a new imbalance
Good on ya Frank,
The whole thing is most certainly there A-Z as you say, it took me a lot of reading, re-reading and cross referencing to get to understand it though, I was learning about tube theory as I went along.
I've been reading up on stability networks and feedback and I'm slowly starting to get the picture regarding the compensation network, I know what it's for but I'm still getting the mechanism down. Anyhow, on the way home last night I was thinking about it and I suddenly realized that there is another imbalance going on, but I haven't had time to look it over yet.
The output section has 2 different output impedances, one for the upper tubes and one for the lower. The output Z for the amp is stated as .4 ohms, but I think that changes with the signal swinging north or south. Now I know that Z out drops as NFB is added, (although I don't know why yet) so I'm thinking that the Z out of just the output tubes is not all that's going on regarding Z out. Perhaps this is one of the other problems with SEPP you were eluding to Frank? Just another bump on the road to understanding.
Yep, while I didn't understand how the whole thing worked when I decided to build it, I was looking for a no nonsense design, that would be a good match for an ESL. I loved what he had written throughout the book in general and when he described his version of the OTL, it looked as a better choice than anything I'd seen so far. I design and build other non audio related stuff, and I work the same way as B.R. so I was easily seduced. I'd been leaning toward a good SS design before reading AR because I thought that running the output through a transformer was just against the principles of really good audio. It was later when I started trying to figure out how ESL's worked that I found out they have step up transformers, oh well.
Now, forgetting for a moment the inductive/non-inductive bit, I'm still trying to figure out why he would say the screen resistors carry only DC.
Hey Steve, I was sure you had signed your name to a post in the past, but I didn't have time to look. If you've got the time, it really is worthwhile figuring out the B.R. patent for yourself. It is quite a little piece of elegent engineering. There are a few more benefits over previous designs that haven't been touched on in this discussion. As the overall forward voltage gain imbalance is greatly reduced in this design, it requires 6db less feedback to get the distortion down than his Futterman version. It also requires less bias for the output as there is more available drive voltage, so it runs cooler, tubes last longer etc.
Something I did touch on way back somewhere is that in the previous art, a negative DC offset emerges at the output as the overall output increases. This pulls the drivers back (I understand that it's the woofer that suffers most in this situation) and kills bass. This is virtually eliminated in his design. I suspect the little bit that is left may be as a result of the non-floating DC bias for the upper buffer, but I've yet to have a good look to see if I can figure out what's going on. Besides he's managed to couple 4 stages with just 2 caps and it's best to leave well enough alone for now, but I am going to give it serious thought. I'd try and get a discussion going with B.R. but as much as he likes to educate through books, he doesn't seem like he much likes to discuss his designs. Perhaps every guy who can bias a tube thinks he can improve the designs, if you know what I mean, and he's just resistant to someone coming out of the blue and telling him all about something he's abandoned long ago.
Lot's to think about,
I still need to understand transmission lines so I can design a little subwoofer to go with these cool little minimonitors a bunch of us are building at work. I also have to dig up some material on winding inductors.
That's me for now,
Chris
Good on ya Frank,
The whole thing is most certainly there A-Z as you say, it took me a lot of reading, re-reading and cross referencing to get to understand it though, I was learning about tube theory as I went along.
I've been reading up on stability networks and feedback and I'm slowly starting to get the picture regarding the compensation network, I know what it's for but I'm still getting the mechanism down. Anyhow, on the way home last night I was thinking about it and I suddenly realized that there is another imbalance going on, but I haven't had time to look it over yet.
The output section has 2 different output impedances, one for the upper tubes and one for the lower. The output Z for the amp is stated as .4 ohms, but I think that changes with the signal swinging north or south. Now I know that Z out drops as NFB is added, (although I don't know why yet) so I'm thinking that the Z out of just the output tubes is not all that's going on regarding Z out. Perhaps this is one of the other problems with SEPP you were eluding to Frank? Just another bump on the road to understanding.
Yep, while I didn't understand how the whole thing worked when I decided to build it, I was looking for a no nonsense design, that would be a good match for an ESL. I loved what he had written throughout the book in general and when he described his version of the OTL, it looked as a better choice than anything I'd seen so far. I design and build other non audio related stuff, and I work the same way as B.R. so I was easily seduced. I'd been leaning toward a good SS design before reading AR because I thought that running the output through a transformer was just against the principles of really good audio. It was later when I started trying to figure out how ESL's worked that I found out they have step up transformers, oh well.
Now, forgetting for a moment the inductive/non-inductive bit, I'm still trying to figure out why he would say the screen resistors carry only DC.
