I had a strange thing happen to my new OTL the other day and was wondering if anyone could shed some light. The design uses the EL509 (I'm using the Ei EL519) with the screen grid tied to the plate with a 100 ohm resistor. Anyhow, during a very loud party, with only about 100 hours of run time I think a tube failed. The 100 ohm resistor got red hot and the -170V rail fuse blew. I can see only one possible failure but am curious to know if is common or even possible. Could the screen grid have shorted to the cathode? I could see the resistor getting hot if the plate connection opened, but it wouldn't have blown the fuse.
Prior to this, there was some crackling coming from the speaker. When I replaced the fuse, the crackling returned, and when I replaced the tube, all seems ok. Any thoughts?
Prior to this, there was some crackling coming from the speaker. When I replaced the fuse, the crackling returned, and when I replaced the tube, all seems ok. Any thoughts?
TUBES CAN FAIL TOO.
Hi,
This can happen with any tube amp...a slight mismatch and the weakest is taking all the heat.
You better measure the 100R resitor that took the heat too and replace it in case its value has drifted.
I don't think the tube you replaced is faulty but it may very well be unfit for that position in the amp.
If you want to find out than you'll need a tube tester or in case of the short you suggest - which I don't think is the case- a DVM can tell you that too.
On a general note, OTLs take time to settle in properly and I wouldn't use them for parties unless they had a couple of hundreds of hours on the counter.
Cheers,
Hi,
This can happen with any tube amp...a slight mismatch and the weakest is taking all the heat.
You better measure the 100R resitor that took the heat too and replace it in case its value has drifted.
I don't think the tube you replaced is faulty but it may very well be unfit for that position in the amp.
If you want to find out than you'll need a tube tester or in case of the short you suggest - which I don't think is the case- a DVM can tell you that too.
On a general note, OTLs take time to settle in properly and I wouldn't use them for parties unless they had a couple of hundreds of hours on the counter.
Cheers,
Thanks Fred,
It was a wire wound resistor, and it's still 100ohms funny enough. I'm going to replace it anyway. The other reason why I thought the tube was at fault was the 'crackle' that went away when it was replaced. While I'm here, if you don't mind my asking, wouldn't the plate be taking all the current since the screen grid has an extra 100 ohms on it? As for letting it settle in for a good long time, well, it's kinda late for that. It was a good party though
I do have access to a couple of tube testers, but neither will do magnoval based sweep tubes.
It was a wire wound resistor, and it's still 100ohms funny enough. I'm going to replace it anyway. The other reason why I thought the tube was at fault was the 'crackle' that went away when it was replaced. While I'm here, if you don't mind my asking, wouldn't the plate be taking all the current since the screen grid has an extra 100 ohms on it? As for letting it settle in for a good long time, well, it's kinda late for that. It was a good party though
I do have access to a couple of tube testers, but neither will do magnoval based sweep tubes.
Bruce says....
me;
"A member has stated that he feels it somewhat bad practice to spec wire wound resistors for in the output section of the T8 without going non-inductive. I said that you would only do so if you felt it made a difference, but that I would ask you.
What say you?
BR;
"The inductance of a 2 ohm resistor is so very low that it is totally insignificant in this circuit. The output stage shows absolutely no
effects of inductive loading. Ask the "critic" what the inductance of these resistors are."
I'll spare you the rest of the details, but I suggested that as I didn't specify, and he defended the 2 ohm cathode resistors (I forgot they were there, so contrary to what I had said earlier, the plate is not unimpeded in this manner at all), that he felt the 100 ohm plate to screen resistors were less important. His answer, those resistors carry no audio, only DC.
I like Bruces approach to design as I said before, he's very pragmatic, no expensive or complicated stuff that doesn't make good engineering sense. That's why I like it.
Have you read his book 'Audio Reality'? He explains his OTL patent and his design choices, that's why I built it. And the preamp too.
