PYE P75 radio conversion to guitar amplifier

[ms660]

Originally Posted by RogerLLL

Thank you! I've also found Uncle Doug's tutorials on YouTube really informative.

No problem, I'd be careful with some of Uncle Doug's explanations:

https://www.youtube.com/watch?v=x5SSKX74DKg

Lawrence.

[RogerLLL]

I've finally made all the modifications. I built it as per my previous schematic, but found the gain with V2 was lower than with an input straight into V3. So I tried a few combinations with the load resistors feeding V2 and the filter capacitors. I tried 10k in place of the 270k, and found that the gain was significantly higher, but there was some sort of oscillation, kind of a pulsing, almost like a tremolo effect, which I took to be ripple due to insufficient filtering. So I replaced the 16uF cap with a 180uF, and that reduced the ripple a lot, but it was still there at higher gain settings. I tried different load resistors, and found that with higher values, up to 100k, the ripple increased, but that values around 15-20k worked perfectly, for both load resistors. I now have high gain without the ripple, and the amp is sounding exactly as I was aiming for.
I've updated the schematic to show what I now have in place. The master volume feeding into V4 works perfectly, and I've routed the negative feedback to this pot, which seems to work ok. TBH there's not really a noticeable difference using the negative feedback instead of ground, really hard to tell.
I've not included the tone stack yet (isolated in the schematic with a dotted box, with the dashed line showing how it's wired currently). I'll experiment with different tone options.

I didn't change any of the existing HT circuit in the end, apart from the filter cap, and haven't yet measured the voltage at the screen of V2. Not sure if this is something to look at. I'll check it tomorrow anyway.
Just to add, both V2 and V3 are incredibly microphonic. Suspect there's not much I can do about this apart from reducing the gain, as it was cited as a potential issue with V2 especially.

[trobbins]

If you get adventurous you can measure the V4 cathode voltage and calculate the idle dissipation in V4, as well as measure the turns ratio of the output transformer and plot the class A load line on the EF41 datasheet.

[Robert Gribnau]

I noticed that the feedback arrangement from the secondary of the output transformer to V4 in your last schematic (post #62) has the same phasing as in your earlier schematics in which the feedback was still supplied to V3.

However, the phasing of V4 is opposite to the phasing of V3. So if the feedback to V3 was indeed negative (like it is in the Pye P75) than without reversing the connections at the secondary of the output transformer, the feedback to V4 would be positive instead of negative.

I also noticed that you lowered the value of the anode resistor of V2 to 18K. With that value, the gain of that stage will be rather low. I suppose that is why trobbins suggested to plot the loadline on the EF41 datasheet. If you would change this resistor back to your earlier 270K, than perhaps an extra screen grid resistor at V2, together with a decoupling capacitor from the screen grid to ground, would be in order so as to not have the screen grid voltage much higher than the anode voltage.

[RogerLLL]

The 270uF load resistor on V2 produced a much lower volume than inputting directly into V3, so I tried different values of the anode load resistor and ended up with a sweet spot at 18k, which gave maximum volume and no ripple with the two 180uF filter caps in place. I've reinstated the negative feedback on V3 which works fine.
I've measured all HT voltages as in the revised schematic. Just to add, the V2 anode voltage with the 270k load resistor was 7.6v which did seem too low. Currently this is measuring at 136v. The V2 screen grid voltage is measuring at 147v which is higher than the anode voltage, which I imagine is not right, so maybe a resistor in series here might make sense.

[Robert Gribnau]

I wouldn't worry about the screen grid voltage of V2 being 147 V while the anode voltage is 136 V. A relatively small difference like that will not be a problem.

You could consider putting some extra capacitance (like 10 uF) at the junction of the anode resistor of V3 and the screen grid of V2. Only 2 uF seems a bit low to me.

[Uncle Bulgaria]

Isn't the reservoir limit of the EZ41 50µF?

