Go Back   Home > Forums > Amplifiers > Class D

Class D Switching Power Amplifiers and Power D/A conversion

Please consider donating to help us continue to serve you.

Ads on/off / Custom Title / More PMs / More album space / Advanced printing & mass image saving
Reply
 
Thread Tools Search this Thread
Old 13th June 2004, 11:40 AM   #21
diyAudio Member
 
destroyer X's Avatar
 
Join Date: Feb 2004
Location: Recife - Brasil Northeast
Default I am reading this post with increasing interest

Very good thread.

Carlos
__________________
SELENIUM!... Brazilian brand of speakers factory tour; https://www.youtube.com/watch?v=v6khSy9HShQ; https://www.youtube.com/watch?v=S52ju-sTwzg
  Reply With Quote
Old 13th June 2004, 11:53 AM   #22
Account Disabled
 
Join Date: Feb 2004
Send a message via AIM to classd4sure Send a message via MSN to classd4sure
Hi everyone,

Excellent driver cookbook you have there Eva. Thank you for sharing it, I'm sure it can serve for class D designs as well, it shows some good techniques. Clearly you know what you're doing.

Nice scope, sure wish I had one like it.

Here is one driver leg of the UCD circuit....I would have included more...but it would be soooooo small you woulnd't see anything...so this will just have to do.

This seems to switch ok...(I'll give you a sample next post). With these resistor values, rise and fall times of the driver transistor match those of the data sheet.

Seeing how tempermental this circuit is to even the slightest changes, I'm having my doubts about prototyping it without a scope. See what happens...I'm not afraid of a little smoke.

Thanks,

Chris
Attached Images
File Type: gif onelegofdriver.gif (6.0 KB, 174 views)
  Reply With Quote
Old 13th June 2004, 12:02 PM   #23
Account Disabled
 
Join Date: Feb 2004
Send a message via AIM to classd4sure Send a message via MSN to classd4sure
Hi,

This is the output at the half bridge, before and after the coil.

Input is 2V AC at 20Khz to make it easy to see. This also clearly shows the next problem I need to work out with it, I think it stems from some sort of imbalance at the input stage that I haven't been able to nail down yet.

Thanks,

Chris
Attached Images
File Type: gif 2v20khzinput.gif (23.6 KB, 144 views)
  Reply With Quote
Old 13th June 2004, 12:50 PM   #24
Pafi is offline Pafi  Hungary
diyAudio Member
 
Join Date: May 2004
Location: Budapest
Chris!

Thanks for UCD!

That driver works well with the right values.

What is the new problem? Is that the frequency falls at high output?
  Reply With Quote
Old 13th June 2004, 01:16 PM   #25
Account Disabled
 
Join Date: Feb 2004
Send a message via AIM to classd4sure Send a message via MSN to classd4sure
Hi,

Pafi, you're welcome, but thank Bruno, it's his design

It's not really a new problem I've just ignored it until now.
When I run THD simulations it tells me "DC component -0.5" or so.

If you look very carefully at the last graph I posted, the PWM wave amplitude sags with the inverse of the input signal amplitude. This happens at all frequencies and input levels that it accepts (100mV to 2 volts AC).

So I guess basically the output rides on an inverse portion of the input. I've tried to match currents through the input stage as closely as I can, and the closer they get the better it gets, but it's still there, and I haven't been able to diminish this by more than .5 volts, even with the comparators currents matched down to the milli-amp range.

With a lower powered version of this exact circuit, the best I've had was THD .2% at 1Khz and 1 volt input.

With the higher powered version I just posted, I get about 6%. This is with no extra low pass filtering of the output.

So it still needs lots of work, but it's comming along.

Btw, the driver transistors I'm using were calculated at 550 ohms...which works but it's ugly. 110 Ohms seems ideal.

If anyone else is working on this I'd love to hear about it.
Thanks,

Chris
  Reply With Quote
Old 13th June 2004, 02:53 PM   #26
Account Disabled
 
Join Date: Feb 2004
Send a message via AIM to classd4sure Send a message via MSN to classd4sure
Hi,

Pafi,
Quote:
- Maybe you should choose faster MOSFET. For example IRFB31N20D, W34NB20, or under 150V: FDP2552...
Just had a chance to check these out. The IRFB31N20D looks alot like the IRFP250 at a glance, you'd think twins. I don't have a model for it though.

The W34NB20, oddly I could find no information on at all.

The FDP2552 looks very promissing, I just might try that model in fact. Is 150volts drain -source a big enough margin of safety for +-50 volt rails?

The two I could find information on look like excellent candidates anyway, thanks for sugesting them.

Had I tripped over that last one sooner my circuit would look different now.

Any comments on my choices of diodes?
Thanks,

Chris
  Reply With Quote
Old 13th June 2004, 05:03 PM   #27
Pafi is offline Pafi  Hungary
diyAudio Member
 
Join Date: May 2004
Location: Budapest
Chris!

I mean thank you for the link!

DC similarity is intentionally, but I tryed to find faster, and lower Qg FETs!

+/-50 V is OK, but because of supply pumping you must watch the supply, and turn off the amplifier, if goes too high! With bridged amp there is not a problem like this. (I use +/-70V supply with FDP2552).

