To Kurt von Kubik:
Hi,
OK, I know ICEpower, has a good sound.
interesting is the second FB, because t is not easy to implement, depends on the architecture of the amplifier (it must necessarily have a very low delay from input to output) so this is already good.
probably have added a filter on ring of FB to stabilize response at low frequencies(HI-damping is necessary on this zone).
Hi,
OK, I know ICEpower, has a good sound.
interesting is the second FB, because t is not easy to implement, depends on the architecture of the amplifier (it must necessarily have a very low delay from input to output) so this is already good.
probably have added a filter on ring of FB to stabilize response at low frequencies(HI-damping is necessary on this zone).
There are many myths around class D feedback schemes that are not necessarily true. Taking the feedback after the output filter does not always result in high output
impedance at high frequencies or instability.
I get around 0.075 ohm output impedance at 10khz with a "single" feedback loop, and this is in the version of my amplifier intended for low frequencies, whose output filter resonates at 12khz and which switches at 125khz, but the feedback loop eliminates any trace of filter resonance. Also, my feedback scheme allows to get -1dB around 15khz, thus extending the usable frequency response of the amplifier well above filter resonance.
UcD exhibits similar performance, but they don't go to the point of extending frequency response past filter resonance. ICEpower suffers from high output impedance at high frequencies. The full range version of my amplifier has everything moved up in the frequency domain by approx. a 2x factor at the expense of some efficiency loss. The low frequency version can also play music quite loud on 4 ohm withou any heatsinking other than the copper tab of the TO-220 transistors, over 100V peak. We are probably using the same latest generation IR or Fairchild MOSFET. This is probably going to become standard performance level within a few years.
Last edited:
to EVA:
Hi,
this is class D module (not public),developed in 2003
300w rms,400 Khz triangle wave, includes:
clip led,alc,pb adj,mute,regulate +/-15 on board,all protections.
continuos power without externalparts.
20Hz-40Khz bw -1 dB
very hi-Resolution. sound vey similar at class A-B
at 400khz, all is very complicate...
Hi,
this is class D module (not public),developed in 2003
300w rms,400 Khz triangle wave, includes:
clip led,alc,pb adj,mute,regulate +/-15 on board,all protections.
continuos power without externalparts.
20Hz-40Khz bw -1 dB
very hi-Resolution. sound vey similar at class A-B
at 400khz, all is very complicate...
Attachments
The two feedback loops on ICEpower are redundant. The same result may be achieved with a single post-filter loop and an output filter and layout having low parasitics. That's UcD, although it may be considered as having two loops because the signal taken from the output follows two different paths. In general, I think that a "single" feedback loop (like that) can do it all, but it may include several poles.
The 60A peak current capabilty may easily be real, although that's not very reliable for one pair of MOSFET with such a small heatsink, unless there is time/temperature dependent current limiting. But then the modules should also be rated at 2 ohm, and they aren't. Also, output coils don't look like capable of handling more than 10A rms or so for some time.
For example, my current project can do around 5kw on 2 ohm for 1 second (with time-dependent true-RMS current limiting to protect not only the amplifier but the power source and the load). Current limit is 75A peak and 25A rms long term, but that's achieved using pairs of parallel MOSFET, and 13 AWG wire on output coils. There are low cost tricks to calculate rms current without expensive multiplier ICs.
It is not UCD eva, and it wouldn't be the same result at all as their performance is not equivalent. Perhaps you should reread the threads where their differences have already been discussed. Why must all your posts turn into a product advertisement for yourself?
Doest it means, it will have lesser catastrophic modes also
BTW:do you guys have good customer support like brenda offers?
Well the Hypex modules uses a different technology than B&O.
You can read all about it @ their website, they actually produced an endless series of academic papers, presented at AES conventions through the last decade.
The designer of the TC module was part of the team behind amongst others this paper http://www.icepower.bang-olufsen.com/files/convention/5197.pdf
This describes in detail some of the essentials in ICE power, which is not at all available anywhere else in any form besides TC.
Through time`I´ve had many visitors coming by showing me their class D amplifiers, and it has been a long row of funny experiences.
Four times did the amps commit suicide just in front of us, all of these burns were due to unloaded condition. This because we compared them to my linear amps, but when unloaded for a few minutes they died.
It was amps from Hypex, Nuforce and Tripath.
ICE power amps such as Acoustic Reality and Bertram were in the contrary unconditionally stable, and they could easily be switched on without load.
Today ICE has sold more than 100 mio channels, and more than 30 engineers are employed in optimizing the technology.
Looking at data and theory of operation of both ICE and Hypex, leaves no doubt on which technology is superior, and in my experience this also goes for the sonic impression.
