Motorboating on EL84 Amp

[The New Steve H]

I got my new amp working. It's basically a 5F6A preamp with four EL84s on the output, which is stolen from a Rocket.

When the amp is turned up full blast, I get a little motorboating. It goes "wup wup wup wup wup." When I turn the PPIMV down slightly, it stops. What's the first place to check? Should I worry about the preamp before looking at the output? I am thinking that if the preamp was the problem, the PPIMV wouldn't stop the motorboating.

Also wondering if I made a mistake by using a power supply which is nearly identical to a Bassman, except that I used 4 33uF caps instead of 20uF caps. This is what I had on hand. Maybe I need more filtering?

I'm not sure it's as loud as it should be. I got the OT leads reversed, as I always do, when I first set it up, but after I switched them, tightened everything up, and did the permanent soldering, it seems somewhat subdued.

[Leo_Gnardo]

Motorboating is usually the sign of insufficient filtering. Your 33 uF's are new? SHOULD be OK but . . . who knows? You can always clip in another capacitor in parallel with what you have & if it stops the motorboating then you know what you have to do. Maybe one of those 33's is duff.

[The New Steve H]

It looks like the Rocket uses a ton of capacitance on the output, so maybe I should cram another cap in that end?

[Leo_Gnardo]

If you're tube rectified, there's a limit to your primary filter cap, usually 50-60 uF. If silicon rectifiers you can go more. In either case, a couple clip leads and an extra cap will prove whether more capacitance is the fix for your motorboat noise. If you have a capacitance meter you could disconnect the caps you have (drat - more work) and see if one of your 33's is sick.

[Structo]

You also can try to add some grid stoppers on the power tubes.

Try 1K5 resistors in series with the grids of the power tubes.

[The New Steve H]

I'll keep fiddling with it. I still have lots of stuff to adjust and tighten up.

[The New Steve H]

I haven't taken all the slack out of the wiring yet.

Is it possible that the motorboating is caused by the resistor I used instead of a choke? I have no idea why I did this or how I chose the value. It was last year. It's a 250 ohm 25W job. It appears to allow quite a bit of current. It gets warm. Wondering if I should increase the resistance or replace it with a choke.

Might as well put up the voltages.

I am trying to get a grip on how 12A_7 tubes work, so I'll understand how to choose my own voltages instead of copying what other people did. It strikes me as weird that the Bassman has 150 volts on the V1 anodes, while the Rocket has around 110. There are some voltages on the Bassman schematic which are 50 volts too high to get with this PT.

I understand how triodes amplify. It's the concept of choosing voltages which is mysterious to me.

For some reason, I have an NOS 5V4 sitting around. Wondering if I intended to use it in this amp.

AC in: 124V

Rectifier (5AR4):

2: 279 V
8: 279
4: 281
6: 281

First power stage: 279
Second power stage (after 250 ohm resistor): 271
Third power stage: (after 4700 ohm resistor): 238

I am getting 46 volts of AC on the first stage of the power caps. Hard to believe that's normal.

V1A (12AY7):

Anode: 151
Cathode: 2.5
Grid: 0

V1B:

Anode: 146
Cathode: 2.5
Grid: 0

V2A (12AX7):

Anode: 110
Cathode: 0.66
Grid: 0

V2B:

Anode: 190
Cathode: 111
Grid: 110

V3A (12AX7):

Anode: 160
Cathode: 26
Grid: 16

V3B:

Anode: 157
Cathode: 26
Grid: 17

The values for all of the power EL84s are identical:

Anode: 276
Cathode: 9
Grid: 0
Screen: 271

The amp sounds very nice. Extremely detailed. But there is a slight crackly hum I can't get rid of. I found a small AC signal on the scope, and it would not go away. I unplugged the amp and turned off the fluorescent lights, and it was still there.

[The New Steve H]

Forgot to say, the amp already has 1.5k resistors on the power tube grids. I used the attached layout and stole everything on the output end.

