4 Tube Amp From Scratch

[R.G.]

if you have nothing to contribute to the project, then why are you here ?

I believe that you said:

Yes, close enough... I am dissapointed some of the others are not helping you with this.... They seem to be off into their "own little worlds"...

I then pointed out that (1) I was watching, (2) was trying to be polite to you based on your earlier comments about my help, and (3) you seemed to have it well in hand.

I have no trouble digging into the technical stuff, as you may remember. I was (politely, I think) keeping my mouth shut out of deference to your sensibilities. Then you flung out that little barb about

They seem to be off into their "own little worlds"...

That seemed a little more caustic than was really justified by the situation.

But back at the problem at hand. I'm confused. I'll help if I can, but would you tell me whether
(1) you can handle it on your own?
(2) you think you can handle it on your own, but you're having a self-confidence crisis, so you want others to chime in and reinforce your ideas?
(3) you can't handle it on your own and need help?
(4) if number 3 is true, then you want anyone to help, including me?
(5) if number 3 is true, then you want anyone but me to help?
(6) you don't want to disclose your tone control secrets for fear of them being appropriated by someone?
(7) you can't disclose your tone control information for security reasons?

Is there one I missed?

[Chuck H]

Is there one I missed?

(8) He won't disclose his tone control secrets for fear of being shown that they are not secrets or are somehow flawed.

[TubeNoob]

Hiya fellas,

Some Ebay bucks just came in with enough to pick up a PT that seems better suited than the honker i have here (thing must be over 15lbs!).
Unfortunately - I think people have caught on to Geloso stuff and the prices were out of my range - drats!

Here's a photo of the PT i grabbed:
http://i.ebayimg.com/00/s/NzY4WDEwMjQ=/$%28KGrHqZ,!loE64s,sRPtBO-F9eG%28vQ~~60_12.JPG

220v Primary
Three Secondaries:
195v .2A
12v 1.5A
12v .5A

It comes with a 100uf / 350v filter cap, and what looks to be two 300uf / 25v ones for the 12v secondaries.
It also says it has 'levelers' (its in Italian) which my guess refers to voltage regulators - but I reckon i would be using a voltage divider and the 6V regulators i have.

I have updated my schematic (with a nod to JMFahey!) to ditch the high-voltage filament tubes for ones i have on hand that i can use in parallel.

So here's a few questions I have:

1) Would i be better off using SS rectification for this to squeeze out a bit more voltage and reduce the heater load? (Wondering how much 'sag' you get on a SE anyway, after what i've read and what i hear on my Princeton build which uses a 5Y3) and would i need something in there to slow down the power surge on start-up?

2) How much overhead in terms of current would I need for the filaments? For example:

12v/1.5A secondary >> 6X4 >> 12AX7 (run at 6v) >> 6AT6 (totals 1.2A)
12V/.5A secondary >> 6V6 (.45A)

Is that cutting it a little too close?

OR i could do...

SS Rectification (I have 1N4007's) and:
12v/1.5A secondary >> 12AX7 >> 6V6 >> 6AT6 (totals 1.05A)

I could also throw the 12AX7 over to the .5A 12v secondary if more headroom is needed.


3) What changes should I make to, for example, plate load resistors given that I'm swapping out the 50B5 for a 6V6 and the 35W4 with a 6X4?


PS. I changed the first two B+ filter caps to 100uf from 40uf.


Updated schematic attached...

[R.G.]

220v Primary
Three Secondaries:
195v .2A
12v 1.5A
12v .5A

It comes with a 100uf / 350v filter cap, and what looks to be two 300uf / 25v ones for the 12v secondaries.
It also says it has 'levelers' (its in Italian) which my guess refers to voltage regulators - but I reckon i would be using a voltage divider and the 6V regulators i have.

Perfect rectifiers and a nominal line will give you about 276V from that (195 * 1.414). But your AC line will vary, and in particular, you may have 240Vac locally, which then varies about +/-10%, so you could be getting 276*240/220*1.1 = 331Vdc (ignoring rectifier loss and ripple) from solid state diodes. You could also be getting 12*240/200*1.1 = 14.4Vac on the heater windings. The Ac power line varies too. It also sags, and the transformer sags.

These are just things to be aware of. Design for nominal, then calculate what happens with variation.

1) Would i be better off using SS rectification for this to squeeze out a bit more voltage and reduce the heater load?

A tube rectifier will lower the output voltage a fair chunk - 30-60Vdc depending on the rectifier, filter cap and loading. Up to you on whether this is a problem for you. Notice that there are problems with the rectifier tube and filter caps as mentioned below; also think about this never being a huge power amp anyway. Small amps which can be massively overdriven give good sound at lower volumes and are highly thought of by some players. "Better off" depends on you and what you want from the amp. If it were mine, I'd use solid state diodes for reliability. You can stick in a power resistor after them for sag if you like.

would i need something in there to slow down the power surge on start-up?

Probably not. Overall amp power too small. That's just a guess.

2) How much overhead in terms of current would I need for the filaments? For example:
12v/1.5A secondary >> 6X4 >> 12AX7 (run at 6v) >> 6AT6 (totals 1.2A)
12V/.5A secondary >> 6V6 (.45A)
Is that cutting it a little too close?

That's fine. Heaters are a constant load, and don't need overhead unless there is some other confounding problem. If nothing messes with the heaters, they just sit there, no pulses or changes to need overhead. If the transformer is overloaded on another winding, you can sometimes cool things down a bit by UNDER loading one winding, but that's playing with fire. If I were you, I would not sweat overhead for heaters. I do think it's a good idea to fuse the heater windings at just above their rating.

SS Rectification (I have 1N4007's)

I like this alternative, but that's a personal choice.

PS. I changed the first two B+ filter caps to 100uf from 40uf.

That's a problem. 100uF is too large an input cap for most tube rectifiers and the 6X4 in particular. Tube rectifiers have a maximum first filter cap because too large a filter cap causes big current pulses through the rectifier, and this degrades the emission surface of the cathode. For the 6X4, the specified maximum first filter cap is 10uF. If you want to use a 100uF first filter, you're pretty much deciding that you'll use solid state diodes to rectify. Well, or spend a LOT of money on replacement rectifier tubes.

[TubeNoob]

Thanks RG for another in depth answer!!!


Sounds like I'll go with SS rectification (without sag resistor and can save that 6X4 for a bigger project).

You seemed to say that 100uf is cool with SS rectification, is there a problem with using a 100uf as the first filter cap and then, say, a 40uf right after it or do they need to match in value?


As for heaters, I'll probably do this then:

12v/1.5A secondary >> Voltage divider to get 6V >> 6v regulator >> 12AX7 (run at 6v) >> 6V6 >> 6AT6 (totals 1.05A). Sound cool?

The 3rd secondary winding (12v @ .5A) i would not use unless there's some benefit to running that 12AX7 off of it (noise, maybe?).


My question regarding any benefits to the higher voltage is that the datasheets for those tubes say normal operation is 315v (plate of 6v6); 100-250v (plate of 6at6) and 250v-330v (plate of 12ax7). So was just wondering if that meant i'd have less headroom at the lower voltage to the point where it sounds lame.

The sound I'm after is definitely leaning more towards the fuzzy-to-distorted sound (and anything i can do to make it less 'Fender' sounding) and I am a fan of cranking up low-watt amps (and not blowing my ear drums out with, say, a Hi-Watt 100 cranked up!).

By the way, I am pretty sure its 220v here, but I will test output voltage once i have the transformer.

THANK YOU AGAIN!

[R.G.]

You seemed to say that 100uf is cool with SS rectification, is there a problem with using a 100uf as the first filter cap and then, say, a 40uf right after it or do they need to match in value?

No problem. With SS rectifiers you can use any amount of capacitance. The pulse currents will not kill SS rectifiers like tubes. The caps don't have to match. They only need to be rated for enough volts.

12v/1.5A secondary >> Voltage divider to get 6V >> 6v regulator >> 12AX7 (run at 6v) >> 6V6 >> 6AT6 (totals 1.05A). Sound cool?

Maybe. Let's think. Heater loads are very predictable in most case. Using a voltage divider to get 6V out of 12V works, but it makes more sense to just use a series resistor. A 6 ohm resistor would "eat" six volts at one amp, and leave 6 volts at one amp for the heaters you have. In addition, that big a series resistor would soft start the heaters, a side benefit. Same as a voltage divider, but uses the heaters as part of the divider. It is quite difficult to drive units-of-amps from a voltage divider and get good results without wasting a lot of power. A single series resistor wastes 6W. That's the minimum that something like this can waste. Another way is to put a resistor in series with EACH heater that eats 6V at the heater current.

The only lower-waste version is to figure out if there is some arrangement of stacking heaters and paralleling resistors etc. That would let you concoct a 12V stack of heaters. That would cut the waste heat, at the price of some more complexity, and uncertainty about what happens when one heater burns out or a tube is pulled. Making 6Vdc and regulating will use more power. The one-resistor-per-heater approach makes things more independent.

The 3rd secondary winding (12v @ .5A) i would not use unless there's some benefit to running that 12AX7 off of it (noise, maybe?).

No advantage/disadvantage, particularly.

My question regarding any benefits to the higher voltage is that the datasheets for those tubes say normal operation is 315v (plate of 6v6); 100-250v (plate of 6at6) and 250v-330v (plate of 12ax7). So was just wondering if that meant i'd have less headroom at the lower voltage to the point where it sounds lame.

Not an issue, because of a quirk of the human ear. The ear is very sensitive to quiet sounds, and gets less sensitive as sound pressure goes up. Most guitar speakers produce something like 98 to 100db sound pressure level for one (!) watt. Getting to a subjectively twice as loud SPL takes not twice as much power but about ten times as much. So a 10W amp is sensed by humans as about twice as loud as a 1W. To get four times as loud needs a 100W amp.

So the slight difference in voltage will not make the difference between great and lame. Different speakers can make a huge difference, though.

The sound I'm after is definitely leaning more towards the fuzzy-to-distorted sound (and anything i can do to make it less 'Fender' sounding) and I am a fan of cranking up low-watt amps (and not blowing my ear drums out with, say, a Hi-Watt 100 cranked up!).

You're going to be very happy when you get done.

By the way, I am pretty sure its 220v here, but I will test output voltage once i have the transformer.

Smart!

[TubeNoob]

RG - thank you yet again.

A 6 ohm resistor would "eat" six volts at one amp, and leave 6 volts at one amp for the heaters you have.

1) Since the secondary is putting out 1.5 amps, does that mean a 6ohm would eat up 750ma and leave 750ma?

2)How is that calculated by the way? (as in, how do i find out the amount of resistance of the tube - being the 2nd part of the divider - i checked the datasheets but didn't see any helpful info).

3) That would need to be a 10watt, wouldn't it? I do happen to have a 6.2ohm 10watt resistor on hand.


4) Just curious - does anyone know any places out there which sell caps and OT transformers? I've had a hard time finding both in one spot where it wasn't overly expensive. (Except Weber Speakers).

5) I am looking at getting a decent OT for this given not only all the time I'm putting in but also all the time you guys have put in!!! :-) What would I need to look for in an OT?


A few questions regarding values:

6) Regarding C14 / 16uf cap off of the B+ line - what range values are ok for this?

7) C1 / .047uf (on the cathode of V2a) - This and C14 are the only caps I don't have on hand....DRATS! So, what range would i have on this cap, if any? (This has to be electrolytic does it not?)

8) @MooreAmps - I added that 100k resistor before the volume pot...what's the story on that? (How does it help and how does it work?)

The sound I'm after is definitely leaning more towards the fuzzy-to-distorted sound (and anything i can do to make it less 'Fender' sounding) and I am a fan of cranking up low-watt amps (and not blowing my ear drums out with, say, a Hi-Watt 100 cranked up!).

You're going to be very happy when you get done.

9) Do you say that because of: the distortion, the 'not so fender' sound, or that i like cranking low-watters?? :-) Either way, I'm stoked you say so...

[R.G.]

1) Since the secondary is putting out 1.5 amps, does that mean a 6ohm would eat up 750ma and leave 750ma?

2)How is that calculated by the way? (as in, how do i find out the amount of resistance of the tube - being the 2nd part of the divider - i checked the datasheets but didn't see any helpful info).

The secondary doesn't put out 1.5A. It has the *ability* to put out *up to* 1.5A without overheating. What the secondary does is make 12V. You connect a load to the "12V" and the load pulls current from it according to Ohm's law, that is I = V/R.

The actual resistance of the tube heaters varies as they heat up, but you can compute what the nominal hot resistance is. A 6V6 for example has a 6V heater that wants 450ma at 6V. So we want to figure out how to get it what it wants from 12V. The "resistance" of the heater is just R = V/I = 6v/0.45A = 13.33 ohms. So if we wanted to run a 6V6 from 12V instead of 6V, we would subtract the 6V that it wants from the 12V we have and get 6V (the fact that this is the same as the heater voltage is a coincidence). We know the heater wants 0.45A, and the same current has to come out of the 12V supply, so we need to put a resistor in series with the 6V6 heater that will cause 6V to appear across it when 0.45A is flowing through it, or R = 6V/0.45A = 13.33. Again, this being identical to the "resistance" of the heater is coincidental.

That's how you do it for one tube. You could do this for each tube separately. To find the "resistance" of the heater winding at full operating temperature, you use Ohm's Law and compute R = V/I.

If you paralleled all of your tubes to get the 1.05A at 6V, then the resistance you need to drop 12V down to the 6V the heaters need is (12V-6V)/1.05A = 5.71 ohms. If you use 6 ohms, this will limit the current in the heaters to a little less, but it will probably work OK.