Hey Steve, I was sure you had signed your name to a post in the past, but I didn't have time to look. If you've got the time, it really is worthwhile figuring out the B.R. patent for yourself. It is quite a little piece of elegent engineering. There are a few more benefits over previous designs that haven't been touched on in this discussion. As the overall forward voltage gain imbalance is greatly reduced in this design, it requires 6db less feedback to get the distortion down than his Futterman version. It also requires less bias for the output as there is more available drive voltage, so it runs cooler, tubes last longer etc.
Something I did touch on way back somewhere is that in the previous art, a negative DC offset emerges at the output as the overall output increases. This pulls the drivers back (I understand that it's the woofer that suffers most in this situation) and kills bass. This is virtually eliminated in his design. I suspect the little bit that is left may be as a result of the non-floating DC bias for the upper buffer, but I've yet to have a good look to see if I can figure out what's going on. Besides he's managed to couple 4 stages with just 2 caps and it's best to leave well enough alone for now, but I am going to give it serious thought. I'd try and get a discussion going with B.R. but as much as he likes to educate through books, he doesn't seem like he much likes to discuss his designs. Perhaps every guy who can bias a tube thinks he can improve the designs, if you know what I mean, and he's just resistant to someone coming out of the blue and telling him all about something he's abandoned long ago.
Lot's to think about,
I still need to understand transmission lines so I can design a little subwoofer to go with these cool little minimonitors a bunch of us are building at work. I also have to dig up some material on winding inductors.
That's me for now,
Chris
Zo.
Hi,
Well, if you must than you could say that Zo_tot = Z_Upper + Z_Lower.
This is to be expected since they have different voltage gain, the one with most gain has the higher Zo.
Everything else being the same.
Hence, when NFB is applied the open loop gain is reduced so inevitably Zo_tot will be reduced by the same token.
So, it follows that if we reduce the gain of the stage driving the upper section and than apply global feedback we stand a good chance of evening out the imbalance and reducing Zo_tot.
Be aware that I'm oversimplifying for claritys' sake.
I'd chalk it up to miscommunicaton for it just can't be true.
Cheers,
Hi,
Well, if you must than you could say that Zo_tot = Z_Upper + Z_Lower.
This is to be expected since they have different voltage gain, the one with most gain has the higher Zo.
Everything else being the same.
Hence, when NFB is applied the open loop gain is reduced so inevitably Zo_tot will be reduced by the same token.
So, it follows that if we reduce the gain of the stage driving the upper section and than apply global feedback we stand a good chance of evening out the imbalance and reducing Zo_tot.
Be aware that I'm oversimplifying for claritys' sake.
I'd chalk it up to miscommunicaton for it just can't be true.
Cheers,
Diversion here.
Bulletin, Bulletin, this just in.
I just saw a tornado some distance to the east. Could see the cloud rotating counterclockwise, moving southeast.
We had a tornado about two weekends ago that came within 1 mile of us, moving northeast. You may have seen it on national news.
The're coming in all directions.
I don't know why, but I get all auxilarated over something like this.
Now back to your regularly scheduled postings.
Bulletin, Bulletin, this just in.
I just saw a tornado some distance to the east. Could see the cloud rotating counterclockwise, moving southeast.
We had a tornado about two weekends ago that came within 1 mile of us, moving northeast. You may have seen it on national news.
The're coming in all directions.
I don't know why, but I get all auxilarated over something like this.
Now back to your regularly scheduled postings.
I must admit.....
I love to see and would like to chase a Tornado. The one two weeks ago was at night, so I could't see or follow it. I heard the winds were estimated at 200mph+ (forgot conversion to kmh).
The only thing I don't like is if anyone is injured or property destroyed, which the one two weeks ago did.
It demolished a business and manufacturing bldg, took the second floor and outer wall off a couple apartment buildings, looked like a large doll house with furniture inside. Knocked some cars off the interstate too.
We had 15 that day in central Illinois; that afternoon and evening.
Well, better get off.
I love to see and would like to chase a Tornado. The one two weeks ago was at night, so I could't see or follow it. I heard the winds were estimated at 200mph+ (forgot conversion to kmh).
The only thing I don't like is if anyone is injured or property destroyed, which the one two weeks ago did.
It demolished a business and manufacturing bldg, took the second floor and outer wall off a couple apartment buildings, looked like a large doll house with furniture inside. Knocked some cars off the interstate too.
We had 15 that day in central Illinois; that afternoon and evening.
Well, better get off.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.