Strange audio output transformer - UK Vintage Radio Repair and Restoration Discussion Forum

valvekits · 24th Jan 2010, 3:51 pm

I was wondering if anyone can explain the output transformer action of this 1958 Murphy stereo radiogram. I haven't come across this arrangement before and it almost looks like part of the output transformer is being used as choke for the preamp section.

Eddie

jjl · 24th Jan 2010, 4:09 pm

Eddie

The output transformers have a hum cancelling winding. This sort of arrangement was quite common in sets with single ended output stages.

John

Leon Crampin · 24th Jan 2010, 7:48 pm

Pioneered I think by Philips before the war. It's a good scheme which works well with the slight disadvantage that the transformer is also loaded by the series resistor at signal frequencies (the smoothing capacitor being a short circuit). As the number of turns is small, the effect is not very significant.

It works by ensuring that equal ampere turns of hum current flow in each portion of the primary, phased so that they cancel rather than sum. So, if i1 of hum current flows through the output valve and its portion of the winding has turns n1, it must be exactly balanced by the hum current i2 flowing through the resistor (remember the smoothing capacitor is a short circuit to hum) in the section feeding the rest of the set, which has turns n2.

Hence if i1 x n1 = i2 x n2 the hum current produces no net flux in the transformer core, and there is no hum induced in the secondary.

The current i1 is goverened by the valve AC resistance ra (not Ra) and the current i2 governed by the series resistor - ignoring winding resistances.

By Ohms Law, i1 is proportional the 1/ra and i2 to 1/R where R is the series resistor.

So, n1/ra = n2/R for hum cancellation or R = ra (n2/n1).

I reverse engineered the same circuit in the Philips 170A when designing a replacement output transformer. For this set, the turns ratio measured was 2708/74 and the value of ra for an EBL31 is given by Mullard as 55k Ohm.

n2/n1 = 74/2708 so R = (74/2708 x 50000) = 1503 Ohms.

Philips fit a 1k5 Ohm resistor in this position.

Circuits of this type will not function correctly if the output valve characteristics (ra particularly) are changed - see the thread on "EBL31 adapters".

Leon.

valvekits · 27th Jan 2010, 11:39 pm

Thanks for the detailed explanation Leon. It all seems so obvious once you know how! I would never have guessed from just looking at the schematics

Eddie

24th Jan 2010, 3:51 pm	#1
valvekits Pentode Join Date: Dec 2008 Location: Nottinghamshire, UK. Posts: 239	Strange audio output transformer I was wondering if anyone can explain the output transformer action of this 1958 Murphy stereo radiogram. I haven't come across this arrangement before and it almost looks like part of the output transformer is being used as choke for the preamp section. Eddie Attached Thumbnails

24th Jan 2010, 4:09 pm	#2
jjl Octode Join Date: Jan 2003 Location: Ware, Herts. UK. Posts: 1,082	Re: Strange audio output transformer Eddie The output transformers have a hum cancelling winding. This sort of arrangement was quite common in sets with single ended output stages. John

24th Jan 2010, 7:48 pm	#3
Leon Crampin Octode Join Date: Nov 2005 Location: Surrey, UK. Posts: 1,874	Re: Strange audio output transformer Pioneered I think by Philips before the war. It's a good scheme which works well with the slight disadvantage that the transformer is also loaded by the series resistor at signal frequencies (the smoothing capacitor being a short circuit). As the number of turns is small, the effect is not very significant. It works by ensuring that equal ampere turns of hum current flow in each portion of the primary, phased so that they cancel rather than sum. So, if i1 of hum current flows through the output valve and its portion of the winding has turns n1, it must be exactly balanced by the hum current i2 flowing through the resistor (remember the smoothing capacitor is a short circuit to hum) in the section feeding the rest of the set, which has turns n2. Hence if i1 x n1 = i2 x n2 the hum current produces no net flux in the transformer core, and there is no hum induced in the secondary. The current i1 is goverened by the valve AC resistance ra (not Ra) and the current i2 governed by the series resistor - ignoring winding resistances. By Ohms Law, i1 is proportional the 1/ra and i2 to 1/R where R is the series resistor. So, n1/ra = n2/R for hum cancellation or R = ra (n2/n1). I reverse engineered the same circuit in the Philips 170A when designing a replacement output transformer. For this set, the turns ratio measured was 2708/74 and the value of ra for an EBL31 is given by Mullard as 55k Ohm. n2/n1 = 74/2708 so R = (74/2708 x 50000) = 1503 Ohms. Philips fit a 1k5 Ohm resistor in this position. Circuits of this type will not function correctly if the output valve characteristics (ra particularly) are changed - see the thread on "EBL31 adapters". Leon.

27th Jan 2010, 11:39 pm	#4
valvekits Pentode Join Date: Dec 2008 Location: Nottinghamshire, UK. Posts: 239	Re: Strange audio output transformer Thanks for the detailed explanation Leon. It all seems so obvious once you know how! I would never have guessed from just looking at the schematics Eddie