The amp is designed for 600 ohm. I've used 250 ohm phones on it without issues and I have built an impedance matching box with transformers if I want to use lower impedance phones, so that's no problem.
The transformer HV winding is 80mA max., I have a 6AS7 pass tube in the PSU but that should be able to handle a lot more.
I guess a 1V input sensitivity would be fine, that would allow iphones etc. to be used a a source.
Right now I have 6SN7, 6BL7 and 6BX7 on hand, the rest of my preamp tubes (mostly 12AX7) are 9-pin. I'd like to keep it based on these types or at least some other octal-based tubes.
If you are happy with your amplifier regarding power output and distortion you should at least bring Uk of the upper tube to a safer level.
Adjust Rk and Ra and/or reduce Ub.
The 200V Ufk rating is a maximum peak rating.
Dc Ufk should never be more than 100V if you want reliable operation.
The damping factor of your amp is on the high side but could be improved with some fb.
It migth be worth to try your amp with 6BX7 tubes only (increase Rk so current does not go through the roof and decrease R4)
If you short out R4 completely you will get a sort of push-pull operation.
A 250ohm load requires heavy lifting and a mu-follower with one 6BL7 at 16mA does not really cut it.
A white cathode follower would be more suitable.
The 6BX7 is the best candidate for that.
A WCF with 2 paralleled sections at Uak 75-100 and 15-20mA per section.
3W anode dissipation per tube would give you long life.
Ub would be ca 170-220V at 30-40mAdc + a few mA for the preceding voltage amp .
The voltage amp could be a single 6SN7 with or without fb wrapped around it.
Without fb a damping factor down to 10 should be possible if you run the tubes with at least 40mA dc
Adjust Rk and Ra and/or reduce Ub.
The 200V Ufk rating is a maximum peak rating.
Dc Ufk should never be more than 100V if you want reliable operation.
The damping factor of your amp is on the high side but could be improved with some fb.
It migth be worth to try your amp with 6BX7 tubes only (increase Rk so current does not go through the roof and decrease R4)
If you short out R4 completely you will get a sort of push-pull operation.
A 250ohm load requires heavy lifting and a mu-follower with one 6BL7 at 16mA does not really cut it.
A white cathode follower would be more suitable.
The 6BX7 is the best candidate for that.
A WCF with 2 paralleled sections at Uak 75-100 and 15-20mA per section.
3W anode dissipation per tube would give you long life.
Ub would be ca 170-220V at 30-40mAdc + a few mA for the preceding voltage amp .
The voltage amp could be a single 6SN7 with or without fb wrapped around it.
Without fb a damping factor down to 10 should be possible if you run the tubes with at least 40mA dc
Last edited:
I really appreciate your suggestions, I see that I have to read up a fair bit... 🙂 I've found a couple of detailed descriptions of the WCF and I've dug out the Morgan Jones books again so that should keep me occupied for a while.
At the moment I'll just try to reduce my B+ down to 310V and listen to the amp while I figure out what I'm reading 🙂
At the moment I'll just try to reduce my B+ down to 310V and listen to the amp while I figure out what I'm reading 🙂
I've installed new 6SN7s now, and a pair of 6BX7 since one of my 6BL7s died. All noise is gone, only a bit of hiss when I turn it up to painful levels, so that's good. I could leave it alone now as everything works fine and sounds good (and I might just do that), but...
I've tried to wrap my head around the circuit, so what I'm thinking (with help from gorgon53 🙂 ) is:
- bigger output cap, 20uF or so for lower bass rolloff
- C1: decrease from 2uF to 0.047-0.1uF
- R3: decrease from 1M to 500K
- R7: increase from 10K to 100K
And here's possibly where I'm on thin ice, I've used the article from Valve Wizard (The Valve Wizard -Mu Follower) and calculated with my voltages and the SN7/BX7 datasheets, so I hope I've done it correctly:
- R2 and R6: increase from 100Ω to 120Ω (or maybe just LED bias the 6SN7s?)
- R4: decrease from 10K to 6.8K (just a 22K in parallel with the 10K)
I'm close to finishing my output transformer box, so the amp will always see ~600Ω.
I've tried to wrap my head around the circuit, so what I'm thinking (with help from gorgon53 🙂 ) is:
- bigger output cap, 20uF or so for lower bass rolloff
- C1: decrease from 2uF to 0.047-0.1uF
- R3: decrease from 1M to 500K
- R7: increase from 10K to 100K
And here's possibly where I'm on thin ice, I've used the article from Valve Wizard (The Valve Wizard -Mu Follower) and calculated with my voltages and the SN7/BX7 datasheets, so I hope I've done it correctly:
- R2 and R6: increase from 100Ω to 120Ω (or maybe just LED bias the 6SN7s?)
- R4: decrease from 10K to 6.8K (just a 22K in parallel with the 10K)
I'm close to finishing my output transformer box, so the amp will always see ~600Ω.
Attachments
Last edited:
good you increased the output cap, I forgot to mention that, 20uF should be ok with 600 ohm.
Do you know why your 6BL7 died?
Heater-cathode breakdown ?
Make sure the same does not happen to your 6BX7, increase the cathode resistor and reduce R4 to get safer voltage levels.
Forget about the "1/3 rules", you cannot go by it with your voltage levels.
Last edited:
Be aware that your tubes are not very happy operating with low anode and grid voltages .
They would like to see the signal peak plus a additional 3V between grid and cathode.
Also, the 6BX7 wants, and can handle, much more current than a 6SN7 can supply.
You should feed additional current to the 6BX7 with a resistor from the R2/3/4 junction to ground.
This resistor is also a life-insurance for your tubes (especially during start up).
They would like to see the signal peak plus a additional 3V between grid and cathode.
Also, the 6BX7 wants, and can handle, much more current than a 6SN7 can supply.
You should feed additional current to the 6BX7 with a resistor from the R2/3/4 junction to ground.
This resistor is also a life-insurance for your tubes (especially during start up).
Last edited:
I got a better idea, instead of decreasing R4 increase the cathode resistor of the 6SN7.
This would increase Ua and -Ug and decrease Ia to levels more suitable.
The resistor from R2/3/4 to ground must be able to handle the power when feeding substantial current to the 6BX7.
This would increase Ua and -Ug and decrease Ia to levels more suitable.
The resistor from R2/3/4 to ground must be able to handle the power when feeding substantial current to the 6BX7.
I found a couple of 47K power resistors in my parts box, would they be suitable to run to ground from the resistor junction?
And something around 150-180R for the 6SN7 cathodes?
Hello,
if you are using a voltage divider to elevate to heater you also should keep in mind that there is not only Uf/k there is also a maximum rating for Rf/k which usually is 20K.
So the lower resitor of the voltage divider should not be higher. I usually use 18K.
Best regards
Martin
if you are using a voltage divider to elevate to heater you also should keep in mind that there is not only Uf/k there is also a maximum rating for Rf/k which usually is 20K.
So the lower resitor of the voltage divider should not be higher. I usually use 18K.
Best regards
Martin
I couldn’t find any Rf/k values for the 6BX7, so I went with the recommendation on the Valve Wizard site (100K max.).
I made some simulations with 320V Ub and a 600ohm load now.
I got the best results when biasing the 6SN7 with 2-3 leds, R4 10k.
2 leds givs a bit more output but the minimum of Ugk is a bit on the low side.
3 leds has more headroom.
A 4,7k!!! resistor connect from the R2/3/4 junction to ground, dissipation 5,8W (15-25 W glazed resistor with brazed!!! connections)
Each 6BX7 cathode connects to its own 120ohm resistor.
This resistors are necessary for current sharing and for buffering capacitive load.
The 120ohm resistors connect together to R2 (220 ohm, diss 0,35W, 1-2Wresistor).
The 20uF output cap connects to this junction.
To reduce lf-overload of the amp use Cin 0,1uF with Rg 220k for the 6SN7input (when 20uF output cap),
this givs 27Hz -1db with 20uF cap but damping gets lousy.
For minimum lf-distortion and for better damping Increase both caps!
2 LED, 1Vrms input, 260mW, 600ohm, 0.84% simulated total harmonic distortion.
3 LED, 1Vrms input, 230mW, 600ohm, 0.6% THD because of lower second harmonic, higher harmonics about the same as with 2 LEDs.
Max output is current limited, around 400mW
Last edited: