Amplifier mod and rebuilding

File: circuit_diagram.png (134KB, 740x952px) Image search: [Google]

134KB, 740x952px

I noticed that there are some mistakes in my circuit: Wrong fuse T1, Capacitors labled in wrong order, tubes (V1/2/3 are ECC83/12AX7 and V4 and V5 are EL84) and diodes miss values..

Ignore those obvious mistakes of mine

File: 1473119952583.png (66KB, 1505x618px) Image search: [Google]

66KB, 1505x618px

>>1050024
>those obvious mistakes of mine

Were you counting these?

>>1050054
You're right, had not seen that one. When I started working my way through the pcb I made that mistake as I was confused and had not seen the center tap.

Now: Can you help me with my actual questions?

>>1049984
>Is my powertransformer supplying enough current?

There is no part number or specification for the transformer on the schematic.
How are we supposed to know its current capabilities?

File: knob.jpg (99KB, 770x578px) Image search: [Google]

99KB, 770x578px

>>1050091
>>1050091
I'm sorry, that was stupid. The rating, as written on the transformer, for SEC1 is 290/48/0/17/290V and 0.08/0.01/0.15/0.08A.

I have estimated that each triode of the preamp valves, which are ECC83, consumes 1mA. The EL84 in push-pull in the poweramp should consume about 45mA (18W/400V).

That's about 51mA. That would be enough headroom, wouldn't it?

What about the cathode bias? Does it work that way?

I will upload a corrected schematic later.

>>1049984
please put dots where wires are supposed to be connected. otherwise it's hard to tell which wires are connected and which are not.

it looks like the amplifier is already setup for cathode bias via R26. though it's hard to tell without the connections clearly indicated, per my point above.

regarding plate voltages: look online for schematics of similar amps and you'll be able to see if your voltages fall into a good range. there are a million working tube amp schematics out there.

File: schem2.png (240KB, 1346x1701px) Image search: [Google]

240KB, 1346x1701px

Just to clarify: I am rebuilding and modding the amp of which I did not find any schematics on the net. Also, I already took the pcb out. I only want to keep switches, both transformers and tubes. The cheap parts remain on the pcb (I can still redo changes at anytime, it is an easy task).

The original amp had two preamp valves and two poweramp valves. It had a fixed bias.

For my mod I'll add another preamp valve, change the whole "inbetween" circuit, use the original powersupply as much as possible and convert to cathode bias.

>>1050429
>it looks like the amplifier is already setup for cathode bias

No, it IS not, but it WILL be. I drew the schematic with my proposed modifications. These include a third preamp valve and cathode bias.

I am not that sophisticated with tubes.
My questions still remain and maybe someone can help me find an answer or educate me on how to find one:
* Will my powertransformer draw too much current? (So much that it gets damaged?)
* Do I need to change anything on the added preamp valves powersupply? (Anything on the left to D 350V?)
* Is cathode bias with this setup (with this powertransformer) possible?

Here is a new,.corrected schematic with added information.o

If anybody feels that I am way too uneducated on this topic, than maybe you can hint me at some webpages with more infos.

>>1050429
>regarding plate voltages

They are ok. At least for the original 2+2 valve configuration. On the net some amps seem to fall in the same range, some are lower, some are higher. I've read that almost any voltages are ok and guitar amps are often running hot.

>>1050188
>45mA (18W/400V)
If 18W is the expected output, then you're assuming 100% efficiency. Something like 60% is more realistic. You're also supposed to derate the transformer based on your rectifier setup, but I don't remember how that exactly goes.
That said, the current drawn by your preamp tube is pretty minimal and if the transformer was sufficient earlier, it will be sufficient now.

Cathode bias is done like your schematics show, but I have no real clue whether your resistor values are correct. That said, all your plate resistors are 100k, which would suggest that the most cathode resistors should be of the same value.

V3b's gate is unconnected, V3a's goes directly to ground and those resistor values around V3 look weird to me.

>>1050762
much easier to read with the dots!

the cathode resistors look a big weird compared what you usually see in marshall circuits. study the JTM45 and 18 Watt TMB circuits for a bit of help there. you really want to lower the cathode resistance on the first stage - this will give you more gain and take your guitar signal farther away from the noise floor.

you could spec the same transformer that people use in the 18 Watt for this build. the way to spec a transformer is first estimate the current draw of each valve including heaters, then use that to determine the output power requirement of the transformer.

what are R20 and R21 supposed to do? something related to S1? it looks like it will just fuck up the push/pull output stage by forcing the bottom tube to work on its own.

File: schem3.png (240KB, 1346x1701px) Image search: [Google]

240KB, 1346x1701px

>>1050775
>you're assuming 100% efficiency
Up to eleven!

So you do think that it will be enough.

>Cathode bias is done like your schematics show, but I have no real clue whether your resistor values are correct.

Yeah, me neither. Well I had yesterday, somehow I calculated it :/ 150 seems to be a good value as I found it on many amps with two EL84s, even if they had variable voltage selectors. I'll change it if it sounds wrong.

>V3b's gate is unconnected, V3a's goes directly to ground
Fuck ups of mine -again- that happened when I changed the aspect ratio of the image.

>and those resistor values around V3 look weird to me.
Got them from 18W Marshalls.

File: picunrelated.jpg (110KB, 1050x490px) Image search: [Google]

110KB, 1050x490px

>>1050779
>the cathode resistors look a big weird
I'll check that. Thanks.

>spec the same transformer...
Nope. I'm out of money. For me, transformers are to expensive at the moment!

>what are R20 and R21 supposed to do? something related to S1? it looks like it will just fuck up the push/pull output stage by forcing the bottom tube to work on its own.
Exactly, great, isn't it? It should work. Just one tube doing all the work equals half the power and just slightly changed tones (?). We'll see.

The resistors... I just decided to stick them in there. Got some 1Ms lying around, while I have to order 470Ks. I should change that for correctness in the next version of the schematic.

Pic unrelated. It's another amp of mine that I bought without case and then built myself. The image was taken when it was unfinished. Unfortunately I do not have better pictures at the moment.

>>1050816
>too poor for new transformer
okay do the same power calculations and see if your existing transformer can supply enough power

>"half-power" switch
that's probably going to sound like dog shit to be honest. these kinds of amps have master volume knobs instead of this for a reason

regarding the missing connection to the "other end" of the phase inverter - ground it with a resistor per the classic circuits, or install a negative feedback loop that plugs in there. though with your shitty half power switch, i'm not sure NFB is going to help with anything.

>>1050829
>that's probably going to sound like dog shit to be honest
Yeah, unless he uses another switch for changing the remaining tube's biasing to class a.

File: 20160827_145634.jpg (844KB, 2048x1152px) Image search: [Google]

844KB, 2048x1152px

>>1050809
The cathode resistor value for the output section will vary. You can start with whatever value you've calculated, but you'll probably need to tweak it once you fire the thing up and check plate voltage and current. For example, old Fender Vibro Champs have a 470ohm cathode resistor. They all run too hot, decreasing tube life. I typically end up putting a 1k or 1k5ohm 10watt wire wound resistor in those.

You may find this helpful
http://www.tedweber.com/webervst/tubes1/calcbias.htm

>>1050829
On a bigger amp, e.g. one with more powerful, fixed bias EL34, it probably will sound like dog shit. But don't forget that this is a guitar amp. After all, in modern days (after 196x) they all pretty much distort in ANY thinkable way: Preamp, phaseinvert, poweramp, powertransformer, speaker. It is more of a synth than a "real" amp. So maybe it will sound good. I'm just gonna try it.

I may add some kind of NFB in the future (well after it is ready...) just for the sake of then installing a presence control. I have not seen one on a EL84 amp yet.

>>1050831
Good idea, maybe in the future.

>>1050779
I checked the JTM45: The resistors are very close in value to mine, which are, as far as I have seen it, identical to later Marshalls like Plexis and JCMs. For example on V2a the JTM45 has a 820Ω and on V2b it's exactly 100K. In fact the whole tone stack and cathode follower on V2 seems to be THE SAME.

>>1050932
So I am going to order some values as I don't have 5 or 10W resistors lying around. Your claim actually surprises me: That's a big jump in voltages (depending on current of course)

File: 20160901_110148.jpg (1MB, 2048x1152px) Image search: [Google]

1MB, 2048x1152px

>>1051308
>So maybe it will sound good.
Read up on push-pull vs single ended output sections and you'll understand why it won't. The switch will keep half of the signal from reaching the output transformer. It's pretty much guaranteed to sound bad. You seem to be on the right track otherwise.

