Sorry. As far as I know, that article is not up on the web. I was one of the 3 Authors of the article, and was involved with all the research.
Later, since the article was fairly unnoticed, I gave a talk about the findings in the article at the 2003 VSAC (Vacuum Tube State of the Art Conference) in Silverdale, WA. VSAC was sponsored by Bottlehead.com.
I was hoping that someone had already found the original article on the web, or perhaps as a CD of Glass Audio articles.
Ed T. Dell was the publisher of Glass Audio. I do not know if he can be contacted. That was the last issue of Glass Audio. Then I believe 3 magazines including Glass Audio were combined into Audioxpress. It would be nice to get permission to scan and re-publish the article, such as a .PDF.
Later, since the article was fairly unnoticed, I gave a talk about the findings in the article at the 2003 VSAC (Vacuum Tube State of the Art Conference) in Silverdale, WA. VSAC was sponsored by Bottlehead.com.
I was hoping that someone had already found the original article on the web, or perhaps as a CD of Glass Audio articles.
Ed T. Dell was the publisher of Glass Audio. I do not know if he can be contacted. That was the last issue of Glass Audio. Then I believe 3 magazines including Glass Audio were combined into Audioxpress. It would be nice to get permission to scan and re-publish the article, such as a .PDF.
Last edited:
Just for a better explain
1- of course this thread is for a SE but if there is a mention on PSE for me is natural to say that the PP is much better, that's all.
2- Regarding A2 that is a consequence (correct) and must be not a standard job for 300B; it is a limit (wrong) that in every case must be taken very far for the normal use.
The solution, in my opinon is to select at the best the 300B then fix the bias with fixed bias ( in my opinion is the best way) close to the max plate power. In some case 400 V dc and 70-75-mA can be good; the neg. voltage is around -87 volt ( depend on 300B) in this way the swing ot the grid of 300B is very large and the power relatively high. In this way is very difficult with a standard music signal to reach the A2.
If the driver stage is coupled in dc when you go into the A2 you get quickly a lot of distortion (non linear, of course and this is the main problem!!!); in the FFT the shape is jagged from harmonics at every frequences with different level; the results in little bit better with RC coupling ( in the same condition).
I have a look if I have recorded that shape in the past on my test. In my case 2500 ohm with a single sec at 6 ohm I can get 8 w rms with a reasonable low THD. Never seen blocking.
3- regarding the driver for 300b I repeat that the Audio Note kit 1 diagram ( cathode bias) is one of the best possible and easy to build; the Miller cap of 300B is not a problem because the 5687 in srpp has around 300 ohm of Zout with a large swing and a reasonable distortion, same for 7044 , with 6463 is not so fine; very good with 6H30 or 6H6, you must adjust the Vdc. The gain are little bit different but in this case is not a problem because the total db of two stage is enough to drive the 300B. Also the frequency answer is larger compared with the response of the OT SE trafo for this purpose. So it isn't a problem.
Walter
1- of course this thread is for a SE but if there is a mention on PSE for me is natural to say that the PP is much better, that's all.
2- Regarding A2 that is a consequence (correct) and must be not a standard job for 300B; it is a limit (wrong) that in every case must be taken very far for the normal use.
The solution, in my opinon is to select at the best the 300B then fix the bias with fixed bias ( in my opinion is the best way) close to the max plate power. In some case 400 V dc and 70-75-mA can be good; the neg. voltage is around -87 volt ( depend on 300B) in this way the swing ot the grid of 300B is very large and the power relatively high. In this way is very difficult with a standard music signal to reach the A2.
If the driver stage is coupled in dc when you go into the A2 you get quickly a lot of distortion (non linear, of course and this is the main problem!!!); in the FFT the shape is jagged from harmonics at every frequences with different level; the results in little bit better with RC coupling ( in the same condition).
I have a look if I have recorded that shape in the past on my test. In my case 2500 ohm with a single sec at 6 ohm I can get 8 w rms with a reasonable low THD. Never seen blocking.
