HELP!! No B+!

[jkey04]

Sorry...the UL tap is connected directly to the SCREEN. I don't have a grid resistor either though. Just a 0.1uf coupling cap between output of preamp and grid of el34. Should I try a grid resistor?

[Tillydog]

... I don't have a grid resistor either though. Just a 0.1uf coupling cap between output of preamp and grid of el34.

Yup, you need one!

It's there to provide a DC reference for the bias voltage and also, there's a tiny leakage current out of the grid that needs somewhere to go, otherwise the grid voltage just builds up and up, turning the tube on and making it run away, which I guess is what is happening.

A few hundred k ohms is needed - anything less than 700k, the datasheet says - how about 470k? Connection is grid (pin 5) to ground.

It wouldn't hurt to have a grid stopper on pin 5 if you haven't got one, either - in series with the grid connection. (A few k ohms, mounted directly on the socket would be my suggestion).

It's worth sketching out a schematic, even for your own benefit as you sometimes spot things that you miss 'in the flesh' (& vice versa).

[jkey04]

OK i put a 470k resistor to ground from grid and a 2.2k grid stopper in as well. You were right, that leveled out the cathode voltage!

BUT...the cathode voltage is low around 28vdc with a 560 ohm resistor to ground bypassed with a 220uf/160v capacitor. The odd thing is the cathode voltage increases slowly as i increase the resistor value. I had it up to 8.8k and the voltage was only 34vdc. I was under the impression that the voltage should drop when the resistance is increased. The voltage needs to be up around 45-50vdc...right?

On a side note...I have output now. The sounds pretty good, but it just isnt very loud at this point. Hopefully, when i get that cathode voltage up it will increase output of the amp.

[rock_mumbles]

...
BUT...the cathode voltage is low around 28vdc with a 560 ohm resistor to ground bypassed with a 220uf/160v capacitor. The odd thing is the cathode voltage increases slowly as i increase the resistor value. I had it up to 8.8k and the voltage was only 34vdc. I was under the impression that the voltage should drop when the resistance is increased. The voltage needs to be up around 45-50vdc...right?

...

The larger the cathode resistor the larger the resistance to ground so the higher the cathode voltage (and plate voltage) ... you should use your plate voltage, cathode voltage and cathode resistor value to calculate the El34 dissipation ...

[Tillydog]

... I was under the impression that the voltage should drop when the resistance is increased...

Nope... As you increase the resistor, the voltage drop also increases (for the same current), but in this case, the voltage also sets the bias of the EL34, so you don't see as much of an increase as Ohm's law would predict.

The voltage needs to be up around 45-50vdc...right?

Where did you get that figure from?

As above, you need to figure out the dissipation in the EL34 and adjust the cathode resistor to suit. (How did you arrive at the initial value?)

28V over 560R is 50mA. Some of that will be coming from the screens, but ignore that for now. From your earlier AC readings, I would guess that you've got about 340V on the plate of the EL34 (and 28V on the cathode), so the EL34 is dropping (340-28 = 312V) at 50mA, so it is dissipating 312 x 0.05 = 15.6W, or about 60% max (lumping screen and plate dissipation together for rough maths).

To get that up, you need to *reduce* the cathode resistor.

You'll need to work the maths out for your actual numbers, though.

Should you wish, I believe you can go up to 100% plate dissipation at idle for a SE amp, but keep an eye on how much current your power supply can safely provide (PT rating, etc.).

[Phil_S]

... I was under the impression that the voltage should drop when the resistance is increased...

Think of the cathode resistor as if it is a drain stopper. (A technically inaccurate metaphor.) The larger the value, the more it plugs the drain, lets less electrons out, forces plate voltage up.

Should you wish, I believe you can go up to 100% plate dissipation at idle for a SE amp, but keep an eye on how much current your power supply can safely provide (PT rating, etc.).

You should look to setup a cathode biased amp at around 100%. They are self-limiting. As you apply more signal to the grid, that affects what is going on inside the tube. Rk is there to manage the balance of things when signal is flowing. It will work great at 90-100%.

[Firestorm]

You never want your signal wrt to the cathode to swing so high that the grid becomes positive nor so low that it shuts off the tube. Biasing close to 100% makes this simpler to accomplish.

[jkey04]

Sorry for the late response...

As of now, I have rebuilt the preamp. I had a single EF86 initially and it wasnt driving the EL34 well at all. Everything was crystal clean and the output volume was very low. So I went back and put a 6021 between the EF86 and EL34. I used the first triode as a cathode follower, put the tone circuit after that, and then used the second triode as a voltage amp. Now the amp is overdriving pretty good. I will also try a 12AX7 and a 5751 in the V2 spot to see how the amp reacts.

As far as the EL34 goes, I still have the 560R on the cathode, but I do plan on swapping for a 200 ohm resistor maybe this evening. I would like to get the power up between 20 and 25 watts. Yes, the plate voltage on the EL34 is in the 350vdc range. Thanks for all the insight about cathode biasing and SE amps! I am learning a lot throughout this process.

