| rmsaudio |
Was wondering if any electronics engineers could offer some technical opinions
on the subject.
Which topology is better, class-d, class-i or even class-h ?
From my limited experience in audio, I've learned that most arguments come
down to application, but since they're all a form of digital, perhaps there is
a "better" choice. When I say better, I'm thinking manufacturing costs,
reliability, sound quality and THD etc. Perhaps Class-I is just marketing hype?
I've been toying with the idea of contracting out some designs, hence the question. Application would be _big_ sound...pubs, clubs, events
and competition car audio. The dream is something along the lines of crown's
i-tech and JBL's 6KW beast http://www.caraudiomag.com/specialf...n/photo_03.html ....probably not quite so big, but along competition lines. |
|
|
| Workhorse |
Well,
Class-D is Classical Pulse Width Modulation switching amplifier...Its output consists of Mosfets in Pushpull manner and a series connected Inductor filter to retrieve the audio wave from the Pulsating signal through averaging effect of inductor.Efficiency is around 90-95%.....
Class-I is just another hyped version of Class-D but it uses two large inductors for storing energy as the both output mosfet conduct at same time, though absence of dead-time, but its high frequency response is not good one.....Its also Known as Balanced Current amplifier [BCA]...
Class-H is actually a Class-AB amplifier whose supply rails were switched from lower level to upper level in accordance with the output voltage required to get efficiency.....2 Tier Class-H is Efficient upto
65-75% with music signals
I have developed both Class-D and Class-H amplifiers......
regards,
K a n w a r |
|
|
| TOINO |
When I look for “class-I” schematics, the only thing I see is two Buck regulators bridge connected…
One positive the other negative, like class-B analog amplifier.
Could some body tell me if I need eyeglasses, please..?:( |
|
|
| jmartins |
Where can I read more about Class-H ?
There is schematics ? websites???
regards,
João |
|
|
| rmsaudio |
Workhorse: thanks for the input, I remember you as ampman ...from a few
years back, glad to see you're still alive and kicking! :)
Perhaps drop me an email when you get a chance, design@rmsaudio.com. I
did try and email you, but the forum didn't allow it. |
|
|
| Workhorse |
Hi RMS,
Nice to see you too..........
email me at
info@workhorsetechnologiesindia.com
for any enquires....
It seems to me that you also changed your Identity as your registration is dated to 2006...and you say you know me from past few years
regards,
K a n w a r |
|
|
| TOINO |
Hi Workhorse
Yes I know all the Crown “pa-ta-ti-bla-bla”, but analysing the schematics: http://www.jblproservice.com/pdf/Ve...0Schematics.pdf
Is just two Buck converters assembled like a class-AB output “linear” power amplifier.
On class-AB you have 1 power transistor for positive part of signal and another for negative.
Positive one could only source current to load=Class-I positive Buck converter.
Negative one could only sink current from load= Class-I negative Buck converter.
For my eyes, class-I looks like a switched class-AB counterpart in which the output transistors are Buck-converters.
Even the modulation scheme is in a way to simulate the linear class-AB amplifier. |
|
|
| Workhorse |
Hi TOINO,
Yes your right, but have you noticed one thing that the two mosfets switch at same time .......but the Pulse width changes according to the signal
Do you have RAM audio schematics....
regards,
K a n w a r |
|
|
| TOINO |
Hi Workhorse
Shure that both mosfets switch with complementary signals, but the output is taken after the coils.
Each switch group, positive and negative, are complementary after the coil.
Please note that at this node (where the two coils join to form the output signal) only one switch group sinks our sources, never both at some time.
Otherwise you have catastrophic cross-conduction (like in class-AB linear counterpart).