3- regarding the driver for 300b I repeat that the Audio Note kit 1 diagram ( cathode bias) is one of the best possible and easy to build; the Miller cap of 300B is not a problem because the 5687 in srpp has around 300 ohm of Zout with a large swing and a reasonable distortion, same for 7044 , with 6463 is not so fine; very good with 6H30 or 6H6, you must adjust the Vdc. The gain are little bit different but in this case is not a problem because the total db of two stage is enough to drive the 300B. Also the frequency answer is larger compared with the response of the OT SE trafo for this purpose. So it isn't a problem.
Walter
This is a non-issue if the direct coupled driver that is used is capable of enough current to feed the grid adequately. A strong enough cathode follower or source follower would be an easy way to go if one wanted to get a very nice grid drive potential.
Proper design is key.
Last edited:
I am afraid Walter I completely disagree with what you say about A2 operation. Why should't be standard? You can only say that for tubes with particular construction where the grid is not up to the task but this is not the case for the 300B as most DHT's. If you get a lot of distortion maybe your driver is not good enough?
I only agree on the fact that the 300B won't give a lot gain in terms of output power going into A2 like other tubes (because it has already good efficiency in class A1) but it is not just for the power.
I only agree on the fact that the 300B won't give a lot gain in terms of output power going into A2 like other tubes (because it has already good efficiency in class A1) but it is not just for the power.
If you design a single ended amp for A2, you should bias the tube to have more current than is usual for A1.
Otherwise, when your large signal draws grid current, then the same large signal in the opposite polarity will cause the output tube to go into cut-off.
Cut-off . . . Ouch to my ears.
Otherwise, when your large signal draws grid current, then the same large signal in the opposite polarity will cause the output tube to go into cut-off.
Cut-off . . . Ouch to my ears.
A2 is a great way to decrease distortion and increase headroom on an amplifier biased deep into class A. Works great for operating points that run a bit higher dissipation that would otherwise have asymmetric swing capabilities. A little lower voltage and a higher current will often put the tube in a nice linear operating range anyways.
With a good direct coupled driver blocking distortion is eliminated too. What's not to like?
With a good direct coupled driver blocking distortion is eliminated too. What's not to like?
x 6A3
6N6/6H6 double triode sister of 6H30, good tube. Regarding the IT trafo for drive the 300B I have a different opinion.
x Merlin
I agree with you; this project with a proper OT (from AN are good) is one of the best for 300B that I have seen, built ( for some friend) and tested & listen.
Walter
6N6/6H6 double triode sister of 6H30, good tube. Regarding the IT trafo for drive the 300B I have a different opinion.
x Merlin
I agree with you; this project with a proper OT (from AN are good) is one of the best for 300B that I have seen, built ( for some friend) and tested & listen.
Walter
Waltube,
American 6H6 is dual diode.
American 6N6 is a special dual triode with the cathode of the first triode directly connected to the grid of the second triode (internal connection).
Perhaps you meant a Russian 6H6 or 6N6? We have confusion on these threads sometimes, perhaps we should always state the country the tube name applies to.
Although I built them, I am no longer a fan of amps with interstage coupling. Part of the reason is I never found/afforded one that I really liked. What was your opinion on an interstage driving a 300B (for SE)?
And by the way, the original thread post #1 was about single ended, and the original author of that post has re-stated that he means SE, not push pull.
American 6H6 is dual diode.
American 6N6 is a special dual triode with the cathode of the first triode directly connected to the grid of the second triode (internal connection).
Perhaps you meant a Russian 6H6 or 6N6? We have confusion on these threads sometimes, perhaps we should always state the country the tube name applies to.
Although I built them, I am no longer a fan of amps with interstage coupling. Part of the reason is I never found/afforded one that I really liked. What was your opinion on an interstage driving a 300B (for SE)?
And by the way, the original thread post #1 was about single ended, and the original author of that post has re-stated that he means SE, not push pull.
Last edited:
As I said in the beginning, a good IT transformer in the Sun Audio circuit would be a good idea because having one 27K resistor at the plate and another at cathode severely reduces the available voltage supply and the driver capability. Probably this is the reason why the 300B is run almost like a 2A3....
Higher output resistance of a driver is better for A1 because clipping does not sound so bad like dynamic distortions caused by bias dependent on signal envelope. At least, you hear clipping on peaks and decrease volume, instead of listening to smeared soundstage. But even better to avoid a coupling cap before the grid of the output tube.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.