Bass Amp for a friend

[martin manning]

Pete, a DC-coupled CF was the idea entertained above in reference to the SSS CF drivers.

I've been spec'ing transformers with Phil from Heyboer. OT will be a Dynaclone A451 at 2k2. Power supply will be SS full wave rectified into a CLC filter. Then B+ to OT. Choke will be rated for 500mA.

I will be running a quad of Winged C 6550s. Max dissipation is 35W. The datasheet specifies the U/L application at either a plate voltage of 560V for 100W or 450V at 70W. Any thoughts on the operating point? (Ian, Tony, Leo, et al)?

For the PT, I am specifying either 400VAC or 410VAC at 450-500mA secondary, a 55VAC bias tap, and two 6.3VAC taps; one for heaters (9A) and the other for accessory stuff (3-4A).

For my bar napkin math using a 400VAC secondary...

With a 400VAC secondary, 400 * 1.3 = 520VDC rectified under load
520VDC across a CLC = ~ 10VDC drop
510VDC B+ to OT centertap
~ 500-500VDC plates
~ 224VDC screens @ 43% screen taps
~ -60VDC bias

Dave, you are specifying the PT voltage and current, so that implies loaded condition. Therefore, I think the 1.3 factor in your voltage calc should be 1.41 for silicon rectifiers.
Your "224VDC screens @ 43% screen taps" does not make sense to me; In UL the quiescent screen voltage is essentially the same as the plate, and the screens are supplied through the OT primary, so you don't need a choke.
You will have to derive a reduced voltage for the preamp, though, and there are a couple of options there. The simplest is a big dropping resistor RC, but you could also use a CLC.

Refer to the snip below from GE 6550A data sheet for two tubes in UL. I would target the operating point suggested there. You can basically double the values for current, except reduce the zero signal current by ~20% due to the 35W vs. 42W Pa max. Note the max signal current (Ia + Ig2) is about 300mA, so 600mA to you. I think your PT should be rated for 325-0-325 @ 600mA (around 350V unloaded) to get ~450VDC under full load. The output power will be around 140W, 2x what's shown there.

If you want to go with the SSS-type drivers, you could have your PT wound with two extra taps at 120-0-120 for the CF driver positive and negative voltages, like a dual voltage secondary.

I haven't come across the driver circuit shown above. Where did that come from, and what are the values of the unmarked resistors?

[Colossal]

Dave, you are specifying the PT voltage and current, so that implies loaded condition. Therefore, I think the 1.3 factor in your voltage calc should be 1.41 for silicon rectifiers.
Your "224VDC screens @ 43% screen taps" does not make sense to me; In UL the quiescent screen voltage is essentially the same as the plate, and the screens are supplied through the OT primary, so you don't need a choke.
You will have to derive a reduced voltage for the preamp, though, and there are a couple of options there. The simplest is a big dropping resistor RC, but you could also use a CLC.

Refer to the snip below from GE 6550A data sheet for two tubes in UL. I would target the operating point suggested there. You can basically double the values for current, except reduce the zero signal current by ~20% due to the 35W vs. 42W Pa max. Note the max signal current (Ia + Ig2) is about 300mA, so 600mA to you. I think your PT should be rated for 325-0-325 @ 600mA (around 350V unloaded) to get ~450VDC under full load. The output power will be around 140W, 2x what's shown there.

Hi Martin, many thanks, as usual for vetting things. I was implying the loaded conditions using the 1.3 factor. I made a bit of a mess with that. I was referencing the GEC KT88 the 70W x 2 U/L application.

Here is the proposed power supply. Values are just placeholders right now. I've incorporated your proposed changes to the PT. It was probably confusing but I meant to say that there would be a CLC off the rectifier. B+1 OT would be fed from that.

EDIT: schematic removed to update to latest specs

I haven't come across the driver circuit shown above. Where did that come from, and what are the values of the unmarked resistors?

I came up with that as a half-assed short cut to using a dedicated CF driver I was thinking of using a 12BH7 for that arrangement. The unmarked values might be something like 20k plate, 750R bias, 470k grid leak, 47-100k grid stopper, and 10k for the CF cathode (half the inverting side's plate load value). I need to run the load line and all that for the cathodyne. Was thinking a B+ of 350V for the cathodyne and follower. DC offset on the heater filaments so as not to exceed h-k.

If you want to go with the SSS-type drivers, you could have your PT wound with two extra taps at 120-0-120 for the CF driver positive and negative voltages, like a dual voltage secondary.

Well, what do you think of going that route? I'm open to suggestions. The goal is a tight, punchy bass amp. I left out the components for the CF driver supply for now.

[martin manning]

I think I would do a CF driver of some kind. You could do a CLC after the rectifier; Sunn did, as did Dynaco on a 4x KT88 amp. It will add some weight, though. If you have the 120-0-120 for the CF drivers you don't need the bias tap. What's the 6.3V @ 4A rectifier winding for?

