Today's stupid question.
I know I can find a uber expensive programmable AC+DC capable power supplies. I just take my $4-6 grand in cash and throw it at a few companies and out pops a nice piece of lab gear. Not happening any time soon.
Not just any unit, either. Something in the ~+2-3 kilowatt range of output.
We know that high power class d sub amplifiers are around for peanuts in cost. even ones that can go up to about +1khz sine, and be about 1500 watts into a 1 ohm load, are available for under $200.
But it needs to be able to do DC at the same time, if required. I know I've got a (potentially) bridgeable but definitely DC capable power MOS-FET amplifier here (literally DC to about 100khz), but I'm not sure I want to stress it and take a chance on it going boom. It's a nice amp!
so, anyone aware of what might fit the bill? Stuffing a few boards is certainly not a challenge, and far preferable than the huge cash outlay. I've got many methods of getting power to such a circuit or board set, so that's not a big deal. I've got a few SMPS 80 amp/12vdc power supplies that can be paralleled, and so on. Meaning, modding a car amp that is suitable to the task at hand.
This is not my wheelhouse, for the most part. I can fake it like the bear dancing, but no tests at the end of the day....thanks...
Realistically, the back end can be a modded out power amplifier of suitable build and capacity, possibly with custom power supply. the front end can be a programmable signal generator. I'm looking or functionality at this time and I'm not too worried about losses.
ie, modded car amp, if one can actually do DC with enough power and efficiency and not catch on fire.....and then a set of SMPS based parallel 12vdc power supplies adjusted up to about 13 VDC (they have sense capacity for sharing, IIRC-- they are 1kw Dell server units)...and then a programmable siggie at the front, with a smattering of metering thrown in for good measure. IF class D amps can be made to do DC, and I'm sure it depends on the individual design..if so, then the whole thing is quite doable with other resources I have in hand. I'd just need to procedure the specific class D amp to beef up and mod out..
Lets just say I'm ignorant about one particular part of the question I'm asking. whether or not class D amplifiers can handle DC output as a norm or in exceptional cases with specific designs, or not..
Since it is about avoiding DC at all costs in audio amplification, one seldom sees any DC ability mentioned, as in never. As in no data, not even in the fine print. What idiot would want DC? (I raises my hand...)
But someone must have some understanding of this aspect ...and I don't want to spend two weeks or even two days.. digging through all the available data to find what I'm looking for. Which is a simple declaration, or clarity.... on what class D can do in the DC+AC department. I strongly suspect Class D is fine with DC but the audio package involved must be beefed up in some specific ways, if it is to be high power class D, and that any controller chips (front end) must also be DC capable and so on. Just a guess. Seems reasonable.
Any ideas?
I know I can find a uber expensive programmable AC+DC capable power supplies. I just take my $4-6 grand in cash and throw it at a few companies and out pops a nice piece of lab gear. Not happening any time soon.
Not just any unit, either. Something in the ~+2-3 kilowatt range of output.
We know that high power class d sub amplifiers are around for peanuts in cost. even ones that can go up to about +1khz sine, and be about 1500 watts into a 1 ohm load, are available for under $200.
But it needs to be able to do DC at the same time, if required. I know I've got a (potentially) bridgeable but definitely DC capable power MOS-FET amplifier here (literally DC to about 100khz), but I'm not sure I want to stress it and take a chance on it going boom. It's a nice amp!
so, anyone aware of what might fit the bill? Stuffing a few boards is certainly not a challenge, and far preferable than the huge cash outlay. I've got many methods of getting power to such a circuit or board set, so that's not a big deal. I've got a few SMPS 80 amp/12vdc power supplies that can be paralleled, and so on. Meaning, modding a car amp that is suitable to the task at hand.
This is not my wheelhouse, for the most part. I can fake it like the bear dancing, but no tests at the end of the day....thanks...
