What gain did you calculate for this circuit?
I've never used this tube before, but the specs look interesting. They're expensive, but not horribly so. I'm curious about anyone's experiences with these as regards microphonics, reliability, tendency toward oscillation, or any other watch-out-fors.
I've never used this tube before, but the specs look interesting. They're expensive, but not horribly so. I'm curious about anyone's experiences with these as regards microphonics, reliability, tendency toward oscillation, or any other watch-out-fors.
I got about 7 (17 db) un-bypassed and about 10.5 (20.5 db) bypassed from Tube CAD.
I was originally looking for a circuit with gain between 3 and 7.
after reading an article claming that un-bypassed cathode resistors degraded tube parameters, I decided to try it both ways.
I am also looking at trying 3 red LED's instead of the cathode resistor. I have never seen it done, and might unbalance the circuit, but it was a thought.
Doug
x-pro said:
That's absolutely correct- see Baxandall's excellent articles, for example- but distortion reduction with degeneration is still true as a general statement.
The art of design is being able to make a judgement as to where the most appropriate use of the feedback tool can be made. If I had, for example, a circuit which is producing 1% of 2nd HD and vanishingly low third and higher, yeah, I'd be hesitant to throw in some degeneration. But more often, one is starting with a circuit that has appreciable amounts of third and fifth to begin with, so the use of feedback can be very appropriate.
Hi folks,
I'm not an expert on tubes but I have designed some tube-stuff.
If you dont need the excess gain at least make options to easily
remove your bypass cap and even make provision for a film type in
parallel with the electrolytic. Then you can listen and compare.
Also I would put in some low pass noise filter at the input to avoid
RF crap into the circuit.
And maybe a small film cap for decoupling near the tube.
What does Tube_Cad say if you pick your signal at the lower halfs anode. ?
See my old post: http://www.diyaudio.com/forums/showthread.php?s=&threadid=35848
/ Mattias
I'm not an expert on tubes but I have designed some tube-stuff.
If you dont need the excess gain at least make options to easily
remove your bypass cap and even make provision for a film type in
parallel with the electrolytic. Then you can listen and compare.
Also I would put in some low pass noise filter at the input to avoid
RF crap into the circuit.
And maybe a small film cap for decoupling near the tube.
What does Tube_Cad say if you pick your signal at the lower halfs anode. ?
See my old post: http://www.diyaudio.com/forums/showthread.php?s=&threadid=35848
/ Mattias
Just want to add.
I did the simulation of my hybrid amp with Synopsys Saber software
and I think I reached the lowest output impedance by picking the signal
at the top of the bottom anode. Otherwise i don't remember any other
difference.
It would be nice to here some opinions on this from you tube-dudes.
/ Regards / Mattias
I did the simulation of my hybrid amp with Synopsys Saber software
and I think I reached the lowest output impedance by picking the signal
at the top of the bottom anode. Otherwise i don't remember any other
difference.
It would be nice to here some opinions on this from you tube-dudes.
/ Regards / Mattias
Good Catch.
Also excellent point. A voltage divider set to 1/3 b+ should do it.
Yes. I will do it.
I'll run it tonight.
My WAG at this point is gain about mu (15) and Zout about Rp (2k) rather than gain 7 and Zout about Rp/2 (1K). But I'll check.
SY, thanks for the sanity check on cathode bypass. I am going to pick up a BlackGate and try it as a learning experience.
The reports of the 6H30 have been positive, but it’s prone to oscillate. I may need to throw a ferrite bead on the upper plate.
I'll report back with my experiences.
Thanks for the responses
Doug
Few thoughts- how would You lower Your gain (since You said that You are seeking for gain between 3 and 7.
Another thing – when picking a Black Gate bypass for the cathode I would suggest using the one with closest voltage range (....if You have bellow 10V on cathode use 10V bypass cap not 50V- personal experience.....).
I am using 5687 in my preamp prototype so I have to fight with the same problem as You do- 5687 works o.k. but it has to much gain and I don't like nfb and voltage dividers- also not bypassing the cathode is not dessirable. So I would try some 27 dht triode in mu follower (with E180F) or something like ccs from mr. Pimm (also with E180F). That would look like half of the gain I have now.
Also – maybe You would go lower with output Z if using mu follower (didn't checked that one)
Best regards and tell us how it works when finnished
daniel
Another thing – when picking a Black Gate bypass for the cathode I would suggest using the one with closest voltage range (....if You have bellow 10V on cathode use 10V bypass cap not 50V- personal experience.....).
I am using 5687 in my preamp prototype so I have to fight with the same problem as You do- 5687 works o.k. but it has to much gain and I don't like nfb and voltage dividers- also not bypassing the cathode is not dessirable. So I would try some 27 dht triode in mu follower (with E180F) or something like ccs from mr. Pimm (also with E180F). That would look like half of the gain I have now.
Also – maybe You would go lower with output Z if using mu follower (didn't checked that one)
Best regards and tell us how it works when finnished
daniel
Hi,
No,no....
That's dangerous for two reasons:
-at startup the voltage on the cathode can go up quite a bit more than what you'd measure statically once all has settled.
- the cap's insulation voltage should also take into account the AC voltage swing form the AC input (the music).
All in all if you measure say, 10VDC on the cathode when the device is properly warmed up, picking a 25V cap would be the wiser choice.
Local NFB AKA degeneration is one method, picking a different loadline is another but either way you can't perform miracles.
Which is one reason we have such a plethora of different tubes.
The standard SRPP already has the mu of the tube divided by 2 with the cathode resistor bypassed.
Removing the bypass cap causes degenerative feedback which will reduce gain and improve linearity but unfortunately the Zout will go up as well.
No free lunches in electronics I'm afraid.