[Colossal]

I think I would do a CF driver of some kind. You could do a CLC after the rectifier; Sunn did, as did Dynaco on a 4x KT88 amp. It will add some weight, though.

It's ok, the CLC up front will keep the amp quiet.

If you have the 120-0-120 for the CF drivers you don't need the bias tap. What's the 6.3V @ 4A rectifier winding for?

Yeah, I was just implying either fixed bias, or the separate taps for the driver at this point. I'll ditch the bias tap then and go with the drivers. The extra 6.3V tap was on a transformer Phil (Heyboer) was suggesting and I'd just run a pilot light LED off of it. That can be done off the primary however. But this is pretty much a custom PT at this point.

[martin manning]

BTW, the CF drivers only need about 10 mA, so the 120-0-120 winding can be sized accordingly.

[Colossal]

BTW, the CF drivers only need about 10 mA, so the 120-0-120 winding can be sized accordingly.

Excellent, thanks for the help Martin, much appreciated. I was thinking 50mA for the winding. I am working on the schematic and power supply drawings now. Are you thinking 12AU7 or 12BH7 for the driver? Adding the driver will change the tube count a little so that adds a dual triode of some kind for the inverter. How about a paralleled 12AU7 cathodyne?

[martin manning]

50 mA would be fine for the 120-0-120. I would do a FWB on this winding, with the CT connected to the center of stacked reservoir caps in typical bipolar supply fashion. 2x 100u at 250V would be more than enough. I think I'd go with a 12AU7 for the drivers... they are easy to find.

[roberto]

I suggest you to read this page too:
https://www.diyaudio.com/forums/tubes-v ... stage.html

Can I ask something to understand some choices?
- has the PI a CCS on the cathode or is simple resistor loaded?
- why CF instead of SF, that will have a lower impedance to drive the power tubes?
- why a 120-0-120V power supply for the CF, instead of using only one side of it (0-120), rectify, filter and use as a negative rail for PI and CF (or IMHO better SF)?
- has been considered any separate winding for screens, to increase plate voltages and so output power?
- has been considered any CFB or Schade?

Thanks
Roberto

[Colossal]

Thanks for the link Roberto. Good reading.

I will let Martin answer your questions! As for my part, I have just been looking to keep the design fairly simple with the goal of a clean and clear, punchy bass sound with fast response and a 3D sound. I wanted to shape the design with input from everyone here as an exercise in a little self indulgence. My friend is very excited about this project. I had to let the secret out early because his wife was talking about him getting an amp and I did not want him getting a pile of garbage.

Best,
Dave

[didit]

Hello Dave -

CF buffering into power tubes seems a sound idea. So far no one suggested LTP PI. Suggest that is also sound, coupled with CF -- better performance all round versus pushing a cathodyne; offering some gain, firm dynamics and ready potential to balance.

Also note that in UL, a pair of KT88 has relatively high primary impedance requirements. The A451 will work. However you'll likely find it works best modelled as 4.4K so each secondary tap winds up at 2x the loading spec - eg. 4 Ohm tap is best with roughly an 8 Ohm load. This may recommend having it wound with spec'ed 2-4-8 versus 4-8-16.

Best .. Ian

[dorrisant]

Roberto, the CF driver requires a plus and minus supply and it can't exceed data sheet rating for 12**7 implemented. Measured across the tube... positve side for the plate, negative side for the cathode. Thus the 120-0-120 winding.

[martin manning]

I think a LTP with CF drivers would be fine, and very SSS-like. SSS had extreme +/- voltages with large cathode resistors. I proposed the bipolar supply (running off the common bias tap) to get away from the potential hazards (heh) of doing it that way. It's been proven out by several TAG members.

If I understand correctly, Colossal wants to keep it all-tube, but FETs are an option.

[roberto]

Roberto, the CF driver requires a plus and minus supply and it can't exceed data sheet rating for 12**7 implemented. Measured across the tube... positve side for the plate, negative side for the cathode. Thus the 120-0-120 winding.

Yes, that's my point. CF are used to easily drive the power tubes with low impedance, so why don't we use mosfets as SF? Higher voltages allowed (so less THD), lower impedances (so better drivers), cheaper, more energy efficient and will never need to be replaced.

[roberto]

If I understand correctly, Colossal wants to keep it all-tube

Clear, thanks!

[dorrisant]

Roberto, the CF driver requires a plus and minus supply and it can't exceed data sheet rating for 12**7 implemented. Measured across the tube... positve side for the plate, negative side for the cathode. Thus the 120-0-120 winding.

Yes, that's my point. CF are used to easily drive the power tubes with low impedance, so why don't we use mosfets as SF? Higher voltages allowed (so less THD), lower impedances (so better drivers), cheaper, more energy efficient and will never need to be replaced.

I am allergic to solid state devices... Lol!