Realistically, the back end can be a modded out power amplifier of suitable build and capacity, possibly with custom power supply. the front end can be a programmable signal generator. I'm looking or functionality at this time and I'm not too worried about losses.
ie, modded car amp, if one can actually do DC with enough power and efficiency and not catch on fire.....and then a set of SMPS based parallel 12vdc power supplies adjusted up to about 13 VDC (they have sense capacity for sharing, IIRC-- they are 1kw Dell server units)...and then a programmable siggie at the front, with a smattering of metering thrown in for good measure. IF class D amps can be made to do DC, and I'm sure it depends on the individual design..if so, then the whole thing is quite doable with other resources I have in hand. I'd just need to procedure the specific class D amp to beef up and mod out..
Lets just say I'm ignorant about one particular part of the question I'm asking. whether or not class D amplifiers can handle DC output as a norm or in exceptional cases with specific designs, or not..
Since it is about avoiding DC at all costs in audio amplification, one seldom sees any DC ability mentioned, as in never. As in no data, not even in the fine print. What idiot would want DC? (I raises my hand...)
But someone must have some understanding of this aspect ...and I don't want to spend two weeks or even two days.. digging through all the available data to find what I'm looking for. Which is a simple declaration, or clarity.... on what class D can do in the DC+AC department. I strongly suspect Class D is fine with DC but the audio package involved must be beefed up in some specific ways, if it is to be high power class D, and that any controller chips (front end) must also be DC capable and so on. Just a guess. Seems reasonable.
Any ideas?
Last edited:
DC power amplifiers aka- PWM- class D-E-F etc take microcontrollers amplify them and control DC motors /Actuators /Servos/ Automobiles with very high current and are already for sale commercially from various US companies .
If I have picked you up incorrectly then please correct me ?
Is it a circuit diagram you are after ?
If I have picked you up incorrectly then please correct me ?
Is it a circuit diagram you are after ?
I'm looking for complex variable waveform generation, a signal which is somewhat akin to a audio signal in complexity. But with DC that is as fast to move as is the AC signal.
Hence the idea of car audio amps, for taking a cheap shot as a first test (first trial run).
What I don't know..is if, in general (depends on the design), class D amplifiers are amenable to DC, as much as they are amenable to AC. Coil saturation, due to the DC then the AC on top..etc. but only half the power as it is all on one side of the dual rail PS, and so on. trace limits, efficiency limits,continual loading, thermal build, impedance shift, all kinds of extra fun.
Do I just remove the blocking cap at the input (if there is no DC protection circuitry) and then feed them signal, within their limits (of various forms of DC and AC)? it's not spec you will generally find for let's say a IRS2092
Hence the idea of car audio amps, for taking a cheap shot as a first test (first trial run).
What I don't know..is if, in general (depends on the design), class D amplifiers are amenable to DC, as much as they are amenable to AC. Coil saturation, due to the DC then the AC on top..etc. but only half the power as it is all on one side of the dual rail PS, and so on. trace limits, efficiency limits,continual loading, thermal build, impedance shift, all kinds of extra fun.
Do I just remove the blocking cap at the input (if there is no DC protection circuitry) and then feed them signal, within their limits (of various forms of DC and AC)? it's not spec you will generally find for let's say a IRS2092
Last edited:
There's no reason why a class D amp would not handle DC.
However if you get a complete amp, it'll have DC blocking capacitors on the input and possibly a DC detector protection circuit to save the speakers in case one FET fails short. If these are implemented in analog circuits, that shouldn't be too difficult to hack. If it's software, though, especially a software highpass filter in a DSP, then, pretty much impossible.
So I'd say a simple cheap class D PA amp without DSP.
Car audio amps are quite complex because they have a big DC-DC converter inside to get high voltage from 12V. There's a lot that can go wrong.
However if you get a complete amp, it'll have DC blocking capacitors on the input and possibly a DC detector protection circuit to save the speakers in case one FET fails short. If these are implemented in analog circuits, that shouldn't be too difficult to hack. If it's software, though, especially a software highpass filter in a DSP, then, pretty much impossible.
So I'd say a simple cheap class D PA amp without DSP.
Car audio amps are quite complex because they have a big DC-DC converter inside to get high voltage from 12V. There's a lot that can go wrong.
Is this potentially what you are getting at---or not ?
