level shifters and mosfet drivers

Status
Not open for further replies.
What do you think about this ?

I added levelshifters.
 

Attachments

  • tl494revised3ob.jpg
    tl494revised3ob.jpg
    64.6 KB · Views: 638
Is it not the case that you will still blow up the 2110?

VS will swing from +25V to -25V with respect to 0V, but VB is being supplied through the diode from +12V. This means that C2 will charge to 37V when VS goes to -25V, which will damage the chip when VS swings back to +25V.
 
Oh Boy!! It's been two weeks that I didnt visit this forum and was surprised that I missed a lot of topic on this thread.

Yo Tekko, I'm so sorry for the pain and sorrow that you encounter in making this circuit but I cant help but to suggest only based on my analysis through your post:xeye: I'm beginning to admire on your courage😉 .Well, that’s what a genuine hobbyist are for. No Pain!! No Gain!!!:clown:
But if you have some time to work with the circuit once again then please try my simple suggestions.

Step 1- Remove the L,C and the fb pot and leave it vacant for a while.
Step 2-Put the fb resistor 22k across pin3 and pin15 of tl494.
Step 3-put the power supply and check the voltage on pin 3, pin15 and 16(should be 2v)if not then tweak the pot.
Step 4-Sample the sound via headphone on pin 8 in series with a 10uf cap @ low volume please..you should hear a nice clean sound. Dont worry to that 100khz pulses, you wont hear it😀
Step 5- Sample the output pulses from the source and Drain of the power fets and you should get approx 50% duty(no music @ this time)... If not then tweak the pot.....50% duty will give you half of your Vcc. If the pulses are ok then….
step 6-sample the sound from the power fets with the headphone again with a 10uf cap (@ lo volume please) and notice the sound. If the sound is like what you heard on pin 8 of 494, then........
step 7- Put 1000uf25v from the vacant slot of the L and connect the subwoofer to it, then listen to the music @ low vol please. dont worry with the 100khz pulses. you wont hear it....
The inductive reactance of the sub is too high @ 100khz so it will suck little current..If you get a clean nice sound, then that’s it!, wait for that beloved toroid suggested by fredos to replace the 1000uf cap. This should work!!😉

Good luck
ledmania
 
I know well how to use them.

I wil NEVER again try to use puter ps filter toroids for class d.

Two T106-2 toroids are ordered.

Ledmania, i actually encountered an error in your schematic, one of the inverters was on the wrong input on your later schemas, but on the right place on your first schema.

Just to let u know.

I´ll try your suggestions.
 
Tekko said:
Ledmania, Y0 heads up. The amp is finally working as it should😀 😀

The feedback was the problem all along. As it is now the fb loop is pin3 and pin15 on the 494.

It pounds loud bass and performs well fullrange.

Well, congratulations!! 😉
If you are happy, then i'm happy too!😀
But I'm going to look for another possible fb to be added @ the output of the fet itself because that is the primary goal of that circuit. Oh well!:clown: Now your smilin!!😉
 
Tekko said:
The problem turned out to be phase. The fb should goto the input on the erroramp where the 2Vref goes, otherwise it is a class d oscillator. Or you need a phase something network on the fb.

Right now i have the fb between pin 3 and 15 on the 494.

Btw join us on the irc channel.

Heres some pics of the amp: http://user.faktiskt.se/Tekko_Tubea...e sided/plugged in/work done/finally working/

Tekko, I'd been trying many times to open your file even inside my own thread about two weeks ago but it say's "Forbiden":xeye:
Please try to use "Image shack hosting" like the way I use to show my schema. You see? Even class d for sure encounter the same fate like me. LOL! LOL! LOL!
Go ahead there now to image shack and practice posting your pics:smash:
 
Status
Not open for further replies.