Unfortunately ICE is not available to DIYérs, but luckily the TC module is. It is a bit pricy though, but compared to linear amps available on the market, you´ll have to spend loads of money to get comparable performance.
You can read all about it @ their website, they actually produced an endless series of academic papers, presented at AES conventions through the last decade.
The designer of the TC module was part of the team behind amongst others this paper http://www.icepower.bang-olufsen.com/files/convention/5197.pdf
This describes in detail some of the essentials in ICE power, which is not at all available anywhere else in any form besides TC.
Through time`I´ve had many visitors coming by showing me their class D amplifiers, and it has been a long row of funny experiences.
Four times did the amps commit suicide just in front of us, all of these burns were due to unloaded condition. This because we compared them to my linear amps, but when unloaded for a few minutes they died.
It was amps from Hypex, Nuforce and Tripath.
ICE power amps such as Acoustic Reality and Bertram were in the contrary unconditionally stable, and they could easily be switched on without load.
Today ICE has sold more than 100 mio channels, and more than 30 engineers are employed in optimizing the technology.
Looking at data and theory of operation of both ICE and Hypex, leaves no doubt on which technology is superior, and in my experience this also goes for the sonic impression.
Unfortunately ICE is not available to DIYérs, but luckily the TC module is. It is a bit pricy though, but compared to linear amps available on the market, you´ll have to spend loads of money to get comparable performance.
UCD committs suicide when driven with NO LOAD??
Recently I saw a smoked ICE-Power amp in active speaker, the speaker coil wire was broken, probablya manufacturing defect, but it created the same open-load condition.
Last edited:
Doest it means, it will have lesser catastrophic modes also
BTW:do you guys have good customer support like brenda offers?
The TC module is completely "LC less" which means it does not share any kind of technology with both ZAP pulse and New Class D.
What do you mean "costumer support"?
I´m not trying to sell any of those modules, they can be ordered at LC audio, but I do not have any interest in that business.
We are building our own support circuitry for it, later when they are closed, you can have schematics.
My interest in this is only, that I became quite impressed by the sound of it, and that was a first time ever for class D.
UCD committs suicide when driven with NO LOAD??
Recently I saw a smoked ICE-Power amp in active speaker, the speaker coil wire was broken, probablya manufacturing defect, but it created the same open-load condition.
I do not remember the name of the module, but it was a Hypex made one.
And it really led out the smoke
The filter coil melted completely.
Of course ICE also can burn, but normally they are very well protected, but a broken voicecoil sounds to me as a DC problem.
You're outright lying, or completely clueless and very imaginative, take your pick.
Class d..Innovations?
Hi,
We speak of defects ... The list is long.
We talk about innovation in the class D!
Hypex is good and very different from other schemes (thanks to Philips R & D)
old but has no innovation.
B & O has a good sound, 500w model does not have much strength in intensive use.
scheme is still older than UCD.
Tell me what company made a class D on real innovations on board?
Best regards
Hi,
We speak of defects ... The list is long.
We talk about innovation in the class D!
Hypex is good and very different from other schemes (thanks to Philips R & D)
old but has no innovation.
B & O has a good sound, 500w model does not have much strength in intensive use.
scheme is still older than UCD.
Tell me what company made a class D on real innovations on board?
Best regards
IMO with the "innovation" statement you are going out on a limb !
Someone mentioned the small-signal bandwidth of UcD. The priciple as such allows an overall first-order lowpass transfer function. AFAIK the current Hypex modules use an additional lowpass at the input to avoid TIM problems.
A post-filter feedback topology can be made such that the small-signal bandwidth equals the unity gain point of the feedback loop. This is not that difficult. It is the designer's job to make wise use of it !
Regards
Charles
It's not advertisement, it's humiliation for the big companies because someone like me, coming from the DIY world, can develop better technology on his home lab without expensive teams of engineers or the help of any big brand...
Of course I saw the datasheets, but I have also reverse engineered the actual products, so my knowledge is not *limited* to advertisements.
Both ICEpower and UcD are phase shift oscillators. ICEpower ruins the beautiful performance that they can provide by using split feedback, a big mistake (and by routing the delicate carrier residual through two inexpensive MC33078). UcD uses almost the full potential of the physical principle and routes the carrier residual to the comparator through passives only (the right way to do it).
My technology tries to use the potential of phase shift oscillation to an even further degree, and at much higher powers whithout giving up on efficiency as much as in UcD2k. My output stage is unique in many ways, it idles at around 15W with 170V DC supply, as opposed to the 35W from UcD2k at only 144V DC, and 2800W/4R efficiency is over 95% from mains input to speaker, as opposed to 92% from UcD2k without including power supply losses, so I'm also one step ahead of the big guys in that field.
