Clearly they have not made this design 100% stable as per below.
AudioFreak Paralleling multiple output devices is one thing that we have yet to completely iron all the bugs out of in this design, it's mostly a matter of increasing the drive current and decreasing the front end impedance but this amp also seems to show stability issues under some situations so we have approached it on a case by case basis thus far.
skaara I know about capacitance, and matching, but I read somewhere that there is a problem with parlelling mosfets in aleph x, but I guess it wasnt true.
I was just adding some valid input, into this design. I am very experienced in amplifier design and I am very interested in helping...
Anthony Holton
AudioFreak Paralleling multiple output devices is one thing that we have yet to completely iron all the bugs out of in this design, it's mostly a matter of increasing the drive current and decreasing the front end impedance but this amp also seems to show stability issues under some situations so we have approached it on a case by case basis thus far.
skaara I know about capacitance, and matching, but I read somewhere that there is a problem with parlelling mosfets in aleph x, but I guess it wasnt true.
I was just adding some valid input, into this design. I am very experienced in amplifier design and I am very interested in helping...
Anthony Holton
Perhaps I'm jumping in a bit late, but I haven't encountered
stability problems with paralleling devices in the Aleph X.
It is perfectly possible to create X circuits which are not
stable, just like anything else, and if you start building
4 or 5 gain stage versions, you will encounter problems
like this. So far, I have maxed out at 3 stages on an X
circuit, and don't see any reason for more. The beauty
and the point of the X circuit is to get high performance
from simple circuits. It doesn't bring much to the table
with high feedback designs.
Paralleling Mosfets is not much of a problem up to a point.
The capacitance we have to drive is mostly Gate to Source,
and as you parallel devices, this goes up proportionately,
but the required drive voltage to fill that capacitance goes
down inversely, largely cancelling out.
As the number of devices in parallel gets large, the Gate
to Drain capacitance starts to get large, and so you either
increase the drive current or start thinking about cascoding.
The Source resistors are calculated as follows: The Aleph
current source is designed to operate the Mosfet so that
at idle the voltage across the Source resistors is anywhere
from about .33V to .66V, although you can extend that
range if you care to. The current value for the each
Mosfet of the current source then becomes I = V/R, and
the idle dissipation becomes either I^2 * R or V^ / R.
Example: .66 volts, .33 ohms. I = 2 A, resistor dissipation
= 1.32 watts.
Use a resistor about 3 times this value, and you'll be fine.
Use parallel resistors to get values and dissipation figures
you might need if you don't have what you want on hand.
stability problems with paralleling devices in the Aleph X.
It is perfectly possible to create X circuits which are not
stable, just like anything else, and if you start building
4 or 5 gain stage versions, you will encounter problems
like this. So far, I have maxed out at 3 stages on an X
circuit, and don't see any reason for more. The beauty
and the point of the X circuit is to get high performance
from simple circuits. It doesn't bring much to the table
with high feedback designs.
Paralleling Mosfets is not much of a problem up to a point.
The capacitance we have to drive is mostly Gate to Source,
and as you parallel devices, this goes up proportionately,
but the required drive voltage to fill that capacitance goes
down inversely, largely cancelling out.
As the number of devices in parallel gets large, the Gate
to Drain capacitance starts to get large, and so you either
increase the drive current or start thinking about cascoding.
The Source resistors are calculated as follows: The Aleph
current source is designed to operate the Mosfet so that
at idle the voltage across the Source resistors is anywhere
from about .33V to .66V, although you can extend that
range if you care to. The current value for the each
Mosfet of the current source then becomes I = V/R, and
the idle dissipation becomes either I^2 * R or V^ / R.
Example: .66 volts, .33 ohms. I = 2 A, resistor dissipation
= 1.32 watts.
Use a resistor about 3 times this value, and you'll be fine.
Use parallel resistors to get values and dissipation figures
you might need if you don't have what you want on hand.
Nelson Pass
Thanks, that was a badly needed post. If the voltage range across these resistors is a maximum 0.66V, what determines this voltage and why do we need to triple the power value? I have a box of 1% 0.5ohm 2W resistors that I would love to use with mosfets drawing 1A each but they won't cut it in this circuit if I have to triple the power tolerance.
