Some KT88's connect the metal shell to pin 1. Grounding it would prevent the possibility of a shock from accumulated charge or an accidental short. I usually connect pin 1 to pin 8 so I can use EL34's in the same amp.
I made several high powered amplifiers out of some old Stromberg Carlson PA amps back in high school. Those amps were intended to use 6L6GB's but we used the old metal 6L6's because we had several hundred new RCA's to destroy in electronics class. I learned two important things in a brief instant one day.
1) The old Stromberg Carlson used pin 1 as a B+ tie point to mount the screen resistor on the tube socket. Pin 1 is connected to the metal shell on a 6L6 and most other metal tubes.
2) The shiny cover of Steppenwolf's second album conducts electricity.
I have one channel done. This thing has WAY too much gain. Aside from that, a bit of hum, I can trace that down. It has about 20 min on it so far. I'm running the KT88's at 60ma each right now. By the way, the Edcor transformer is under volts by about 25! I'd much rather deal with the other problem. And yes, my line voltage is correct. I'm not sure what tap the feedback resistor is intended to connect to, I have it on the 8 ohm tap. Square waves look pretty sloppy. My comp cap is 82pf for now. Without feedback and without the comp cap I was picking up a nice AM radio station. Those who build this should probably plan in a shield for the EF86. I will probably try triode wiring the thing.
Nice progress!
What's the resistance of the power transformer primary and secondary? I'll fine-tune PSUD model with those numbers and we can see how close to reality the model is.
The important spec for a mosfet follower is the Reverse Transfer Capacitance (Crss). This is the gate to drain capacitance which appears directly across the driver tubes output. It should be low and constant in the range of voltage seen in the circuit.
The gate to source capacitance is bootstrapped since the source follows the gate, and the source to drain capacitance is shunted by the channel resistance (a few ohms) and driven by the source.
The AOT1N60 from Alpha and Omega Semiconductor has a 1.8 pF Crss, a 600 volt rating and is 72 cents each at Digikey. I bought 100 ($43) but haven't had the time to try them yet.
12bh7 is a great tube. Not everyone here will agree, but I liked it the best. The 12AT7 in the LTP is the only one I played with in that slot, I have had no reason to change it.
If you look back in the OPUS a modern Mullard thread you can see the tube complement there. You can play with the GNFB to your liking. I think depending on the tube I was 7 dB to 15 dB of feedback depending on the input tube.
How is the hum?
If you look back in the OPUS a modern Mullard thread you can see the tube complement there. You can play with the GNFB to your liking. I think depending on the tube I was 7 dB to 15 dB of feedback depending on the input tube.
How is the hum?
Hum is gone, but I'm not getting nearly the power I should. Right now I start clipping at 15v out, about 28 watts. My gain w/o feedback is 314 (0.043 v in = 13.5 out) and with feedback it is 15.8 (0.95v in, 15 out). I don't remember how to compute feedback in db, help? Also, any thoughts on why the power is so low? I'm running the KT88's about 75 ma each. B+ under load is 445, B+ to the LTP is 442 and I have about 170v across the 50k plate resistors. Any ideas?
Stupid me!! I just realized I don't have a 16 ohm tap, only 4 and 8. So I was making my measurments from the 4 ohm tap into an 8 ohm load. When I do it CORRECTLY, I just hit 50 watts with the KT88's running at 55ma. That's about what I figured, based on using a 5k primary OPT. The remaining question at this point is, knowing open and closed gain, how do I convert that into db of feedback?
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.