Blah blah blah...
You learn much more about how things work by tracing PCBs than by reading the documents that you are expected to read
Sometimes, playing around with oscilloscope, you can even discover circuit mistakes that the original designers overlooked because they didn't care to look at every waveform on a production sample ICEpower is slowly dying. Their technology has almost not changed at all since they began, and they are slowly losing customers as suggested by their various price increases during 2009.
People is just switching to other technologies because they are a bit fed up of them and their continuous price increases, and the lack of protections on the non-ASP amplifiers, their unfriendly support, etc...
My amplifier is not available for the DIY market either, I was just hired to design it for someone else
So talking about it in a place like this can't be regarded as advertisement of any product My purpose is to tease other people, whose products are targeted at the DIY market, and that are trying to advertise themselves here for free and to persuade everyone to buy their product at their store because it has the best performance, when they are not that good... The method: Comparing their achievements with mine
ICEpower is a special case bacause, although it is not available to the DIY market, it has a lot of positive hype around that it does not deserve at all.
Last edited:
Blah blah blah...
You learn much more about how things work by tracing PCBs than by reading the documents that you are expected to read
ICEpower is slowly dying. Their technology has almost not changed at all since they began, and they are slowly losing customers as suggested by their various price increases during 2009.
People is just switching to other technologies because they are a bit fed up of them and their continuous price increases, and the lack of protections on the non-ASP amplifiers, their unfriendly support, etc...
My amplifier is not available for the DIY market either, I was just hired to design it for someone else
My purpose is to tease other people, whose products are targeted at the DIY market, and that are trying to advertise themselves here for free and to persuade everyone to buy their product at their store because it has the best performance, when they are not that good... The method: Comparing their achievements with mine
ICEpower is a special case bacause, although it is not available to the DIY market, it has a lot of positive hype around that it does not deserve at all.
Well you better bring the news to the stockmarket, you could make a fortune by shortselling B&O shares, with that knowledge.
To my knowledge though, the ICEpower division is very succesfull, i.e. Samsung adobted it for their TV sets and for their mobile phones.
I don´t want to go into a discussion whether hypex or ICE is the best class D on the market, Hypex is simply of no interest to me what so ever, just as ICE is not either.
I put my money on the TC module, i´ve already got one, and my pal and I did already build a buffer and a decent PSU for it, just as we investigated what the module is sensitive to.
So by now, in my experience, class D became somewhat interesting, which I´ve never found it before incl. Hypex, Nuforce, Tripath, LC and New Class D.
I think it is due to their special feedback circuitry, because I´ve never heard any amplifier capable of the control and 3D, which the TC module is.
I did not test all different Hypex modules, as well as I did not check out all Nuforce amps, because the first impression did not at all make me curious. This one did, and later on I´ve heard of very respectable audiophiles saying the same.
To me this is real world competition to even very high end class A amplifiers, each of them having their strengths, none of them having all of them.
Btw. if anyone should doubt the AES papers scientific accuracy, I think AES would be thankfull recieving advice and correction.
TO PHASE_ACCURATE:
I would be out of limbo why?
I said that on hypex and B & O is not innovative architecture.
... this is fake?
General digram of UDC is different from others, ex. driver stage, pre-drive and FB entry-Point.
repeat ... Hypex is a great amp but does not contain any new architecture(innovations), it sells good.
PSU on amplifier? ... Maybe 30 engineers are still working on the amplifier!
I would be out of limbo why?
I said that on hypex and B & O is not innovative architecture.
... this is fake?
General digram of UDC is different from others, ex. driver stage, pre-drive and FB entry-Point.
repeat ... Hypex is a great amp but does not contain any new architecture(innovations), it sells good.
PSU on amplifier? ... Maybe 30 engineers are still working on the amplifier!
TO PHASE_ACCURATE:
I would be out of limbo why?
I said that on hypex and B & O is not innovative architecture.
... this is fake?
General digram of UDC is different from others, ex. driver stage, pre-drive and FB entry-Point.
repeat ... Hypex is a great amp but does not contain any new architecture(innovations), it sells good.
PSU on amplifier? ... Maybe 30 engineers are still working on the amplifier!
ICEpower themselves states, that 80% of their 50 employees are working with R&D.
To me that is pretty real, because the modules are just a tiny bit of their business, ICE IC´s is a much larger business area, which is used in mobile phones, iPODs, computers and much more.
The modules were just the beginning, but never mind, they still hold patents on 2 feedback loops, which others would break a leg to aquire.
Look at their data sheets, ICE rules that world by a comfortable factor
But the TC is even better, just try it out.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.