RAM is robust, but conventional…
It is like this one, http://www.audio-circuit.dk/PDF/Sch...0LYNX_v2-1B.pdf
but with class-G rails. |
|
|
| SpittinLLama |
The Class I just seems to me to be what is also called PWM Class BD. Both TI and National make 5V versions using that modulation class. Rockford Fosgate got a patent on some improvements to Class BD. Here is the patent. It is an interesting read and talks about Classs AD and BD.
Hmm, can't attach patent because the file is too big. The patent number is: 6,097,249, dated Aug 2000
-SL |
|
|
| TOINO |
Hi SpittinLLama
Class-BD is just the modulation scheme.
I am talking about the Crown “class-I” switch arrangement because it seems to me as a regular Buck converter (or two).
Anyway apparently nobody is interested in deeply comment this subject.:bawling: |
|
|
| rmsaudio |
TOINO: Maybe start another thread regarding Class-I ?
I know nothing about electronics, but I'm told by a very reliable source that Class-I _is_ the business..hence the idea of it's own thread.
Perhaps it's time for some Class-I diy modules :) |
|
|
| rmsaudio |
| SpittinLLama: Thanks for the patent number, any idea of the patent number for Crowns Class-I BCA ? |
|
|
| rtarbell |
I read in one of their articles on their website that:
"A class-I amp handles reactive (inductive) speaker loads easily and gracefully. It re-uses the energy returned from the speaker rather than dissipating it as heat or forcing the amp into premature current-limiting."
I wonder how this is accomplished? |
|
|
| Pafi |
| quote: | | I am talking about the Crown “class-I” switch arrangement because it seems to me as a regular Buck converter (or two). |
You are right! Its only 2 paralelled (and 2 sets bridged) Buck converter. But this fact has consequences!
- dead time distortion can be avoidable
- switching loss and EMI is less, because of lower Qrr of discrete diode
- possibility of using higher voltage MOSFET
- possibility of 5 level modulation (in bridge)
Hence this is the best arrangement for very high power. |
|
|
| Pafi |
| quote: | | I wonder how this is accomplished? |
This is simply due to ClassD principle. |
|
|
| TOINO |
Hi rmsaudio
this is the "class-I" patent
http://www.freepatentsonline.com/5657219.pdf
Hi Pafi
Thanks for your comments.
I need some time to think about it, as I am a learner in these things.
Anyway, a first glance, I honestly can’t see nothing on “class-I” patent who could be really patented.
My meaning is that the Buck-converter and BD-modulation are in fact public domain at the time the patent was granted.
See for example this one: http://www.freepatentsonline.com/5025360.pdf is the some arrangement...
About the topological similitude’s I found in “class-I” / class-AB(linear) the “class-I” has also “bias adjust” to compensate for crossover distortion…?! |
|
|
| SpittinLLama |
I guess it is a pet peve of mine but I hate when companies make a new class of amplifier when, in fact, it is just some improvement to an existing class. Class I and Class T to me are nothing more than marketing, just more BS. Sure they might do some things other Class D's haven't done but there are tons of Class AB designs that do some things different. Drives me crazy. Who determines a Class? Problem is no one, just name it if you want., take out a trade mark out and BOOM, new class of amplifiers that are not really much new to the basic idea/theory of an existing class.
I will read their patent but I think they have just a different way to accomplish a Class BD design that is still Class BD.
Stupid marketing.
-SL |
|
|
| Pafi |
| quote: | | Anyway, a first glance, I honestly can’t see nothing on “class-I” patent who could be really patented. |
Do you know the story about the egg of Columbus?
Basically every good invention is simple... if somebody have already invented it. You can combine 2 known things, and if the result is an arrangement with unprecedented feature, then it can be patented.
| quote: | | About the topological similitude’s I found in “class-I” / class-AB(linear) the “class-I” has also “bias adjust” to compensate for crossover distortion…?! |
Yes, but it can be done without shoot trough current, and you have to adjust only one delay, not two. |
|
|
| classd4sure |
Hi,
Is there not still a possible timing error (even if not your typical shoot through causing deadtime) between each switching half?