I'm not sure I need to torture my amp with such a test as you describe, but I might. I know somewhere he says it's stable while driving a square wave into a capacitor, I'll try and dig up the details.
Chris
me;
"A member has stated that he feels it somewhat bad practice to spec wire wound resistors for in the output section of the T8 without going non-inductive. I said that you would only do so if you felt it made a difference, but that I would ask you.
What say you?
BR;
"The inductance of a 2 ohm resistor is so very low that it is totally insignificant in this circuit. The output stage shows absolutely no
effects of inductive loading. Ask the "critic" what the inductance of these resistors are."
I'll spare you the rest of the details, but I suggested that as I didn't specify, and he defended the 2 ohm cathode resistors (I forgot they were there, so contrary to what I had said earlier, the plate is not unimpeded in this manner at all), that he felt the 100 ohm plate to screen resistors were less important. His answer, those resistors carry no audio, only DC.
I like Bruces approach to design as I said before, he's very pragmatic, no expensive or complicated stuff that doesn't make good engineering sense. That's why I like it.
Have you read his book 'Audio Reality'? He explains his OTL patent and his design choices, that's why I built it. And the preamp too.
I'm not sure I need to torture my amp with such a test as you describe, but I might. I know somewhere he says it's stable while driving a square wave into a capacitor, I'll try and dig up the details.
Chris
Frank says...
Hi,
Still ticked off over nothing, are we?
Let's look at some facts here:
Ask yourself what would be the better ( not necessarily the cheapest) component in that position?
A non-inductive resistor, right?After all in audio we try to avoid non-linear components, not?
Ask my advise and I invariably will give you the best component I know of unless you specifically ask me what the most economical component would be.
It is not good practice IMHO to add inductive components that are invariably non-linear...same goes for adding stray capacitance and what have you.
As you state it, BR replies besides the point at hand but nonetheless I still disagree with him if he claims that that resistor is not in the signal path...let that be a disagreement between him and myself.
Those 100R resistors tie plate and suppressor grid together,making it part of the plate, hence the "triode mode", if you or anyone else think the plate carries no signal (AC) than I let me know what does.
Rest assured, I like those designs too...
But I can always find ways on how to improve them...
The design published in GA back in 1990 was in some ways more appealing to me and in some respects better and more developped.
I hold some copies of the articles regarding this design and some others.
It may sound arrogant from my part but I don't buy a book that's not going to teach me anything.
There are others out there that are far more interesting to me at least.
His GA article OTL passed it (admittedly that was my 80W version of it) , so does my own OTL.
And I tested it with real music, not just some sine waves.
If you like to look at the GA article regarding that OTL and the one offered in the book you will see where corners have been cut.
BTW, I am still wondering what exactly was patented...but that's another issue.
Now, if you think you can teach me something than please go ahead, somehow however I feel I have a 25 year worth headstart.
Cheers,
Hi,
Still ticked off over nothing, are we?
Let's look at some facts here:
Ask yourself what would be the better ( not necessarily the cheapest) component in that position?
A non-inductive resistor, right?After all in audio we try to avoid non-linear components, not?
Ask my advise and I invariably will give you the best component I know of unless you specifically ask me what the most economical component would be.
It is not good practice IMHO to add inductive components that are invariably non-linear...same goes for adding stray capacitance and what have you.
As you state it, BR replies besides the point at hand but nonetheless I still disagree with him if he claims that that resistor is not in the signal path...let that be a disagreement between him and myself.
Those 100R resistors tie plate and suppressor grid together,making it part of the plate, hence the "triode mode", if you or anyone else think the plate carries no signal (AC) than I let me know what does.
Rest assured, I like those designs too...
But I can always find ways on how to improve them...
The design published in GA back in 1990 was in some ways more appealing to me and in some respects better and more developped.
I hold some copies of the articles regarding this design and some others.
It may sound arrogant from my part but I don't buy a book that's not going to teach me anything.
