Tweed Bassman / Deluxe w/ reverb. any opinion?

[MWJB]

"Thanks for that suggestion MWJB, I think I understood that. (See attachment for what I think you mean't)" - Almost, but tap the reverb from before the 270K mixer resistor, not after it.

[tubeswell]

yep there is plenty of current capacity on the HT and the heater, but how much will all those little bottles pull down the voltage? With 325-0-325, will I still get a B+ of around 430V or not?

[MWJB]

12A#7s shouldn't have a marked affect on B+ voltage, not as much as say rebiasing the power tubes, or brand of power tubes. I wouldn't get too fixated on 430vdc exactly, PT ratings are nominal & there are a host of factors that will affect final voltage...+/- 15-20v or so might be a reasonable target?

[tubeswell]

Thanks MWJB.

I'm working on a layout drawing at the mo'. I'll post for feedback it in a couple of days once I get a version that I think will do the job

:-)

[txstrat]

Haven't been home for two days.
The rectified voltage is the vac times squareroot of 2 (=1.41). (well, roughly)
325vac will result in 458vdc under NO LOAD.
Pluggin' the tubes in will drop the voltage (depending on the idle current) as well as a tube rectifier (look here: http://www.300guitars.com/articles/r...ge-drop-chart/).
In my latest build I used a PT with 300vac rectified by a bridge recto (1N4007).
Has 423vdc under no load and 395vdc with tubes plugged in.
As long as the tubes wouldn't argue with a few volts more or less, at 325vac I'd recon with a value of around 420vdc to 430vdc depending if you use a tube recto or not.

[MWJB]

It's not quite that simple, a lot depends on the current ratings of the PT (some are quoted as loaded, others unloaded) as well as the voltage ratings, in a fixed bias amp even the loaded B+ voltage can swing a great deal...40-50v in some cases.

With a GZ34/5AR4 the 1.41 mutiplier you mention, drops to more like 1.3 (1.26 for a 5V4, 1.2 for a 5U4, 1.1 for a 5Y3), so 422v-ish - IF you really have 325AC exactly coming off the PT secondary and IF you actually have the specified AC at the wall socket (varies with time of day), at a reasonable current draw - change brand of rectifier or power tubes and things change again. I'd perhaps be concerned if I had more than +/- 15-20v difference to expected voltages.

[txstrat]

in a fixed bias amp even the loaded B+ voltage can swing a great deal...40-50v in some cases.

Yes thats's right but under playing conditions, wouldn't it?

I'd perhaps be concerned if I had more than +/- 15-20v difference to expected voltages.

Right again. I'd change the sort of rectifier, brand of rectifier or power tubes if the voltages wouldn't be as what I expected.

[tubeswell]

Haven't been home for two days.
The rectified voltage is the vac times squareroot of 2 (=1.41). (well, roughly)
325vac will result in 458vdc under NO LOAD.
Pluggin' the tubes in will drop the voltage (depending on the idle current) as well as a tube rectifier (look here: http://www.300guitars.com/articles/r...ge-drop-chart/).
In my latest build I used a PT with 300vac rectified by a bridge recto (1N4007).
Has 423vdc under no load and 395vdc with tubes plugged in.
As long as the tubes wouldn't argue with a few volts more or less, at 325vac I'd recon with a value of around 420vdc to 430vdc depending if you use a tube recto or not.

hi txstrat

I already know about the difference between 'loaded' and 'unloaded' B+ supplies. And I already knew about the 'rule of thumb' ratios MWJB subsequently gave for estimating tube rectifier voltage drops. What I was wondering about was the effect of adding another eight preamp gain stages. However it appears from the information I have provided that MWJB is confident it should'nt drop much. However what I haven't learned properly yet is how to work out loads based on all the stages. I suspect is it a complicated matter that involves calculating the impedance and/or DC resistance of each stage (based on how each stage is set up (e.g.; whether it is a conventional triode, or a CF stage, or a pentode and so on) and the figuring out the sum total of the effect of all of these together on the DC supply. So it probably involves complicated theory which I can only get to by more reading the valve wizard for the billionth time, or in the Radiotron Designers Handbook or similar, and no amount of me asking questions here is going to reveal these mysteries any faster. However thanks to the replies I have had about this particular design from my many friends here, I am confident that on this occasion, the amp I am building now will have aproximately a B + of at least 410-430 loaded, which will do.

[MWJB]

We're concerned with B+ at idle.

Add up idle current per power tube, screen current per power tube, say 1.5mA per 12AX7 triode, whatever your 12AT7 and 12AY7 draw per triode (2mA-ish)...and now add a margin for current rise under signal. How big a margin? 50-100%?

