Marshall 100 high voltage AC question

[cortezthekiller]

Hey,
This is my first post, but I have been building amps for a few months now with the help of this forum.

I'm building a Marshall "Plexi" 100 using the Hoffman and ceriatone design ideas and referencing the Super Lead 100 schematic on schematic heaven. I'm recycling a TCM head (made in Ottawa Canada) which seemed to use the same tubes (4 El34's, 3 12ax7's). My question is this:

I omitted the 2 .22uf caps connected to the high voltage AC right before the rectifier, and am using 1n4007 diodes. I have 600vdc caps in this range and was wondering if they will be ok? I know the caps are supposed to smooth out the AC "spikes", and the diodes seem to drift until AC leaks through and a fuse blows. I am getting higher voltage handling diodes, and hopefully this will help as well.

Thanks.
-Dave

[Chuck H]

If your 1n4007 diodes are "drifting" they are bad componants. I doubt that is the case. Probably 99% of the 100 watt Marshall's and their clone cousins have used that diode in that application. I think your problem is somewhere else.

Don't worry about the .22 caps yet. Most amps don't use them. Get the power supply working properly first and then add them if you like.

If your blowing fuses there is something wrong. I'm sure you've checked and rechecked your work, but you need to keep checking until you find the problem.

Pull the tubes and take some voltage readings so we can see whats going on better. B+ after the rectifier, screen voltage and plate voltage on each el34 socket. Do you know what the PT voltages are? Does the fuse still blow with no tubes in?

Also, draw a schematic of what you've done so far based on what you find in YOUR build. That can help you to find errors. Or at least you can post it here so we might be able to.

Chuck

[cortezthekiller]

Well, i bring the voltage up on the variac, and the two first filter caps in series start to hiss and heat up quite a bit. The total AC voltage from red to red is over 1000 volts, and the voltage of the B+ after the rectifier is around +550 volts. And the two caps were rated at 390 volts each with a 100k resistor across both of them, so that should give 780 volts of handling (right?).
The center tap of the high v is connected to the middle of these 2 caps.
Either way, after doing the power formula i've calculated that perhaps these two resistors should be about 5 watts for some leeway.
I'm not sure if the caps need to be formed or not, but i could do that by bringing up the voltage slowly over a period of time.

BTW, the amp did work for about 20 mins or so until an arc seemed to have its way with it and the diodes seemed to have drifted lower.

thanks.
-Dave

[cortezthekiller]

it seems that a half wave rectifier with the center tap grounded and new caps at the start of the B+ did the trick. I will post updates as they arise.

[Bruce / Mission Amps]

Wordy and we need artwork

Hmmm... something fishy there.
You must supply a drawing because I think you have misunderstood or miss-explained the layout...

If you are using the full 1Kv with the center tap connected to the +- junction of two main filter caps, then you have to have a full wave bridge rectifier. The B+ would be astronomical.... 1200vdc-1300vdc easy.
I don't think it ever did but if the center tap is grounded then I'm thinking you might not fully understood the layout and the relationship of the two filter caps, the Hi-V center tap, the actual ground point with respect to the rectifier (no grounded anodes), those two filter cap's negative leads and the bleeder/balancing resistors.
I suspect each cap has a resistor across it which is grounded on one end with the center tap of the Hi-V.
A full wave rectifier would have both negative ends grounded and your grossly under voltage rated caps could be now burned up trying to handle +600vdc.

... if your total "high voltage" from end to end is over 1Kv... that would be over 500v from the center tap to either end of the full wave rectifier. Your filtered B+ (with known good main filter caps) should be around 625vdc to maybe as high as 675vdc, but that would be if the four power tubes are not idling too high and the power tranny has a proper ma rating.

Regardless, I still suspect the filter caps in this amp are now shot and shorting out so I'd disconnect them asap.

Look up the difference between a full wave rectifier and full wave bridge and see how the negative ends of the filter caps could be drawn as connected in series but back to back, with their common negative leads actually grounded with the center tap.

Get rid of it all after the cathode ends of the 1N4007s and try using a pair of 22uF to 47uf @350v in series The top one's negative lead in series with the positive end of the lower one and use a pair of 100K-220K balancing/bleeder resistors. You know the drill.... one resistor across each cap.

[cortezthekiller]

Thanks Bruce.
I've currently replaced the filter caps and added the bleeder resistors which seemed to do the trick. The B+ seems to be quite normal with nothing connected to the rectifier (about 600 v), then jumps up to around 800 with the first filter cap connected, and finally moves down to about 650 or so once the standby is engaged. The two caps i used were a 100/100uf @ 500v (the two + pins connected, and a 100uf @ 350v finally connected to ground with the resistors across both.

I am monitoring it for any weird behavior and sounds or overheating.

[cortezthekiller]

the schematic i've referenced is on the third page

[pappie]

Power Supply Problem

You should use (6) 50+50uf 500v caps and (4) 56k 2 watt resistors.
2 for mains (wire in series, 56k across + & - on both, ground second cap.
2 for screens (wire in series, 56k across + & - on both, ground second cap.
1 for PI (wire as 100 like the mains and screens).
1 for Preamp (split this one 50-50).

Should be cool after these changes!

Pappie

[cortezthekiller]

Hey,
I've been running tests with the transformer out of circuit, and it is simply producing too much AC to work with. Even in the amp, with the caps replaced and setup properly. The amp was working fine, but the B+ would damage the tubes in the long run.

I had an old Hammond iron with slightly larger dimensions, with a high volt of about 420vAC. I replaced it and now the tubes are biasing properly with the B+ well within their rated range.

I have a few questions:

I've also recycled a large output transformer with a 5K primary with 4,8,16 ohm taps.
I know this is off from the 3.6k the tubes want to see, what are the issues that result from a mismatch like this.

The amp sounds fine and runs up to about 80 watts clean, but what will changing it to the exact value do to the sound/tubes etc.?

Would using the 16 ohm tap into an 8 ohm load balance the ratio of the OT?

I'm not in a financial state to buy one, but would it be worth it?


Thanks
-Dave