Ultra Clean Push-Pull EL84 design?

[printer2]

Amplification factor.

How does that give the 6V6 more headroom with both clipping at 15 watts?

If you want a really clean EL84 amp, build half a Mullard 5-10. It's an old hi-fi design that delivers 10 Watts RMS. Use an overdrive to get an overdriven sound and the amp to do the lifting.
The 6p1p-EV can be subbed in with only a few wiring changes and a change of cathode resistor, but take care with the screen resistor choice. I found through trial and smoke that they prefer a higher value.

So reduce the gain and then use a pedal to get it back? Why not just use the amp's volume control?

I wanted to maintain the PT. Is there really enough juice for a Harvard or Princeton? The schematic shows ~315 B+

You have enough juice for a 15W amp. Open up two tabs of this link. Select EL84 for one and 6V6GT for the other. Set both V+ for 315V, Iq at 35 mA and screen at 300V. Power out at 20W and 21W, the values do not take into account the power lost in the OT, knock off a few watts.

The EL84 needs -12V bias and the 6V6 -12V. So you can turn the volume control a little lower or you can stick a 12AY7 in the front spot (you may need to turn up the volume though).

Both tubes leave about 25V on the table as they can not get closer than that in saturation. Otherwise the power is the same for the same voltage and output transformer impedance.

 

[SerpentRuss]

How does that give the 6V6 more headroom with both clipping at 15 watts?


So reduce the gain and then use a pedal to get it back? Why not just use the amp's volume control?


You have enough juice for a 15W amp. Open up two tabs of this link. Select EL84 for one and 6V6GT for the other. Set both V+ for 315V, Iq at 35 mA and screen at 300V. Power out at 20W and 21W, the values do not take into account the power lost in the OT, knock off a few watts.

The EL84 needs -12V bias and the 6V6 -12V. So you can turn the volume control a little lower or you can stick a 12AY7 in the front spot (you may need to turn up the volume though).

Both tubes leave about 25V on the table as they can not get closer than that in saturation. Otherwise the power is the same for the same voltage and output transformer impedance.

You're right. If the volume is between the two 12A*7 stages, you can just turn down. I think it would be hard to drive the input stage into distortion, even with a hot pedal chain.

The pre-made Pro Jr. Hoffman board is very tempting. Then perhaps run a 12AY7 in the first slot and bias the output to something a bit more sane. If you plug the documented fender values into a load line calculator, they're broiling those EL84s in the Pro Jr.

 

[2L man]

In same circuit EL84 produce about 20% more powes when distortion is the same what 6V6 produce because EL84 produde more anode current when g1 voltage meet zero volt grid line. Simply saying it is newer design where knowledge from previous tubes was used

 

[Lowerleftcoast]

a lot of clean headroom and likes pedals

One of my friends has a Carvin Vintage 16. He installed the Hasserl mods for clean. He uses it as a clean platform and it takes pedals well. Yep, it has a pair of EL84s in it. It has two gain stages before the PI. Besides the tone stack it is not much different than the Pro Jr with the mods I suggested. I agree the EL84 does not have the same character as a 6V6. They can do clean though. Imo, you have said you want to use the EL84s... use them.

The suggestions that contain a paraphase inverter will not have the same clean of a long tail pair. Just different cleans. The pathway to Twin type clean is with a long tail pair and more voltage on the tubes, especially the preamp tubes.

 

[SerpentRuss]

Re the 6p1. Ignore me. I've gone back over what I did and it's on me... I got confused and I apologise if I've mislead.

Re the EF86. Vox used it in the past, and it's fun. Pentode front ends do sound different to triodes. Perhaps a pentode/triode such as the 6f1p-EV?. The way Musikding implement it in the Madamp G3 Blues is unlike anything you currently have ( probably ) Changing the screen voltage via P1 alters both gain and compression. You've then got a triode to play with.
The 6k4n-EV, 6j1 and 5654 ( and equivalents ) could possibly sub in for the EF86.

I'm going to squirrel this away for a later look. Like the DC coupled Pentode-Triode pre-amp. I don't think it's a good fit for this project though.

 

[Phrygian77]

How does that give the 6V6 more headroom with both clipping at 15 watts?

Distortion numbers aside, what I was talking about was that in this circuit 10Vpk-pk in is close to 230Vpk-pk out. With a 6V6GT it's 120Vpk-pk out with 10Vpk-pk in, which doesn't even get you out of the class A line by the way.

 

[printer2]

Distortion numbers aside, what I was talking about was that in this circuit 10Vpk-pk in is close to 230Vpk-pk out. With a 6V6GT it's 120Vpk-pk out with 10Vpk-pk in, which doesn't even get you out of the class A line by the way.

So put 20V in, not hard to do with tubes. Turn the volume up from 3 to 4. The point is the power supply and the output transformer mainly sets your power out, the EL84 gets you there before the 6V6 but when one clips the other will clip more or less putting out the same power. Even if there was a 20% difference between them it will not give you an audible increase in headroom out of the speaker. The EL84 was designed as a more sensitive 6V6 replacement, it does this at a cost of more heater current though.

 

[andrewRneumann]

If you plug the documented fender values into a load line calculator, they're broiling those EL84s in the Pro Jr.

