Good points dude...
As far as shunt reg's are concerned, play around with the math... convert or burn.
Conduction angle modulation was about the only game 'round for awhile when it came to big (I mean huge) variable power supplies and such. Not sure about this... but I believe in the early days big SCR's were easier to build...certainly easier to "drive"
PFC will remove line variations... but so will "current-mode" control at a lower cost. Interesting you bring it up though because we will be seeing more of it. Harmonic line degradation is becoming a serious probem. There is an"other" reason for it as well...
As far the "other" reason switchmodes came into being... think about airplanes. The reason is not the weight... well, not the weight of the supply that is...
As far as shunt reg's are concerned, play around with the math... convert or burn.
Conduction angle modulation was about the only game 'round for awhile when it came to big (I mean huge) variable power supplies and such. Not sure about this... but I believe in the early days big SCR's were easier to build...certainly easier to "drive"
PFC will remove line variations... but so will "current-mode" control at a lower cost. Interesting you bring it up though because we will be seeing more of it. Harmonic line degradation is becoming a serious probem. There is an"other" reason for it as well...
As far the "other" reason switchmodes came into being... think about airplanes. The reason is not the weight... well, not the weight of the supply that is...
When I say "current-mode control" that's not engineering english, but rather a very specific name to describe a control and feedback strategy for SMPS.
Enough mysteries already... it was learned pretty early on that a bunch linear supplies drove generators nuts. You can't run 10, 100 Watt supplies on a 1 kW generator. 400 Hz helped... three phase helped...
Linears take big gulps of current for a brief time during the peak of the sine. The rms currents are simply higher than a casual glance at input power would suggest. The currents would over- load the generator, requiring a larger unit.
Now a PFC unit is really just a class-D amp running backwards... flip a schematic around if you don't believe me... use your imagination.
Aircraft latched on to PFC qucikly because for every pound of supply you cast off, you could jetison some generator and wire too.
Enough mysteries already... it was learned pretty early on that a bunch linear supplies drove generators nuts. You can't run 10, 100 Watt supplies on a 1 kW generator. 400 Hz helped... three phase helped...
Linears take big gulps of current for a brief time during the peak of the sine. The rms currents are simply higher than a casual glance at input power would suggest. The currents would over- load the generator, requiring a larger unit.
Now a PFC unit is really just a class-D amp running backwards... flip a schematic around if you don't believe me... use your imagination.
Aircraft latched on to PFC qucikly because for every pound of supply you cast off, you could jetison some generator and wire too.
poobah said:When I say "current-mode control" that's not engineering english, but rather a very specific name to describe a control and feedback strategy for SMPS.
Sorry, I wasn't at all enticed into playing the old "let's test your background knowledge, here's a hint" jockohomo style guessing game. It's for suckers. I share what I share and that's about it.
I find your quoted comment somewhat condescending, thereby ending my participation in this discussion. I just don't have the patience today for where this seems to be headed.
Hey Chris, no ill intended... I only brought up "current mode control" as a clarification... as I thought, perhaps mistakenly, that we were headed toward apples and oranges.
The bit about generators and line harmonics might have been a teaser... but not a la homo. It is also anecdotal history, as I would guess your a bit younger than I, not something I would expect you to know. I wouldn't have gotten into it had you not shown an interest; and demonstrated an ability to make correct and insightful remarks in the first place.
The bit about generators and line harmonics might have been a teaser... but not a la homo. It is also anecdotal history, as I would guess your a bit younger than I, not something I would expect you to know. I wouldn't have gotten into it had you not shown an interest; and demonstrated an ability to make correct and insightful remarks in the first place.
Hi,
I can appreciate all that, thanks. Sorry if I took anything the wrong way.
My comments about current mode control were because it seems to me you'd need to regulate based on the point of load, or the output current.
I suppose you could also regulate based on constant ripple current, but then I'm uncertain as to whether or not that would complicate matters when you take load interractions into account.
So to me at this point of my understanding, it seems to make more sense to do away with the supply in such a case, and just regulate load current directly, as per the audio reference.
I'm honestly uncertain as to what people seem to dislike about this as an option.
I'd like to try it though.
I can appreciate all that, thanks. Sorry if I took anything the wrong way.
My comments about current mode control were because it seems to me you'd need to regulate based on the point of load, or the output current.
I suppose you could also regulate based on constant ripple current, but then I'm uncertain as to whether or not that would complicate matters when you take load interractions into account.
So to me at this point of my understanding, it seems to make more sense to do away with the supply in such a case, and just regulate load current directly, as per the audio reference.
I'm honestly uncertain as to what people seem to dislike about this as an option.
I'd like to try it though.
I have heard SMPS maligned for audio use simply because they had FEEDBACK... and in the words of Mr. Macki, feedback is bad, umkay. Suspicious at best because no mention was was made of real problems like: noise, radiation etc...
I will admit that most SMPS I build would probably not have the sheer energy storage or response time to deal with needs of an audio amp (of any class). Big boy caps would solve all that... but then, where's the advantage? Yes, an SMPS would provide efficient regulation... as you point out, most amps are somewhat oblivious to regulation anyway.
BTW, if you run into a plausible explanantion (or pure BS for that matter) for these "open loop" switchers... post it or PM. I'm curious I guess... wanna figure out if it is really something new or some audio "NO-FEEDBACK" dip.
Cheers -
I will admit that most SMPS I build would probably not have the sheer energy storage or response time to deal with needs of an audio amp (of any class). Big boy caps would solve all that... but then, where's the advantage? Yes, an SMPS would provide efficient regulation... as you point out, most amps are somewhat oblivious to regulation anyway.