Thanks, that was a badly needed post. If the voltage range across these resistors is a maximum 0.66V, what determines this voltage and why do we need to triple the power value? I have a box of 1% 0.5ohm 2W resistors that I would love to use with mosfets drawing 1A each but they won't cut it in this circuit if I have to triple the power tolerance.
The voltage across the Source resistor will be a function
of the PN junction of the NPN transistor on the current source,
which is about .66 V. By injecting +/- current into the base
junction, you can raise or lower the value appearing across
the Source resistor.
The 3 X margin comes from the fact that at full output, the
average power across the resistor will be twice that of idle.
of the PN junction of the NPN transistor on the current source,
which is about .66 V. By injecting +/- current into the base
junction, you can raise or lower the value appearing across
the Source resistor.
The 3 X margin comes from the fact that at full output, the
average power across the resistor will be twice that of idle.
yldouright,
If you had correctly read this thread thru, you would know that Grey had declared his circuit stable and free from RF but noted that if you were to change output devices or add more or you live near a radio station, you may very well run into problems. Grey's circuit is also free from the usual NPN compensation cap. Adding this would help stability issues some what. But needless to say that some of the simulators out there have shown twitchy behaviour to say the least.
As to my continually posting incorrect information, that is simply not true and many here on the board would agree. Sure I make mistakes, some of them blatantly obvious but I'm not afraid to admit a mistake when I'm caught out (or even when i'm not). Given the amount of time I spend here, I say the occasional typographical error should be forgiven as was the case 2 days ago in the back to back regulator thread when i accidently reversed the resistor numbers.
I would very much appreciate it in future that if you have a problem with me, please take it up in private via email (see that little dinky email button just there ....
) instead of making accusations that you can not prove.
Finally, for what it's worth, my background is in computer hardware / programming / electrical / electronics / engineering.
Now how about you apologise for that outburst and we can move on.
If you had correctly read this thread thru, you would know that Grey had declared his circuit stable and free from RF but noted that if you were to change output devices or add more or you live near a radio station, you may very well run into problems. Grey's circuit is also free from the usual NPN compensation cap. Adding this would help stability issues some what. But needless to say that some of the simulators out there have shown twitchy behaviour to say the least.
As to my continually posting incorrect information, that is simply not true and many here on the board would agree. Sure I make mistakes, some of them blatantly obvious but I'm not afraid to admit a mistake when I'm caught out (or even when i'm not). Given the amount of time I spend here, I say the occasional typographical error should be forgiven as was the case 2 days ago in the back to back regulator thread when i accidently reversed the resistor numbers.
I would very much appreciate it in future that if you have a problem with me, please take it up in private via email (see that little dinky email button just there ....
) instead of making accusations that you can not prove.Finally, for what it's worth, my background is in computer hardware / programming / electrical / electronics / engineering.
Now how about you apologise for that outburst and we can move on.
yldouright,
I suggest you cease this continuing line of insults and sarcasm immediately or you will not like the consequences. I am also still waiting for that apology. While you are at it, I suggest you apologise to the entire community here because judging by the response we have gotten (via private email), there are many people in the community that are not very happy with you or your comments right about now.
Just to dispell any doubt about this being a threat, no this is not a threat; it is a promise.
I suggest you cease this continuing line of insults and sarcasm immediately or you will not like the consequences. I am also still waiting for that apology. While you are at it, I suggest you apologise to the entire community here because judging by the response we have gotten (via private email), there are many people in the community that are not very happy with you or your comments right about now.
Just to dispell any doubt about this being a threat, no this is not a threat; it is a promise.
What the heck happened here?
I find it odd that there is so little going on with the board these days... when did it up and die? HarryHaller is one of the few left that is adding anything worth reading these days! Where did everyone else go?
That being said, let me apologize on behalf of the US for yldouright... his attitude is typical New Yorker. I find rehashing old points of this discussion with a know-it-all newbie to the board to be tiresome, so I have decided not to participate. It has already been stated more than once that the design in practice is stable, as originally built by Grey with one pair per side of output devices. That being said, other people have found ways to make it closer to being unconditionally stable with that setup. When we start throwing more devices on the output, there is a greater chance for instability (although Nelson, who obviously knows best, dosen't seem to think it's a problem, at least with his version of the design. But we don't know how close Grey's version is to his design). I have found that the low rail voltages seem to be a problem with the input capacitance of the output stage- not so much with stability, but with generating a lot of 2nd harmonic. I have decided to use as few devices on the output as possible in my version, and go to higher rail voltages when I have the money to build a super-power version, because of this.
