My "audiophile" LM3886 approach

Jump to Latest
#2,702
Hi All,

I want to report that my LM4780 paralel current pump gives very good result. The sound is more powerfull than single 3886 current pump, the bass is deeper, soundstage more wide and the head room is more dynamic. but the offset voltage is a little bit bigger than single 3886 (single 3886 output has +- 30 mV offset, the 4780 paralel has about 60mV offset) but I think is still acceptble. Paralel 4780 has higher temperatus idle (after running for about 1 hour) rather than single 3886. I will send the pic later (I still try to find out digital camera, my old dig camera damaged), but overall I very satisfying with the result....


regards,


lukio,,,
 
#2,704
The short answer is yes, but I would not do it (unless you roll your own current pump PCB and protection) for a number of reasons. The first being that the protection circuit is not designed for that application, the other is that the amp itself is not very well suited to driving a lot of current on its own. But provided you have relatively high impedance speakers it actually might work OK, but the speaker protection circuit will not work at all.

Cheers!
Russ
 
#2,709
Should work great

Putting them in series, watch the polarity, will make a 1000 VA +24, 0 -24 VOLT AC. This will be perfect, 34 volts or so dc.
The monos will work fine with a single tranformer. Take a trick from the original, and reverse the ac going to the rectifiers on one channel.

George
 
#2,712
My fault

What I meant to say is that the two single secondary transformers can be connected to make a single dual secondary transformer.
The resulting transformer will work fine to power either the mono or stereo My_Ref circuit boards. Mauro Penasa actually recommends using a single power transformer.
This way the polarity of the ac leads going to the rectifiers can be reversed on one channel. He feels this works better than connecting up the usual way.
I have built versions with one power transformer and others with two. Cannot say for sure that two power transformers made any difference.
One thing for sure, the two 500 va 24 volt transformers will make a 1000 va +24/0/-24 transformer when connected properly in series. This will power multiple channels of My_Ref.

George
 
#2,713
Hi,

Finally I have completed one chanel of my My-Ref RevC monoblock (my second RevC). I swapped some components and did some minor changes:

1) 250VA, EI type transformer for each monoblock.
2) Snubberized diode rectifiers: Epcos polipropilene 100nF + 1R.
3) R12 swapped for 3K7 Riken-Ohm (input LP).
4) C1-C2 are now 470uF/50V instead of 220uF.
5) R2 is 10k instead of 33K.

These mods/swaps are aimed at making the amp even sweeter and more musical.

I have not compared yet to the stock RevC. This one is also totally silent despite very near location of EI TX 😎 (local brand, cheaper than toroid)

I would say it has more punch than the stock but will wait for direct comparison.

The only black point is a big turn-off "TUMP".

Can this be produced by the increased capacitance in C1-C2?

I remember this effect in some occasion with my first RevC, but now it has totally desapeared.

Any ideas welcome :angel:

I promise picks of my awful box and detailed sound description.

Again, this amp rocks!

Cheers,
M
 
#2,715
Offset Issues

What you are hearing is the effect of dc offset. When the LM3886 heats up the offset drifts a little. Check your offset when cold, and then after an hour of play.
All mine have done this, usually letting them cool down for a minute will reduce it. