New build 5F1 & AA764 champ, 2 in 1

[donluca]

As expected, I messed it up.

Looking again at the schematic, even when the switch S1-A is lifted, the signal still goes into the tone stack, which means I'll have to lift the tone stack's output as well. I'll need a 3PDT to do that as I also want to switch the NFB resistor.

This guy here got it right:



Although I'm not sure why he put a 5MΩ resistor across the V1A output and the volume pot's input. Is it to avoid a bump when switching from one circuit to the other?

I'm still debating whether to keep the grid resistor on the 6V6, lower it to 1.5K like in more modern Champ designs (the Champion 600 has it) or remove it altogether.

[donluca]

Alright, done some work today and added a switch (waiting for my 3PDT switch to arrive from China) to completely bypass the tone stack and I've figured out the source of hum and hiss which are due to poor grounding choices I've made when rewiring the thing.

When all the components will be here I'll open it up again and make all the final mods and that will be (hopefully) it.

I'm still wondering about that 5MΩ resistor the guy put there, I guess for safety reason? Should the switch go bad and DC would destroy the next 12AX7 stage due to a coupling cap missing to block the DC?

Also pondering on leaving the 5.6K grid stopper on the 6V6 there or removing it. Or maybe put a 1.5K like in the Champion 600.

Anyway, those things will have to wait.

Meanwhile, the amp sounds great both in 5F1 and AA764 mode so far. When in 5F1 mode the P90s really push the amp hard, IMHO they sound way better in AA764 mode, more smooth and less gritty.
With single coils, I really prefer the 5F1 mode, more dynamic and somehow more sensitive to my touch.

Anyway, that'll be it for now, I'll update the thread when all the components are here and the mods in place.

[g1]

I don't see the 5meg on any schematic. If it's there, please enlarge the area and re-post.
If it's on the switch wiring but not the schematic, please draw it in.

[The Dude]

The resistor might be there to abate some switching noise. Just a guess without seeing where it is.

[g1]

That's what I think too, and don even mentioned it. But it may make more sense to anyone looking if it can be seen on the schematic.
There was a thread recently where Chuck H pointed out a similar thing where (without a resistor) the switch would leave a grid un-terminated while the switching was in process. I think this is probably similar.

[donluca]

.

Lower right corner, there's the switch wiring with the 5M resistor wired across the middle terminals.



Chuck's explanation would be fitting, although I'm not sure what harm could come to leaving one tube's grid unterminated, especially for such a low time frame.
The second grid would still be connected to the volume pot, so you just have to be careful to don't make the switch while the volume is maxed out to prevent shorting the grid to ground.

[Helmholtz]

I take "INPUT V1A" to mean output (plate) of V1A.

Anyway wired as shown the 5M is no good idea. It provides an unwanted DC path between the plate potential of V1A and the grid of V1B, actually applying up to 33VDC to the V1B grid depending on the vol pot setting. So V1B plate current will strongly shift with vol setting. Also the DC on the pot is likely to make it scratchy.

The tubes are already well terminated without the 5M resistor.

I don't expect the resistor to have a positive regarding switching noise.

[donluca]

Following the guy's schematic:

Input V1A = V1A Plate output (pin 1 of 12AX7)
Output to Volume Pot = Volume pot external lug

And yeah, I'm with you on this one, the 5M resistor looks like a really bad idea. Despite the high resistance it would still let some VDC pass which is a big no-no in my (very small) book and it doesn't make any sense even going with something higher.

I'm 99,9% sure it was put there to avoid a "bump" on the speaker when switching the tone stack in or out when the volume is wide open (which I just tried and *IT DOES* make a bump with the volume at max, but it's perfectly quiet with the volume even half way down).
It's a non-issue for me since I'll never switch from one to the other while playing.

On a side note, today I went and took some measurement and found out that the 12AX7 is getting a 225VDC on its plates and 1.5V on the first half cathode and 1.6 on the second half.

Looking at the AA764 schematic, it says that the plates should be getting 200VDC. The 5E1 schematic (which is nearly identical to the 5F1) instead reports either 150V or 160V (can't read well, the scan is pretty bad: https://imgur.com/7rebkhR ).

Should I drop the plate voltages a bit? It would be quite easy, I just need to increase R5 (refer to the VHT schematic). This should give me a little less clean headroom which I'm ok with so I don't have to crank it all the way to get some distortion.
Besides, as per datasheet, 12AX7 can have up to 250VDC on their plate which means I'm still within spec.

