D AMP is back !!!

[luka]

Hi

But if I would swap pin12 and pin9 with each other too?

 

[mehrdad]

Hello

I Have Downloaded

http://www.diyaudio.com/forums/showthread.php?postid=1174802#post1174802

And

http://www.diyaudio.com/forums/showthread.php?postid=1177100#post1177100


But I Cant Open This Zip Files.

Can Anyone Extract And Place Them Here Please?

 

[mehrdad]

ok!!!

I have Solved The Problem.
Thanks To All.

 

[Bender.ru]

Hi, here is schematic (last version):

 

[mehrdad]

Thank You Very Much.

But Would You Place Here The PCB Of Your New Version Amp?

I Mean:
http://www.diyaudio.com/forums/showthread.php?postid=1189673#post1189673

Thanks.

 

[TroelsM]

Hi.

Does anyone know who sells IR20124 and PHP28NQ15?

IR20124 seems to be available but the MOSFET is not on the current Philips (NXT?) site? Could someone recommend a replacement? Perhaps one that could withstand 200V instead of 150V.

Thanks

Regards TroelsM

 

[IVX]

Hi Bender, your level shifter seems superfluously complicated. Zener+diode etc. Did you try simply dc856 +two res? BTW, BJT could be less warm, if its base =319_pin#3, emitter =319_output node, and 2kOhm from 319_output to GND. What is the goal of the R17;18? Try replacing R10 to 10kOhm, for 319 input current matching.

 

[Bender.ru]

Hi IVX, you are right. Levelshifter came from early experimental proto (where i use low voltage BJT & experiment on rise and fall slope🙂),
So there is no any reason for such complicated schematics, 1 BJT+2 res function perfectly normal. Thanks for advice, i'll try to do it.
To be honest, i didn't think about R17R18 .

mehrdad, unfortunately i haven't free time for this project, i'll post all schematics & results later.

 

[IVX]

This is simple shifter working picture (2 input 74hc load + 5pF probe) 20nS/div.

 

[Bender.ru]

Too fast, too real

 

[IVX]

Too fast, too real

What you mean? Too fast, to be real? 🙂 Actually it's absolutely normal speed for cascode.

 

[Bender.ru]

I mean, that figure really nice.
For Trise=Tfall i use extra diode+res, i think it isn't necessary, but it was interesting for me🙂.

 

[IVX]

I mean, that figure really nice.
For Trise=Tfall i use extra diode+res, i think it isn't necessary, but it was interesting for me🙂.

Indeed, this shifter will produce time assymmetry about 10nS (rising edge reached 1/2 Vcc for 10nS, falling edge for 20nS), that almost uncomparable to the total delay, e.g. lm319 already gave 80nS etc.

 

[mehrdad]

For Bender.ru:

I Have Downloaded The Spring Layot And Sow Your Pcb.🙂
I Can Only Say That`s Very Very Good.
But The Number Of Components Are Not Specified On The Layout. 😕


Thank You Very Much.

 

[Bender.ru]

Hi mehrdad, you are right, SLayout simple editor, so there is no relation between schematics & pcb project.

Important note: central pin connected at TOP layer (GND) , but NOT connected at BOTTOM layer (TOTEM POLE).

 

[Gyula]

Hi Bender!

Where can your layout be downloaded? It must be so good by the scope images.

 

[Bender.ru]

Hi Gyula, here it is http://www.diyaudio.com/forums/attachment.php?s=&postid=1177100&stamp=1175843731

 

[Gyula]

Good design. I also would like to design, but H-bridge with paralleled Schottkies at each FET, on two layers. But the track layout is still not so fine. I also would like to measure the current of the bridge, so current ringing must be minimal. Your design, with the ground plane behind the fets is so good. In the v2 design you doubled the fets, or put something new component to the board?

 

[Bender.ru]

in v2 i use ir2110 instead irs20124s, it required external OCP circuit.

 

[luka]

Hi Bender

You use IR2110?? Hope to see schematic if you have it, and board someday...

I am compliting supply for my ClassD which also has IR2110, but isn't stable yet, making it now in SMD