[AMRadio] Class AB and B audio XFMRS

W5OMR/Geoff w5omr at satx.rr.com
Sat Mar 11 12:16:22 EST 2006

John E. Coleman (ARS WA5BXO) wrote:

>	I have been doing some experimenting here on the side and have found
>a interesting thing that I've seen before but forgotten about.  I thought I
>might share it here with you guys.
>	When using a Class B or AB push pull output stage into an audio XFMR
>such as Modulation XFMR or output XFMR for Speakers, there is of course a
>cross over problem that must be contended with.  What many don't understand
>is that the XFMR is the deciding factor for determining the lowest frequency
>at which some distortion will NOT occur.  If the tubes are driven so that
>peak current level is held constant for all frequencies, you will notice, as
>the frequency of the driving signal is reduced you will find a frequency
>point at which the output measured at the secondary begin to drop.  The
>distortion that occurs in the shape of the signal is at this point is a
>result of XFMR saturation and cross over distortion.  Reducing the bias and
>drive will correct the distortion, at the expense of less output, but as the
>frequency is reduced it will come back again.  Trying to pass frequencies
>that are too low for your XFMR can cause intermodulation and harmonics, as
>well as cause RF oscillations in the output of an audio amplifier or
>modulator stage.  The distortion when viewed on a scope will appear as cross
>over distortion.  It looks like a flat spot between the peaks.  It is there
>at say 20 CPS but not at 100 CPS.  This is because the XFMR can hold and
>release it magnetic flux energy over 1/100 of a second but it can not hold
>it for 1/10 of a second.  It releases early and changes the shape of the
>induced output wave on the secondary at the lower frequency.
>More iron is better, have no or little gap in the E-I sections of the XFMR
>and keep the current out of the secondary and balanced in the primary.
>Hope this helps some one.

I can't believe that no one else has replied to this, on the reflector.

Certainly, it helps me understand more about what's going on in my 
transmitter.  I wonder if reversing the phase on the grids of the 
modulator would help eliminate some of the cross-over distortion?  I 
know that the peaks won't be as high (and this might be a -good- thing, 
in my case).

What would the effects be, John, if someone were to introduce some 
negative feed-back around the whole of the modulation transformer as a 
couple of WA1HLR mods suggest?

Tim says this of the Johnson Ranger:
"Negative feedback around modulator is retained, but fed into the 
cathode of the 3rd audio stage. Modulator resting current should be set 
for 65 - 70 mA for 6L6 tube types, 80 - 90 mA for 6CA7 and 90 - 100 mA 
for the 6550 and 8417.

Once the modulator has been reworked and properly set up with all 
modifications in place you will get good clean audio and resonance from 
80 - 15,000 WITH THE STOCK IRON! The only factor limiting deep bass 
response is the undersized modulation transformer. A better trasformer 
could be used but much relocating of components to be done - to be 
covered later."
(ref: http://www.amwindow.org/tech/htm/rangtron.htm)

In my case, of course, I need to maintain a peaceful accord with the 
neighbors.  Therefore, I only run 'high' power at night-time.  That is, 
sometime after Leno and Letterman have gone off the air, and most folks 
go to sleep, so they can get up and go to work in the morning.  So, 
around 12:30 or 1am, and then I reduce back to 'low' power at 6am, so 
that blenders, microwaves and toasters are not doing the St. Vidas Dance 
on the counters in the neighborhood (or so they claim, but that's 
another discussion...)

By High power, I mean 1500V on the plates, at 250mA, for 375w DC Input.
1500v @ 250mA also = 6000ohms of impedance.  By using the 'rule of 
thumb' for 8Hy for 1000ohms of Impedance, to get the total number of 
Henries required to shunt the DC off the secondary of the Modulation 
transformer, I come up with a figure of 48 Henries, @ the current 
required.  In my case, I use 500mA chokes.  300mA would do, but I don't 
have any of those, sooo...

however, when I'm running LOW power, the final plate voltage stays the 
same, and I pull the link out, and now the transmitter is down to 100mA 
@ 1500V in the final, makes for a Final Z of 15,000 ohms.  Well, 42 Hy 
just ain't gonna cut it.  NOW, I need something more like 120Hy.  
Certainly, it's -much- easier to leave the reactor in place, regardless 
of how little power I run.

That being said, I do still wonder if reversing the phase on the input 
to the modulators would smooth out some of the cross-over distortion?

I'd like to see -lots- of discussion on this, here on the reflector ... 
Don/KYV?  Bacon? 

73 = Best Regards,

More information about the AMRadio mailing list

This page last updated 18 Oct 2017.