Philco Stereo Radio - Transformation to SE Guitar Amp

[Jay]

That is utterly terrifying!

... but makes for compelling reading

Hey, you are back I am honoured you are reading this humble thread.

Wasn't there a dude in Christchurch who's hobby it was to convert old radios to guitar amps? Haven't seen him posting for a while now. Might need his wisdom...

[MiniForklift]

Yes, great guyI I took my 1974 HH Electronics Tape Delay to him to fix, he also serviced my Laney head and did a great job on both. His username on here is PaulWD although unfortunately I don't seem to have his contact number anymore

[codedog]

That is utterly terrifying!

... but makes for compelling reading

Hey, you are back I am honoured you are reading this humble thread.

Wasn't there a dude in Christchurch who's hobby it was to convert old radios to guitar amps? Haven't seen him posting for a while now. Might need his wisdom...

I think I have his email address. PM me if you're interested.

[Jay]

Remember that I thought the current iteration has TOO MUCH GAIN....

I have now updated the circuit diagram to reflect the changes I made and document actual resistor component values, regardless of what the colour bands tell me they should be. Two resistors jump out

Power supply changes: A, B, C voltages feeding the amp stages now much higher. The EL84 grid2 now is a bit lower than plate voltage, wasn't before. Extra stage D added for the first pre-amp. This stage effectively killed off the motorboating oscillations.

Re the two 6AV6 valves in the pre-amp section - one is a Pope brand, the other a Mullard. I couldn't find any of their datasheets so settled on a General Electric 6AV6 datasheet to find out where the pre amp valves are biased.

GAIN: Look at the V1 and V2 plate resistors - they should be 270K but are actually 400K and 470K respectively which is the reason for so much gain in the circuit. Drawing the anode & cathode load lines, you can see they drop quite a bit from what it should be (green) and hence the bias points are too far left IMO. I will redesign it such that the bias points move towards the blue circled area.

I hope that the redesign doesn't change the sound too much as I quite like it as it is at lower volume. Anyway, we shall see.

20240910_182558.jpg (1.85 MiB) Viewed 2108 times

20240910_182903.jpg (2.14 MiB) Viewed 2108 times

[sizzlingbadger]

Shouldn't that Treble control wiper be on the other end of the 50nF Cap ?

Screen Shot 2024-09-11 at 11.30.59 AM.jpg (99.88 KiB) Viewed 2101 times

[RectifiedAmps]

If you like the sound then leave the anode resistors alone and maybe just fiddle with the cathode resistors on v1&2. But there’s nothing that says you have to be centre-biased and your current settings aren’t hurting anything.

[Jay]

Shouldn't that Treble control wiper be on the other end of the 50nF Cap ?


Screen Shot 2024-09-11 at 11.30.59 AM.jpg

I shall try that but I was thinking of putting a Fender/Jansen 4/12 tonestack in between V1 and V2. I find the treble too shrill at the moment and the bass not enough. I calculate the bass 6db cutoff point to be at about 42 Hz. I prefer around 20Hz.

[Jay]

If you like the sound then leave the anode resistors alone and maybe just fiddle with the cathode resistors on v1&2. But there’s nothing that says you have to be centre-biased and your current settings aren’t hurting anything.

There are some unruly clipping sounds occasionally, bit like a rattle which I like to remove somehow. Also if I were to leave this amount of gain in the circuit, I think I need some negative feedback.

[RectifiedAmps]

The rattling clipping is probably blocking distortion from all that gain. You might tame it by just adding a 22k grid stopper to v2. Also, you can ditch the 10nf cap on the volume wiper - it might be contributing to the blocking and it’s not needed, unless you want some sort of high-pass RC filter with the 1M resistor?

Another option is a 100-220k between the other 10nf and the top of the volume control, to attenuate things a touch. This is how Marshall did it on the JCM800.

[Jay]

The rattling clipping is probably blocking distortion from all that gain. You might tame it by just adding a 22k grid stopper to v2. Also, you can ditch the 10nf cap on the volume wiper - it might be contributing to the blocking and it’s not needed, unless you want some sort of high-pass RC filter with the 1M resistor?

Another option is a 100-220k between the other 10nf and the top of the volume control, to attenuate things a touch. This is how Marshall did it on the JCM800.

