Re-design of Pye P75 amp project

[RogerLLL]

I converted a Pye P75 chassis to a dedicated guitar amplifier as in this thread:

https://www.vintage-radio.net/forum/showthread.php?t=198257

I use the amp almost daily, and its been amazing. However, I feel it needs a little more in terms of overdrive options and tonal control. The original radio had rectifier, power valve and three RF/pre-amp valves. My current amp has rectifier, power valve and two EBC41 pre-amp valves, and I was thinking of adding a third.
I was also looking at modifying the current NFB set-up, which TBH doesn't seem to have any impact on the tone of the amp. I can't distinguish between the sound with it in or out of the circuit.
I've attached the proposed schematic. An easy modification, but thought I'd share just to see what other's thoughts might be on:

• Where the additional stage sits relative to the tone stack
• Whether the NFB circuit would be better placed on the first stage rather than the third
• Whether the power transformer is likely to struggle with the extra gain stage - there were 3 valves originally but worried that the current draw will be too high

[CambridgeWorks]

Don't know what you have done with the EZ41. It is drawn as a single diode wired across the entire secondary of mains transformer with ct going to chassis. That final part is correct.
The two ends of secondary should go to the two anodes of the EZ41 and the cathode to your ht rail.
Rob

[RogerLLL]

There wasn't a symbol for a rectifier tube, so I've created one as in the attached, and shown what I believe to be the correct connections as per the original P75 schematic.

[Gabe001]

How much more gain do you need?. An extra ebc41 with a voltage gain if, say 40, will dramatically increase your amplifier's sensitivity, up to 40 fold depending on a number of factors

Easier to switch the front end ebc 41 to a pentode. The ef40 uses the same base and will increase the amplifiers sensitivity about four fold. So, for example, you will need 100mv to overdrive the amp instead of 400mv at the input. Just wire it up as per figure 4 pg 337 on the datasheet (r_type website).

If the nfb isn't doing anything for you, and you need the gain, you could just drop it. Valves are used in guitar amps for their distortion characteristics, and nfb in guitar amps is quite uncommon.

If you're increasing the amps sensitivity, you may see an unacceptable increase in hum and noise depending in how good your grounding scheme is.

Hope this helps
Gabriel

[CambridgeWorks]

Post #3, that rectifier symbol showing connections looks fine and makes sense.
Rob

[RogerLLL]

Originally Posted by Gabe001

How much more gain do you need?. An extra ebc41 with a voltage gain if, say 40, will dramatically increase your amplifier's sensitivity, up to 40 fold depending on a number of factors

Easier to switch the front end ebc 41 to a pentode. The ef40 uses the same base and will increase the amplifiers sensitivity about four fold. So, for example, you will need 100mv to overdrive the amp instead of 400mv at the input. Just wire it up as per figure 4 pg 337 on the datasheet (r_type website).

If the nfb isn't doing anything for you, and you need the gain, you could just drop it. Valves are used in guitar amps for their distortion characteristics, and nfb in guitar amps is quite uncommon.

If you're increasing the amps sensitivity, you may see an unacceptable increase in hum and noise depending in how good your grounding scheme is.

Hope this helps
Gabriel

Interesting! I hadn't considered looking at the output valve. Makes sense and I've ordered a NOS Mullard EF40. Wiring looks pretty straightforward.
That said I have a spare B8A socket on the chassis wired with filament voltage, and I already have an EBC41 so will also try that extra gain stage as a non-permanent set-up just to see how that works.
Re the NFB, the original way this was wired is different to how this is done on most guitar amps. Uses a voltage divider and feeds into the grid as in the original schematic on the link in the first post. For whatever reason this really had no impact on tone. I'm trying an adjustable set-up feeding into the cathode - following an Uncle Doug tutorial. Reason being to provide an option for reducing gain for a cleaner tone, while still having the higher gain option. I'll probably make it fully switchable.

[See_Mos]

Is the resistor in the first valve anode really 18K? I would expect 220K.

I made a similar mistake over fifty years ago when I first started out thinking higher voltage to the anode would give more gain but that is not the case

I would have thought the two triodes would give enough gain without the third one and I agree that the EF41’s would be better.

[RogerLLL]

Originally Posted by See_Mos

Is the resistor in the first valve anode really 18K? I would expect 220K.

I made a similar mistake over fifty years ago when I first started out thinking higher voltage to the anode would give more gain but that is not the case

I would have thought the two triodes would give enough gain without the third one and I agree that the EF41’s would be better.

I tried various values and 18k was the sweet spot which produced the highest gain.
I'll probably try the additional gain stage tomorrow and feed back how it goes.

[Gabe001]

So, what's the final verdict?

[Marconi_MPT4]

Originally Posted by RogerLLL

There wasn't a symbol for a rectifier tube, so I've created one as in the attached, and shown what I believe to be the correct connections as per the original P75 schematic.

With the original triple EBC41 lineup, you should be able to at least double overall gain by adjusting cathode bias resistors for each stage to a value that optimises gain with selected anode load. Check out suggested use case https://frank.pocnet.net/sheets/030/e/EBC81.pdf page 2. In terms of triode characteristics EBC41, EBC90, EBC81, 6AT6, 5751 and EABC80 are the same or similar.

Replacing 18k anode resistor with 100k and cathode bias resistor changed to 1.2k should give extra gain.
Rich

[RogerLLL]

I did quite a lot of research into small signal pentodes as pre-amp valves and came across a very informative webpage by Merlin Blencowe (the Valve Wizard) https://www.valvewizard.co.uk/pentode.html which is specifically geared towards using an EF86 (electrically the same as an EF40) in a guitar amplifier. He advocates for using it in a later stage of the amp which minimises microphony and maximises the tonal advantages of a pentode. So I've re-designed the circuit along these lines, and used the majority of the values he specifies in the schematic at the bottom of the page - I'm using 250v so thinking they should be about right. I did try it as V1 and it was quite microphonic and quite harsh sounding.
The only thing I've changed is the value of the two bypass caps - I found the low values produced too harsh a tone for what I'm after, so ended up with 33uF for both. This produces a rich tone with a smooth overdrive, but is a little bass heavy so at some point I'll try tweaking to improve that. That said, reducing the bass pot in the tone stack does pretty much do that anyway! Its more of an issue playing chords and I tend to play more lead so it does work ok as is.
I'll also try changing those resistor values for the EBC41 as suggested by Marconi_MPT4. Sounds like that should have quite an impact, so thanks for that!