D AMP is back !!!

[luka]

Hi alexclaire

What was your input voltage peak-to-peak to get 144Vpp?

 

[alexclaire]

Hi luka,

Voltage amplification was about 17*3.6 so 61
144vpp/61 gives 2.36Vpp at the input...
could you post pics of clipping of output signal ?

 

[luka]

Hi

Could you measure it with scope so I will be 100% sure how much your gain is.

Output (no load) just before clipping. It is that small because generator clips lower period before id does upper one. 10v/div

And this one is with max input that I can give

Bender was right, those resistors were inductive, a lot. This is output 250kHz, 0.5v/div

 

[Bender.ru]

That's much better , my congratulations!

 

[luka]

Hi

Much to my surprise spikes on 0.1 resistors are the same size, that I didn't expect.

Now will be time to go to /-50v, and see what will wait for me there. Maybe some destroyed speaker

 

[Bender.ru]

Now will be time to go to /-50v, and see what will wait for me there. Maybe some destroyed speaker

i don't think so, i hope everything gonna be alright.

 

[luka]

Hi

Just if you want to see this are both gates. Both looking from -Vcc

I'll see if this will change any with higher supply voltage.

Bender where do you use your amp and with what load ?
Does it heat up any at full power?
What is your supply (Trafo)?

 

[fredos]

Add decoupling capacitor accross your mosfet supply. A good 0.47uF 250V MKT will cure all ringing you have on your squared wave! Some 0.1uF ceramic with 4.7 ohms resistor to ground help a lot too...

Fredos

 

[luka]

Hi fredos

I have 2x 0.22uF caps, but I don't think they are good for this. Will try to get better one.

Some 0.1uF ceramic with 4.7 ohms resistor to ground...

From output to ground?

 

[Bender.ru]

Hi, Luka.
I built this amp for subw, but it applicable for full range. It's only light (small power) version, now i assemble next version of this amp. Effeciency = 93..94%, you can easily calculate heating power.
My PS is classic 250W (~220V 50Hz) torroid with 20000uF capacitor per rail (48V*2 at idle), planned replace it with SMPS.

 

[luka]

Hi bender

Effeciency is really high in my too since I can drive 4 ohm speaker with ~70w with no heatsink and fets heat up only about 10C

How do you get other voltages for amp(+/-10v if you have, 12v for fet driver...)

I will skip trafo and go directly on smps, just have to make new board which will have synchronization on it, everything else is already ready.

 

[fredos]

0.1uF with 4.7 ohms form V+ and V- to ground.....

Fredos

 

[alexclaire]

Hi Luka,

voltage gain is about 36.1dB (so 64.17 voltage amplification...)

Dangerous to push an amp to clipping without any load !

It doesn't mean a thing as well because clipping depends on the load connected...

bye.

 

[luka]

Hi alexclaire

Dangerous to push an amp to clipping without any load !

Why is that?

If with no load doesn't count, I will have to out it on bigger supply, because eveytime there was bass, supply shut down. I think I have burned LM319 while testing current limit.

 

[Bender.ru]

Here is output stage without PW modulator (next version):

 

[Bender.ru]

..with UcD modulator:

 

[luka]

Hi

Wow now that looks scary . So what are the specs? Is it full bridge?

 

[Bender.ru]

No, it's classic half bridged amp, i didn't test and measure this amp, maybe tomorrow.
I'm gonna use couple in bridge mode to get 500w RMS at 8Ohm driver.

 

[luka]

Hi

I am sure that you could get that power right now with this modul, just bigger supply voltage, so my question is why don't you? I mean why would you use two moduls if you can get it from one?

 

[Bender.ru]

Because i need 90V per rail minimum, to get such power at 8Ohm load, in that case each of ballast resistors must dissipate ~2.4w..

With couple and 60V per rail, i can easily get 500w of continuous power, and ~900W of peak power at 8Ohm driver.

ps: one module provide ~500w at 4Ohm load, i think enough for home or car subwamp.