After several push-pull pentode amps that rely on lots of feedback, I decided it was time to go to the other extreme: how low could I get THD with no feedback whatsoever?
The result is this push-pull 300B amp. Details on my web site: "Fatboy" 300B Push-Pull Amp
It's not too complex, and it gets to 0.08% THD at 1W and 0.2% at 10W. Very much old school, choke loaded input tube direct-coupled to DHT driver, IT coupled to a pair of 300Bs. Point-to-point wiring with the exception of a couple of power supply PCBs inside.
And yes, it does sound pretty good, too!
Pete
Thanks!
Those are some digital "panel" meters I designed. They're built inside a standard octal tube envelope.
There is info on them here: SiTubes
Pete
Clever. Another way to do this is a CCS instead of Rk (an LM334 works well) bypassed with a cap and a resistor load. Vp is constant in that case.
That looks great. Hats off to you 🙂 !
Must look very nice in the dark when it's all lit up and sound fantastic too.
Must look very nice in the dark when it's all lit up and sound fantastic too.
Excellent design and a beautiful amp inside and out! I see you went with the ultrapath connection on the driver. Was that a design decision made from the outset, or did it come about by trying it against the more common config of the cathode bypass to ground? If so, could you briefly share your thoughts on this?
Beautiful amp!!! And Tango iron🙂
I really like the isolated DC for the driver tube heaters, the little boards directly at the sockets are neat.
I would like to know how much worse your measurements were using the triode strapped 6SJ7.........I have some I would like to use but if the results were terrible I don't want to bother.
I really like the isolated DC for the driver tube heaters, the little boards directly at the sockets are neat.
I would like to know how much worse your measurements were using the triode strapped 6SJ7.........I have some I would like to use but if the results were terrible I don't want to bother.
There was not a big difference, maybe from 0.08% to 0.10% overall THD. In general 6SJ7 and its relatives are quite good both as pentodes and as triodes...
Pete
That's very good performance for a no feedback transformer coupled design. Are you keeping it as it is or will you be making any modification?
Good to know
Thanks. I see on some old Fender amps they used to use them but stopped, they seemed to have replaced them with the 12AX7 and never looked back. I was intending to use them as pentodes for a guitar amp and now seeing some plate characteristic curves for triode mode, and your reassuring field data I might try them in some audio playback applications.
Looking at the machine work on the Landfall chassis I really like the work done around the 300b's. Did you do that yourself?
I am also curious to why you chose to go with the hybrid bridge
You had to punch the holes for the SiTubes PWR socket, why not just use 4 6CL3 in a bridge? I am guessing the 6.3v winding wouldn't tolerate the 4.8A of heater current😀
Last edited:
Zout is not low... looks like about 3 ohms on the 8 ohm tap, 1.5 ohms on the 4 ohm tap.
Pete
I just did an amp with no feedback around the output transformer and I learned that it turns out to be pretty difficult to get the Zout really low without a feedback loop around the transformer.
I just built a Unity-Coupled amp with a Plitron transformer and was able to get it down to ~1.1 Ohms but I think that represents something close to the practical lower bound for what is possible for an amp of this size without feedback around the transformer. The Plitron transformer's winding resistances were lower than I had ever seen and the driving impedance on the transformer was really low due to the heavy cathode feedback.
I would say that a 3 Ohm Zout on the 8 Ohm tap and no local feedback on the 300Bs is quite good. That's just the nature of zero feedback amps. You just wouldn't want to pair them with speakers that demand a lower Zout than that for good performance.
I designed it in AutoCAD, then Landfall did the machining for me.
I am also curious to why you chose to go with the hybrid bridge
Partly because of the heater power, partly because I didn't want any more diode drop than I had to have. The 6CL3's are there as much for slow turn-on as anything. I added the B+ delay relay just in case the amp gets switched while the 6CL3s are still hot.
Pete
- Status
- Not open for further replies.