I'm sorry, have you looked at the specifications and the technology background information? On first looks it doesn't look that appetizing at all to be honest.
That's a bit on the high side, also no other values are given, unlike Hypex that publishes full THD, IMD and power bandwidth figures. Furthermore they use a clock-oscillator driven design which has basically been obsoleted for HiFi use when self-oscillating amplifiers were introduced a few years ago.
Best regards,
Sander Sassen
http://www.hardwareanalysis.com
That's a bit on the high side, also no other values are given, unlike Hypex that publishes full THD, IMD and power bandwidth figures. Furthermore they use a clock-oscillator driven design which has basically been obsoleted for HiFi use when self-oscillating amplifiers were introduced a few years ago.
Best regards,
Sander Sassen
http://www.hardwareanalysis.com
Well, that's the quoted THD figure at full power. You know that distortion must be fixed at a given value in order to determine maximum power. 0.1% or even 1% are common values. If you look carefully, you will see a much smaller value for low power. It would be nice to see the complete THD vs power graphs, anyway, but the quote figures are not bad at all.
Personally, I like synchronous designs and they are not obsolete at all IMHO. Plus they offer several things that hypex nor Lcaudio offer, such as the notch filter and clipping indicator. And the modules look as a much more robust unit .
Not that I have nothing with them but when I first saw the page I got interested. Let's hope they release the SMPS soon also...
Personally, I like synchronous designs and they are not obsolete at all IMHO. Plus they offer several things that hypex nor Lcaudio offer, such as the notch filter and clipping indicator. And the modules look as a much more robust unit .
Not that I have nothing with them but when I first saw the page I got interested. Let's hope they release the SMPS soon also...
Pierre,
I'm sure they'll be well suited for PA or other applications, but I'd be hesitant to use them for HiFi purposes. Without powerband and THD vs. power and bandwidth figures those numbers are a bit meaningless anyway, or at least to me. Synchronous designs have mostly been used for subwoofer applications thusfar, hence I usually raise an eyebrow if I see another synchronous design try its hand at full-range. UcD and other self-oscillating topologies are far more elegant and flexible in that respect in my opinion.
Best regards,
Sander Sassen
http://www.hardwareanalysis.com
I'm sure they'll be well suited for PA or other applications, but I'd be hesitant to use them for HiFi purposes. Without powerband and THD vs. power and bandwidth figures those numbers are a bit meaningless anyway, or at least to me. Synchronous designs have mostly been used for subwoofer applications thusfar, hence I usually raise an eyebrow if I see another synchronous design try its hand at full-range. UcD and other self-oscillating topologies are far more elegant and flexible in that respect in my opinion.
Best regards,
Sander Sassen
http://www.hardwareanalysis.com
Indeed Ssassen, the THD figure you quote should be interpreted as:
If THD reaches 0.5% then you are putting 400Watts in a 4Ohm loads.. it's a measure of power output not of THD.
It doesn't say anything about the THD at 1Watt .. but if you start reading the datasheet, you'll find out that it's a very nice kit...
If THD reaches 0.5% then you are putting 400Watts in a 4Ohm loads.. it's a measure of power output not of THD.
It doesn't say anything about the THD at 1Watt .. but if you start reading the datasheet, you'll find out that it's a very nice kit...
Tschrama,
I realize that, but that figure is kind of meaningless in my opinion without seeing a plot of power vs. THD, much like Hypex provides.
Best regards,
Sander Sassen
http://www.hardwareanalysis.com
I realize that, but that figure is kind of meaningless in my opinion without seeing a plot of power vs. THD, much like Hypex provides.
Best regards,
Sander Sassen
http://www.hardwareanalysis.com
Well, at the risk of seem biased, that I am not, I personally like synchronous design more. Ssassen, I understand that you defend Hypex, but it is good to have more options out there, isn't it?
I don't like my output ripple to increase with amplitude. That happens in a lot of self-oscillating amplifiers. And what if you have interchannel whistles? With a sync. design the solution is inmediate, but what happens with the self-osc ones?
Furthermore, coldamp seems to have made an excellent work in the carrier suppression by adding a notch filter tuned at the sw frequency (that can't be done with a self-osc one, right?) At least that's what they claim. 500mVpp of residue (max) at +/-60V is quite a nice figure! (from the datasheet).
I personally built a sync design for full range obtaining quite low THD figures so this architecture, although simple and straigthforward, can and is used succesfully for the whole band. (is something bad in doing things the simple way). Ah, there is something: you need really good and expensive components for the triangle generator.
tschrama: if you download the datasheet you will see that the quoted THD is 0.015% at 1W.
I don't like my output ripple to increase with amplitude. That happens in a lot of self-oscillating amplifiers. And what if you have interchannel whistles? With a sync. design the solution is inmediate, but what happens with the self-osc ones?
Furthermore, coldamp seems to have made an excellent work in the carrier suppression by adding a notch filter tuned at the sw frequency (that can't be done with a self-osc one, right?) At least that's what they claim. 500mVpp of residue (max) at +/-60V is quite a nice figure! (from the datasheet).
I personally built a sync design for full range obtaining quite low THD figures so this architecture, although simple and straigthforward, can and is used succesfully for the whole band. (is something bad in doing things the simple way). Ah, there is something: you need really good and expensive components for the triangle generator.
tschrama: if you download the datasheet you will see that the quoted THD is 0.015% at 1W.
Pierre,
It so happens that I'm just completing a self-oscillating class-D design which is simple and straightforward and yields 0.015% at 100-watts at 1KHz into 8-ohms. The component count (excl. power supply components) is below 25 parts for this particular design. I have yet to see a synchronous design with a similar or lower component count yield such figures. So considering the component count on the ColdAmp I'm not impressed, from what I see it is a textbook synchronous design, no more no less.
But as always I'm keeping an open mind, assumption is the mother of ... as they say.
Best regards,
Sander Sassen
http://www.hardwareanalysis.com
It so happens that I'm just completing a self-oscillating class-D design which is simple and straightforward and yields 0.015% at 100-watts at 1KHz into 8-ohms. The component count (excl. power supply components) is below 25 parts for this particular design. I have yet to see a synchronous design with a similar or lower component count yield such figures. So considering the component count on the ColdAmp I'm not impressed, from what I see it is a textbook synchronous design, no more no less.
But as always I'm keeping an open mind, assumption is the mother of ... as they say.
Best regards,
Sander Sassen
http://www.hardwareanalysis.com
Pierre,
I've looked at the datasheet again, the 0.015% is quoted for 1-watt into 5-ohms at 100Hz that's cheating in my book, at these low frequencies the synchronous designs excel and a THD figure is meaningless. See if it manages the same low THD at 1KHz and at 10KHz.
Best regards,
Sander Sassen
http://www.hardwareanalysis.com
I've looked at the datasheet again, the 0.015% is quoted for 1-watt into 5-ohms at 100Hz that's cheating in my book, at these low frequencies the synchronous designs excel and a THD figure is meaningless. See if it manages the same low THD at 1KHz and at 10KHz.
Best regards,
Sander Sassen
http://www.hardwareanalysis.com
..
it does have some meaning... it specifies a maximum power output.. ...
Rod isn't saying it's the holy grail of HiFi, but he suggests that it is descent enough for a good sub-amp, for which it is intended anyway...
for a class-D amp, the THD versus power should be more or less constant ....
In my view it'll be a nice kit, Rod nerver tries to be Hi-end.... buy a gainclone for that
whahahaa
it does have some meaning... it specifies a maximum power output.. ...
Rod isn't saying it's the holy grail of HiFi, but he suggests that it is descent enough for a good sub-amp, for which it is intended anyway...
for a class-D amp, the THD versus power should be more or less constant ....
In my view it'll be a nice kit, Rod nerver tries to be Hi-end.... buy a gainclone for that
whahahaaWell, I haven't seen in any place of the article that Rod is saying that it is intended for subwoofer use. With the datasheet at hand, I would say it can make a good full-range hi-fi amplifier...
In fact I think I am going to ask price and buy a pair of modules just to try...
In fact I think I am going to ask price and buy a pair of modules just to try...
Hi Pierre,
The heterodyning possible with self oscillating amps hasn't turned out to be that restrictive an issue even for less than audiophile setups that pack the amps tightly next to each other.
Simple clock based amps using feedback can have a THD curve just like a bad class B amp with too much feedback and a lack of bandwidth. It pushes the noise in the lower frequencies up to the higher frequencies. The simpler and traditionally "sub use only" clocked amps had distortion ratings well beyond 10% at full power, and were highly band limited with low switching frequencies at that. Without any level of pulse edge correction to account for distortions in the time domain/delay of the output stage I don't think you can get a full range audiophile class d amp.
Self oscillating designs brought this to be as their delay error is an "intrinsic part of the control loop" in that the feedback signal is the total integrated error of everything inside the loop and so all delay errors are automatically accounted for and in the simplest, most robust way. That's well worth the risk of heterodyning which if well built still isn't a very restrictive problem.
It can't be done in a simpler or more robust way. While coldamp may be a cheap alternative I don't think it's an alternative to a "Hi Fi" amp. Also even with a notch filter to surpress carrier residual it still throws off enough EMI to blank out the nearby radio, that kind of emmission still has to couple to their carrier which must remain ideal and I can't imagine that being good for a low noise floor or what it would do for microdetail. I have my UCD180s sitting on a plate, no case, in the same room as an AM radio that is tuning into over 50 odd stations at this hour of the day and just as cleanly as if the amps were turned off.
Have you ever heard a self oscillating amp that you didn't like the sound of? This thing might be a cheap alternative but to what? I wouldn't say an audiophile amp.
Regards,
Chris
Regards,
Chris
The heterodyning possible with self oscillating amps hasn't turned out to be that restrictive an issue even for less than audiophile setups that pack the amps tightly next to each other.
Simple clock based amps using feedback can have a THD curve just like a bad class B amp with too much feedback and a lack of bandwidth. It pushes the noise in the lower frequencies up to the higher frequencies. The simpler and traditionally "sub use only" clocked amps had distortion ratings well beyond 10% at full power, and were highly band limited with low switching frequencies at that. Without any level of pulse edge correction to account for distortions in the time domain/delay of the output stage I don't think you can get a full range audiophile class d amp.
Self oscillating designs brought this to be as their delay error is an "intrinsic part of the control loop" in that the feedback signal is the total integrated error of everything inside the loop and so all delay errors are automatically accounted for and in the simplest, most robust way. That's well worth the risk of heterodyning which if well built still isn't a very restrictive problem.
It can't be done in a simpler or more robust way. While coldamp may be a cheap alternative I don't think it's an alternative to a "Hi Fi" amp. Also even with a notch filter to surpress carrier residual it still throws off enough EMI to blank out the nearby radio, that kind of emmission still has to couple to their carrier which must remain ideal and I can't imagine that being good for a low noise floor or what it would do for microdetail. I have my UCD180s sitting on a plate, no case, in the same room as an AM radio that is tuning into over 50 odd stations at this hour of the day and just as cleanly as if the amps were turned off.
Have you ever heard a self oscillating amp that you didn't like the sound of? This thing might be a cheap alternative but to what? I wouldn't say an audiophile amp.
Regards,
Chris
Regards,
Chris
Hi Chris
I agree that it it is more difficult to make a good carrier based switching amp than a self-oscillating one. But saying that it is impossible to make a good one is quite a bold statement.
And the RFI emission of a switching amp has definitely nothing to do with the working principle of its modulator but how it is implemented.
Regards
Charles
I agree that it it is more difficult to make a good carrier based switching amp than a self-oscillating one. But saying that it is impossible to make a good one is quite a bold statement.
And the RFI emission of a switching amp has definitely nothing to do with the working principle of its modulator but how it is implemented.
Regards
Charles
Hello all.
As some of you know, I am one of the old members of DIYAudio and also one of the founders of Coldamp. After years of experience with amplifiers in gneral, I started around 2-3 years ago in the world of Class-d and after building a lot of prototypes finally reach a very good sounding design about one and a half year ago. I was encouraged by a lot of people to keep working on it and develop a commercial version and that's what I have done, in partnership with a company here, in Spain.
