Yey - LM3886 works like a charm

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
I've been reading about chipamps for quite some time now, making sure I'll have everything set up nice and sound when I start working on my own. I've had the parts in my desk for quite some time now, but today I finally got the soldering gun and got to work to finish them up.

I've a kit from chipamp.com, with a few tricks here and there to make it perfect.

So, the setup is:

2x120VA torroidal transformers with 2x24V secondaries. Those are wired trough 2A slow-blow fuses on the primary side (1 on each rail). This is plugged into 2 rectifier+smoothing boards from chipamp.com, with 3300pf WIMA caps paralleled across each AC input to lower the ringing frequency (optimum snubbing to be determined later, maybe - we'll see). I get perfect +/-34.7V on each rail. I connected and measured and nothing blew up! Awesome.

Those go through 3.15A fast-blow fuses (on each power rail) and plug into the two boards from chipamp, set up with the feedback cap on. I measured the DC offset with no source, no load and it was 5mV. So far so good, so I plugged in a 10ohm 15W resistor and measured the offset without and with source signal. Same story - awesome. So, confident, I plugged in my 8ohm, low-power test speaker and let it dance. 5mV offset on both channels, sound without any audible artifacts, everything just great!

I mean - of course, with a small fullrange speaker, without enclosure you cant hear much, but it did dance pretty well. I've high hopes for the sound of this thing.


Next is to set up a mains filter, with two thermistors for power-on surges as well. Which reminds me - should I substitute the mains fuses with 2A fast-blow, if I do that? Sounds safer to me.

After that, if everything is allright, it's enclosure and heatsinking time, while in the meantime I get hold of 2 speaker protection circuits, just in case.


I was so happy, I wanted to post right away, so no pics yet. Not like you guys don't know how those things look anyway.

I just wanted to say - thanks, for answering my noobish questions and for all the info that has been shared here. I rememberd quite a lot about simple electronics and learned quite a bit more. Cheers !
 
So - after a little more hardware was salvaged or purchased, I did more tests and here goes:

Custom-made mains filtering. It will get neater, I promise. Or I'll scrap it, alternativelly:

An externally hosted image should be here but it was not working when we last tested it.


Chips in testing, mounted on the heatsinks of the case I got. Wires everywhere, I'm going crazy:

An externally hosted image should be here but it was not working when we last tested it.



Heatsinks. I've my doubts about 'em. I hope they perform, cause I love the case

An externally hosted image should be here but it was not working when we last tested it.



3" cheapo 8ohm test speaker. If that guy is the judge - the amp kicks serious bottoms:

An externally hosted image should be here but it was not working when we last tested it.


Those guys mean more business. They're also apparently 4ohm, cause the right channel got a lot warmer than the left while testing:

An externally hosted image should be here but it was not working when we last tested it.



It's ON, baby:

An externally hosted image should be here but it was not working when we last tested it.
 
The heatsink could be okay for 8 Ohm loads, if you have one per IC. I wouldn't be so sure about 4 Ohm loads, especially since your supply voltage is quite high and you are using the isolated package. Consider adding a fan.

You should absolutely add thermal grease, otherwise the heatsinks won't help at all.
 
Mm, yes - one heatsink per chip. It could hold. It worked even without thermal grease - this was just a temporary setup for testing. I've got a whole lot of thermal paste, I won't forget!

I didn't plan on making it run 4ohm loads to begin with, that's why I got the higher voltage trafo. If it ever has to do that, we'll have to use it at very low volume.

Fans are indeed an option, but only if things go really bad. I don't like noise.
 
I think it all will work just fine, I've used two of those IC on much smaller heat sink than yours, and it worked -> still works just fine. It never becomes too hot so you can't touch it.

Remember that when it feels too hot it is only a little over 40 Celsius and for Si that is not too much.
 
Remember also that there will be ~55 W to dissipate with those rails and 4 Ohm speakers. The LM3886TF has a thermal resistance of 2 K/W, hence will be 110 K hotter than the heatsink.

With 8 Ohm speakers all will be well.

been using mono lm3886tf on very small heatsink and 4 ohm speaker on 38.5v rails ,i think 200va toroid maybe more,, spike is heard after few minutes on max volume but not gonna burn lm no matter how hard i try to burn it..i think
voltage drop kills it when people take 120va trafo on ab class amp almost 70 watts rms

sry on bad langugage

edit:and sorry i didnt saw 2009 year
 
Last edited:
It will work ok on the case, but see that you mount these not too close, so that the case is shared individually by each IC. The ICs should be isolated from the case metal if you have the non-isolated version.

Moreover if things start getting too hot, the SPIKE protection in the chip will kick in, which you will hear as distortion.

Gajanan Phadte
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.