Tweed Style SE Build

[lkrasner]

Alright, I asked for ideas about this some time ago, but now I've got a schematic. What do y'all think? I'm going for a tweed champ style sound with a few extra features. I started with a 5F1, converted the PS to solid state (no point of a valve rectifier with SE), and added a tone control and MV. Then I made the NFB variable, and added a "fat" switch to change the cathode bypass resistor on the first preamp stage.

[2deaf]

The fat control won't work with R17 in series with C11.  Hook R17 to ground and the switch to the junction of R17 and C11.  The MV is inside the feedback loop, so it changes the feedback ratio as you adjust the MV.

[lkrasner]

The fat control won't work with R17 in series with C11.  Hook R17 to ground and the switch to the junction of R17 and C11.  The MV is inside the feedback loop, so it changes the feedback ratio as you adjust the MV.

Thanks, the Fat switch was a dumb mistake, I'd missed the MV. I've removed it now, not really that necessary IMO.
How's this look now?

[kagliostro]

Nice project

as to avoid confusion, add a connection dot when two lines on the schematic cross each other (if a junction is planned)

if you put random connection dots, like under node B+3 and not under node B+1 and B+2, confusion can came

See R8 C3 junction and R10 R11, someone can read the schematic in a wrong way

K

[VMS]

Looks good, but you need a grid-leak resistor to ground where you had your master volume.

[Fresh_Start]

The circuit is missing a grid return (or "grid leak") resistor for the 6V6. If you put it between the .02 coupling cap and the 1K grid stopper, you won't create a voltage divider. Not that 1k over 220K would knock the signal down much.

I'd also suggest different cathode bypass resistors for your Fat switch. 2.2uf and 22uf would be a better starting point IMHO.  Assuming this is a guitar amp, 22uf paralleled with 1.5K will boost the full frequncy range. 470nf paralleled with 1.5K wont let through anything but the highest frequencies. 680nf may be the smallest cathode bypass resistor I've seen, but that's in an amp with multiple gain stages and a lot of frequency sculpturing.

Cheers,

Chip

[lkrasner]

Nice project

as to avoid confusion, add a connection dot when two lines on the schematic cross each other (if a junction is planned)

if you put random connection dots, like under node B+3 and not under node B+1 and B+2, confusion can came

See R8 C3 junction and R10 R11, someone can read the schematic in a wrong way

K

Agreed, the program I used for this is kind of stupid about it, I will fix that.

Looks good, but you need a grid-leak resistor to ground where you had your master volume.

Yeah, that would probably be a good idea. Forgot to add it back.

I'd also suggest different cathode bypass resistors for your Fat switch. 2.2uf and 22uf would be a better starting point IMHO.  Assuming this is a guitar amp, 22uf paralleled with 1.5K will boost the full frequncy range. 470nf paralleled with 1.5K wont let through anything but the highest frequencies. 680nf may be the smallest cathode bypass resistor I've seen, but that's in an amp with multiple gain stages and a lot of frequency sculpturing.

I agree that 47 is probably too big, I will drop that down. For the lower one, I was essentially going for no bypass at all. The 5F1 has no bypass, the 5E1 does. This was kind of meant to be a switch between the 2.

Alright, here's a new schematic. Fixed the grid leak and lowered the bypass cap value. Also fixed the input resistors, those were messed up. Could I also bypass the second stage, or would that mess with the NFB? Someone suggested that I add a presence and/ or resonance control to the NFB. is this a good idea, or will it just not have much affect? If I were to add the MV back, how could I do it in a way that wouldn't mess with the NFB?

[Fresh_Start]

Note that I typed most of this before seeing your third revised schematic

OTOH a 22 or 25uf bypass cap on the 470 ohm power tube cathode will boost every frequency you can hear and then some.

On the input jack, you need a 1 meg grid return resistor, not 68K. The net effect of Fender's traditional input jack wiring is a 1m grid return and a 33K grid stopper(actually 34K but 33K is a standard value). But the grid stopper can go down to 10K according to taste and how close you are to a radio tower.

