www.classdaudio.tk

I have AMP2, which is based on tk2350 chipset. The mosfet driver is the bigger IC that needs to be cooled. For this it has soldering slug under it. Since this is under IC you can't solder it with iron so you have one or two options.

First one is to use heat transfer paste under IC and use heatsink on top of this IC, or second you have to solder that slug to pcb.

Since second one is better for cooling is not the easiest one. But I went for this one anyway in hope that it was the right thing to do.

For this you need soldering oven or hot plate or in my case standard electric oven. I first made test with some remaining pcb that wasn't useful for anything. I did what AMP2's pdf said. And this is what came out from all that.

Day 1:

To220 broken fet soldered to PCB

So this thing is working in normal oven too, since it can heat up a lot more then you would need. I timed this to get some figures, to see what to expect.

After doing that it was time to go with the real thing and this is what came out this time

After soldering IC and all bottom SMDs and some on top

Day 2:

All SMD components are soldered + output diodes

End of day2:

Next day I soldered other IC and as writen in pdf had to test it already...

Day3: Having problems with testing resistance between Vpp and Pgnd and Vnn...
Found out that small piece of copper that came off one leg of fet when putting fets into pcb has shorted D and S of one fet. Holes are a bit too close together, so that fets would go in without force. Now resistances are ~450k and ~660k, and 505 ohm for 5v which is ok

Now I must test board

Day3:

5v line with led in draws 28.4mA from 7805.
LED turns off when supply voltage is around +/-36v
So far everything is working as it should, so now I will place last components on PCB

End of day3:

All components but toroids have been mounted, toroids will be mounted tomorrow. It is starting to look like amp now, can't wait to hear the sound, will it be what I expact it to be? Only one way to find out...

PS: with no fets and amp in mute current in Vpp rail was 0.25mA, so everything is working in this path too.

Day4, finall day:

Today toroids were first on the list... After winding them, I found out that this wires insulation wasn't coming off with lavacom (paint and wire insulation remover among other things), which meant making my own litz wire. I have some same wire size so this was great. With kit came wire with 15 stands, I used 18 so inductors should be even better. This took me about 3-4 hours.

Now the testing went great also. First I put supplys ON, check that there wasn't any shorts. After that I put amp into active mode, set offset of one channel to 0.01mV and other to 0.05mV, well below 10mV that I should . Next was signal, and speaker connections... and I was ready to play some music.

......YES IT IS ALIVE

Before this amp I was listening to LM3886, but when I played music with this amp I was : This amp is just something else, so great sound that you can't even begin to describe it.

Thank's to Tripath and 41Hz for this super amp...

Some pic...

Files:

Amp2 visits:

How to make it loud - real loud. (230 watts rms into 8 ohms without clipping from 4 ICs. How to match LM3886s for parallel operation, then bridge the matched parallel banks. If you are not familiar with bridging amplifiers (BTL configuration), read the Application Note on bridging listed left in article Application Note on bridging audio amplifiers.