Output transformer - UK Vintage Radio Repair and Restoration Discussion Forum

ITAM805 · 27th Jul 2014, 12:32 pm

hi all

I have a 60W Marshall amp here that kept blowing HT fuses, had a 100Hz buzz and "wasn't very loud". I carried out various repairs on it (new o/p pair, o/c screen resistor) which seems to have cured the fuse popping and buzz and the amp now sounds ok - except it's only producing around 28W before the onset of clipping? All voltages appear ok so I wondered the o/p tx is faulty?

The primary resistance measures around 40Ω between either side to the centre tap. I injected an AC signal into the secondary at 1V @ 1kHz, this produces only around 7V on either primary to c/t ?

Any thoughts please?

Many thanks

vidjoman · 27th Jul 2014, 2:52 pm

Are you measuring your 28 watts the same way that they measured theirs? If theirs is music power that will equate to about 30 watts RMS.

ITAM805 · 27th Jul 2014, 4:18 pm

hi vidjoman,

Marshall quote 60W RMS for this amp. I'm wondering if the 7V out for 1V in sounds right for an valve o/p tx? I thought it was rather low

ukcol · 27th Jul 2014, 6:13 pm

Agreed, very low. The voltage ratio, in a healthy transformer, will be the same as the turns ratio.

Cobaltblue · 27th Jul 2014, 6:38 pm

Don't know much about Marshall but if the Output Valves are a pair of EL34's then the Anode to Anode load could be 3 to 4Kohm and if it's a 15 Ohm speaker output the the turns Radio could be as low as 7+7:1

I think more information is needed

Cheers Mike T

GrimJosef · 27th Jul 2014, 6:45 pm

... which will be the same as the ratio of the square roots of the impedances. So if the full primary (anode-to-anode) has 14 times as many turns as the secondary and, say, the secondary is loaded with a 4ohm speaker then there will be 14-squared times 4ohms between the output valve anodes. That's less than 800 ohms which does indeed sound too low.

EDIT: as Mike says, it would be more-or-less OK with a 15ohm speaker though. How did you measure the 1V fed into the secondary and the 7V on the primary sections ?

Cheers,

GJ

Ed_Dinning · 27th Jul 2014, 7:36 pm

Hi Gents, I have seen several output transformers where the sound level has been disappointing, but voltage readings appeared OK. This was often due to shorted turns on the primary. I'm not sure if the "ring test" will apply to audio transformers.

If many turns are shorted it will show up as resistance variations between each pri half, but there is normally a variation here with the "outer" winding having a higher resistance due to the longer mean length of turn.

Working backwards, it is possible to calculate the voltage into an 8R speaker running at 30 watts. Apply this voltage at 50Hz to the sec winding and look for balanced and undistorted outputs from each pri half of the correct voltage level (turns ratio)

Ed

ITAM805 · 27th Jul 2014, 7:37 pm

Thanks for the replies chaps

Mike T, what you say is interesting as the amp is a TSL601 and it indeed uses a pair of EL34's and outputs into either 8 or 16Ω. HT is around 475V.

Just to throw a little confusion into the mix though, I have another 65W amp here (Proamp VSQ65) that uses a pair of 6550's, and doing the same test I get the same 7V results

Just to clarify the process, I'm using an valve Advance J2 genny on the 5Ω output, plugging this into the output jack socket of the amp, then increasing the output until I get 1V on the DMM - then connecting one meter probe onto the c/t, t'other on either of the anode taps.

This youtube clip shows the process

ITAM805 · 28th Jul 2014, 3:01 pm

hi chaps

On further investigation I discovered that the new 1K screen resistor (R48) that I fitted was reading 250R? When I pulled it I found it to be good but that the traces were still reading 340R? Then I found that C26, a 22pF was the culprit and if I waggled it it dropped to a dead short

So this means that the signal at V5 anode was shorted to the screen and was being bypassed through the filter caps C28/34.

So I'm hoping this would account for the amp being so gutless and that the tx is in fact ok?

Omegaman · 28th Jul 2014, 4:32 pm

Hi,

I've had several of these through the workshop with both of C26 and C27 dead short. It would seem that this effectively puts the output valves in 'triode' mode thereby reducing the power.

Hope this helps

G6Tanuki · 28th Jul 2014, 4:38 pm

Quote:

Originally Posted by Omegaman

I've had several of these through the workshop with both of C26 and C27 dead short. It would seem that this effectively puts the output valves in 'triode' mode thereby reducing the power.

What's the voltage-rating of the capacitors used here? I can easily imagine how a guitar-amp driven from a fuzzbox, and with the volume turned up to '11' [so it's essentially being expected to amplify near-square-waves] could make the OPT generate some seriously high-voltage and capacitor-life-shortening transients.

ITAM805 · 28th Jul 2014, 5:10 pm

hi Howard, thanks for the info - I'll order a couple 1kV replacements and hope that's the fix!

G6, they were 500V, but as you say, it's a pretty hostile environment inside a guitar amp at full tilt

turretslug · 28th Jul 2014, 5:45 pm

Quote:

Originally Posted by Omegaman

I've had several of these through the workshop with both of C26 and C27 dead short. It would seem that this effectively puts the output valves in 'triode' mode thereby reducing the power.

And resulting in R48/R49 shunt damping the primary, effectively now between anode and AC ground (C28/C34).

G6Tanuki · 28th Jul 2014, 7:18 pm

Quote:

Originally Posted by ITAM805

G6, they were 500V, but as you say, it's a pretty hostile environment inside a guitar amp at full tilt

Indeed - you need to spec these sorts of components with lots of leeway.

As a bare minimum, think that in the anodes/screen-grids of the P-P output-stage your capacitors should be rated to handle the HT-supply plus the potential audio-voltage swing [which will be +/- the HT-rail]. I'm honestly thinking 2.5Kv components here!

You've got 500-odd worst-case HT volts DC. And a push-pull AC audio-voltage of the same order as long as everything's remaining linear. Add in clipping and the P-P transformer's inductance becomes a horrible voltage-multiplier.

It may seem odd to massively over-specify but in customer-relationship and reputational terms a couple of pounds spent on bomb-proof parts is a lot cheaper than having to pay the carriage/bench-time to replace parts that failed after a few months of Heavy-Metal 110% load.

trobbins · 28th Jul 2014, 10:24 pm

The half-primary leakage inductances can really let the anode voltages flying a guitar like amp, both above 2x HT, and below 0V (measurements were made on another forum). Unless some other form of anode voltage containment is used (eg. commonly found old favourite styles were RC's, and diode to 0V, and nowadays solutions using MOVs) then 1kV would be bare minimum and in an X style if possible (not sure if X rated caps ever came in values like 22pF !), but preferably up to 2.5kV for surety.

27th Jul 2014, 12:32 pm	#1
ITAM805 Nonode Join Date: Jul 2010 Location: Folkestone, Kent, UK. Posts: 2,172	Output transformer hi all I have a 60W Marshall amp here that kept blowing HT fuses, had a 100Hz buzz and "wasn't very loud". I carried out various repairs on it (new o/p pair, o/c screen resistor) which seems to have cured the fuse popping and buzz and the amp now sounds ok - except it's only producing around 28W before the onset of clipping? All voltages appear ok so I wondered the o/p tx is faulty? The primary resistance measures around 40Ω between either side to the centre tap. I injected an AC signal into the secondary at 1V @ 1kHz, this produces only around 7V on either primary to c/t ? Any thoughts please? Many thanks

27th Jul 2014, 2:52 pm	#2
vidjoman Dekatron Join Date: Jun 2013 Location: East Sussex, UK. Posts: 3,326	Re: Output transformer Are you measuring your 28 watts the same way that they measured theirs? If theirs is music power that will equate to about 30 watts RMS.

27th Jul 2014, 4:18 pm	#3
ITAM805 Nonode Join Date: Jul 2010 Location: Folkestone, Kent, UK. Posts: 2,172	Re: Output transformer hi vidjoman, Marshall quote 60W RMS for this amp. I'm wondering if the 7V out for 1V in sounds right for an valve o/p tx? I thought it was rather low

27th Jul 2014, 6:13 pm	#4
ukcol Rest in Peace Join Date: May 2003 Location: Harlaxton, Lincolnshire, UK. Posts: 3,944	Re: Output transformer Agreed, very low. The voltage ratio, in a healthy transformer, will be the same as the turns ratio.