[ms660]

Originally Posted by Uncle Bulgaria

Isn't the reservoir limit of the EZ41 50µF?

The reservoir is 32uf in the schematic.

Lawrence.

[Uncle Bulgaria]

I see Lawrence - thank you. I was confused by the rectifier being at the bottom and followed the HT line!

[ms660]

No problem, the EZ41 is 32uF max. and the EZ40 is 50uF max. in the valve data.

Lawrence.

[RogerLLL]

Hi again, just when I'd finalised the amp and was thinking about the tone stack, I was playing through the amp and the volume suddenly started to increase, with a clicking sound from somewhere, and then no sound. I switched it off immediately. No visible blown components. Then turned it back on to troubleshoot. All voltages have increased as shown by the blue voltages in the attached schematic. Thinking this is probably an issue with the power transformer, and in turn maybe the rectifier tube as the maximum voltage of the EZ41 has clearly been exceeded (350v max with a voltage of 418v now going into it). But I could be wrong of course. Anything to check to pin down where the issue is?

[ms660]

The 390 V (AC) quoted in the Trader sheet would be the anode to anode voltage, the actual anode voltage WRT chassis would have been 195 V (AC)

418/2 is 208 V (AC)

Which in both cases is well within the rectifiers maximum input voltage limits.

Lawrence.

[Gabe001]

I reckon v4 has failed. Not your fault as long as you didn't exceed maximum anode dissipation.

Check if there is any voltage drop across its cathode resistor.

Gabriel

[RogerLLL]

I suppose if the rectifier had failed there's be no DC, which clearly there is, so that's good!
When it "blew" I was using an overdrive pedal so was definitely driving the amp with a higher level of input than just the guitar alone, so was definitely pushing the tubes harder. Could that have caused this? The voltage across the 220ohm cathode resistor of V4 is 7.1v. What would be a typical reading?

[Robert Gribnau]

7.1 V over the cathode resistor of V4 is a typical reading, which indicates that V4 and the primary of the output transformer are OK. V2 and V3 are also passing current, which indicates that they too are OK.

I have no explanation why the voltages went up though. I would check if the connections between the stages are in order (making sure that the signal can pass through the amplifier). And check the loudspeaker.

[Gabe001]

That's actally spot on for an el41. You can use ohm's law to calculate the current draw by that valve

The voltage usually goes high when something isn't drawing as much current as it should. Try doing the same test on the other valves, although the drop across their anode resistors suggests they're working as they should. Not too sure to be honest. The only other possibility I can think of is a shorted output transformer. Can you measure the resistance of the output transformer primary?

Crossed posts with robert

[RogerLLL]

The speaker's fine.
Cathode resistor for v3 (EBC41) is reading 1.3v, cathode resistor for V2 (EF41) is reading 13v. Seems high. That tube has a very mottled look to the silvering on the inside walls as in the pic.
Total resistance across the primary of the output transformer is 7.5 ohms. This is split as 6.2 & 1.3 ohms.

[Gabe001]

Ef41 looks like it's OK to me, drawing 2.8mA in total. Max is around 7mA with a bias of 2 to 22v, so looks happy to me

Can you confirm that 7.5 ohms is a+b on your circuit diagram (the valve end, not the speaker end)? If so, the output transformer is kaput. This should be around 150ohms. It would explain why there is no voltage drop across it

[RogerLLL]

I can confirm that the resistance measurements were taken across the primary of the OT ie the amp side not the speaker side. Sounds like that's the culprit then! Could this have blown through driving the tubes too far? It was much louder than just driving the tubes using gain setting, though this is common practice with guitarists using tube amps. Much more volume also than what the radio would have ever generated, so I do wonder if I need a higher rated OT for my intended use of the amp?

[Gabe001]

Just for completeness could you remove/snip off c23 and re-measure?

If you want to play loud, I would suggest more powerful output valves and suitably rated transformers.