Pafi odd you mention that, I was watching Discovery last night and I believe I heard something about sliced bread. They said how everyone always say's "greatest thing since sliced bread", but the guy who invented sliced bread didn't invent bread, nor did he invent slicing, just the combination of two existing technologies. |
|
|
| Pafi |
| quote: | | I think they have just a different way to accomplish a Class BD design that is still Class BD. |
I agree! I rather call it on its previous name: BCA. |
|
|
| Pafi |
| quote: | | Is there not still a possible timing error (even if not your typical shoot through causing deadtime) between each switching half? |
Yes, but if it makes ton1+ton2>T, it causes only a slowly increasing DC current (typically 1 A/cycle) through inductors. It can be controlled much easyer then dead time. |
|
|
| Workhorse |
You call it robust but i have seen some serious drawbacks in that design , such as current limiting is only sensed from only one output device pair not through the every device at output with summed resistors..
RAM a Switched rail Class-H, not a Class-G design....
Also in their BUX7.0 the SMPS is half-Bridge with only one pair of IGBT's which means they cannot drive the full 2 ohm stereo load for extended periods over the showtime.....
regards,
K a n w a r |
|
|
| TOINO |
Hi Workhorse
You are right! It is class-H, not class-G. My keyboard has neighbour G and H letters… sorry.
By robust, I mean on practice; must confess that only know the “insides” of BUX5 model and is a reliable amplifier on the “street” driving sub-woofers.
Could not recommend it for nothing more because I don’t like the sound on Mid-Hi frequencies.
At electronics level, I must agree with you: nothing special… old classic topologies and some simplifications as the ones you refer in protection circuits.
Please stop criticize your own competition. :cuss: It looks like a kind of publicity for your own commercial products…:clown:
Pshiu…! This is of topic and Mr. Moderator will put us on Texas!:xeye: |
|
|
| classd4sure |
| quote: | | Please stop criticize your own competition |
NNoooo, that's most valuable information for those who'd otherwise be willing to spend all kinds of time reproducing what they think is good based on pure marketing and no real knowledge. Good work, horse. |
|
|
| TOINO |
OK! I agree with you but we must be very rigorous in this kind of criticism
:( Workhorse just forget to point out that the BUX7 has one PSU per channel...
Surely this LITLE additional detail will change the 2 ohm performance?
Surely Workhorse doesn’t deliberately omit the second power supply :rolleyes:
Surely Workhorse is not here declaredly as a concurrent company owner and doesn’t lose a single opportunity of publicity for their own products without make real proof of their quality.:eek:
Surely I will go to Texas or Sin Bin because of all this off topic!..:dead: |
|
|
| Workhorse |
| quote: | Originally posted by TOINO
Hi Workhorse
Please stop criticize your own competition. :cuss: It looks like a kind of publicity for your own commercial products…:clown:
Pshiu…! This is of topic and Mr. Moderator will put us on Texas!:xeye: |
I am not criticising them , but only pointing out some of the drawbacks in thier designs
| quote: | Originally posted by classd4sure
NNoooo, that's most valuable information for those who'd otherwise be willing to spend all kinds of time reproducing what they think is good based on pure marketing and no real knowledge. Good work, horse. |
Yeah, thanks Chris...| quote: | Originally posted by TOINO
OK! I agree with you but we must be very rigorous in this kind of criticism
:( Workhorse just forget to point out that the BUX7 has one PSU per channel...
Surely this LITLE additional detail will change the 2 ohm performance?
Surely Workhorse doesn’t deliberately omit the second power supply :rolleyes:
Surely Workhorse is not here declaredly as a concurrent company owner and doesn’t lose a single opportunity of publicity for their own products without make real proof of their quality.:eek:
Surely I will go to Texas or Sin Bin because of all this off topic!..:dead: |
I think you love RAM audio....:D :D ;)
K a n w a r |
|
|
|
