Please express your opinion about my design.
Chris
Chris
An externally hosted image should be here but it was not working when we last tested it.
Hi Chris,
If you don't mind, there are a few details that seem rather odd to me.
Keep in mind I didn't do any math on your design, just hunches...you know the kind of things that jump at you when you're used to look at circuit diagrams.
So here goes:
- The filter chokes seem low in value.
-I'd feel happier if you could fit some 500VDC caps for the B+ iso
the series arrangement.
- The filter caps in the heater supply seem extremely low.
On a more "tweaker" level:
-I'd up the first coupling cap to 0.47uF and the second to 0.68uF.
-I'd put some extra bypass caps in // with the 10uF Mundorfs,
even thinking of tripling their value in the process. (the Mundorfs
may not need bypassing)
- I'd increase the filtering, reservoir caps substantially.
in the B+ supply.
-If money no object I'd even endow the ECC99 with a dedicated
regulated B+.
-And have a dedicated supply for the heaters of the ECC99 and
300B, the ECC99 needs 6.3V and the 300B 5V so I reckon the
heater supply drawing isn't quite finnished?
-You can probably drop or lower the value of the gridstopper
resitor on the second half of the ECC99 and use the 1K on the
300B instead. If you have a scope you may find the exact value
for the job.
-I see you use a ripple cancelling technique on the output which
is fine if implementig correctly, again a scope will help to fine
tune the values of the caps.
-The 100K volpot, if yor aiming for a high-end design get the best
you can do there as well, I personally like stepped attenuators
with a constant impedance.Particularly in that position.
Passive component choice is a matter of personal preference, I like as honest to source as I can come best but YMMV.
If you like what this amp does I heartily recommend the Loftin-White for an SE amp as a next step up, you could of course use a choke iso a resistive load or even a CSS all of which should, in theory at least, bring more linearity and better sound.
Subjectively that's not always the case so listening test may be the decisive factor, as they should be anyway.
Hope it helps,
P.S. Hope I haven't missed out anything blatantly obvious...I usually do. nad sorry for the long rant.
If you don't mind, there are a few details that seem rather odd to me.
Keep in mind I didn't do any math on your design, just hunches...you know the kind of things that jump at you when you're used to look at circuit diagrams.
So here goes:
- The filter chokes seem low in value.
-I'd feel happier if you could fit some 500VDC caps for the B+ iso
the series arrangement.
- The filter caps in the heater supply seem extremely low.
On a more "tweaker" level:
-I'd up the first coupling cap to 0.47uF and the second to 0.68uF.
-I'd put some extra bypass caps in // with the 10uF Mundorfs,
even thinking of tripling their value in the process. (the Mundorfs
may not need bypassing)
- I'd increase the filtering, reservoir caps substantially.
in the B+ supply.
-If money no object I'd even endow the ECC99 with a dedicated
regulated B+.
-And have a dedicated supply for the heaters of the ECC99 and
300B, the ECC99 needs 6.3V and the 300B 5V so I reckon the
heater supply drawing isn't quite finnished?
-You can probably drop or lower the value of the gridstopper
resitor on the second half of the ECC99 and use the 1K on the
300B instead. If you have a scope you may find the exact value
for the job.
-I see you use a ripple cancelling technique on the output which
is fine if implementig correctly, again a scope will help to fine
tune the values of the caps.
-The 100K volpot, if yor aiming for a high-end design get the best
you can do there as well, I personally like stepped attenuators
with a constant impedance.Particularly in that position.
Passive component choice is a matter of personal preference, I like as honest to source as I can come best but YMMV.
If you like what this amp does I heartily recommend the Loftin-White for an SE amp as a next step up, you could of course use a choke iso a resistive load or even a CSS all of which should, in theory at least, bring more linearity and better sound.
Subjectively that's not always the case so listening test may be the decisive factor, as they should be anyway.
Hope it helps,
P.S. Hope I haven't missed out anything blatantly obvious...I usually do.
nad sorry for the long rant.
New design
Hi,
thanks for your advice. Mainly at the suggestion of Vuki I have updated my design.
Regards
Hi,
thanks for your advice. Mainly at the suggestion of Vuki I have updated my design.
Regards
An externally hosted image should be here but it was not working when we last tested it.
Hi Chris,
It's getting there........
May I suggest you buy another 10Vac transformer and a pair of LT1085's per channel. Run a similar Shottky bridge, cap, then LT1085 as a voltage reg, then a second 1085 as a current reg. You'll need to tweak both to get the voltage at the 300B filament to 5V. It's much quieter and less grainy, and I've now used it with success on GM70, 813 and 300B filaments, and sounds much better than even an LCLCLC-CMC passive filament supply. A common mode choke after the current reg, and a small cap (~100n) soldered across the filament pins are icing.
Similarly, if you can find another PS choke per channel, an CLCLC supply can improve the sonics, esp in a SET. I even use them in my PP amps now.
Because I'm commenting so late in the game, I'll just say that I wouldn't use the ECC99 - ECC99 as drivers. They'll make more distortion than the 300B, and definitely be the limitation in the amp IMO/IME, and the ECC99 is one of my fave tubes.
It's getting there........
May I suggest you buy another 10Vac transformer and a pair of LT1085's per channel. Run a similar Shottky bridge, cap, then LT1085 as a voltage reg, then a second 1085 as a current reg. You'll need to tweak both to get the voltage at the 300B filament to 5V. It's much quieter and less grainy, and I've now used it with success on GM70, 813 and 300B filaments, and sounds much better than even an LCLCLC-CMC passive filament supply. A common mode choke after the current reg, and a small cap (~100n) soldered across the filament pins are icing.
Similarly, if you can find another PS choke per channel, an CLCLC supply can improve the sonics, esp in a SET. I even use them in my PP amps now.
Because I'm commenting so late in the game, I'll just say that I wouldn't use the ECC99 - ECC99 as drivers. They'll make more distortion than the 300B, and definitely be the limitation in the amp IMO/IME, and the ECC99 is one of my fave tubes.
Hi,
And that would be the ultimate tweak for Jack Eliano's DRD amps as well...
Which for me at least are the most clever SE design to date and we ( meaning Brett and I) ain't even fans of SE but it would great to implement it in PP design as well...
Very clever thinking but you need to know how the currents flows to understand it.
Glad to see you back posting, Brett.
And that would be the ultimate tweak for Jack Eliano's DRD amps as well...
Which for me at least are the most clever SE design to date and we ( meaning Brett and I) ain't even fans of SE but it would great to implement it in PP design as well...
Very clever thinking but you need to know how the currents flows to understand it.
Glad to see you back posting, Brett.
I'd use a 12GN7A, 12HL7, 6CH6, C3m/o or a EL84. The first two are my faves, running pentode (12GN7A), Rl=4k, Ik=36mA, Va=250, Vg2=150 which will give you a gain of about 40, low distortion (almost all 2H). Tweaking the voltages and loads can give you up to about 100x gain.hugoland said:
Sounds so good, it made an ECC99 or 6H30 into 300B - IT - GM70 (low and high voltage) sound broken. The only place I think I'd use a two stage now would be an 845 needing 120+V of drive.
fdegrove said:
I'm not sure I'd try it today. Too tired to work it out, but I'm sure it can be done. The HV cap requirement puts me off the DRD. Plus I don't like 300B's and doing it with a bright emitter seems a bit pointless.
Glad to see you back posting, Brett.
Thanks, I'm still too busy to be back here much. But I will admit the air smalls a lot cleaner.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.