Diodes: in driver they seem to be OK, but the 1N4003 on output I don't see any denefit on them. You'd better to use a MOSFET with faster dody-diode! The mentioned FDP2552 is excellent, since its Qrr is only 300 nC! APT makes amazing MOSFETs too. (For example 20M36BFLL.) Buy from them if you can!

W34NB20 is made by ST, search at www.st.com!

P.S.: Be careful! IRFP250, and IRFP250N are not the same! 250N is superior in speed!
  Reply With Quote
Old 13th June 2004, 10:18 PM   #28
Account Disabled
 
Join Date: Feb 2004
Send a message via AIM to classd4sure Send a message via MSN to classd4sure
Hi,

Pafi, I'm seriously impressed...you're like a walking mosfet inventory

I didn't even know about the IRFP250N, I don't have a model of it though.

It seems I would be foolish not to go with the FDP2552...I ...feel it...beckoning me...pulling me in..closer....

I'm still pretty far from the point where I have to worry about supply pumping, but you are of course correct.

Thanks,

Chris
  Reply With Quote
Old 14th June 2004, 08:37 AM   #29
ekaerin is offline ekaerin  Sweden
diyAudio Member
 
ekaerin's Avatar
 
Join Date: Jun 2004
Location: Stockholm
Hi Chris and everyone else,

To my experience you can't beat a driver IC. Although EVA showed that
it is possible to do with discretes.
When driving large die FETs at high speed the losses in the drivers will
be significant and to my believe it is hard to find discrete components
that can be fast enough and have the dissipation you need.
Also, the layout is hard to keep tight even with the ICs.

Regarding the tailing effects maybe some volts negative drive at turn off
may improve. I have never used that on FETs however and don't know
if it takes the tail off.

Last, I get inspired by your courage. No scope doing switch-stuff, thats brave!
At home I have some old Philips and TEK but they don't do much good either.

I append a scope-shot from a dc/dc using Warp 2 IGBTs and IR2113s driver that shows
some tailing effect for thoose that have not seen this live.

The IGBTs are thrown out now.

/ Regards Mattias
Attached Images
File Type: gif tail.gif (13.8 KB, 107 views)
  Reply With Quote
Old 14th June 2004, 09:20 AM   #30
Account Disabled
 
Join Date: Feb 2004
Send a message via AIM to classd4sure Send a message via MSN to classd4sure
Quote:
Originally posted by ekaerin
Hi Chris and everyone else,

To my experience you can't beat a driver IC. Although EVA showed that
it is possible to do with discretes.
When driving large die FETs at high speed the losses in the drivers will
be significant and to my believe it is hard to find discrete components
that can be fast enough and have the dissipation you need.
Also, the layout is hard to keep tight even with the ICs.

Regarding the tailing effects maybe some volts negative drive at turn off
may improve. I have never used that on FETs however and don't know
if it takes the tail off.

Last, I get inspired by your courage. No scope doing switch-stuff, thats brave!
At home I have some old Philips and TEK but they don't do much good either.

I append a scope-shot from a dc/dc using Warp 2 IGBTs and IR2113s driver that shows
some tailing effect for thoose that have not seen this live.

The IGBTs are thrown out now.

/ Regards Mattias
Hi there,

Do you find your at home scopes too slow? If I did get one it would be have to be the cheapest I can find, but I guess a 1Mhz scope wouldn't cut it would it.

Do those tail currents contribute to "supply pumping"? I just saw a diagram/pic of supply pumping in action and it looks strikingly like what I have going on...which I find curious as I have no capacitors to be pumped as I'm only using a VDC source.

If I waited until I had a scope alot of time would be going down the drain when I could be learning something..

I've gotten alot better with spice lately anyway, learnt alot about driving mosfets, no parts have gone up in smoke..

I have wired a few class D circuits just for kicks, powered them +-12 and 1 amp regulated...Just to see how it sounds.

I found qualities in the sound I liked alot, every note jumps out of the speaker with alot of life and transparancy, I can only dream of having that at 500W, for at the same time, lots of crackle, which is where I leave them at, with no scope handy.

I also shame the designs I test by just grabbing whatever parts I have around...including 741's and "best fit" caps that were cut from an old motherboard or PSU.

Gladly I dont' have these limitations in the land of spice, however it's plagued with it's own.

So as I said, I've learnt alot, it's yet to cost me a dime. I can however assure you, while I'm willing to let the smoke out of the odd component, I will protect my Cerwin Vegas at all cost, I do plan on buying a used scope off of ebay some day, but they are far from cheap aren't they.

Until then, on the brighter side of things, even a black 100 Ohm ...is still 100 Ohms.

Thanks,

Chris
  Reply With Quote

Reply


Hide this!Advertise here!
Thread Tools Search this Thread
Search this Thread:

Advanced Search

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off


Similar Threads
Thread Thread Starter Forum Replies Last Post
Poor Man's Strad peace brainerd Multi-Way 75 14th August 2012 07:53 PM
poor man reciever v1d9uy Solid State 4 6th January 2006 02:28 PM


New To Site? Need Help?

All times are GMT. The time now is 08:27 PM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2014 DragonByte Technologies Ltd.
Copyright 1999-2014 diyAudio

Content Relevant URLs by vBSEO 3.3.2