There are others out there that are far more interesting to me at least.
His GA article OTL passed it (admittedly that was my 80W version of it) , so does my own OTL.
And I tested it with real music, not just some sine waves.
If you like to look at the GA article regarding that OTL and the one offered in the book you will see where corners have been cut.
BTW, I am still wondering what exactly was patented...but that's another issue.
Now, if you think you can teach me something than please go ahead, somehow however I feel I have a 25 year worth headstart.
Cheers,
My my, where to start.
Ticked off? Not at all. As I said earlier, before I'd go off replacing 48 resistors, which had been spec'd by the designer, I would have done the research and come to my own conlcusion.
Only if it makes a real difference. Either in a DBLT or on the measurements. If a component doesn't make a difference, and it costs more, then in this case it is not "better" at all. We're not building radars here. So in this case, I've asked twice, and now B.R. has asked as well, what is the inductance of those resistors, and further, how would it affect the circuit?
Well I'm sure Bruce could explain his postition to me, can you explain yours?
No one said the plate doesn't carry audio, what he said was the 100ohm resistor carried DC only. I had made the same assumption that you have,(that the resistor makes the screen part of the plate), but based on Bruces reply, there's something I missed, I'm going to try figure out what that was.
Well don't keep us in suspense I'm sure that if you have some useful improvements, Bruce would most certainly appreciate your input, I know I would. I'm serious here, what do you propose? I am assuming topology changes, as component changes are within pretty much anyone's ability assuming they can read a spec sheet, and unless they make a difference at the speaker terminals, or add to reliablility or such they are not improvements at all.
As for the GA vs AR version of the OTL, I don't see any cut corners at all, perhaps you can point them out. Since you photocopied parts of the book, there was obviously something you could learn, therefore your statment not only sounds arrogant, it is, in fact, arrogant. Furthermore, if you can't see what the patent was about, (hint: 3 zener diodes) then either you didn't understand what you read, or, you didn't read it. Me, I've read it 7 or 8 times now, and I think I'm finally beginning to get it. I, however am apparently 25 years behind, so I have an excuse.
Square waves actually, and I did dig it up, it was a 2uf cap in parallel with an 8ohm resistor. Tougher than music I'd say.
I never said I had something to teach you. I do, however think you've got lot's to learn.
Chris
Ticked off? Not at all. As I said earlier, before I'd go off replacing 48 resistors, which had been spec'd by the designer, I would have done the research and come to my own conlcusion.
Only if it makes a real difference. Either in a DBLT or on the measurements. If a component doesn't make a difference, and it costs more, then in this case it is not "better" at all. We're not building radars here. So in this case, I've asked twice, and now B.R. has asked as well, what is the inductance of those resistors, and further, how would it affect the circuit?
Well I'm sure Bruce could explain his postition to me, can you explain yours?
No one said the plate doesn't carry audio, what he said was the 100ohm resistor carried DC only. I had made the same assumption that you have,(that the resistor makes the screen part of the plate), but based on Bruces reply, there's something I missed, I'm going to try figure out what that was.
Well don't keep us in suspense I'm sure that if you have some useful improvements, Bruce would most certainly appreciate your input, I know I would. I'm serious here, what do you propose? I am assuming topology changes, as component changes are within pretty much anyone's ability assuming they can read a spec sheet, and unless they make a difference at the speaker terminals, or add to reliablility or such they are not improvements at all.
As for the GA vs AR version of the OTL, I don't see any cut corners at all, perhaps you can point them out. Since you photocopied parts of the book, there was obviously something you could learn, therefore your statment not only sounds arrogant, it is, in fact, arrogant. Furthermore, if you can't see what the patent was about, (hint: 3 zener diodes) then either you didn't understand what you read, or, you didn't read it. Me, I've read it 7 or 8 times now, and I think I'm finally beginning to get it. I, however am apparently 25 years behind, so I have an excuse.