Let's go with 100% & play real safe:
70+70+10+10+6+6+6+6+6+8+8 = 206mA. 200mA would do in most circumstances, more might be nice, say 300mA? Be aware many reliable amps have been built with less current capacity than that (Bandmasters & Vibroluxes might only be rated at 150-160mA, many DR's run 5881/6L6 and only have one less preamp tube than you are anticipating & PT is rated at 120mA). BUT under load/extremes of bias voltage drop will be greater. The more current headroom, theoretically the more stable the voltages. You can also see how, even when doubling the current draw on the 12A#7 that 56mA is small potatoes when compared to maybe whacking in some JJ's and biasing to 45-50mA per tube?

Same for the 6.3VAC winding:
900mA+900+300+300+300+300+300+300+300= 3.9A. I would be expecting you to use a 6A winding at least, 8A would be rock solid. As it's only another 300mA over a Super Reverb that would "work" (4-5A). With 6A plus you would have no worries if you wanted to sub in EL34/6550 etc, whereas with the lower ratings you might need a auxilliary 6.3VAC transformer.

Secondary voltages affect each other, so skimp on one winding and you might be fine with 6L6 at 30mA, sub those for EL34 at 40mA and the biggerer draw on B+ & 6.3VAC will pull down voltages.

It's really not that complicated, but some of the rating "ideals" will be hard to find in an off-the-shelf tranny.

[tubeswell]

Thanks MWJB - I'm going to use the 325-0-325 200mA, 6.3V 5A 5V 3A, 50V 1A PT I got

I have done a preliminary layout (forgive the crude drawing, but its intended to be conceptual as to where to position stuff, and I don't have very flash computer software)

The dark green wires are where I think shielded cable should be. I wouldn't mind some feedback on this.

The light green wires are Output tube grid wires.

The red wires are the power rail. I wouldn't mind some feedback on this as well; esp w.r.t. how I have the decoupling caps arranged w.r.t. to the pre-amp (The layout differs slightly from the schematic in that V1 shares a 10uF filter cap with the reverb input and bypass stage, and the DC Pair has its own 10uF filter cap, as do the reverb recovery stages.

I also wouldn't mind some feedback on the overall layout and positioning. I tried to base the main amp on the 5F6A layout, with the trem from the 6G16, and the reverb a bit of a cross between a 6G15 and a BF style. The reverb circuit sits between V1 and the DC pair, then the LTP, and then the LFO circuit nearest to the Filter caps. (The overall width of the chassis I am aiming to build for this is 65cm x 12cm high and 78.5cm deep) I think the overall layout is about right, but if anyone has some wise ideas, then feel free to jump in.

I have simplified the input socket drawing by omitting the 1M to ground (but I will put these on)

I have also omitted the Trem power supply connection, and the Reverb Tranny power supply (both should go to the screen node filter cap)

I have also omitted the rectifier protection diodes that I included on the schematic (but I will put these in)

I realise the drawing of the board is a bit distorted scale-wise (limitation of the computer software I am using), and in reality I expect there will be a bit of wiggle room, as well as clearance from the PT.

Edit - I realised afterwards that the second pre-amp filter cap dropping resistor (going to the direct-coupled pair) is laid out wrong - I will fix this later

[MWJB]

Firstly, you want the PI to be the last tube before the power tubes, keeping the grid wires to the power tubes as short as possible.

Make life easier for yourself, stick the filter caps in a pan, even with a 25 1/5" chassis, your still going to be pushed for space. The preamp filters can probably stay on the board, leave enough room to upgrade to 20uf or more in case you have filtering issues.

To get a better idea of layout, I'd make up a dummy board an physically lay out the components to ensure that you have the space.

[Tubenit]

Bassman w/reverb

I have used this successfully on a number of amps. Not my design. I got it from some of the D'lite builders.

With respect, Tubenit

[Tubenit]

DE3 with reverb

I have used this with an amp similar to the Deluxe 5E3 topology but not specifically on a 5E3. And it worked just fine.

With respect, Tubenit

[txstrat]

Tubenit
Did you use just one 12AX7 for the whole reverb section in these circuits?
I mean this looks quite interresting. Which reverb pan and transformer did you use?

[Tubenit]

one tube reverb

I use one 12AX7 and a standard Fender replacement trannie and reverb tank like one would find in a Princeton Reverb. If I am remembering correctly I have used this on about 7-8 amps all of which were successful.

I usually have the dwell on around 7 and the reverb pot on about 3 ...... maybe as high as 5 on a slow blues. It has waaayyy more reverb than I need but I don't do surf stuff.

With respect, Tubenit