3) That would need to be a 10watt, wouldn't it? I do happen to have a 6.2ohm 10watt resistor on hand.

For any device dissipating power, the power is the product of the current and voltage. So P = I*V. And we know the additional resistor has 6V across it and 1.05A through it, so that's 6V * 1.05A = 6.3W. Resistors get to their full rated power with something like 100C to 200C on their surface, so a 6W or 7W resistor might work but would get HOT. A 10W would be cooler, maybe enough not to be a burn hazard to bare skin.

7) C1 / .047uf (on the cathode of V2a) - This and C14 are the only caps I don't have on hand....DRATS! So, what range would i have on this cap, if any? (This has to be electrolytic does it not?)

General electrolytic caps are used only when a suitable capacitance and voltage are not available in a NON-electrolytic type. Generally, under 1uF or so there will be no need for electrolytic. For 0.047uF, I'd probably use a film type with the proper voltage rating.

9) Do you say that because of: the distortion, the 'not so fender' sound, or that i like cranking low-watters?? :-) Either way, I'm stoked you say so...

I was responding to the "cranking low-watters" but also to some pride you should have when you get it finished.

[TubeNoob]

Thanks RG for a great education on that stuff. I know the basic principle/formula of Ohm's law but applying it has kinda been like knowing a car has a gas pedal, brake and steering wheel and being overwhelmed by everything else involved in driving! That really broke it down for me in a way I can apply in the future. Very cool.

Didn't know that about electrolytics - that's great, and I do have a .047 film cap that's rated for 600v (tho I know only 50v or so is needed there).

And yes, major pride over this, even now!



A few remaining questions i had:

1) How does this transformer look for this project?
https://taweber.powweb.com/store/lopo_ot.jpg
("Flea OT" - 4k and 1.5k primaries, 8 and 4 ohm outs, 3 watts)

2) I do have the OT that came on the radio, it's only about a 1/4" smaller than this Weber one. Is it suspected that it won't sound good given that it was for a cheap old radio and not a guitar amp or some other reason? Just curious, I'm still looking at getting a new one.

3) Regarding C14/16uf cap - what range of values are ok for this, and how do you determine that, by the way?

4) Regarding the 100k resistor on the volume pot - MooreAmps - what does that do?



Thanks again to all, I'm going to take the first letter from all your names and try to find an anagram from them to give this amp a name! :-)

[mooreamps]

Nope, I've been watching.

It was just a typo..

[TubeNoob]

Heh! C'mon MooreAmps! Tell me what's happnin' with that 100k resistor!
I would really appreciate your input. I think the Fender tone control is pretty lame on my Princeton.

[R.G.]

It was just a typo..

Oh, hey, no problem. Me too. Pretty much everything I've ever typed that someone didn't like was a typo.

[R.G.]

Tell me what's happnin' with that 100k resistor! I would really appreciate your input. I think the Fender tone control is pretty lame on my Princeton.

It isolates the tone control parts from being "shorted out" by the portion of the volume control between the wiper and the top lug. It keeps the "shorting" from messing with the tone control time constants and rolloffs so much and the tone effect from being changed a lot by the position of the volume control. It's got its own set of issues, but for a big enough impedance after the volume control, it works OK.

[TubeNoob]

Ok cool, I will give that a shot. I had guessed some posts back that what was happening is that it creates more resistance going into the volume pot forcing more signal through the tone controls, if i'm not too far off on that understanding (since the vol pot at 100 would just completely bypass the tone controls, nay?)

Well, i'm in the parts-sourcing stage now (!) and will be ordering missing parts Monday, so if anyone has any ideas on those few questions I had in my last post, that'd be killer, especially #1 and 2 regarding the OT.

Also, attached is the latest schemo with my 'guesses' on the different resistor wattage's and capacitor voltages. If any of you see any probs, please do let me know.
(BTW, If you see caps with '600v', its cause i have 600v caps in that value on hand)

Next up i will post my power supply schematic and the layout diagram i just did.

Then, as soon as this cast is off my leg in a week or so, the sweet smell of solder ensues....! Rockin!

Thanks y'all...

[überfuzz]

Slightly off topic
Ever consider a spice software to simulate your beast? They wont clone your amp, but they will give you a fair idea of voicing and break up. Most of the problematic issues mentioned in the ever so polite and objective discussion in this thread would have been revealed. I design all my amps, not many, this way. An analogy would be writing a letter with a pen licking a suspiciously tasting stamp etc, or typing, shuffling the words and clicking SEND e-mail.