BTW, if you run into a plausible explanantion (or pure BS for that matter) for these "open loop" switchers... post it or PM. I'm curious I guess... wanna figure out if it is really something new or some audio "NO-FEEDBACK" dip.
Cheers -
Saying that feedback is inherently bad for amplifiers or for SMPS is, IMHO, a very incorrect simplification usually coming from the old-school valve guys.
A well designed amplifier with feedback has lower THD than the same amplifier without feedback, lower output impedance, etc. Obviously, it must be designed with care to avoid problems such as inestability, etc, but there are good examples out there that show that it can be done without too many difficulties. In fact, 99.999% of amplifiers out there (both linear and switching ones) have feedback.
About feedback in SMPS, it is only added in order to have voltage regulation. A regulated linear PSU has feedback as well, in that sense. However, nobody has gone against that, curiously ;-) There seems to be a certain prejuice agains SMPS in the audio market. Again, a well designed SMPS can provide a more than suitable source of power for audio amplifiers, even high power ones, ruling the best linear PSUs out in terms of performance, weight, size and cost. That's our experience, anyway.
A well designed amplifier with feedback has lower THD than the same amplifier without feedback, lower output impedance, etc. Obviously, it must be designed with care to avoid problems such as inestability, etc, but there are good examples out there that show that it can be done without too many difficulties. In fact, 99.999% of amplifiers out there (both linear and switching ones) have feedback.
About feedback in SMPS, it is only added in order to have voltage regulation. A regulated linear PSU has feedback as well, in that sense. However, nobody has gone against that, curiously ;-) There seems to be a certain prejuice agains SMPS in the audio market. Again, a well designed SMPS can provide a more than suitable source of power for audio amplifiers, even high power ones, ruling the best linear PSUs out in terms of performance, weight, size and cost. That's our experience, anyway.
Hi,
I dont' at all question the feasiblity of them, I simply question what seems to be the common misconception as to them being hard to regulate, and for what reasons. So far I've seen instability mentioned, and some mention of it effect the sound adversely.
Sergio, how well regulated is your supply?
Regards,
Chris
I dont' at all question the feasiblity of them, I simply question what seems to be the common misconception as to them being hard to regulate, and for what reasons. So far I've seen instability mentioned, and some mention of it effect the sound adversely.
Sergio, how well regulated is your supply?
Regards,
Chris
Hi, Chris.
From the start we have tried to get the best of both worlds with our series of SMPS for audio. Our supply is regulated but has not too high a gain, so it keeps the output voltage quite stiff, but at the same time doesn't show any stability problems with any kind of load. We haven't skimped on capacitance, either. For example, our SPS80 PSU has 6600uF per rail plus a big amount of capacitance in the primary side, as well. We have seen other PSUs out there that claims 1KW power capabilityD ) with, say, 470uF at the input and 470uF/rail at the output. Just imagine the stiffness of that, or the level of feedback needed to keep voltage reasonably constant under load.
As compared to a 1KVA transformer + 18000uF/rail capacitance, SPS80 holds the voltage much better under heavy loading, bass, etc, and that is definitely very noticeable in the sound.
From the start we have tried to get the best of both worlds with our series of SMPS for audio. Our supply is regulated but has not too high a gain, so it keeps the output voltage quite stiff, but at the same time doesn't show any stability problems with any kind of load. We haven't skimped on capacitance, either. For example, our SPS80 PSU has 6600uF per rail plus a big amount of capacitance in the primary side, as well. We have seen other PSUs out there that claims 1KW power capability
D ) with, say, 470uF at the input and 470uF/rail at the output. Just imagine the stiffness of that, or the level of feedback needed to keep voltage reasonably constant under load.As compared to a 1KVA transformer + 18000uF/rail capacitance, SPS80 holds the voltage much better under heavy loading, bass, etc, and that is definitely very noticeable in the sound.
That makes sense.
Alot of these other supplies also claim such power, yet it is for for a quick pulse..... and after that, anything goes (usually the protection circuitry I think).
One thing I have come to hate about the unregulated raw DC supply I have, based on line variation it seems that (with my luck) when I go to show it off, nobody is home..
Anyway, am I ok to assume that with that level of capacitance on the output you need not concern yourself with balancing the rails in case of pumping?
Alot of these other supplies also claim such power, yet it is for for a quick pulse..... and after that, anything goes (usually the protection circuitry I think).
One thing I have come to hate about the unregulated raw DC supply I have, based on line variation it seems that (with my luck) when I go to show it off, nobody is home..
Anyway, am I ok to assume that with that level of capacitance on the output you need not concern yourself with balancing the rails in case of pumping?
Good point, Chris.
As you said, a typical unregulated linear PSU can show a 10% variation of output voltage, so you cannot usually go to the highest voltage allowed by the amplifier or you can fry it in high-line conditions. That doesn't happen in regulated PSUs.
About your other question:
We have used that PSU with 2 bridged BP4078 on a BIG PA subwoofer with a total impedance of 4 ohm (more than 1.2KW power), and observed no pumping effect.
With stereo 400+400W amplifiers no pumping either
As you said, a typical unregulated linear PSU can show a 10% variation of output voltage, so you cannot usually go to the highest voltage allowed by the amplifier or you can fry it in high-line conditions. That doesn't happen in regulated PSUs.
About your other question:
We have used that PSU with 2 bridged BP4078 on a BIG PA subwoofer with a total impedance of 4 ohm (more than 1.2KW power), and observed no pumping effect.
With stereo 400+400W amplifiers no pumping either
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.