Tweed Twin Design

[Sir Cuitous]

Can anyone tell me something about the first pre-amp stage of the Fender Twin 5D8? It looks like a local feedback loop is tied back to two of the input jacks. How exactly is that wired up?

[tubeswell]

Looks like part feedback loop, part parallel stage when the cable is plugged into the other input jack. AFAICT it would add an inverted signal to the initial input stage via the the other input stage, which would help 'clean up' the input signal (and maybe fatten up the tone?). I've never heard one of these but the concept seems interesting.

[Sir Cuitous]

Howdy tubeswell. Yes it does seem interesting. I got an amp in trade for some guitar work and I am thinking about converting it to the 5D8 twin.

I'm still a little hazy on exactly what is happening with that input network. It looks as if two of the inputs do indeed incorporate the FB loop, but the other two leave it out. Otherwise the rest of the amp is pretty straightforward. OK, my amp building has been on hiatus, but I think I'm going to have to start getting some parts together soon.

[tubeswell]

Hi Sir Cuitous

To me it looks like, when the cable is inserted into the jack that doesn't have the 'LFB' switch, it goes straight into the grid of that stage, gets amplified and then AC coupled to the 'LFB' loop which is connected to the opposite stage's grid by way of the switch on the opposite input socket. This amplified signal is then further amplified by the opposite stage (and is also inverted from the signal that the 1st stage puts out), and so would become the dominant signal I guess. Being inverted from the 1st stage's plate there is some kind of LFB type function going on, although it would seem to be a case of the tail wagging the dog.

On the other hand, if the input cable is inserted into the jack that has the "LFB" switch, this disconnects that "LFB". But there would still seem to be some kind of LFB going on even in that case because the signal swing on that stage's plate resistor would I guess induce a smaller (kind of 'backwobble') swing on the opposite stage's plate resistor, which maybe acts as a 'brake' on the signal of the former stage?? However, this would seem to be mitigated a wee bit by the 100k mixing resistors. (But I could be wrong about this bit).

Of course jumpering the channels would produce all kinds of crazy results I guess

[loudthud]

To me it looks like the intent of the feedback is to reduce noise from unused inputs. With no cables connected, the top tube's output is connected back to the input making a virtual ground at the output. This also kills the noise from the other half of the 12AY7 that gets through the mixing resistors. If you plug into the top input of the pair, you disconnect the feedback and the virtual ground is removed from the output. If you use additionally the next input down, the signal from that input gets added as it normally would via the mixing resistors. If you use only the lower input, not much signal will get through because the upper triode will be fighting the signal trying to cancel it at the virtual ground.

It's interesting that this amp does not have the 4.7 meg or 10 meg feedback resistor around the next stage like the later amps did. Also it has a text book James tone control network except for one extra resistor.

[Robert M. Martinelli]

Hi all,
I'm with Loudthud on this one, I think that connections are there to simply reduce the noise coming from the unused inputs.

JM2CW

Best regards

Bob