Spot on Jaret . It appears the rattle has gone after I soldered in the 22K grid stopper and removed the cap. Haven't tried at full bore yet, but looks promising. Still want to reduce gain so going to lower the plate resistors but I'll increase the V1 bias voltage first to something between -1V and -1.5V and see how that sounds.

[Jay]

Shouldn't that Treble control wiper be on the other end of the 50nF Cap ?


Screen Shot 2024-09-11 at 11.30.59 AM.jpg

I tried that and did not hear any difference.

[sizzlingbadger]

Shouldn't that Treble control wiper be on the other end of the 50nF Cap ?


Screen Shot 2024-09-11 at 11.30.59 AM.jpg

I tried that and did not hear any difference.

Maybe not but it will stop you having the high dc voltage on the Treble pot, which is likely how it should of been.

[Jay]

Few staps backwards I think...

Stripped all the radio bits out so looks a bit cleaner on the right side. Left side is still a shocker

I replaced the original tonestack with the Fender 5F6-A Bassman Tone Stack. That tonestack was copied by many amp builders, including Jansen. You can find this tonestack in the 4/12 and 6/16 amongst others. I really like it in the 4/12 hence trying it here between V1 and V2.

While it does work, its load on V1 destroys the raw character the amp had with the original tonestack. It is too clean now at relatively low volume and I want that growl back.

So, reverting to original tonestack and perhaps experimenting with some small changes to its components.

I think I might tame the gain in the circuit via a three way switch over the V2 cathode resistor. Setting 1 - just cathode resistor, setting 2 - adds parallel cap to 1, setting 3 - add NFB to 1. (Setting 2 doesn't tame it, quite the opposite).

20240917_125631.jpg (4.09 MiB) Viewed 1180 times

[Jay]

Amp gurus, please help...

Having some trouble biassing the old EL84 valve with respect to the circuit posted earlier showing the original bass tone control. That bass pot and the 47K resistor to ground, effectively constitute Rg1 (grid leak resistor). So I have called the two together Rg1 in the table below.

With the cathode resistor being 284 ohms (avo reading) and Rg1 set to 220kOhm, I measure a Cathode bias voltage of 13.2V (column 5). On pin 2 (Vg1) though I measure 6.87V, effectively the actual negative bias equals 13.2 - 6.87 = 6.42V.

So by varying the bass pot I get a whole range of Cathode bias voltages as per the table below and the voltage on pin 2 also changes.

I always thought pin 2 would sit at ground voltage level and would only be positive if a negative grid 1 current exists (which is bad in my understanding).
So I swapped out the EL84 for two other ones, one fairly new, and all three exhibit this 'feature'.

Column 3 shows the calculated current through Rg1 as about 30 uA which is about 100 times more than the datasheet states. I have also tried a few of the old Suzuki different coupling capacitors as I suspect these may be the source of the current leak. I haven't got any brand new ones though atm.

Question - should Vg1 always be around 0V? If so then the coupling cap must be leaking

Thanks

Biassing Effects Rg1.jpg (127.61 KiB) Viewed 1074 times

[AiRdAd]

Amp gurus, please help...

Having some trouble biassing the old EL84 valve with respect to the circuit posted earlier showing the original bass tone control. That bass pot and the 47K resistor to ground, effectively constitute Rg1 (grid leak resistor). So I have called the two together Rg1 in the table below.

With the cathode resistor being 284 ohms (avo reading) and Rg1 set to 220kOhm, I measure a Cathode bias voltage of 13.2V (column 5). On pin 2 (Vg1) though I measure 6.87V, effectively the actual negative bias equals 13.2 - 6.87 = 6.42V.

So by varying the bass pot I get a whole range of Cathode bias voltages as per the table below and the voltage on pin 2 also changes.

I always thought pin 2 would sit at ground voltage level and would only be positive if a negative grid 1 current exists (which is bad in my understanding).
So I swapped out the EL84 for two other ones, one fairly new, and all three exhibit this 'feature'.

Column 3 shows the calculated current through Rg1 as about 30 uA which is about 100 times more than the datasheet states. I have also tried a few of the old Suzuki different coupling capacitors as I suspect these may be the source of the current leak. I haven't got any brand new ones though atm.

Question - should Vg1 always be around 0V? If so then the coupling cap must be leaking

Thanks

Biassing Effects Rg1.jpg

Have you tried turning it off and on again?