6V6 ECC83 push-pull amp project

[robinshack]

If you read my post #5, it is explained there.

With that part number, IT IS A MODULATION TRANSFORMER!
The 94 ohms winding is for modulating the high voltage dc supply to the vhf tx pa stage.

The low z loudspeaker winding you are using here in your amp would have originally been designed to work with a low z public address horn loudspeaker mounted on a vehicle.
Eg. Water board vehicle warning the water will be cut off shortly.
Rob

[Gabe001]

You can get the turns ratio by following the instructions here: https://how-to.fandom.com/wiki/How_t...rs_turns_ratio

Then use an online calculator to convert this to impedance ratio.

The hardest part is finding a low voltage ac source. Having said this, I struggled to get very accurate measurements when I tried this myself with a multimeter. Remember to put the multimeter on the AC setting.

If you are tempted to put 6.3v AC in the secondary, remember that it will translate to quite a high voltage on the primary.

[RogerLLL]

Originally Posted by robinshack

If you read my post #5, it is explained there.

With that part number, IT IS A MODULATION TRANSFORMER!
The 94 ohms winding is for modulating the high voltage dc supply to the vhf tx pa stage.

The low z loudspeaker winding you are using here in your amp would have originally been designed to work with a low z public address horn loudspeaker mounted on a vehicle.
Eg. Water board vehicle warning the water will be cut off shortly.
Rob

Gotcha! I checked out that pdf and that absolutely is the same part! Sorry for missing your post. Described as:

Modulation Transformer (9 K anode to anode and 3.5 K ohms secondary at 75 mA DC)

The amp is a bit of a hotch potch of components - which does work! But definitely scope for upgrades, and like I said, Jim told me the output transformer was a little under-specced which is why the bias is lower than it would otherwise be. Could that explain why those grid leak resistors are wired in the way that they are?

[RogerLLL]

Originally Posted by Gabe001

You can get the turns ratio by following the instructions here: https://how-to.fandom.com/wiki/How_t...rs_turns_ratio

Then use an online calculator to convert this to impedance ratio.

The hardest part is finding a low voltage ac source. Having said this, I struggled to get very accurate measurements when I tried this myself with a multimeter. Remember to put the multimeter on the AC setting.

If you are tempted to put 6.3v AC in the secondary, remember that it will translate to quite a high voltage on the primary.

Thanks Gabe, I will give that a go. I have a 6v ac supply which should work ok, albeit with caution.

[GrimJosef]

Originally Posted by Gabe001

... The hardest part is finding a low voltage ac source ... If you are tempted to put 6.3v AC in the secondary, remember that it will translate to quite a high voltage on the primary.

General purpose mains transformers quite often used to have a 0V-4V-5V winding for driving 4V-heater or 5V-heater rectifiers. You get a little over 1V between the 4V and 5V terminals when the transformer's running off-load.

Cheers,

GJ

[Gabe001]

That's a great idea GJ. Thank you

[Robert Gribnau]

Did you take notice of posts #17, #23, and #33?

For use as a guitar amplifier the first stage has to have cathode bias which it doesn't have in your schematics (in them the cathode of the first stage is grounded).

[RogerLLL]

I've incorporated a number of changes in the attached schematic. These include:

• cathode bias for V1a

• 6V6 grid leak resistors now connect to ground rather than cathode

• master volume added pre-phase inverter, as depicted in the link in the post from Gabe001

• removed low gain input to V2a

• added a proposed switchable low gain input to V1b, wasn't sure if I needed to take the V1a signal to ground, also not sure about the reason for the 470k resistor after the gain pot

• identified the tone stack as a Baxandall type, mostly used in HiFi amplifiers but also used in some Orange & Ampeg amps https://monster.partyhat.co/article/...cks/#baxandall This does work well, but would be keen to compare with the same high gain Fender tone stack I used for the Pye75 amp conversion, which worked extremely well

So far I have added the V1a cathode bias and moved the 6V6 grid leaks to ground, and tested the amp. Definitely more volume! The crackle sound is still there. I received an OT from Gabe001 so many thanks to him for that! I will try swapping out the modulation transformer for that and see what effect it has.
6V6 ECC83 p&p v3.pdf

[trobbins]

The MV pot is inside the feedback loop - which won't work as you intend - suggest you place it in front of the feedback stage (ie. after the treble pot).

Do you have a scope and a signal generator? That can help check when the output signal gets clipped or asymmetric or blocked from overloading - which may be related to the crackling sound effect. Also may be worthwhile swapping the 12AX7's around to see if that changes the crackling effect.

[Gabe001]

It may be worth disconnecting the negative feedback loop from the secondary and seeing whether this makes things better or worse. I had an amp that sounded like yours once and the problem was instability. It could be that the negative feedback loop is connected the wrong way round (positive feedback). Given that the feedback amount is tiny anyway it may not be enough to cause overt oscillation.

Another quick test is to measure the voltage (AC) on the secondary using your dmm and a resistor load and no signal and seeing if you get a reading - also suggestive of oscillation.