It really sucks that we have to resort to so much moderation around here... it clearly has killed off the board. Either that, or the people who are the reason for the moderators have killed off the board. Either way, it's sad to see the result.
SteveG
I find it odd that there is so little going on with the board these days... when did it up and die? HarryHaller is one of the few left that is adding anything worth reading these days! Where did everyone else go?
That being said, let me apologize on behalf of the US for yldouright... his attitude is typical New Yorker. I find rehashing old points of this discussion with a know-it-all newbie to the board to be tiresome, so I have decided not to participate. It has already been stated more than once that the design in practice is stable, as originally built by Grey with one pair per side of output devices. That being said, other people have found ways to make it closer to being unconditionally stable with that setup. When we start throwing more devices on the output, there is a greater chance for instability (although Nelson, who obviously knows best, dosen't seem to think it's a problem, at least with his version of the design. But we don't know how close Grey's version is to his design). I have found that the low rail voltages seem to be a problem with the input capacitance of the output stage- not so much with stability, but with generating a lot of 2nd harmonic. I have decided to use as few devices on the output as possible in my version, and go to higher rail voltages when I have the money to build a super-power version, because of this.
It really sucks that we have to resort to so much moderation around here... it clearly has killed off the board. Either that, or the people who are the reason for the moderators have killed off the board. Either way, it's sad to see the result.
SteveG
AudioFreak
I have been made aware that as a moderator you have been given the power to temporarily suspend my activities on this forum. You have rightly admitted that you have made errors. I have witnessed them and made The Saint aware of the potential unreliability of your information. I hardly think a gentle jibe pointing out errors should warrant such discipline but perhaps you wish to create an environment where you can get away with posting mistakes or erroneous information and not be brought to task about it. Many governments operate this way but their constituents don't like it much and many when given an alternative will take it up immediately. It has been my experience that the pathetic and cowardly are the most ready to use any power granted to them in order to compensate for the inadequecies they know themselves to have. In short, if I am to be made an example let it be to show how inept you are at your given responsibilty because an apology is not forthcoming. Perhaps it will highlight what kind of moderator you really are and prompt the appropriate actions against you.
Steve G
dude, I'm sorry you find my posts tiresome but if I don't speak up about this who will. We need a moderator who isn't so quick on the trigger.
I have been made aware that as a moderator you have been given the power to temporarily suspend my activities on this forum. You have rightly admitted that you have made errors. I have witnessed them and made The Saint aware of the potential unreliability of your information. I hardly think a gentle jibe pointing out errors should warrant such discipline but perhaps you wish to create an environment where you can get away with posting mistakes or erroneous information and not be brought to task about it. Many governments operate this way but their constituents don't like it much and many when given an alternative will take it up immediately. It has been my experience that the pathetic and cowardly are the most ready to use any power granted to them in order to compensate for the inadequecies they know themselves to have. In short, if I am to be made an example let it be to show how inept you are at your given responsibilty because an apology is not forthcoming. Perhaps it will highlight what kind of moderator you really are and prompt the appropriate actions against you.
Steve G
dude, I'm sorry you find my posts tiresome but if I don't speak up about this who will. We need a moderator who isn't so quick on the trigger.
Why are you making this such a slinging match!
This Audio forum is ment to be a friendly place where we can exchange ideas and have fun, but not to make fun of each or sling insults.
I am appalled at where this thread has gone. Let it be said that I only joined this forum to help others here. I am an expereanced technician in Audio and in computing/electronics. I hope in the future that this lack of grace and understanding will not be displayed again.
Otherwise we should all just get our bat and ball and go home!
Anthony Holton
This Audio forum is ment to be a friendly place where we can exchange ideas and have fun, but not to make fun of each or sling insults.
I am appalled at where this thread has gone. Let it be said that I only joined this forum to help others here. I am an expereanced technician in Audio and in computing/electronics. I hope in the future that this lack of grace and understanding will not be displayed again.
Otherwise we should all just get our bat and ball and go home!
Anthony Holton