27th Jul 2014, 6:38 pm	#5
Cobaltblue Moderator Join Date: Dec 2010 Location: Exeter, Devon and Poole, Dorset UK. Posts: 6,865	Re: Output transformer Don't know much about Marshall but if the Output Valves are a pair of EL34's then the Anode to Anode load could be 3 to 4Kohm and if it's a 15 Ohm speaker output the the turns Radio could be as low as 7+7:1 I think more information is needed Cheers Mike T __________________ Invisible airwaves crackle with life or at least they used to Mike T BVWS member. www.cossor.co.uk

27th Jul 2014, 6:45 pm	#6
GrimJosef Dekatron Join Date: Sep 2007 Location: Oxfordshire, UK. Posts: 4,311	Re: Output transformer ... which will be the same as the ratio of the square roots of the impedances. So if the full primary (anode-to-anode) has 14 times as many turns as the secondary and, say, the secondary is loaded with a 4ohm speaker then there will be 14-squared times 4ohms between the output valve anodes. That's less than 800 ohms which does indeed sound too low. EDIT: as Mike says, it would be more-or-less OK with a 15ohm speaker though. How did you measure the 1V fed into the secondary and the 7V on the primary sections ? Cheers, GJ __________________ http://www.ampregen.com

27th Jul 2014, 7:36 pm	#7
Ed_Dinning Dekatron Join Date: Jan 2004 Location: Newcastle upon Tyne, Tyne & Wear, UK. Posts: 8,194	Re: Output transformer Hi Gents, I have seen several output transformers where the sound level has been disappointing, but voltage readings appeared OK. This was often due to shorted turns on the primary. I'm not sure if the "ring test" will apply to audio transformers. If many turns are shorted it will show up as resistance variations between each pri half, but there is normally a variation here with the "outer" winding having a higher resistance due to the longer mean length of turn. Working backwards, it is possible to calculate the voltage into an 8R speaker running at 30 watts. Apply this voltage at 50Hz to the sec winding and look for balanced and undistorted outputs from each pri half of the correct voltage level (turns ratio) Ed

27th Jul 2014, 7:37 pm	#8
ITAM805 Nonode Join Date: Jul 2010 Location: Folkestone, Kent, UK. Posts: 2,172	Re: Output transformer Thanks for the replies chaps Mike T, what you say is interesting as the amp is a TSL601 and it indeed uses a pair of EL34's and outputs into either 8 or 16Ω. HT is around 475V. Just to throw a little confusion into the mix though, I have another 65W amp here (Proamp VSQ65) that uses a pair of 6550's, and doing the same test I get the same 7V results Just to clarify the process, I'm using an valve Advance J2 genny on the 5Ω output, plugging this into the output jack socket of the amp, then increasing the output until I get 1V on the DMM - then connecting one meter probe onto the c/t, t'other on either of the anode taps. This youtube clip shows the process

28th Jul 2014, 3:01 pm	#9
ITAM805 Nonode Join Date: Jul 2010 Location: Folkestone, Kent, UK. Posts: 2,172	Re: Output transformer hi chaps On further investigation I discovered that the new 1K screen resistor (R48) that I fitted was reading 250R? When I pulled it I found it to be good but that the traces were still reading 340R? Then I found that C26, a 22pF was the culprit and if I waggled it it dropped to a dead short So this means that the signal at V5 anode was shorted to the screen and was being bypassed through the filter caps C28/34. So I'm hoping this would account for the amp being so gutless and that the tx is in fact ok? Attached Thumbnails Last edited by ITAM805; 28th Jul 2014 at 3:19 pm.

28th Jul 2014, 4:32 pm	#10
Omegaman Hexode Join Date: Dec 2009 Location: Leeds, West Yorkshire, UK. Posts: 250	Re: Output transformer Hi, I've had several of these through the workshop with both of C26 and C27 dead short. It would seem that this effectively puts the output valves in 'triode' mode thereby reducing the power. Hope this helps __________________ Howard G7AJN/M3OCL "How hard can it be?" - Jeremy Clarkson

28th Jul 2014, 5:10 pm	#12
ITAM805 Nonode Join Date: Jul 2010 Location: Folkestone, Kent, UK. Posts: 2,172	Re: Output transformer hi Howard, thanks for the info - I'll order a couple 1kV replacements and hope that's the fix! G6, they were 500V, but as you say, it's a pretty hostile environment inside a guitar amp at full tilt

28th Jul 2014, 10:24 pm	#15
trobbins Heptode Join Date: Oct 2009 Location: Melbourne Australia Posts: 901	Re: Output transformer The half-primary leakage inductances can really let the anode voltages flying a guitar like amp, both above 2x HT, and below 0V (measurements were made on another forum). Unless some other form of anode voltage containment is used (eg. commonly found old favourite styles were RC's, and diode to 0V, and nowadays solutions using MOVs) then 1kV would be bare minimum and in an X style if possible (not sure if X rated caps ever came in values like 22pF !), but preferably up to 2.5kV for surety.