SSS KT88 schematic and info?

[martin manning]

The power amps in ODS's and SSS's are more Hi-Fi-like than most guitar amps, so using Hi-Fi tubes shouldn't be a bad thing, as long as they are set up properly. Redplate had a KT88 amp at the Nashville show that killed.

[talbany]

The power amps in ODS's and SSS's are more Hi-Fi-like than most guitar amps, so using Hi-Fi tubes shouldn't be a bad thing

I agree!!... if you like to play mostly clean and like the big tone the HI FI tone is pretty cool.. The problem is if you have to use some distortion (which we all need sometime) Where do you get it? ..If you go to a pedal then the distortion is very transistorized/sterile and you can really hear it because the output section is so clean/articulate.. Some kind of preamp (tube distortion) is a little better but can get sometimes a bit buzzy on the top!.. Although I am sure Henry makes a great amp but I can kind of hear it in this clip with Lance Demoing the amp when he really starts to push the gain..All pre amp distortion.
http://www.youtube.com/watch?v=PTjl3XR5Z8Y

IMO There is no way around this and one of the rather BIG drawbacks I have with my SSS (and other UL amps)..Getting a good overdrive tone! (and mine is not even UL)..
I am not saying it's a bad thing and some might not notice and think it's really cool with a pedal or tube pre..But it is something we need to think about when we build these very clean amps (especially in UL)..This IMO is why these more Hifi amps never really made it in the guitar department too much preamp distortion and not enough output saturation to smooth/round compress things out on the way to the speaker..IMO

Tony

[Luthierwnc]

Let's circle back to the idea of a separate screen supply. With the transformer I mentioned (not married to it -- just handy) I've got two HV taps at 0-400v 500ma each. In the usual configuration, one of them will put out around 530 B+. With a swinging choke followed by an RC filter into a 1k dropping resistor (12k) load I get 380 volts on that unit. There are probably easier/better ways to dumb-down the second winding but is that do-able?

[martin manning]

The screen supply could be done that way, but your driver negative voltage supply won't work with a FWB. There's no negative voltage to speak of available.

[Max]

The power amp of this amp http://personalpages.manchester.ac.uk/s ... 20%230003/ might be another option.

Cheers,

Max

[Luthierwnc]

I was working off my Gemini II schematic. In this case, I replaced the last resistor with a resistor/pot but I don't remember the voltage after the diode. I'm open to ideas, sh

[talbany]

The power amp of this amp http://personalpages.manchester.ac.uk/s ... 20%230003/ might be another option.

Cheers,

Max

Not really... Ultralinear Taps

Tony

[martin manning]

I was working off my Gemini II schematic. In this case, I replaced the last resistor with a resistor/pot but I don't remember the voltage after the diode. I'm open to ideas, sh

Notice the capacitor coupling the bias supply into the HT and its ground reference in the Gemini II schematic... You need that to make some negative voltage available, but I don't think you can get the current you'll need for the CF driver using that technique.

Looks like a par of 42W Pa max KT88's would work at ~425V plate and screen with a load of around 2k. Just for grins I can try that using my ODS when I receive the ones I ordered.

[Luthierwnc]

Anytime I look at KT88 schematics, this comes up. I don't fully understand how the two FWBRs work together but this does get the volts right on plates and screens between the two. If I get this right, the higher voltage winding runs positive DC into the back of the lower voltage winding and all nodes downstream from the plates. The bias would have to be something else but this seems to get the rest. sh

[martin manning]

The Hiwatt is just two FWB supplies stacked one on top of the other. With a pair of identical secondaries you'd be looking for VAC's of ~220 on both to hit ~600 plate and ~300 screen. That's all good but the question then would be: is 300v at the screen node enough to get the required swing out of the PI. That may be the reason that the Hiwatt power supply secondaries are not matched.

[tictac]

Another thing is the negative supply needs to be able to deliver more current than your standard run-of-the-mill neg bias supply.

I don't think cap coupling would be advisable; also that 4K7 resistor should be rated about 10W to be safe.

TT

[Luthierwnc]

Since the D method remains misty, I'll keep groping. The SVT power supply comes up again. Getting one wound in a smaller version would probably be pricey but maybe the best option. The UL choice seems easiest but I'd want to leave a few extra lugs on the board for plans B and C.

I've thrown a couple of negative voltage supplies in the attachment. One I sketched for a two-wire supply. The one below I filched from the net. With a center-tapped winding, that seems to handle both the driver and bias needs rather neatly. Food for thought.

Thanks again to all for your comments. If I find the definitive 6550/KT88 supply schematic surfaces, I'll post it. If anyone else does and it doesn't come at the expense of a production model, there is plenty of room to turn this thread into a file.

In the meantime, good soldering on your projects, Skip

[mhartman]

So, it looks like most of the KT88/6650 designs out there are UL. What happens if you just disconnect the UL taps? What factors need to be considered other than just adding another node for the screen grids?

[talbany]

So, it looks like most of the KT88/6650 designs out there are UL. What happens if you just disconnect the UL taps? What factors need to be considered other than just adding another node for the screen grids?

Because the screens should be supplied from a voltage source of good regulation
You can read about about this approach in the RDH4 on page 562 (Sec 13.3.iii.E). It shows a pretty good picture of how to graph it, as well as how to calculate screen current..

http://www.tubebooks.org/Books/RDH4.pdf

Tony

[martin manning]

So, it looks like most of the KT88/6650 designs out there are UL. What happens if you just disconnect the UL taps? What factors need to be considered other than just adding another node for the screen grids?

The problem is that if you run the screens at close to the plate voltage, the load line will be in the wrong place, crossing the Vg1=0 grid line far below the knee, leading to excessive screen current at high power. If you lower the primary impedance to reposition the load line, then plate dissipation will become a problem. Lowering the screen voltage is the right approach, but it will need to be reduced to nearly half of the plate voltage. This you could do with a branched screen node, to avoid lowering the PI and preamp supply voltages.