I do have another question...

I got the amp functioning this morning with 352vdc on the plate of EL34, 310vdc for B+ on the 6021 (250vdc on the plate of second triode), and 250vdc B+ on the EF86. I think the EF86 voltages are too low (about 85vdc on the plate and 65vdc on the screen), but i didn't have time to try another value power resistor to raise the B+. The amp sounds decent, but if i turn the volume knob up past 10 oclock and strum a chord, the volume of the output starts to really attenuate...BAD. The volume almost drops to nothing, but as the chord fades, the volume comes back. If the volume knob stays below 10 o'clock this is not a problem. SO....a large input signal is effecting the amp somehow. Any thoughts?? My thought is that it is something with the EF86 output.

[Tillydog]

I wondered about the single EF86 stage before the EL34 - I figured you might be using EMGs, or something!

The problem sounds like it might be 'blocking distortion' which would be related to coupling cap & grid stopper size, but without knowing the detail of your design, it is impossible to say. It would really, really help if you can post an accurate schematic (doesn't need to be neat - pencil & paper is fine!).

The voltages don't seem too bad for the EF86, but I think the way of getting them up is to look at how it is biased, rather than upping B+ necessarily.

Glad that you're making progress.

(Oh and 560R to 200R is one hell of a change for the cathode resistor - tread carefully)

[Firestorm]

Yeah, I can't guess at the attenuation issue without seeing the circuit. I also agree that 200Rk might be too low at your voltages. The Mullard datasheets have 260R at 375V. You should be able to coax about 10W out of it with the UL connection. You can't get 20-25 in SE; half your power dissipation is pretty much heat.

[jkey04]

No EMGs here...ive got Fralins in my Tele!

Im not currently at home so i dont have access to the schematic that i have drawn up, but my i stuck a 0.1uF coupling cap between the preamp and power amp. I wasn't sure what value to use. Several schematics that I had been studying used different values so I just went with one. Seems to have been a bad choice maybe...

To solve this issue, should I use a 2.2k grid stopper and a coupling cap of proper value? If so, what is a good ballpark value to start with?

Also, the amp is very BASS heavy...could this problem be causing that as well?

Currently, I have a switch to select the cathode resistor for EL34 or KT88, and another switch to select either UL or tetrode operation. Hopefully in the end it will be a very versatile amp....that's the goal anyway!

[jkey04]

Here is a picture of the schematic. This is how the amp is currently wired.

With a 6201 in V2, the amp craps out around half volume. With a 7058 in V2, the amp will barely function due this attenuation, or blocking distortion. I am assuming I need some kind of voltage divider between the preamp and EL34?? I increased the grid stopper from 2.2k to 5k. It helped a little with the 6201, but didn't notice any difference with the 7058 installed.

I removed a capacitor that I previously had installed in the tone circuit and it seemed to have removed some of the excess bass, but it could still use a little more upper mids and treble clarity.

Again, thanks for the help so far! This is a huge learning experience for me. I guess the best way to learn is to get your hands dirty, heat up the soldering iron, and troubleshoot the your issues.

[Tillydog]

That helps!

I can't make sense of your tone citcuit as drawn, but that might just be me - where did it come from?

Apart from that, I would reduce the coupling capacitor between V1 and V2A by about a factor of 10 (somewhere around 2n2). Lose the 0.0068uF 'bright' capacitor for now and fit a grid stopper to V2A - whatever you can get away with as a compromise between blocking and losing out on the high end - maybe try about 33k and experiment either way. This will reduce the bass and also help with the blocking distortion (which I'm pretty sure is your problem).

I'd be reasonably confident that the above will make a dramatic improvement the sound, but I'm not sure what effect your tone circuit will have, so the next step would be to check that it is correct (maybe using Duncans 'TSC' or similar).

If you're happy that the tone circuit is OK, then you can tweak the value of the V1->V2A coupling cap up or down one value to trim the bass. Then add a 'bright' cap again if you feel the top end deteriorates at part volume. It will likely be a lot smaller than the one you have (a few hundred pF, probably).

I'd be happier with a (small) grid stopper on V2B, but I don't think that is causing a problem at the moment.

[jkey04]

The tone circuit is a derivative of the Dr Z Z28 tone circuit. It interested me because it is supposed to add gain to the selected frequencies. I thought I would put it in this amp and then play around with component values until it sounds good...if possible.

Thanks for the advice! I will try to implement your ideas this evening or in the morning. I will post my findings.

[Tillydog]

The tone circuit is a derivative of the Dr Z Z28 tone circuit. It interested me because it is supposed to add gain to the selected frequencies.

OK - Judging from a quickly found Z28 schematic, there should be an additional capacitor (250pF) between the input of the tone stack (junction of 150k & cathode of V2A) and the wiper of the treble pot. The circuit then makes sense to me. Without this capacitor, you will have (effectively) no treble.

Good luck!

ETA: No passive R/C circuit can 'add gain' - they are all lossy to some extent.