Power Supply Help

[TIMBO]

Hi guys, I powered up the little Booga with some success.I'll start with the Power supply.I was trying to get close to the voltages that Mesa had on their schem. I was able to achieve a good match for the preamp/FX loop by using the SS rect and the 300-0-300 PT but the voltages for the power amp i think are a bit high for the EL84s ( wanting to be in the marshall 18w area). Would i be able to add a dropping resistor before Node A to reduce A & B node it by about 50v ( what would that approx be and would it need any filtering) and if that is done i'll need to change the resistor between B & C to keep C in range of Mesa voltages. At the moment i have 366/344 at node C is this OK for V4 reverb driver.When fired up the EL84s were showing that they weren't a great match. Would cause it to produce a sound like a low note from a bass guitar. Thanks

 

[kagliostro]

Assuming you want to drop a bit the SS DC you can put a resistor in series with the DC path between the SS output and the standby switch (blue resistor in the example)

this resistor (without filtering cap) will also add a sag effect to the SS DC

I think no filtering cap is needed there also if you want you can add one, but if you add one, pay attention that with your arrangement the tube max filtering cap value is not exceed

if DC voltages are too high in both situation (SS and Tube rectify)

you can put drop resistors in series between AC and rectifiers (no filtering cap there) (red resistors in the example)

Kagliostro

[TIMBO]

Thanks Kagliostro, The red resistors could be the way to go but i would put them after the preamp rect before the SS/ Tube switch to keep the preamp with the full 300v.What size resistor would i require to bring it down to 290-0-290 (typical marshall voltage).Would anyone have a voltage chart for a 18w marshall power amp stage to compare. Thanks

[jojokeo]

Timbo, my thoughts are that I'm not a big fan of resistors in the power section like that "causing sag". You'll get enough of and a certain amount of sag when you crank up the amp creating it's natural load condtions. Getting "extra" isn't something I prefer as I like a punchy/fast attack and sensitivity when playing faster phrases and passages and power chord strumming. Sag seems to slow down my playing in certain respects but for slow blues stuff it's fine/fun, although that's not the only thing I play. If that's something you like then it would be a good suggestion and also lower your B+ at the same time. A better way of dumping some voltage is w/ zener diodes or a VVR circuit to name a few. Also, if you bias the power tubes right on the edge of thier power dissipation, you'll use more current which also has a lowering of voltage effect too (depending on the size of your tranny's capacity rating). Lastly, the more important issue here is to make sure that the screens are not exceeding their power dissipation limit as this is more apt to lead to the tube's distruction than the higher plate voltages. As long as they aren't red plating your okay there, but this could be due to the screens being exceeded too. You may have to use 22k resistors or more on the screens to keep them happy w/ the high voltage.
I'm just shooting from the hip w/out data sheets but if the screen uses 5mA for example and you have 400V and want to drop to 300v, then you calculate the voltage drop amount you want at your guess-timated current and use ohms law: 100v/.005A=20k resistor to start with. Put it in, warm the amp up a couple of minutes and measure what you get. Also know that under actual use there will be more current running through the screens than the static condition you're testing on the amp at idle.
Hope this helps? Joe

[sluckey]

Here are my voltages...
http://home.comcast.net/~seluckey/amps/18w/18w.pdf

[TIMBO]

All good info, Thanks.I like the idea to increase the screen resistors as node C is slightly lower than i want. The OT i am using is a Magnetic Components 18w marshall with 10.8k ohm primary @ 4ohm load,9.2k ohm @ 8ohm load and 8.4k ohm @ 16ohm load and using JJ el84s so you are saying that the voltages for node A might be OK? As for the sag i was trying to have as little as possible, hence the two rect tubes (although i am only using one at the moment)The idea that sag might slow my guitar playing  I WISH!!!! Thanks

[TIMBO]

Hi guys, Did some more testing. I am concerned that the voltages for the JJ EL84s are too high. So this is what i have done, i changed the screen resistor from 1k to10k with all voltages taken with SS rect.
                         pin 2    pin3    pin7   pin9
V7 Cath bias       -14.5    16.6    419    395
V8 Cath bias       -14.5    16.6    419    395

V7 Fixed bias      -15.5     27.5    424   401
V8 Fixed bias      -15.5     24.5    424   402

I would like to change the 10k to 20k to see what would happen.Doing this doesn't reduce the voltages on the plates. Is this a concern?
I tried a 10k/3w resistor on the centre tap of the OT but soon relized that was a mistake  :shocked: no damage done.
At the present voltages there was no signs of RED PLATING (not having experienced it, i thinking the metal plates inside the tube start to glo red)
The amp does produce sound :-
Hum in large amounts - that could be due to the rough layout and long runs of wires
NF not working with fog horn sound - May have fixed it by swoping OT wires
Sound to 3 on volume any louder produces squeal -
Reverb level produces Squeal after 3 on the dial   - Thinking these two are related due to the joint plate resistor as shown on the schem.
Not sure how the FX switching is used?
Volts on Plates of V4 @ 360v - Would like to reduce 10k resistor to 6.8k to increase volts on PI but this will increase volts on V4, is this OK?

Thanks

[RicharD]

19 pots & 13 switches!   

[sluckey]

                         pin 2    pin3    pin7   pin9
V7 Cath bias       -14.5    16.6    419    395
V8 Cath bias       -14.5    16.6    419    395

V7 Fixed bias      -15.5     27.5    424   401
V8 Fixed bias      -15.5     24.5    424   402

I don't understand your voltages for pins 2 and 3. If switched to cathode bias, pins 2 should be zero volts. If you ground out the bias supply when in cathode bias, where can the -14.5V come from?

And when switched to fixed bias, pins 3 should be nearly zero volts, except the 1Ω resistors will show a small voltage in the two digit milli-volt range. Maybe you meant 27.5mV and 24.5mV?

That's a busy layout. Lot's of places for hum and squeals to sneak in. Is this a prototype experimental layout, or do you want this to be a workable gigging amp?

[Willabe]

19 pots & 13 switches!   

Yeah, wow and I think 10 tubes.
 
 
               Brad           

[TIMBO]

Sluckey,

                         pin 2    pin3    pin7   pin9
V7 Cath bias          0       16.6    419    395
V8 Cath bias          0       16.6    419    395

V7 Fixed bias      -15.5   27.5mV  424   401
V8 Fixed bias      -15.5   24.5mV  424   402

I don't understand your voltages for pins 2 and 3. If switched to cathode bias, pins 2 should be zero volts. If you ground out the bias supply when in cathode bias, where can the -14.5V come from?

And when switched to fixed bias, pins 3 should be nearly zero volts, except the 1Ω resistors will show a small voltage in the two digit milli-volt range. Maybe you meant 27.5mV and 24.5mV?

Where did i get the -14.5 I don't know. I took new voltages and as you said it was "0"and the 27.5 & 24.5 are mV.
This build is purely learning and fun 
I'm not too concerned with the hum at the moment, hopefully in a proper chassis and better wiring this can be addressed.
The squealing has gone after putting a separate power resistor on V1a plate.
The next thing is the reverb at low LEVEL works OK but turn it up LOTS OF BUZZ.
Thanks

[DummyLoad]

jeez! one size fits all.   

buzz is likely grounding scheme related or poor PS regulation. try using parallel taps instead of series chain.

nice prototyping.   

[TIMBO]

Hi guys, i haven't had time to fully investigate the BUZZ,but i did notice that when the XF loop is ON it disappears. I thought it may have been too many volts on the 12at7 so i found another power source, NO change. All grounding for the reverb is to one point then connected to the earth buss on the circuit board. Thanks