File: layout.png (63KB, 1417x496px) Image search: [Google]

63KB, 1417x496px

Just a quick update:
I am working on the layout (pic related, not complete yet). I tried using EDA software, but that stuff sucks if you want a quick layout without a step learning curve :/

>>1051345
I did my best and read a lot. I tend to disagree. It is not guarenteed to sound bad - per se. However, I discovered more interesting things, most I have not understood yet. For example on amps made by "Carr" there is an option for a variable output section that is completly different from other manufacturers and it really seems to really sound really well.

File: 1468920799177.jpg (70KB, 640x427px) Image search: [Google]

70KB, 640x427px

Sorry I can't give any advice, but I enjoyed reading through this thread and wanted to bump it off the bottom of the board. Cool project OP!

>>1052604
>It is not guarenteed to sound bad
It really is unless you drop your plate voltages and redesign it for push pull class A so both tubes amplify 360deg of your signal. The way you have it now will basically create as much crossover distortion as possible and you'll end up with 3rd and 5th order harmonics. It will sound like an overdriven cheap solid state amp, aka shit. Take a working push pull AB amp, pull one power tube and play through it. That's what you'll get. If you want to reduce output with that circuit, look into pentode/triode switching.

>>1050188
>about 45mA
times 2 (tubes). Look at the anode current on any data sheet.

I'm not clear on the reasoning behind solving for your PT demand. Are you using a new PT? If you are simply reconfiguring for cathode bias, then the original transformer of your fixed bias push-pull amp would be suitable. Bias is just a means of creating a negative potential from the cathode to the grid with the outcome of idle current control. If you're running the same power tube current from the original design then an additional preamp tube or two is unlikely to give you trouble.

V3b's grid needs to tie to C10 and R18. I see you got this covered.

R21,25 and R20,22 are doing the same job. Just go with 1M or 500k and anex 22 and 23.

I've unintentionally forgotten to couple the grid of one of my el84s in a similar design. This is pretty much your S1 switch. It didn't sound good, but it didn't sound solid state. My theory is the shared power tube cathode probably creates a make shift el84 phase inverter. There's no harm in seeing what your ears think.

I saw one pencil traced Carr schematic. It had cathode and fixed bias switching. Not worth the trouble imo.

OP here
>>1053409
Thanks for bumping! I am still waiting for the delivery of the goods. So no update from my.
I could not even update my layout as I was working every day. Maybe I will have some spare time tomorrow. ;)

>>1053617
>>1053808
Thank you for both of your replies and your help. However, I want to make something clear that I am convinced of: Tube amps don't always sound good and transistor amps don't sound bad. Not even the cheap ones. Therefore - only to me, other people may feel different about the topic - I don't think that any amp sounds "tube" or "solid state". There are many examples of great solid state tone in the guitar world: The Rockman (from Tom Scholz) and the Roland Jazz Chorus are the ones that come to my mind. Some of my personal (cheap transistor) amps sound very good running them through eminence speakers. My homebuild pedals (e.g. a pedal based on the circuit "Dr. Boogie", a Mesa Boogie simulator) sound insanly well if you connect them to a computer and run a cabinet/speaker impulse response on the output.

Therefore I simply don't believe in the "transistor" or "tube" sound. I DO believe in (simple) circuits and speakers.

>>1053617
Well, the thing is that I am NOT pulling a tube. I am just muting the input signal of one. That's a different case.

>>1053808
>There's no harm in seeing what your ears think.
That's why I am going to try it!


>>1053617
>>1053808
If it sounds like shit, what I am probably going to is: Switch the phaseinverter out off the circuit, rebias and run single ended as Class A. That way I am going to get more different sounds.

Tomorrow, I will post pictures of the other amp that I already posted a picture of.

>>1055101
You seem to be quite the expert for someone that needs help with such a simple circuit. Have fun getting electrocuted.

>>1055210
Thank you, will check your advice as soon as possible. Thank you for providing such a sophisticated reply. I am unworthy

>>1055284
There's some merit to anon's potshot. It wouldn't be unwise to get the above schematic in working shape before you dive into wild switching options.

Just a couple other items.
You have an extra preamp gain stage on a pretty bassman style design. It will be unstable with VR1 up if not built cleanly. I'd also consider anexing VR5 for R20 and 21 t become a dual gang PPIMV.