I didn't want to advertise in this forum and I won't do it anyway, but after Rod's article it has been a lot of interest on the modules (although we are receiving requests from six months ago or so). Some of the posts I have seen here are based on nothing more than speculations so I wanted to clarify some points...
The amp is based on a clocked design as most of you know. But the level of linearity of the triangle and a very, very careful design and layout allow us to reach quite low THD levels. Perhaps not as low as Hypex or LC Audio, that are very good amplifiers, but low enough to be unaudible by several orders of magnitude so it doesn't matter too much anyway. It is true that the datasheet doesn't show a THD vs power / freq. at this moment but we are working on it and will publish that data a.s.a.p.
With BP4078, we have tried to make a compact and self-contained module, very flexible and including several features such as protections, clip indicator, etc, in order to allow very easy integration for DIYers and also serve as a good solution for professional audio makers.
About RFI, the amp behaves quite well in fact. Rod's problem is only an advertence (he tends to speak under a policy of sincerity that we are decided to follow as well) but the modules shouldn't create any problems when properly enclosed, and anyway what he had were prototypes that have been improved in this aspect since he tested them (about 6 months ago).
For that who say that coldamp isn't an hi-fi amp or that it's intended only for subwoofer use, I would suggest not to speak without having at least tested them. They perform surprisingly well for full-range and in fact that has been its main application to date. We are even receiving requests from local studio monitor makers after some demos.
Having said that, I will try not to involve too much in discussions because these forums are mainly intended to learn and share knowledge. It was very useful for me and I have always tried to be useful for others. I won't use it to advertise or promote commercial wars.
Best regards,
Sergio
As some of you know, I am one of the old members of DIYAudio and also one of the founders of Coldamp. After years of experience with amplifiers in gneral, I started around 2-3 years ago in the world of Class-d and after building a lot of prototypes finally reach a very good sounding design about one and a half year ago. I was encouraged by a lot of people to keep working on it and develop a commercial version and that's what I have done, in partnership with a company here, in Spain.
I didn't want to advertise in this forum and I won't do it anyway, but after Rod's article it has been a lot of interest on the modules (although we are receiving requests from six months ago or so). Some of the posts I have seen here are based on nothing more than speculations so I wanted to clarify some points...
The amp is based on a clocked design as most of you know. But the level of linearity of the triangle and a very, very careful design and layout allow us to reach quite low THD levels. Perhaps not as low as Hypex or LC Audio, that are very good amplifiers, but low enough to be unaudible by several orders of magnitude so it doesn't matter too much anyway. It is true that the datasheet doesn't show a THD vs power / freq. at this moment but we are working on it and will publish that data a.s.a.p.
With BP4078, we have tried to make a compact and self-contained module, very flexible and including several features such as protections, clip indicator, etc, in order to allow very easy integration for DIYers and also serve as a good solution for professional audio makers.
About RFI, the amp behaves quite well in fact. Rod's problem is only an advertence (he tends to speak under a policy of sincerity that we are decided to follow as well) but the modules shouldn't create any problems when properly enclosed, and anyway what he had were prototypes that have been improved in this aspect since he tested them (about 6 months ago).
For that who say that coldamp isn't an hi-fi amp or that it's intended only for subwoofer use, I would suggest not to speak without having at least tested them. They perform surprisingly well for full-range and in fact that has been its main application to date. We are even receiving requests from local studio monitor makers after some demos.
Having said that, I will try not to involve too much in discussions because these forums are mainly intended to learn and share knowledge. It was very useful for me and I have always tried to be useful for others. I won't use it to advertise or promote commercial wars.
Best regards,
Sergio
Hi Charles,
I didn't mean to imply it was impossible to make a good carrier based amp only that it has to be far more complex and less reliable given the added circuitry required simply to achieve what we already seemingly get for free with a self oscillating amp. In that respect the carrier based one will always be inferior.