I’ve noticed this before. All I can think is that the load line calculator is a little off.

IMHO, clean power will always be controlled by the last tube in the signal chain. There’s no way around it. If you want more clean output headroom you need to max out the EL84s. Minimizing power supply sag and keeping the highest voltage on the plates and screens possible should be your priority. With this in mind, maybe 12AX7s in the preamp are the way to go with their low current operation.

You might find ways of rolling off the high end in the preamp as those are the typical frequencies to break up first—a pedal platform is probably better suited with an even frequency response.

In the end, I don’t think there is much more that can be wrung out of this setup because it‘s already pretty hot. It‘s interesting to look at an existing design and see what could be done with the constraints given.

 

[David Barnett]

Cathode biased does not equal *a lot of clean headroom*.

Same goes for Open Loop.

Go to fixed bias, and put a feedback loop around the power amp stage, and you'll stand a fighting chance of getting some apparent headroom out of it.

 

[David Barnett]

I'd be interested to know what the initial (too low) value of screen resistor was on your 6P1P build. I have an SE project I'm tuning and tweaking now and if I recall, I'm using 1.8K, 2 watt.

I just looked at the Mullard 5-10, it's interesting. I don't think a EF86 pre-amp would be all that pedal friendly, although I have no experience, in this case, it has to be running at very low V+ since it's power is downstream of a 150K resistor.


View attachment 944263

Keep in mind that's a hifi amp, intended to be used with a preamp. You wouldn't want to use it as is for a guitar amp, especially with the tone controls on the input.

 

[radiocaster]

If you overspec the output transformer and speaker, you might get a little more clean sound.

 

[SerpentRuss]

If you overspec the output transformer and speaker, you might get a little more clean sound.

To keep costs under control the PT, OT and Speaker are being re-used and seem to be in decent shape. The main issue with the board has been corroded traces. The speaker is an Eminence Legend Alnico that doesn't have a model number, but I think is the 35 watt. I think they made 35s and 50s.

 

[dan40]

I will agree that you can get a bit more out of the EL84's with a bump in plate voltage and a larger OT. I built a Friedman JJ Jr clone last year with a large 40 watt/8k OT and a B+ of 380vdc. Even with the EL84's, this amp has plenty of headroom and grunt and it gives some of my larger builds a run for the money. It is quite hard on the EL84's at that plate voltage and dissipation level but so far the JJ tubes seem to be handling it with no sweat.

 

[printer2]

It would help running the tubes at a higher voltage and lower idle current. When you are biased hot there is only so much remaining room for the signal. If you did have higher voltage and lower idle you can make more power. But with the original PT, OT and speaker there is not a lot more you can get. What I would do with the PJ is change the circuit between the first and second triodes, like the 1-Watt Marshall JTM.

I would use different values to start with, 500pF, 47k, 8.2nF, 150k, 47nF, 8.2k. This gives a mild scoop like the Bassman/Marshall, it will reduce gain a little but not a bad idea here. You can play around with the amount taken out, it may sound cleaner as it takes some of those pesky mids out.

 

[SerpentRuss]

I've decided to order the Hoffman Pro Jr. conversion turret board. I'll do that Monday. I'll spend some time this weekend looking at my parts inventory. With shipping being a large percentage of any order of little pieces and parts, it will probably make sense to get everything I'm missing and any parts I may want to tweak with from Hoffman when I place that order.

I'll keep everyone posted and take photos as I make progress. It should go relatively quickly.

 

[2L man]

So put 20V in, not hard to do with tubes. Turn the volume up from 3 to 4. The point is the power supply and the output transformer mainly sets your power out, the EL84 gets you there before the 6V6 but when one clips the other will clip more or less putting out the same power. Even if there was a 20% difference between them it will not give you an audible increase in headroom out of the speaker. The EL84 was designed as a more sensitive 6V6 replacement, it does this at a cost of more heater current though.

6V6 does not produce same power than EL84 when driven AB1 class. When Phase Inverter is changed capable to AB2 drive the 6V6 produce more power but also more distortion. Obviously Screen current also increase to level it shorten tube life?

 

[printer2]

6V6 does not produce same power than EL84 when driven AB1 class. When Phase Inverter is changed capable to AB2 drive the 6V6 produce more power but also more distortion. Obviously Screen current also increase to level it shorten tube life?

How many guitar amps have you seen lately running AB2? I doubt too many people here would want to go that route. And you can only pull out of the amp what the power supply will give. In this case there is not a lot more to draw out of it.

 

[Lowerleftcoast]

I've decided to order the Hoffman Pro Jr. conversion turret board.

I have not looked at the Hoffman board carefully. That said, I would order a few more turrets in case a little different ground scheme etc might be considered mid-build. Better to have them in hand than wishing you had ordered them.

 

[printer2]

I have not looked at the Hoffman board carefully. That said, I would order a few more turrets in case a little different ground scheme etc might be considered mid-build. Better to have them in hand than wishing you had ordered them.

Too many turrets never hurt anyone. Having too few can be a pain in the ass.

 

[andrewRneumann]

How many guitar amps have you seen lately running AB2? I doubt too many people here would want to go that route. And you can only pull out of the amp what the power supply will give. In this case there is not a lot more to draw out of it.

One.