soppelsa said:
How about SOA?
Ciss?
Coss?
Crss?
Power vs. Temp de-rating?
Id vs Vgs transfer characteristics?
Saturation voltage vs drain current vs. temp?
There's more, but I don't have time to list it. A complete semiconductor device specification sheet is a very good idea. I mentioned Renesas because they have one.
Scott
soppelsa said:
It's heat.
Nothing is 100% efficient. The amount of temperature rise over ambient has to do with the amount of heat capacity of the thermal sink. Even with low power dissipation a super tiny heatsink will get hot.
Class-D amplfiers that specifiy a single number for efficiency are incompletely specified. Efficiency over the full power output range is what is needed for a minimum picture. I'd rather see a plot of power dissipation vs. power output vs. load impdeance, which would be more help in selecting the heatsinking requirements for the module.
Scott
Class-D Design said:
It was a Tripath TA 0104A based amp on an evaluation board that I bought from friend and originally it was built by another friend.
I think it was to much modified by all of us and it broke when I raised the supply voltage as I thought 2*200 W wasn´t enough for my subs.
The new supply voltage was within the limits in the application papers but both the chip and some output devices needed replacements. Ended up with buying a RA0105 chip from Profusion (?) in England and "bigger" MosFets (recommended by Tripath) Eventually I miss aligned the chip because of fatigue, things got a bit complicated, and the amp is now a memory of "not how to do it". I agree to that efficiency should be stated at different output levels. I read an article by D. Self* on the subject many years ago.
Class A is definitely "worst case" when no signal is present as all energy has to be dissipated as heat (disregarding for the moment the heat that is caused in the voice coils of ordinary dynamic speakers while driven). Heat is a waste of energy but some say that if it warms up your house in the winter it´s OK. But I believe this to be too much of a simplification. It depends e.g. of the nature and hysteresis of your warming system.
A Grand piano is said to have an output capacity of one to two acoustic watt. With all losses in the chain of reproduction and thus also considering the often very low efficiency of dynamic speakers the very problem seems to be in the dynamic capabilities of a system, even if compression is used. By modern computing methods (delaying and examining the signal) I think it should be possible to adapt the system to the conditions that is to be encountered. In a home system maximum output capability isn´t usually required from an amp if you are supposed to listen at low SPL. If you adapt the supply voltage of a Class D amp so it is in the vicinity of the level of clipping IMHO something will be gained.
* Probably: Cool audio power 1999 at this link;
http://www.dself.dsl.pipex.com/ampins/library/ampartew.htm
"Efficiency over the full power output range is what is needed for a minimum picture."
Yes.
IMO idle dissipation is the most important parameter.
Even where peaks of 100's of W for satellites and 1000's of W for subs are needed, the average power is 10's of W if that for home use), so IMO the need for high efficiency at full output is much less important.
Yes.
IMO idle dissipation is the most important parameter.
Even where peaks of 100's of W for satellites and 1000's of W for subs are needed, the average power is 10's of W if that for home use), so IMO the need for high efficiency at full output is much less important.
The golden mean said:
It was a Tripath TA 0104A based amp on an evaluation board that I bought from friend and originally it was built by another friend.
I think it was to much modified by all of us and it broke when I raised the supply voltage as I thought 2*200 W wasn´t enough for my subs.
I agree to that efficiency should be stated at different output levels. I read an article by D. Self* on the subject many years ago.
Class A is definitely "worst case" when no signal is present as all energy has to be dissipated as heat (disregarding for the moment the heat that is caused in the voice coils of ordinary dynamic speakers while driven). Heat is a waste of energy but some say that if it warms up your house in the winter it´s OK. But I believe this to be too much of a simplification. It depends e.g. of the nature and hysteresis of your warming system.
A Grand piano is said to have an output capacity of one to two acoustic watt. With all losses in the chain of reproduction and thus also considering the often very low efficiency of dynamic speakers the very problem seems to be in the dynamic capabilities of a system, even if compression is used. By modern computing methods (delaying and examining the signal) I think it should be possible to adapt the system to the conditions that is to be encountered. In a home system maximum output capability isn´t usually required from an amp if you are supposed to listen at low SPL. If you adapt the supply voltage of a Class D amp so it is in the vicinity of the level of clipping IMHO something will be gained.
* Probably: Cool audio power 1999 at this link;
http://www.dself.dsl.pipex.com/ampins/library/ampartew.htm
Tripath are of course no longer in business these were what I call frequncy chasing amps basically controlled by varying he frequency ok if you have 1 amplifier problems if you have 2 or more add a switching power supply you have a nightmare.
I simply have to agree with most of your remarks except about taking a class d amplifier to clipping. This is one of the misunderstood parts of class d amplifiers if you drive it near clip it is effectively in excess of 90% modulation this is where some of the problems lie you need to limit to about 85% if you go further the output stage starts misfiring shoot through occurs this just produces more transients and misfires and now you have lots of heat and an immenent explosion. The other issue is dead time now that is a minefield.
richie00boy said:
Richie we work most weekends this one has been difficult major transport company went bust on friday causing issues.
Thelittle amp will be back on shortly we are just finalising a switching PSU to go with it.
might be easier to phone me you are only 100 mikes away mobile is 07980 600373.
You obviously make amps is it a hobby or profession
richie00boy said:
http://www.classd.ltd.uk/home.php?cat=293
Richie00boy,
You can find the 'cute little amp modules' here
Regards,
Sam.
Class-D Design Ltd.
There seems to be something a bit odd going on here that needs clarifying.
If I look at the plots on the Aussie Amplifiers NXV200 here:
http://www.aussieamplifiers.com/nxv200.htm
and then look at the plots on the Class-D CD-NX200 here:
http://www.classd.ltd.uk/Datasheets/Specifications for the CD NX's.pdf
They appear to be identical.
I thought they were using completely different transistors??
Actually, if you read the spec from both sites it all appears to be the same apart from Aussie Amplifiers uses the wording "VFET-MOSFET" and Class-D "Alfet Lateral Mosfets".
col.
If I look at the plots on the Aussie Amplifiers NXV200 here:
http://www.aussieamplifiers.com/nxv200.htm
and then look at the plots on the Class-D CD-NX200 here:
http://www.classd.ltd.uk/Datasheets/Specifications for the CD NX's.pdf
They appear to be identical.
I thought they were using completely different transistors??
Actually, if you read the spec from both sites it all appears to be the same apart from Aussie Amplifiers uses the wording "VFET-MOSFET" and Class-D "Alfet Lateral Mosfets".
col.
Above link for the ClassD.Ltd doesn´t work for me- so go this way;
http://www.classd.ltd.uk/product.php?productid=16269&cat=293&page=1
and then Specification sheet "click here".
http://www.classd.ltd.uk/product.php?productid=16269&cat=293&page=1
and then Specification sheet "click here".
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.