6G3 build, switchable 6V6 to 6L6

[centervolume]

Just wrapping up a brown deluxe build. Used a oversized OT to allow for 6L6 operation. http://www.triodeelectronics.com/tf65wscta48opaper.html

To make the bias switchable, I ran the bias board ground over to a spdt switch I put in the "ground" switch location on the back panel.

I set up the new spdt switch to take the bias board ground line, then switch it out to either a) ground or b) a 27K resistor that goes to ground.

The thinking is that when operating 6L6s, the 2nd 27K resistor is added to give 54K.

The amp sounds great with 6V6 but 6L6s create a weird hum/signal distortion that increases/decreases with the "intensity" control.

Should I just try another set of 6L6s or am I completely whiffing on something a more seasoned builder would know?

thanks in advance

[tubeswell]

Given that the intensity control affects the signal distortion you're getting with the 6L6s, it's most likely the way the bias-vary trem affects the bias of the 6L6s. You haven't shared any details about your schematic or anything about the operating voltages. Can I assume that you're running the 6L6s at low voltages? And if that's the case, is the distortion you're getting like a slightly out-of-phase warble?

[ToneMerc]

Given that the intensity control affects the signal distortion you're getting with the 6L6s, it's most likely the way the bias-vary trem affects the bias of the 6L6s.

I concur

TM

[martin manning]

What happens to the positive lead of the bias supply filter cap when you switch in the 27k? It needs to stay connected to ground, not lifted to the junction of the two resistors.

[centervolume]

the 6L6s run about 430 on the plates, the rest of the voltages are within 10% of published 6G3 schematic.

when I flip the switch to add the 2nd 27K resistor, the switch requires the ground coming off that smaller bias board by the pilot light assembly to be deferred until it goes through the 2nd resistor.

I am not sure how to better explain it, all I did was to run a longer insulated conductor from the same position that the bias board uses for its ground, along the side of the chassis (next to PT) back to the single pull of the spdt switch. That switch lets me throw ground directly to ground (6v6) or require it to pass through the 2nd resistor on its way to ground (6L6 setting)

[sluckey]

You can't do it like that. You need to separate the resistor from the positive side of the cap.

***THE CAP POSITIVE MUST CONNECT DIRECTLY TO GROUND AT ALL TIMES.***

When switching, switch only the resistor(s)

[martin manning]

The capacitor on the bias board has to stay connected to ground with the switch in either position. It might be easier to put a 56k resistor on the board and use the switch to place another 56k in parallel, which will make 28k.

[centervolume]

ok I see the problem now, thanks for your patience on that.

as for solutions, yes changing values to 56K and allowing for the switch to divide the values is good.

I'll try that, in the mean time, any other solutions welcome

thanks much you all, will report back with results

[martin manning]

Another way, which might be a little more flexible:

[centervolume]

ok - cool, let me check consider these options

and THANKS for helping me realize the tonal possibilities of this cool circuit

[centervolume]

[centervolume]

ok - got back to this amp last night and got the bias switch properly wired, now just need to fine tune some values based on the tube selection, so that is under control. Really happy with the results so far, she is quiet with good power


I am trying out speakers and voicing with tubes now. My OT has multiple 4, 8 and 16 ohm secondaries. I'm interested in comparing some 4 Ohm 12s to the 8 Ohm ones.

Looking for a way to hook up the external speaker jack to the 16 ohm OT tap, while the 8 Ohm tap is hooked up to the standard speaker jack.

I tried it but both channels were much quieter. ran common ground tying both jacks together, separate neg feedback conductors to tip of both. The external spkr jack currently installed however is not self shorting, unlike the standard speaker jack.

[martin manning]

Does your main speaker jack have two solder lugs or three? If it's two you need a new jack. if it's three, disconnect the switching contact that grounds the tip when nothing is plugged in.

Connect one OT lead to the tip terminal on each jack, and the NFB lead to whichever jack is the standard output impedance. Connect the OT ground lead to the sleeve terminal on both jacks and ground one of them to the chassis.

This leaves the output in an open circuit condition when no speaker load is connected, and you can damage the OT if you accidentally try to play through it in that condition. A safer option is to get two switching jacks (Switchcraft type 11A), and connect the switching contact terminals together. This will short a portion of the secondary and protect the OT from a no-load situation if no speaker is plugged in, and open the short when either jack is in use.

[centervolume]

ok - I did all of this except for uncoupling the switching contact from the shield on the main (8ohm) jack. Wonder if this is why I got low power output from the other (4 Ohm) jack? I am also going under the assumption that the sleeve terminal is a ground connection as the jack assembly is mounted to the chassis, please let me know if separate hard solder to the chassis is needed.

the main speaker jack is a self-shorting jack when nothing is plugged in (3 terminals). It remains tied to it's sleeve/ground terminal, but will disconnect those 2 terminals to use the 2nd speaker jack as a 4 Ohm output.

this 2nd jack (labeled ext speaker, intended for 4 Ohm out) is currently not self-shorting, but I have an extra self shorting one for that application, as I am aware of danger to OT if it has nowhere to dump its load.

Regarding the NFB line - once I do the comparison, and decide on the speaker I want, I should dedicate the line to whichever I decide is the primary/main spkr output right? and for the purposes of doing the A-B comparison, I should move it from jack to jack since it is working to provide a cleaner signal iirc. Am I on track? or is it ok to simply run a duplicate line from the NFB resistor on the board to this other jack (tip).

[martin manning]

ok - I did all of this except for uncoupling the switching contact from the shield on the main (8ohm) jack. Wonder if this is why I got low power output from the other (4 Ohm) jack?

Yes. it's shorting out part of the secondary unless something is plugged into that jack. Stick a plug with no cable, or a pare speaker cable in there and see what happens.

I am also going under the assumption that the sleeve terminal is a ground connection as the jack assembly is mounted to the chassis, please let me know if separate hard solder to the chassis is needed.

It's better to have a solid ground to a solder lug, but not necessary. Lots of amps don't.

the main speaker jack is a self-shorting jack when nothing is plugged in (3 terminals). It remains tied to it's sleeve/ground terminal, but will disconnect those 2 terminals to use the 2nd speaker jack as a 4 Ohm output.

this 2nd jack (labeled ext speaker, intended for 4 Ohm out) is currently not self-shorting, but I have an extra self shorting one for that application, as I am aware of danger to OT if it has nowhere to dump its load.

Wire it like the drawing below. Most people leave them open-circuit, but this is better, IMO. The optional resistor could be 2x the value of the lower of the two impedances, 8R, 5W.

Regarding the NFB line - once I do the comparison, and decide on the speaker I want, I should dedicate the line to whichever I decide is the primary/main spkr output right? and for the purposes of doing the A-B comparison, I should move it from jack to jack since it is working to provide a cleaner signal iirc. Am I on track? or is it ok to simply run a duplicate line from the NFB resistor on the board to this other jack (tip).

I'd start with the NFB on the tap that it normally goes on, and experiment with moving it to the other tap. Treat that as a separate decision for each speaker.