Breadboard Class D!!

[kartino]

Chis's UCD without pre IC

Hi,

I have fun with Chris's UCD.

I already create PCB but it is very DIYer version, single layer with non SMD component. I force the size to the smallest one.

I remove the pre IC since I lazy to create. You know I create it with AutoCAD, how can?

I tried this amp but maybe because I got compnonent as I have, it was work until I hold the before filter area. Ough.... it is very sensity, the high side mosfet and the charging gate BJT blow followed by the other mosfet.

It is very fun and I will try with new component.

have fun!

Kartino

 

[kartino]

The component side

Please ignore some diode at gate mosfet, it is for provision only.

 

[kartino]

Guide to fix the component

This is the component view with track. Easy to guide you to fix.

I create the pcb with special paper and using soldering iron after the pcb printed on the paper. The paper is easy to find here.

 

[kartino]

Sorry, miss attachement

I miss the attachment, here is!

 

[kartino]

This is the simulation

And the last, this is the simulation.

 

[kartino]

Hi,

I am forget with R5 and R10, that shall at least 10ohm. I use 15 ohm for first trial. Beware that if you use 5ohm, the mosfet will blow.

br
kartino

 

[classd4sure]

cool

What values did you end up with in the feedback loop, and I see you made some minor changes to the driver. Those diode drops could end up holding the gate source voltage higher than the turn off transistor on its own, what power level did you go for?

Nice work. How does it sound?
Chris

 

[nitrate]

Hello everyone,

Last night i had my first attempt at a self oscillating design. I just created the output stage ( 12 single supply for now ) and the driver. I sent this into oscillation to see what happend. The thing oscillated but it was like 30/70% ratio and it went so fast that it destroyed the last of my maxim gate drivers I knew i shouldn't have put it in yet, they are rarther delecate. I tried again with a half duff driver ( works with small loads ) and slowed down the oscillation by inserting a cap in the feedback path. I got it to tick at around 600KHz but it was very unstable. Just putting my hand near the circuit caused it to self modulate its frequency and the amp seems to go through a warm up stage. If anyone knows of any good articles or schematics on self osc input/controll stages please let me know.

Have fun all, and try not to waste as many component's as i do

Regards
Mad.P

 

[Kuzmenko]

I can see bcp, silkscreen and no scheme..
So where i can get schematic of these UcD in viewable format i.e. bmp/jpg/pdf/etc.?

 

[nitrate]

Eh? UcD?

 

[Kuzmenko]

Yes, that UcD, which pcb was shown some msg before..

 

[rfasic]

To Kuzmenko:

Just go there

http://www.linear.com/company/software.jsp

and download this amazing little simulator For free.
LTSpice/Switcher CADIII
Thanks Mr Linear!
Your will be then able to open Kartino's file .asc
and enjoy both schematics and simulation.

Regards

Eric

 

[classd4sure]

I just hope everyone realizes the intent of that circuit is for learning experience, fun, ETC. Don't try selling it!


Kartino, any chance you'd want to post those files in the thread the circuit came from? Don't want to hijack this one. Maybe a mod would be so kind as to move it to the DIY UCD thread?

Still haven't told us how it sounds. You know I haven't heard it myself yet?

Chris

 

[nitrate]

....Still trying to stablise main stage of self oscillating amp whilst listening to nice music on flawless ( in my opinion )carrier amp. I must admit after reading up in depth about sig delta modulation its looking very tempting to try... decisions decisions...,

I'll try and post the schematics of the self oscillator this weekend. Gotta get rid of the max driver i.c though, it's just too delicate at high frequency.

Long live the self builders!

Regards
Mad.P

 

[classd4sure]

Long live the self builders!

 

[kartino]

Beware! This is DIY forum, do not sell it!

Hi all,

I think my post will realize who want to feel how "a bad class d is still better produce good sound", like Chris said.

Remember that the concept is belong to someone here. I am sorry for that. I also give message al board as "Non commercial use only. So lets share your experience with this knowledge.

Anyway for the feedback I use 30K for 24V supply. With this value this amp will have high gain and will clip when input voltage over approx. 2V.

I still try to fine tuning the gate driver. But the value shall be OK. I choose the secure valu for both.

Esspecially for high side driver because the voltage refer to negative supply is 12V plus positive supply, it it very high.
As I said the high side driver and mosfet easy to blow. I think the gain for the high side driver BJT must be lower than the low side driver BJT. What do you think Chris?

I still have time to buy new component. But the sound is very good, even I only use a bad filter, about 50 turns small wire at a piece of broken ex. PC power supply transformer core. But beware do not touch the coil or the area before filter when the amp still play. The mosfet simplify to blow.

The sound feel feel very natural and very fast. It is like ferrari!!

And hi, Nitrate. Sorry for using your thread to post it. Because I see some people want to feel a class d.

I don't want to post at my own thread because I have no much time to take care the thread. You know that I live at third country. I have connection when I work at office.

br
kartino

 

[kartino]

Interest? Please create new thread!

Hi,

As I said before, because this is Nitrate's thread so there is no discussion for my post at this thread. Also I will not create new thread, because I only create the PCB and minor modification. The original design is belong to Chris.

Thanks,
kartino

 

[classd4sure]

Re: Interest? Please create new thread!

Hi,

As I said before, because this is Nitrate's thread so there is no discussion for my post at this thread. Also I will not create new thread, because I only create the PCB and minor modification. The original design is belong to Chris.

Thanks,
kartino

Hi,

You based this off some of my work, my schematic, and made your own schematic. The original design belongs to Bruno.

Regards,
Chris

 

[Kenshin]

Re: Re: Re: Re: The latter one

Thank you, I have plenty of MOSFET and Shottky diodes.

High power MOSFET is not suitable to low power applications because of their high gate charge. The gate drive current would be as big as the output current.

The reason I didn't use Baker clamp in my previous version is that I forgot it (also, I'm not aware of satuation and storage at that time ).

Low power MOSFETs are available at some big cities, but neither easy to get nor cheap. Since I want to do something easy to mimic, it's not suitable.

You can use BJT's if you're really stuck but they're far from efficient, at low power it should be OK. If you can get your hands on old computer SMPS, Monitors... anything else you can think of, you'll probably find enough MOSFETs to play with.

I see no reason you can't use a high power mosfet in a low power circuit, I am.

Samples are good too if you can get them.

If you can't get Shottky diodes for your Baker clamps, let me know when you get there, I have a two diode alternative for you, seems to work just fine.

Regards,
Chris

 

[classd4sure]

Re: Re: Re: Re: Re: The latter one

Thank you, I have plenty of MOSFET and Shottky diodes.

High power MOSFET is not suitable to low power applications because of their high gate charge. The gate drive current would be as big as the output current.

The reason I didn't use Baker clamp in my previous version is that I forgot it (also, I'm not aware of satuation and storage at that time ).

Low power MOSFETs are available at some big cities, but neither easy to get nor cheap. Since I want to do something easy to mimic, it's not suitable.

I ran old industrial IRF460's just fine. In fact I recommend them for initial testing due to their rugedness.

They run just a little warmer with the exact same driver transistors used for more efficient mosfets. Try it you'll be surprised. It might add a little delay but not enough to care about in a self oscillating amp.