Re: Transformer Frequency Response Problem

mr electron <schmillivolt@xxxxxxxxx> wrote:

I am attempting to plot the frequency response curves of several UTC
interstage transformers. The problem is there is a nasty ~15dB dip in
the frequency response curve centered at around 16KHz which takes up
the entire last octave (10KHz - 20KHz). I tried several transformers
with identical results.

So, it's an actual dip and not a shelf?

My setup is as follows:

HP 200 Oscillator (unbalanced)>15K series resistor> A18 primary(rated
at 15K imp)

WHAT if you remove that series resistor, and drive it from a lower
output impedance? Does that fix the problem?

A18 secondary(rated at 80K imp)> 90K resistor across secondary> HP
400F AC voltmeter

Sounds good.

You should know that if you have any old UTC types with an 80K secondary,
you should consider yourself lucky to get decent response at 10 KC. It's
very hard to make such a high-Z winding, and the interwinding capacitance
kills you.

Here is a copy of two units I plotted, UTC calls a frequency response
of 20 - 20Khz +/- 2dB:

http://i186.photobucket.com/albums/x233/schmillivolt/a18.gif

What am I missing here. seems like there is some resonance comming
into play. What am I doing wrong?

You're using a transformer that was designed for efficiency rather than
best possible response. Try and drive it with a lower-Z source and
notice what that does to the response.

You should also note that the 200CD output isn't necessarily constant
with level, so try and measure both the input and output levels when
you do your measurements.

Looking at 1 KC square wave response is a good way to judge the rough
quality of a transformer and what it's bandwidth is. You can see high
end and low end rolloff issues easily but you can also see ringing and
other impulse response issues.
--scott
--
"C'est un Nagra. C'est suisse, et tres, tres precis."
.

Relevant Pages

  • Re: Transformer Frequency Response Problem
    ... I tried several transformers ... it's an actual dip and not a shelf? ... you should consider yourself lucky to get decent response at 10 KC. ... Makes me wonder how UTC derived at their response curve. ...
    (rec.audio.pro)
  • Re: Neve "clones"
    ... Well I was responding to your comment about transformers being less coloured ... better frequency response, distrortion etc. ... All of the old core materials are gone. ... market for commodity transformers is gone or going away, ...
    (rec.audio.pro)
  • Re: Transformerless microphone - why?
    ... Also of course if they have a powered microphone, ... use said power for the circuit. ... What advantage over ones with transformers ... the high frequency response will ...
    (uk.rec.audio)
  • Audio transformer questions
    ... I notice that inexpensive audio transformers have limited low frequency ... response and are invariably offered as either step-down or 1:1. ...
    (rec.audio.tech)
  • Until Candy costs the advertisings sneakily, Ayub wont insure any disappointed woods.
    ... You won't dip it. ... last in general the mainland, if Bill boldly manages the republic? ... Get your overall accelerating portion in response to my rebellion. ... If the optimistic horizons can ...
    (rec.arts.drwho)