[Tillydog]

Is it possible that the motorboating is caused by the resistor I used instead of a choke?

Very possibly - you asked in the first post whether the cause was more likely to be the preamp or the poweramp - It's neither (or both, depending on how you look at it)... 'Motorboating' is usually caused by one section of the amp 'stealing voltage' from another. To prevent this, the power supply (and grounding) needs to be designed to keep each section of the amp in its own little world (i.e to have sufficient filtering).

Likely what is happening is that as the voltage at the screens of the EL84s (and so back to the PI) is not well enough isolated from any sag on the main B+ when the EL84s start drawing serious current. Changing that 250R for a choke will improve this dramatically, as would increasing the 250R resistor, but this would also drop your voltages (it would be a 'suck it and see' exercise - maybe 1k?).

It strikes me as weird that the Bassman has 150 volts on the V1 anodes, while the Rocket has around 110.

Just looking at a couple of schematics:

Bassman has 100k anode resistor, the equivalent of a 1.64k cathode resistor (shared 820R) & 320V B+

Rocket has a 220k anode resistor, a 1.5k cathode resistor & ~180V B+

Therefore the rocket has a lower anode voltage for similar bias conditions.

I think the sequence goes:

1) Choose the anode resistor for the gain that you want to achieve (gain depends directly on the anode resistor value).

2) Choose the cathode resistor using the curves on the datasheet to give you the most headroom (or deliberately offset the bias warm or cold, as desired) - this gives you a bias current.

3) Figure out what the bias current from (2) would mean in terms of the voltage dropped across the anode resistor (bias current x anode resistor).

4) Arrange for a B+ of (very roughly) twice the voltage drop across the anode resistor.

5) Return to (1) and repeat, tweaking your choices until you get something that fits

6) Build it, listen to it, return to (5) until you have something that pleases you...

Good luck!

[ToneMerc]

I would like to see the layout, sounds like its oscillating to me.

TM

[The New Steve H]

I should really sit down and draw a real layout, but I can't do it today.

A bigger resistor would be easy. I guess I should try that before using a choke.

It occurred to me to ask: would a diode between power stages help? I am guessing that if the power caps are stealing from the preamp caps, there would have to be waves of positive current going from the preamp caps to the power caps, and a diode would shut that down. But it's just a guess.

Thanks for the info on the correct order for working with tubes.

[The New Steve H]

I could not resist the temptation to stick a diode in there. It made a big difference. Now the motorboating goes away by itself, and it doesn't start as easily. But I still can't play the amp at top volume. I guess I'll have to get a choke or increase the size of the resistor.

[Tillydog]

It sounds like you're homing in on the cause - You could try increasing the amount of capacitance on the preamp side of the resistor (not sure what you have there at the moment).

[The New Steve H]

I know I'm going to bring some well-deserved pain on myself by saying this, but I can't find big axial-lead caps around here, so I used whatever was available. I found some 33 uF radial-lead caps locally, so I rigged four of them up with wires that ran to the ground side of the turret board. Two go together. Then there's a 250-ohm resistor and one more cap. Then there's a 4700K (now with a temporary diode) and the last cap, which feeds the preamp.

Increasing the capacitance on the power tube side didn't help. I haven't tried it on the preamp side.

I'm going to run out Monday and get two or three power resistors to see if a bigger value will fix this. I can't get chokes around here, so I want to try resistors first.

I don't know if I'll like this amp or not. There was something about the tone of Fender EL84 amps I liked, so I built this one to see if I liked it better than the same preamp with 6L6s. I won't really know until I get done adjusting it.

It's very, very detailed until you turn the volume up pretty high, and then there's a great deal of distortion. Doesn't seem as warm as I had hoped.

[cbass]

Fwiw I like 6v6's better .I built a 2 el84 rocket and I want to like it but I just cant.I don't think I have the right speakers for it though