Marshall 18watt, el34's

[isaac]

Hey guys. I built a Marshall single channel 18watt clone, but I built it with el 84's. Now I found Transformers that should work. The OT works fine, and I'm pretty goldarn sure of the power transformer. Anyway, I built it just like the el84 version otherwise, but I only got 30W out of it, and it didn't even sound as good. But the problem is that I had a Bridge Rectifier (not sure of part # or whatever) but it worked great. The problem with that is that the power tubes would redplate, and that the preamp would distort really bad the the voltages were so high. So I replaced the bridge rectifier with two in4007's and it worked for a while but it was only 30w... So I went to put in a real bias supply, and I have a 100V winding on my PT. I used two in5400 diodes (big guys) and I KNOW I put the components in correctly, but I'm barely getting any negative voltage. All I'm getting is -2, and for some reason, I actually got upwards of +50VAC after the diodes. I can't figure that out. So anyway, the circuit I tried is like this. Winding/both sides connected to reverse diodes/filter cap (also reversed, +to ground)/at the junction of the cap and diodes there is the top lug of a 25kb pot/ then the wiper of that pot goes to the two 220k resistors which connect to the grid stoppers. the bottom of the pot goes to ground. I didn't use a dropping resistor since I drew too much current through the 2W I had used (which blew up) when I tried drawing the voltage from right before the normal rectifier, which is two in4007's running in parallel which works fine. But this amp has always had that problem of drawing WAY too much current through the rectifier. I had THREE 1k 9W resistors in parallel for about 300ohms and 27w capability because the bridge rectifier (i had at the time...) had too much V. But it blow those resistors, and once they got SO hot that the solder holding them in place was as wet as if I had held the soldering iron there. But now it resides with the funny bias supply and two in4007 diodes. as the two rectifiers in the amp. Something that just came to mind is that the bias winding on teh PT does NOT have a Center tap, so I'm wondering if it needs one to function, since I only was getting -2Volts, which I Think were basically "ghost" volts. If I got a center tap would that work? (although you can't really "get" a CT) but would fashioning a "fake" CT with two 100k resistors to ground work? Like many filaments do?
Also, I read that to prevent redplating on the power tubes you can just raise or lower the cathode resistor (when running in cathode bias mode) is that true? Could it be possible for me to run the I believe around 450V through EL34's as long as the cathode resistor is +or-'ed and as long as the screen voltage is lower than the plate voltage of course? Hope these ramblings made enough sense for someone to help me. I need the help. I really want to get this to work.

[Diablo]

I have trouble following your description of the bias circuit. If you could post a drawing of the schematic you used, it would help a lot.
You could certainly use cathode bias and prevent redplating with the selection of a larger bias resistor.
Take a look at other schematics, and I'm sure you can find one that shows EL34s using cathode bias.
You may find one with similar plate voltage to your amp, and this may give you a starting point for bias resistor value and screen resistor value.

[isaac]

Thanks, That's what I thought on the cathode bias issue. But if I use a adjustable bias, like say a trainwreck or many fender amps, that was my problem right now, but if I can get it to run more than 30watts in cathode bias mode, that's awesome. I love cathode biased amps. Easy and simple, no bias adjusting or anything. ;-) The bias circuit, as I left it yesterday, is like this. There is a 100V winding on the Power tranny which I put two in5400 diodes (basically BIG 4007's is what they look like) reversed of course and then the leads of those connect together, like in a typical rectifier, where they run into a junction of the top lug of a 25kb pot, and a reversed 8uf filter cap. the wiper of that pot goes to the bias resistors right before the 1500 grid stoppers on the power tubes, and the bottom lug of the pot goes to ground. But I'm only getting -2volts after the diodes, which I think are just kinda ghost volts. I'm wondering also if, since that winding doesn't have a center tap, do I need one, and if so, can I make an artificial one with two 100 resistors to ground? I've also tried using one diode off of the top of that winding, and the other wire of the winding going to ground, I believe I had about the same result. Can't remember all of what happened there. Also, I don't remember if the preamp tubes were redplating before or not with the high voltages. Can I do the same cathode resistor raising to fix this, or do preamp tubes just need a bigger voltage dropping resistor going from the filter caps and where they branch off to the plate resistors. I have the schematic software to draw one, but I haven't drawn one yet and I don't have time right now, maybe tomorrow.

[pdf64]

From your description my guess is that the bias supply is missing a ground reference.
Unless the 100V winding has a (grounded) center tap, only 1 diode is needed (for half wave rectification) or 4 diodes (full wave bridge).
I can't see how the 2 diode arrangement you describe could work.
There's no need for schematic software, just sketch things freehand, take a photo / scan it, post to a image hosting site and send a link.
Pete

[isaac]

It's easier for me just to draw it. Here it is. this should get you there. https://lh3.googleusercontent.com/-g...93/PT+bias.bmp
How would you change that. The winding, when measured with your multimeter probes on each of the wires from the tranny, has 100vac, now when I measure voltage coming off of the diodes, i get -2 volts. Could this problem be due to no CT on the tranny? or do I have other problems. I got a similar problem with a half-wave rectifier there. I also measure upwards of 45vac out of the pot. I'm sure that's wrong. The diodes are IN5400's which seem to be basically like huge IN4007's. I'm guessing that filter cap could also be a little small. I'm not sure.