EMI may be somewhat related to method of control as it can dictate the implementation of it, like an impliverter being soft switched would be less prone to EMI I'd think? Aside from that I know UCD does nothing special as it's still a hard switched inductive load which is the worst case scenerio for EMI related switching artifacts, and with good layout etc as you say, it's a non issue, but it an amp where it still blanks out a radio that's placed in the same room
You're right about the RFI I should have been specific to the UCD because the others likely aren't benefiting from as good an implementation. My point that even with their notch filter EMI is still high for their amp. It doesn't have a good implementation and the notch isn't enough to save it. If that high level of EMI should couple with the carrier it will result in a loss of performance via increased noise floor/ loss of microdynamics. Hopefully they don't run the carrier all over the board but it could still benefit from a better layout, dont' see how one might claim this is hi fi.
Regards,
Chris
I didn't mean to imply it was impossible to make a good carrier based amp only that it has to be far more complex and less reliable given the added circuitry required simply to achieve what we already seemingly get for free with a self oscillating amp. In that respect the carrier based one will always be inferior.
EMI may be somewhat related to method of control as it can dictate the implementation of it, like an impliverter being soft switched would be less prone to EMI I'd think? Aside from that I know UCD does nothing special as it's still a hard switched inductive load which is the worst case scenerio for EMI related switching artifacts, and with good layout etc as you say, it's a non issue, but it an amp where it still blanks out a radio that's placed in the same room
You're right about the RFI I should have been specific to the UCD because the others likely aren't benefiting from as good an implementation. My point that even with their notch filter EMI is still high for their amp. It doesn't have a good implementation and the notch isn't enough to save it. If that high level of EMI should couple with the carrier it will result in a loss of performance via increased noise floor/ loss of microdynamics. Hopefully they don't run the carrier all over the board but it could still benefit from a better layout, dont' see how one might claim this is hi fi.
Regards,
Chris
ssanmor said:
Hi Sergio,
Not to speculate it can't sound great in full range, comments to that affect are geared towards the best topology for the job though and that's self oscillating.
As far as speculation you're right, giving the track record of this kind of amp no one has been very optimistic about it. When we see your graphs we'll be able to make an informed opinion.
I'm sure it can still sound very good but I wonder if it's competitive with the cheap self oscillating alternatives, if you can give us a price guestimate for them? I'm sure that won't be a forum issue.
Regards,
Chris
Chris, I am sure that your critics were not with bad intention, just suggested that before throwing opinions on coldamp modules one should be very well informed or at least have tried one.
About complexity, it is not much more complicated than a self-osc one, the main difference is the triangle generator, that have proven to be very robust by the way. In fact you don't need some circuitry such as cranking for self-osc designs such as IRaudamp, etc.
You may get the impression of a great number of components in the board, but have in mind that the vast majority of them are mounted on top side and that you must take into account the protections (overvoltage, undervoltage, overcurrent) and added features (turn on/off control, protection latching, clipping detector), etc.
So it is not less reliable than a self-osc design.
About pricing, I am not going to contaminate the forum with commercial information. Interested people should read our web (prices are not shown currently but we will happily answer enquiries to the given e-mail address). If you are outside Europe please talk to Mr. Rod Eliott.
Best regards,
Sergio
About complexity, it is not much more complicated than a self-osc one, the main difference is the triangle generator, that have proven to be very robust by the way. In fact you don't need some circuitry such as cranking for self-osc designs such as IRaudamp, etc.
You may get the impression of a great number of components in the board, but have in mind that the vast majority of them are mounted on top side and that you must take into account the protections (overvoltage, undervoltage, overcurrent) and added features (turn on/off control, protection latching, clipping detector), etc.
So it is not less reliable than a self-osc design.
About pricing, I am not going to contaminate the forum with commercial information. Interested people should read our web (prices are not shown currently but we will happily answer enquiries to the given e-mail address). If you are outside Europe please talk to Mr. Rod Eliott.
Best regards,
Sergio
Hi Sergio,
good to hear from you....point taken ... fair enough.. really interested in the kit btw...
grz,
Thijs
good to hear from you....point taken ... fair enough.. really interested in the kit btw...
grz,
Thijs
ssanmor said:
Hi Sergio,
Very nice to have comments from the "father" of these modules on the board. Would be good to see some comments of the first users, could give good feedback to you.
Succes with these products, sorry, I won`t order some as I already have loads of UcD amps lying around and some of them actually being used 🙂
Best regards
Gertjan
- Status
- Not open for further replies.