That 120 ohm resistor between the first two filter caps is suboptimal in terms of creating an effective pi filter. Something in the 1k to 5k range will create a much better filter, plus the voltage drop won't hurt a bit if you're going for a Tweed tone.

Bottom line:  you really need to start with a "known good" circuit. Build that, then tweak to taste. If you built the circuit as drawn, you'd drive yourself crazy trying to figure out why it didn't work or had very low output or only amplified the very top end of your signal. Print out a 5F2-A schematic. Get a highlighter and confirm every component and connection on BOTH schematics. Grid return resistors aren't optional.

Sorry if I seem like a buzzkill here.

Hope this helps,

Chip

[lkrasner]

Thanks for the input. I fixed most of those things in the last schematic.

How do you think this would take to using the 40x20x20x20 JJ can cap sold on Hoffman for the power supply instead of my current setup? seems like it would be fine to me and a good space saver.

[VMS]

I agree that 47 is probably too big, I will drop that down. For the lower one, I was essentially going for no bypass at all. The 5F1 has no bypass, the 5E1 does. This was kind of meant to be a switch between the 2.

You could use an on-off-on switch here so that in middle position you only have the 1k5 resistor in the cathode.

That 120 ohm resistor between the first two filter caps is suboptimal in terms of creating an effective pi filter. Something in the 1k to 5k range will create a much better filter, plus the voltage drop won't hurt a bit if you're going for a Tweed tone.

I think the 120 ohm is plenty because the resistor is before the plate of the power tube.


-v

[2deaf]

The circuit in Reply #6 will work fine.  The JJ can cap. will work fine.

The frequency response of V1a is the same with a 22uf bypass capacitor or with no bypass at all, but the gain is a lot different.  I don't see any value in changing the gain at this point in a low-gain amp unless that's what you mean by "Fat".  If you only want to vary the amount of lower frequencies, then don't use a center-off switch.

In actual practice, the difference between 2.2uf and 22uf as a cathode bypass capacitor is a little bit subtle.  The difference between 0.47uf and 22uf is blatant.  I've gone as low as 0.22uf, but that was on a Metal amp that needed to severely limit low frequencies.  Starting with 1.0uf, as you have in Reply #6, would be a good starting point.   

[jjasilli]

I agree that 47 is probably too big.
Probably, but you can try different values. YMMV

Could I also bypass the second stage, or would that mess with the NFB?
Yes you can put a bypass cap there, BUT ONLY IF you split the cathode resistor, etc.  Use the VibroChamp AA764 Schematic for that.  (Otherwise the NFB signal will simply bleed to ground through the bypass cap.) 


Another spot for more gain is to increase the 6V6 grid leak resistor to 470K (the 1M pot was too much.  If you want  a pot there, use 500K).


The elevated heater circuit is probably overkill (but can't hurt).

[Fresh_Start]

Yes, the circuit in post #6 will work. However, #2 would not and that's what I was looking at when I typed #7. 

EDIT2EDIT:  now I realize that I suggested the OP visit AX84 myself in an earlier thread.  Sometimes I just need to hit myself in the head. Apologies to Mr. Krasner

 

EDIT: I understand wanting to get as much input as possible, but posting the same questions and schematics on two forums simultaneously can be annoying (here and AX84).

I suggested a bigger resistor between the reservoir cap and the power tube plate node because it would improve the filtering all the way down the power rail. Necessary?  No. But I ended up with a choke there like the 5E2 and 5F2 in my Tweed Princeton derivative after a lot of experimentation.

Hope that helps,

Chip

[chocopower]

For a 3-6dB boost you can add a switch to the V1b cathode with a 22uf cap.


You can make it footswitch-able without any relay.

[jojokeo]

For more Tweed tone & increase of signal gain try un-grounding the Tone at some point so that only the vol is grounded. If you like it you could put it on switch.