If everything fails and you are stuck, try disconnecting your opt and running the signal through the opt I sent you. You can do this easily using croc clips without physically having to mount it. Obviously be careful regarding the high voltages. I've had a opt oscillate at 50khz once for no apparent reason (I think it was underrated for current) and the only option was to replace it

[RogerLLL]

Originally Posted by trobbins

The MV pot is inside the feedback loop - which won't work as you intend - suggest you place it in front of the feedback stage (ie. after the treble pot).

Do you have a scope and a signal generator? That can help check when the output signal gets clipped or asymmetric or blocked from overloading - which may be related to the crackling sound effect. Also may be worthwhile swapping the 12AX7's around to see if that changes the crackling effect.

I have a scope on order, just a cheap digital one but it should do what's needed. Can just use a signal generator app on my phone. Was wondering if what I'm hearing is crossover distortion?
I'll relocate the MV as you have suggested 👍

[RogerLLL]

Originally Posted by Gabe001

It may be worth disconnecting the negative feedback loop from the secondary and seeing whether this makes things better or worse. I had an amp that sounded like yours once and the problem was instability. It could be that the negative feedback loop is connected the wrong way round (positive feedback). Given that the feedback amount is tiny anyway it may not be enough to cause overt oscillation.

Another quick test is to measure the voltage (AC) on the secondary using your dmm and a resistor load and no signal and seeing if you get a reading - also suggestive of oscillation.

If everything fails and you are stuck, try disconnecting your opt and running the signal through the opt I sent you. You can do this easily using croc clips without physically having to mount it. Obviously be careful regarding the high voltages. I've had a opt oscillate at 50khz once for no apparent reason (I think it was underrated for current) and the only option was to replace it

I'll try disconnecting the NFB as you suggest, and like the croc clip idea for the OT!

[RogerLLL]

Had a very busy few days, and finally had time today to make some mods to the circuit, as in the attached. The aim was to simplify the circuit by removing the low gain input and incorporate just the master volume for now. However, I'm getting no output. I've attached the schematic. Have I made any errors in the circuit? I've checked voltages and all seem ok, around 130v HT and 90-100v at the plates. Any ideas?
6V6 ECC83 p&p v4b.pdf

[RogerLLL]

OK so that was a bit embarrassing, I missed the small ground wire on the master volume pot, and was using that tag for the signal... Now reversed and all is as it should be. Master volume works well and allows more control of the amp. Still have the crackle, but a better system to test the amp with.
One thing I'm not sure about is whether the MV should go to ground or to the junction between the 100k resistor of the NFB and the 2.7k to ground, as this is what is depicted in this link:
https://robrobinette.com/5e3_Modific..._Master_Volume
Next will be the OT swap and trying disconnecting the NFB.

[Radio Wrangler]

You have the grid DC potential of some valves coming entirely via pot sliders. That's a strong recipe for crackles. DC blocking capacitors and fixed resistor for bias paths combats this.... even if the DC being blocked looks like ground volts. Similarly with switches which cause floating or jumps in thermoelectric DC. Sounds silly or trivial, but it's enough.

David

[RogerLLL]

I tried disconnecting the NFB, but the volume dropped considerably, so I'll just leave it connected.
I swapped out the OT. Immediate improvement in the overall quality of sound. Break up is much more even, smoother and less harsh sounding.
The "crackle" sound is still there. I'm convinced this is actually crossover distortion, having listened to a couple of YouTube clips demonstrating this. I know for a fact that the bias is set quite low due to limitations of the power transformer, and this could be why. I've ordered a digital oscilloscope so will check the waveform when this arrives.

[Gabe001]

Volume dropped significantly when you removed nfb? It's connected the wrong way round then (providing positive feedback) which makes the amp unstable.

You could try increasing the current through the output valves and see how hot the transformer gets after 20 mins

[RogerLLL]

Could it be that the NFB was wired this way intentionally to compensate for the power transformer? Can it be wired differently so as to provide negative feedback? Would this likely also reduce the volume? I know very little about NFB loops so will research this.

[RogerLLL]

Another test I've just done was to try plugging into the 12" speaker of my Sessionette 75 amp, which is a great solid state amp in itself. It's a Celestion G12H-100 8 ohm, so a pretty decent speaker. No more crackle. I just didn't imagine it could have been linked to the speaker. The crackle sound certainly wasn't the coil rubbing or anything else to indicate a bad speaker. I have read that the right speaker can make a world of difference, makes sense, but wasn't expecting that. The speaker I was using was from a Fender Bullet 15W 8 ohm, so in theory up to the job, but there you go.

[Gabe001]

Originally Posted by RogerLLL

Could it be that the NFB was wired this way intentionally to compensate for the power transformer? Can it be wired differently so as to provide negative feedback? Would this likely also reduce the volume? I know very little about NFB loops so will research this.

Yes, just swap the ground on the secondary from the output transformer. Most guitar amps use little to none global negative feedback (wrapped around the output transformer) so I'd leave it out. But if you want to try it the other way round, just swap the grounded lead on the transformer secondary ie take the nfb from the other wire to speaker

If you take the feedback out and you're not satisfied with the gain, with the feedback removed, you can add a bypass capacitor across the 2.7k cathode resistor of v2a (as per v1a)