AC+DC Programmable Power Sources up to 67kVA - N4L
AC+DC Programmable Power Sources up to 67kVA - N4L
As Doc Brown said, "It's a science experiment!"
When one searches, that is the first thing they come across, is that particular brand and make of ac+dc power supply. And it is in the area of what I'm looking for.
But, it's in the costly category. and way out of line in doing an initial foray or test into a given thing. Unless one has the funds to splash on such a thing. Which would be corporate or academic level of expense. A privateer does not have that mass behind them.
This is more like teenagers with a bag of weed and a $200 turbo on a $500 civic, staying with the Lambo for about 2 minutes - until the 4 banger explodes.
Anyway, thanks gentlemen, this thread answered my question, and confirms what I thought was true about such devices. The kind of things that seem obvious (with enough knowledge about the basics of the DUT) but assumptions should be checked by wiser and more experienced heads... where such a 'check' can also help outline some of potential pitfalls in the whole mix of what it takes to get to the desired end point.
When one searches, that is the first thing they come across, is that particular brand and make of ac+dc power supply. And it is in the area of what I'm looking for.
But, it's in the costly category. and way out of line in doing an initial foray or test into a given thing. Unless one has the funds to splash on such a thing. Which would be corporate or academic level of expense. A privateer does not have that mass behind them.
This is more like teenagers with a bag of weed and a $200 turbo on a $500 civic, staying with the Lambo for about 2 minutes - until the 4 banger explodes.
Anyway, thanks gentlemen, this thread answered my question, and confirms what I thought was true about such devices. The kind of things that seem obvious (with enough knowledge about the basics of the DUT) but assumptions should be checked by wiser and more experienced heads... where such a 'check' can also help outline some of potential pitfalls in the whole mix of what it takes to get to the desired end point.
Last edited:
I believe IRS2092 Class D driver works fine with AC and DC so an amp based on that should be no trouble.
Yep, that is the answer I got back from them just today... that the chip (IRS2092) or developed system (for that chip) itself has no inherent DC blocking that cannot be easily defeated. But that they have no white paper on implementation of such a scheme.
So, if one wanted to, for example, create a low impedance ac+dc power supply, a programmable one, for product testing etc.... that this might be doable with a IRS2092 implementation of some sort. for the sake of being lab grade, it might be better to use a bash or pure analog amp, tha has a PWM modulated power supply, which I think that the Netwons4th gear is probably built in that direction or area. From their paperwork, PWM on the internal power supplies in their associated gear, for sure.
for the sake expedience, simplicity, build, and costs ..and ignoring output filter effects .. I'm turning the equation around and putting the PWM on the output tied to a high current high storage analog power supply. The 'supposed higher quality' audio version of a class D implementation.Evertin' bein' debatable, as per normal...
Bash amps might be the perfect direction to go in. Or at least a better implementation of BASH being key, that is.
So, if one wanted to, for example, create a low impedance ac+dc power supply, a programmable one, for product testing etc.... that this might be doable with a IRS2092 implementation of some sort. for the sake of being lab grade, it might be better to use a bash or pure analog amp, tha has a PWM modulated power supply, which I think that the Netwons4th gear is probably built in that direction or area. From their paperwork, PWM on the internal power supplies in their associated gear, for sure.
for the sake expedience, simplicity, build, and costs ..and ignoring output filter effects .. I'm turning the equation around and putting the PWM on the output tied to a high current high storage analog power supply. The 'supposed higher quality' audio version of a class D implementation.Evertin' bein' debatable, as per normal...
Bash amps might be the perfect direction to go in. Or at least a better implementation of BASH being key, that is.
Last edited:
looking around there seems to be some decent high powered and capable units out there ..with some units obstinately rated at 2000 watts. Ahem. "Hello output filter my old friend, it's good to buzz with you again....
Anyone know where they can buy such things without waiting 60 days for something to show up? double the price is not the issue. the 30-60 days is a huge issue. any thoughts?
Anyone know where they can buy such things without waiting 60 days for something to show up? double the price is not the issue. the 30-60 days is a huge issue. any thoughts?