balanced

STEPS was calibrated so we see that the transformer was driven by ~ 1Vrms




We can see here as well that 3rd order is dominating – here all the way up to 10kHz where 2nd takes over


Comparing to Gary's Lundahl we see that this one performs ~30dB worse at around 1kHz – I guess it costs only 1/10 the Lundahl tho...
http://www.diyaudio.com/forums/multi...ml#post2042289

Michael

[SQLGuy]

Am I reading correctly? 1% distortion at 20Hz?

[mige0]

Hmmm, actually 10% distortion @ 20Hz, I'm afaid.


Michael

[Gary P]

Hi Michael,

Have you run a baseline loop back test with Steps? The rise in distortion above 4K is an artifact of my sound card so the distortion of the Lundahl is better than my measurement system...

With the noise floor where it shows in Steps, I suspect that Steps is running in 16 bit mode. I've not had time to look and see if the resolution is configurable in Steps.

Gary

[Pano]

Thanks for the measurement Michael.
I have seen a rise in 3rd order distortion in the bass for small transformers. Thats one reason you can't go too small.
But otherwise the harmonics track well. Yes, odd order is dominant, but everything seems to track well together.

I wish I could get Steps to work for me. I always get glitches, but never at the same place. This results in nasty distortion spikes. No other software does that on my setup. Might have to try a different computer.


Gary. I did try balanced into the DUO soundcard and got somewhat better results - thanks! I was going to post them, but then saw that I had done different scales and FFT values. =( Will do over. Mostly what I'm seeing is that the transformer output is much cleaner in the top end than the DCX active output. Harmonically speaking.

[mige0]

Originally Posted by Gary P

Have you run a baseline loop back test with Steps? The rise in distortion above 4K is an artifact of my sound card ..

Well I did - but looking twice sometimes isn't enough.
Thanks for pointing there.

I checked and found that my setting in STEPS was on 32 bit – no good - inserting line drives feeding the adjustable mic-input plus checking ground connects brought another improvement...


###########

Repeated my measurement under optimized conditions.

To allow for some further interesting insights I added measurements on –20dBV and on –40dBV


Michael

[mige0]

Michael

[mige0]

Michael

[mige0]

Michael

[mige0]

Looking at this row of measurements we see that the increase after the slope clearly have been from artefact's.

What else we see is that the 3rd order - obviously form saturation – fully dominates distortion.
Interestingly this plays a big role up into the voice region – depending on signal level.

Also it becomes clear that this very transformer is of no adequate performance on line levels – but possibly could be OK for mic levels

Michael