[mac dillard]

Well, I kinda got lost in what you did and what you want to do. But, if you built a 18 watt marshall pre amp and are using it with a PS using El 34's instead of the EL 84's then I think you will have to make some changes to the PI circuit. Not sure, been a long time but I do remember that there are some mods to make to the PI when you convert that amp to 6V6's. As for the cathode bias resistor to use with the El 34's, 200 ohm 10 watt works good with a plate voltage of about 370 VDC. Go to Home_Page and take a look at the schematic for the dd-30 amp.

[pdf64]

That rectifier can't work, it would need a center tap.
Half wave is fine; as you say you tried it and it didn't work, could you provide a sketch of the circuit you tried? Maybe there was a mistake in there?

Why choose beefy diodes? The bias supply only requires very low current.
Pete

[frus]

It's easier for me just to draw it. Here it is. this should get you there. https://lh3.googleusercontent.com/-g...93/PT+bias.bmp

it won't work that way because your winding is also referenced to the ground. just remove the lower diode and increase the cap to some 100uF 150V. Also, never trust a wiper to supply the bias voltage, I'd rather put a series dropping resistor and wire the pot to the ground like a variable resistance. something like this:

http://www.davidsonamp.com/sf/images/super5f4.gif

only you replace the 56k resistor to ground with a pot (or better yet a pot and a resistor in series, say 50k pot and 22k resistor)

[isaac]

What do you mean remove the lower diode? Take it out and the. Solder that wire to ground? Remember I have a 100V winding on the transformer for this. I'm not just taking voltage from the Main high voltage winding. So one side of the wind goes to the diode, and the other to ground. Then the other side of the diode goes to the top of a pot and a cap to ground. Also you said not to use the wiper. So use it as a variable resistor to ground in series with another resistor. Bigger cap also. Also, I use the large diodes since I have only a few in4007's on hand. They seem to work just as well. Now if I made an artificial center tap, could it work as it is? My main high voltage rectifier in this amp is just like that except the bias one doesn't have a CT to ground. Right now I'm using mobile, so no sketch of my first half wave try, but its the same as the other sketch except lower diode is removed and that wire is sent to ground. I'll hopefully get to try this stuff tomorrow. If all else fails, I might go back to cathode bias with the bridge rectifier. Just then I'd have to increase the voltage dropping resistor going to the preamp and PI plate resistors. Say from 8.2k to 15 or 20k 6-10W resistor. But that's for another time.

[frus]

What do you mean remove the lower diode? Take it out and the. Solder that wire to ground? Remember I have a 100V winding on the transformer for this.

is your winding totally separate from the main hi voltage winding, or is it similar to fender's scheme I linked?
do you have a spec or something so we can see?

If it's separate, you ground the "other" side of the winding, yes.
if it's not, then the other side is center tap, which is grounded already

you can use one of these circuits. you'll probably have to adjust the 15k resistor to bring the voltage to exact range you need, I just copied these numbers from the bassman. note that in the second circuit you need smaller caps, but two of them

[isaac]

It's a Totally different winding. Yea maybe I did something wrong the first time. I'll try the half-wave style. The very first time, I did tap off the main high voltage winding kind of like the super, but not on a tap off that winding, just the top wire. The dropping resistor I used there was a 470 2W and it blew almost instantly, so I shut it off, and tried the half-wave style on the other winding that I'm using now, I'm not sure what happened that time, that 470 2w just started smoking like crazy so I shut it off. I'm also going to try a IN4007 instead of the big guys. Even though I'm not sure what the difference may be, Everyone uses them so...

[frus]

I don't quite understand what did you do the first time, what do you mean just the top wire?

Of course you can use the beefier diode if you got it on hand, it's not gonna cause any problem

[isaac]

I mean the top wire as in, two wires from the xformer winding. one is on top in the schematic and one is on the bottom in the schematic.
I know diodes have threshold voltages (?... or something like that) so would those have anything to do with this?

Here's what I described as the very first thing I tried, that blew the 470 2w resistor almost instantly. https://lh5.googleusercontent.com/-B...Bias+schem.jpg

[frus]

of course, the main HT winding's voltage is too high , maybe he meant 470k
typical Marshall bias circuit shows 220k from HT winding:

http://www.drtube.com/schematics/marshall/1987u.gif

[isaac]

No i tried just 470 but thats ok.
So I just finished messing with teh amp, and i reverted to teh half-wave style there, adn put a much higher filter cap in. Now the bias works great. and it had weird voltages on the amps normal rectifier, so i put in a bridge rectifier, and it sounds awesome now. The output tubes were redplating so I adjusted teh bias and now they are working great. I calculated it out, and I'm getting almost 75W so I think for two el34's I did pretty well. It didnt sound that loud when dimed, but I was using a blown speaker so... idiot me. Thanks for the help. I've been trying to figure this otu for a while.