Square waves actually, and I did dig it up, it was a 2uf cap in parallel with an 8ohm resistor. Tougher than music I'd say.
I never said I had something to teach you. I do, however think you've got lot's to learn.
Chris
Well the best sounding,
I dropped four different resistors on the floor, and I liked the sound the wire wound made the best.
Seriously, I'm not sure if you've been following the "Blind Listening Test" thread, but this little discussion has spilled over from that one. I am one who maintains that in most cases, one resistor actually sounds the same as another. (ie when you no longer know which resistor you are listening to ,you won't be able to tell a difference) although I'm open to any evidence to the contrary.
I have been doing a little research, "little" being the operative word here, and so far I have to agree with you and Frank on the other point though. It still seems to me that that resistor would be carrying audio. According to what I've read, about 1/10th of what the plate carries. Still, B.R. must have solid grounds for saying so, so my research continues.
In any event, even if that resistor was to be attenuating signal in a non-linear manner,(which in this application I don't beleive it is) as 10 times the current is going through the plate, the 'sound' should not be altered. Thus, the best looking resistor would be just as valid a choice as the best sounding.
Chris
I dropped four different resistors on the floor, and I liked the sound the wire wound made the best.
Seriously, I'm not sure if you've been following the "Blind Listening Test" thread, but this little discussion has spilled over from that one. I am one who maintains that in most cases, one resistor actually sounds the same as another. (ie when you no longer know which resistor you are listening to ,you won't be able to tell a difference) although I'm open to any evidence to the contrary.
I have been doing a little research, "little" being the operative word here, and so far I have to agree with you and Frank on the other point though. It still seems to me that that resistor would be carrying audio. According to what I've read, about 1/10th of what the plate carries. Still, B.R. must have solid grounds for saying so, so my research continues.
In any event, even if that resistor was to be attenuating signal in a non-linear manner,(which in this application I don't beleive it is) as 10 times the current is going through the plate, the 'sound' should not be altered. Thus, the best looking resistor would be just as valid a choice as the best sounding.
Chris
To The Shredder...
Hi,
Obvoiusly so.
You're being stubborn or downright stupid?
Go educate yourself before jumping to conclusions.
Learn how to read replies, it may help.
Sure, his is economical, mine isn't.
Well, yes you did.And in the same sentence you contradict yourself a few times already.
It would really come as a surprise if Mr.Rosenblit said any such thing for it it is absolutely incorrect.
Really a novice to all things electronic or what?
The inductance of a 5W 100R will vary from one manufacturer to another.
More importantly, it was your mistake to assume all WW resistors had to be replaced in the first place.
How on earth can you jump to such conclusion from my reply...mind you your conclusion IS correct inductive components do not have a place in audio...unless you can put them to your advantage.
I can, but I'm not inclined to given your "friendly" attitude.
See above.
The only thing I learned from it is that in the US you can patent the way how you tie your shoelaces and that you don't have a clue what the patent is about.
Zener diodes my *ss.
Great, now you really got the message...ever looked at the impedance correction network at the output?
Need more hints?
No, you don't for you're smart and educated an rely on everything Mr.Rosenblit puts on paper, right?
Sorry, but you got yourself a very KIND reply both technically and personally.
You figured so many things wrong, hard to believe really...and I'm the bad guy?
Cheers,
P.S. And I always had the impression Canadians were open minded, hospitable people...
Sorry. cheap shot.
Hi,
Obvoiusly so.
You're being stubborn or downright stupid?
Go educate yourself before jumping to conclusions.
Learn how to read replies, it may help.
Sure, his is economical, mine isn't.
Well, yes you did.And in the same sentence you contradict yourself a few times already.
It would really come as a surprise if Mr.Rosenblit said any such thing for it it is absolutely incorrect.
Really a novice to all things electronic or what?
The inductance of a 5W 100R will vary from one manufacturer to another.
More importantly, it was your mistake to assume all WW resistors had to be replaced in the first place.
How on earth can you jump to such conclusion from my reply...mind you your conclusion IS correct inductive components do not have a place in audio...unless you can put them to your advantage.
I can, but I'm not inclined to given your "friendly" attitude.
See above.
The only thing I learned from it is that in the US you can patent the way how you tie your shoelaces and that you don't have a clue what the patent is about.
Zener diodes my *ss.
Great, now you really got the message...ever looked at the impedance correction network at the output?
Need more hints?
No, you don't for you're smart and educated an rely on everything Mr.Rosenblit puts on paper, right?
Sorry, but you got yourself a very KIND reply both technically and personally.
You figured so many things wrong, hard to believe really...and I'm the bad guy?
Cheers,
P.S. And I always had the impression Canadians were open minded, hospitable people...
Sorry. cheap shot.
ND I'M STILL P*SSED.
Hi,
Well,well we do come around?
It's not just one resistor but more than half a dozen of'em.
I grant you this much, I want things to be as perfect as I can get.
My audiobandwidth runs well into the hundreds of kiloherz...
Ever asked yourself why?
And, I'll be the last to accept unnecessary commercial compromise.
You know, it's all about roll off and there's little point in having an OTL amp if you don't take that brownie along...
Cheers,
Hi,
Well,well we do come around?
It's not just one resistor but more than half a dozen of'em.
I grant you this much, I want things to be as perfect as I can get.
My audiobandwidth runs well into the hundreds of kiloherz...
Ever asked yourself why?
And, I'll be the last to accept unnecessary commercial compromise.
You know, it's all about roll off and there's little point in having an OTL amp if you don't take that brownie along...
Cheers,
Resistor, screen, plate
Chris, I hope you don't mind as I am lazy.
I respect your opinion about resistors sounding the same, but If there was a sonic difference, that sonic difference will show up in full force on the plate.
The screen grid has great, nearly total control over the plate current, and how the screen goes (sonically), so goes the sonics at the plate.
Chris, I hope you don't mind as I am lazy.
I respect your opinion about resistors sounding the same, but If there was a sonic difference, that sonic difference will show up in full force on the plate.
The screen grid has great, nearly total control over the plate current, and how the screen goes (sonically), so goes the sonics at the plate.
Ok let's see...
Hi Steve,
According to what I've read, there is about 10% of the overall current going through the screen grid in this configuration. Are you saying that connected in triode mode as the screen goes so goes the plate? Or were you speaking in pentode mode? I must admit I still can't figure out why Bruce would say that it is only DC going through this resistor, as I see no mechanism for this to be true, I may end up asking him.
However, in answer to the overall impact of inductive vs non-inductive I have learned the following. Anyone feel free to correct any assumptions I've made as this is pretty free form.
I had a really hard time finding any hard numbers, and I have a friend who has a good inductance meter who is going to do a measurement for me, but in the mean time...
I was able to find one data sheet that said the <5ohm values were <1uh. A quick calculation shows that at1uh and 1MHZ the reactance is ~6ohms. So that's 3ohms at 500Khz 1.5 at 250 and so on. So if you are really uptight about how linear your cathode resistors are at 250khz you do have a somewhat valid concern. Keep in mind this inductance in no way will limit the bandwidth, just change it's linearity a bit. Now from 5 to 50K(published BW of this design) these cathode resistors will change about .3 of an ohm. I would suppose that the natural variation in current throught tube would be greater than what the rise and fall of .3 ohms on the cathode would create. Further, since removing them altogether would lower the output impedence by about 10%, adding another .3 of an ohm changes that a max of .6%(in this case about 25 milliohms). Yes I know it's non-linear, but the important thing is what effect will this have on the performance of the amp? I can't see a non-linear swing in output impedence in the fractions of percent to be even remotely important when you consider how non-linear the load is in comparison. As for the 100ohm resistors, well given that it's 10% of the plate and even a 6 ohm swing is only 6%, numbers get pretty insignificant pretty fast.
More to come...
Chris
Hi Steve,
According to what I've read, there is about 10% of the overall current going through the screen grid in this configuration. Are you saying that connected in triode mode as the screen goes so goes the plate? Or were you speaking in pentode mode? I must admit I still can't figure out why Bruce would say that it is only DC going through this resistor, as I see no mechanism for this to be true, I may end up asking him.
However, in answer to the overall impact of inductive vs non-inductive I have learned the following. Anyone feel free to correct any assumptions I've made as this is pretty free form.
I had a really hard time finding any hard numbers, and I have a friend who has a good inductance meter who is going to do a measurement for me, but in the mean time...
I was able to find one data sheet that said the <5ohm values were <1uh. A quick calculation shows that at1uh and 1MHZ the reactance is ~6ohms. So that's 3ohms at 500Khz 1.5 at 250 and so on. So if you are really uptight about how linear your cathode resistors are at 250khz you do have a somewhat valid concern. Keep in mind this inductance in no way will limit the bandwidth, just change it's linearity a bit. Now from 5 to 50K(published BW of this design) these cathode resistors will change about .3 of an ohm. I would suppose that the natural variation in current throught tube would be greater than what the rise and fall of .3 ohms on the cathode would create. Further, since removing them altogether would lower the output impedence by about 10%, adding another .3 of an ohm changes that a max of .6%(in this case about 25 milliohms). Yes I know it's non-linear, but the important thing is what effect will this have on the performance of the amp? I can't see a non-linear swing in output impedence in the fractions of percent to be even remotely important when you consider how non-linear the load is in comparison. As for the 100ohm resistors, well given that it's 10% of the plate and even a 6 ohm swing is only 6%, numbers get pretty insignificant pretty fast.
More to come...
Chris
closed minded and stupid, but I try...
A bit of both I guess, I just figured you would know off the top of your head, and I'd have to do a tonne of work, and since you maintained it was a bad practice, you could support that with some basic engineering, I suppose that was stupid of me to assume. See the above post, I answered the question to the best of my ability, feel free to correct the math or the assumptions or point out anything I may not have taken into account. I'm here to learn.
Sorry, I wasn't clear in my question, what I meant was if Bruce says only DC and you say AC then I suspect he feels he can support that, or are you saying that he says it only carries DC for economic reasons?
Well, yes, well not quite, but it's already been established that I'm not strong in ciruit theory, but I'm learning, and I do have a fairly good capacity for understanding when someone explains something new to me. And I did say roughly and typically(when I first asked on the other thread that this didn't spill over from) Now one would assume that if you figure the screen resistors should be non-inductive, then for god sake man, get them off the cathodes! And that's all of 'em in this case. Vary from one manufacturer to another? Really, that's not much help. You could have done a worse case scenario, at least it's a place to start.
Ok, now that's funny. Just so we're all clear on this, you don't think the zeners have anything to do with what makes this OTL special (or at least on par with a new shoelace tie) BTW it is a rather common misconception that anything can be patented. But that's another discussion.
No, but I have asked myself how.
If Bruce publishes the BW to be 5 to 50Khz, you must have a few pretty nifty tweaks up your sleeve.
Rolloff? I've put my best effort out there for all to review/critique, how do you suppose the choice of these resistors affects the rolloff? If you are willing to explain it I'm willing to learn.
and finally.....
All things in moderation I suppose, but as for taking all my info from Bruce? Hardly. So far, he is one of the few I've read in this buisness that talks any sense at all. I have to admit, that the DC comment really caught me by surprise, and so far all the research and critical thinking I've done certainly doesn't support it. One of the reasons I'm so often willing to take him at face value, is I've wondered about all kinds of little things that seem to make no sense (cable direction, cable choice, biwiring and on and on) and he comes along and say he thinks it's dung and then goes on to support it with some pretty solid arguments. That just immediatley endears a guy to me. OTOH when someone just comes along and says 'no that's wrong' and then backs it up by saying, 'and if you weren't so stupid and uneducated you wouldn't be saying such a thing' well I tend to be suspicious of anything he has to say.
Compensation network? Are you saying he did something with the AR version that makes it less stable than the GA version(into inductive loads), or there was a way to stabilize the amp without using the network but he couldn't be bothered? I'm sorry but figuring that out for myself would take about a week, and even then I'm not sure I'd be right. BTW there's something I've been just dying to ask, do you biwire?
Oh yeah and one thing I did learn while I was trying to get inductance numbers for the WWs. These things kick ***! They are rugged, have the best temperature coefficient, last forever, take more heat, drift over time is virtually non-existant. Pretty good stuff for a piece of gear yes?
So far, the WW looks good to me.
Chris
A bit of both I guess, I just figured you would know off the top of your head, and I'd have to do a tonne of work, and since you maintained it was a bad practice, you could support that with some basic engineering, I suppose that was stupid of me to assume. See the above post, I answered the question to the best of my ability, feel free to correct the math or the assumptions or point out anything I may not have taken into account. I'm here to learn.
Sorry, I wasn't clear in my question, what I meant was if Bruce says only DC and you say AC then I suspect he feels he can support that, or are you saying that he says it only carries DC for economic reasons?
Well, yes, well not quite, but it's already been established that I'm not strong in ciruit theory, but I'm learning, and I do have a fairly good capacity for understanding when someone explains something new to me. And I did say roughly and typically(when I first asked on the other thread that this didn't spill over from) Now one would assume that if you figure the screen resistors should be non-inductive, then for god sake man, get them off the cathodes! And that's all of 'em in this case. Vary from one manufacturer to another? Really, that's not much help. You could have done a worse case scenario, at least it's a place to start.
Ok, now that's funny. Just so we're all clear on this, you don't think the zeners have anything to do with what makes this OTL special (or at least on par with a new shoelace tie) BTW it is a rather common misconception that anything can be patented. But that's another discussion.
No, but I have asked myself how.
If Bruce publishes the BW to be 5 to 50Khz, you must have a few pretty nifty tweaks up your sleeve.
Rolloff? I've put my best effort out there for all to review/critique, how do you suppose the choice of these resistors affects the rolloff? If you are willing to explain it I'm willing to learn.
and finally.....
All things in moderation I suppose, but as for taking all my info from Bruce? Hardly. So far, he is one of the few I've read in this buisness that talks any sense at all. I have to admit, that the DC comment really caught me by surprise, and so far all the research and critical thinking I've done certainly doesn't support it. One of the reasons I'm so often willing to take him at face value, is I've wondered about all kinds of little things that seem to make no sense (cable direction, cable choice, biwiring and on and on) and he comes along and say he thinks it's dung and then goes on to support it with some pretty solid arguments. That just immediatley endears a guy to me. OTOH when someone just comes along and says 'no that's wrong' and then backs it up by saying, 'and if you weren't so stupid and uneducated you wouldn't be saying such a thing' well I tend to be suspicious of anything he has to say.
Compensation network? Are you saying he did something with the AR version that makes it less stable than the GA version(into inductive loads), or there was a way to stabilize the amp without using the network but he couldn't be bothered? I'm sorry but figuring that out for myself would take about a week, and even then I'm not sure I'd be right. BTW there's something I've been just dying to ask, do you biwire?
Oh yeah and one thing I did learn while I was trying to get inductance numbers for the WWs. These things kick ***! They are rugged, have the best temperature coefficient, last forever, take more heat, drift over time is virtually non-existant. Pretty good stuff for a piece of gear yes?
So far, the WW looks good to me.
Chris
OKIDOKI.
Hi,
Not that many people here understand OTL amps which is probably why you mainly depend on me and quite likely SY...
Sorry if I left someone out.
'Course not and it ain't that simple either...basically you need to avoid stray inductance for it creates resonant circuits.
Whether these resonate within or outside "your" audioband or not is irrelevant...thing is they resonate.
Just keep in mind it's an additive effect, the more the worse it gets.
That resistor carries both AC and DC, where you derive that 10% figure from is a mystery to me.
Economics only come into play when the type of resistor is concerned, after all when we consider 5W'ers a run of the mill inductive one is cheaper.
Oh, and there are no cathode resistors in the output stage...they are only there for measuring purposes and to protect the valves from back EMF...
Again, using inductive WWs is going to falsify measuements to some extend.
Fine, I have no problem with that...just keep in mind that there's is nothing mysterious about OTLs other than you always have an innate imbalance between the anode and cathode followers...
Valves just don't come in complementary pairs, you see?
Other than that, a lot of semiconductor amps rely on the same basic circuitry.
They don't have anything to do with the patent, they are likely to degrade the sound ...hence my remark that the GA circuit has much more potential.
And I can only confirm that the latter sounds good, has TPs built in and is a good basis for you to understand how an OTL works.
Not to mention the fact that zeners are noisy and to add insult injury the whole shebang is in the feedback loop...
It's not only that handful of resistors, it's the entire design that contributes to it...
50kHz is not bad but you can easily have an OTL running linearly into the almost 1/2 Mhz range...helps phase linearity alot.
Seems you're vulnerable to the easy way out kind of explanations for most of that makes sense to me...and I do involve in comparitive listening...
Those topics are really very old hat to me anyway.
Yes and no, it's an impedance compensation network and no it won't necassrilly compromise the amps stability...it will, however make the amp change response with different loads.
And, if I may say,make your tests seem odd if you're unaware of its' presence.
As a matter of fact I use tri-wire....what about it?
Yep, nothing against them here...if it weren't for their...well, you guessed it...
May you rock for ever,
Hi,
Not that many people here understand OTL amps which is probably why you mainly depend on me and quite likely SY...
Sorry if I left someone out.
'Course not and it ain't that simple either...basically you need to avoid stray inductance for it creates resonant circuits.
Whether these resonate within or outside "your" audioband or not is irrelevant...thing is they resonate.
Just keep in mind it's an additive effect, the more the worse it gets.
That resistor carries both AC and DC, where you derive that 10% figure from is a mystery to me.
Economics only come into play when the type of resistor is concerned, after all when we consider 5W'ers a run of the mill inductive one is cheaper.
Oh, and there are no cathode resistors in the output stage...they are only there for measuring purposes and to protect the valves from back EMF...
Again, using inductive WWs is going to falsify measuements to some extend.
Fine, I have no problem with that...just keep in mind that there's is nothing mysterious about OTLs other than you always have an innate imbalance between the anode and cathode followers...
Valves just don't come in complementary pairs, you see?
Other than that, a lot of semiconductor amps rely on the same basic circuitry.
They don't have anything to do with the patent, they are likely to degrade the sound ...hence my remark that the GA circuit has much more potential.
And I can only confirm that the latter sounds good, has TPs built in and is a good basis for you to understand how an OTL works.
Not to mention the fact that zeners are noisy and to add insult injury the whole shebang is in the feedback loop...
It's not only that handful of resistors, it's the entire design that contributes to it...
50kHz is not bad but you can easily have an OTL running linearly into the almost 1/2 Mhz range...helps phase linearity alot.
Seems you're vulnerable to the easy way out kind of explanations for most of that makes sense to me...and I do involve in comparitive listening...
Those topics are really very old hat to me anyway.
Yes and no, it's an impedance compensation network and no it won't necassrilly compromise the amps stability...it will, however make the amp change response with different loads.
And, if I may say,make your tests seem odd if you're unaware of its' presence.
As a matter of fact I use tri-wire....what about it?
Yep, nothing against them here...if it weren't for their...well, you guessed it...
May you rock for ever,
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.