Speaking of cathode voltages: AA764 has 1.8V on first cathode and 1.7V on second one, while 5E1 (again, the 5F1 doesn't report voltages) has 1.5 for both. Those looks close enough to what I have and I'll probably leave them alone.

Unfortunately I don't have an oscilloscope, which means that the best I can do is just making sure that all the voltages are ok and there's no VDC on the tube's grids.

[g1]

Like I said, it would be nice to see it drawn into the schematic. Schematically.
(no offence intended, those who've been here awhile know I'm a real nitpicker about actual schematic vs layout or wiring diagrams )

[g1]

Just so no one thinks I'm being difficult.

[Chuck H]

Yep. Like Helmholtz said, that 5M resistor is going to kludge the bias on the V1B grid relative to the volume setting. That was no idea of mine I can assure you. That switch should be prefaced by a 100n cap from the V1A connection to decouple it from DC. Or... You could just move the "bass capacitor" to the plate, run the tone stack AND switch connection from that and omit the bass cap from the tone stack arrangement. That way you don't have to have an additional capacitor in the circuit and the switch is decoupled from DC. Boom!

[g1]

That was no idea of mine I can assure you.

No, now that I can see it, it's not at all like the one of yours I mentioned. In case you missed it though, maybe a coincidence with the name on this schematic?

[Chuck H]

No, now that I can see it, it's not at all like the one of yours I mentioned. In case you missed it though, maybe a coincidence with the name on this schematic?

If that's our Bob it's a rare miss for him. Maybe an early effort where it got past him? We haven't seen him here since March of last year here on the forum. I've heard from him once in that time and he said all was well. I surely hope so.

[donluca]

Like I said, it would be nice to see it drawn into the schematic. Schematically.
(no offence intended, those who've been here awhile know I'm a real nitpicker about actual schematic vs layout or wiring diagrams )

My apologies, I misunderstood your previous post and thought you weren't able to locate it at all, so I pointed it out more specifically where it was instead of drawing it in the schematic as a whole.
No offense taken, actually, my bad on not asking more specifically what you meant before :P

Or... You could just move the "bass capacitor" to the plate, run the tone stack AND switch connection from that and omit the bass cap from the tone stack arrangement. That way you don't have to have an additional capacitor in the circuit and the switch is decoupled from DC. Boom!

That's a very nice and elegant solution! I'm trying to picture how would I wire it to keep leads as short as possible to avoid interferences.
Another thing that kind of worries me is having a capacitor twice the intended size as decoupling cap, maybe I should put a resistor in series after it? I think a 1.5K-3K should do the trick to avoid the additional surge from the bigger capacitor discharging (and, as a bonus, it would also work as a nice grid stopper for the next stage).

EDIT: just to make 100% sure I've correctly understood your plan... You're suggesting of "repurposing" the "bass cap" (which is 0.047uF) to decouple the V1A plate from the grid of the next stage by cutting it from the tone stack and inserting it where "my" additional 0.022uF cap is through a switch.

To put it in simpler words, the switch would cut the 0.0047uF cap from the tone stack and put it in series between V1A plate output and V1B grid. Correct?

[Chuck H]

My apologies, I misunderstood your previous post and thought you weren't able to locate it at all, so I pointed it out more specifically where it was instead of drawing it in the schematic as a whole.
No offense taken, actually, my bad on not asking more specifically what you meant before :P



That's a very nice and elegant solution! I'm trying to picture how would I wire it to keep leads as short as possible to avoid interferences.
Another thing that kind of worries me is having a capacitor twice the intended size as decoupling cap, maybe I should put a resistor in series after it? I think a 1.5K-3K should do the trick to avoid the additional surge from the bigger capacitor discharging (and, as a bonus, it would also work as a nice grid stopper for the next stage).

EDIT: just to make 100% sure I've correctly understood your plan... You're suggesting of "repurposing" the "bass cap" (which is 0.047uF) to decouple the V1A plate from the grid of the next stage by cutting it from the tone stack and inserting it where "my" additional 0.022uF cap is through a switch.

To put it in simpler words, the switch would cut the 0.0047uF cap from the tone stack and put it in series between V1A plate output and V1B grid. Correct?

I made some changes to the Champ mod diagram. It should explain it. Note that I removed the .02 cap so there's actually one less cap in the circuit than before (I just love an economy of parts). But this means the coupling cap in the switched position would be your .047 cap (.100 in the diagram). The difference is NOT significant in this circuit. About .02dB increase at 82Hz. The tone stack will operate a tad differently, but actually for the better IME.