Sometimes you guys are really hard to follow...
On one hand you complain on having too much and on the other you pick a topology that squeezes the last ounce of gain out of a tube...
All this for a situation calling for no gain whatsoever, just a decent buffer capable of driving a few meters of interconnect cable capacitance seeing at worst an input impedance of what? 10K?
Oh, BTW, with the 6N30-P in SRPP you won't need to bias the heater up unless you'd use a B+ way past 400VDC.
You could use a plate stopper resistor iso the ferrite bead.
Cheers,
No,no....
That's dangerous for two reasons:
-at startup the voltage on the cathode can go up quite a bit more than what you'd measure statically once all has settled.
- the cap's insulation voltage should also take into account the AC voltage swing form the AC input (the music).
All in all if you measure say, 10VDC on the cathode when the device is properly warmed up, picking a 25V cap would be the wiser choice.
Local NFB AKA degeneration is one method, picking a different loadline is another but either way you can't perform miracles.
Which is one reason we have such a plethora of different tubes.
The standard SRPP already has the mu of the tube divided by 2 with the cathode resistor bypassed.
Removing the bypass cap causes degenerative feedback which will reduce gain and improve linearity but unfortunately the Zout will go up as well.
No free lunches in electronics I'm afraid.
Sometimes you guys are really hard to follow...
On one hand you complain on having too much and on the other you pick a topology that squeezes the last ounce of gain out of a tube...
All this for a situation calling for no gain whatsoever, just a decent buffer capable of driving a few meters of interconnect cable capacitance seeing at worst an input impedance of what? 10K?
Oh, BTW, with the 6N30-P in SRPP you won't need to bias the heater up unless you'd use a B+ way past 400VDC.
You could use a plate stopper resistor iso the ferrite bead.
Cheers,
Frank,
Thanks for the reply.
I actually came up with this when the 6N1P SRPP thread was on the board. How to lower the gain? Find a different tube. This one seems to have potential. If I really need less gain, I am going to go to a Grounded cathode and transformer load with a 4:1 step-down. For the most part, the circuit is well within what I was hoping. Now if it sounds good too ...
Ekaerin,
My simulation shows there isn't a big difference between the top cathode and the bottom plate. No big surprise, they are 6v apart, separated with 232 ohms.
Un-bypassed gain was still 7, Zout was 2.7K rather than 1k. since my goal was the ability to drive 100k loads, the difference doesn't seem significant. I may try it in listening tests. Thanks.
Again, Thanks for all the comments.
Doug
Thanks for the reply.
Good to know.
I actually came up with this when the 6N1P SRPP thread was on the board. How to lower the gain? Find a different tube.
This one seems to have potential. If I really need less gain, I am going to go to a Grounded cathode and transformer load with a 4:1 step-down. For the most part, the circuit is well within what I was hoping. Now if it sounds good too ...Ekaerin,
My simulation shows there isn't a big difference between the top cathode and the bottom plate. No big surprise, they are 6v apart, separated with 232 ohms.
Un-bypassed gain was still 7, Zout was 2.7K rather than 1k. since my goal was the ability to drive 100k loads, the difference doesn't seem significant. I may try it in listening tests. Thanks.
Again, Thanks for all the comments.
Doug
fdegrove said:Hi,
No,no....
That's dangerous for two reasons:
-at startup the voltage on the cathode can go up quite a bit more than what you'd measure statically once all has settled.
- the cap's insulation voltage should also take into account the AC voltage swing form the AC input (the music).
All in all if you measure say, 10VDC on the cathode when the device is properly warmed up, picking a 25V cap would be the wiser choice.
Local NFB AKA degeneration is one method, picking a different loadline is another but either way you can't perform miracles.
Which is one reason we have such a plethora of different tubes.
The standard SRPP already has the mu of the tube divided by 2 with the cathode resistor bypassed.
Removing the bypass cap causes degenerative feedback which will reduce gain and improve linearity but unfortunately the Zout will go up as well.
No free lunches in electronics I'm afraid.
Sometimes you guys are really hard to follow...
On one hand you complain on having too much and on the other you pick a topology that squeezes the last ounce of gain out of a tube...
All this for a situation calling for no gain whatsoever, just a decent buffer capable of driving a few meters of interconnect cable capacitance seeing at worst an input impedance of what? 10K?
Dear mr. Fdegrove
I guess I didn't thought about a.c.- please correct me if I am wrong- d.c. would not go up much if the cathode is hot enough- and if the cathode is bypassed with the cap than You have no a.c. component added to the d.c. voltage on cathode. If we don't have no bypass cap than we would have d.c. voltage and music(that is added on d.c.) on cathode and if the cathode is on say 10V d.c. than the voltage will be sometimes higher because of the added music on it.Please correct if wrong.
About mu-follower....
I originally intended to use 5687 (since I like the sound of that tube) in two stage line amp. First stage- grounded cathode, second- cathode follower. I didn't like it- so I have (after few other circuits I have tried with this tube) settled down with simple grounded cathode. That works o.k. but has to much gain. Now, 27 iht has mu around 9
which is o.k. for my purposes (I intend to use is as line for my transistor power amplifier and also as line amp for my future project using 211). In fact gain of the line amp around 9 would be perfect for these 2 purposes, I think (little bit to much for my power amp but perfect for the 211).
So I have thought- how to make it easier for the 27 to work? With mu follower or with ccs from mr. pimm 27 would work better- sound is not the question here because I have to decide that when I made it- only thing that bother's me here is - should I make mu follower or ccs from mr. pimm
(the question is now ). All in all I would probably try both and if it doesn't work right then I can allways use inductor in anode of 27.Hope I am right with everything......
Sorry for the long mail and for taking place and time in DougL's thread
Best regards
daniel
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.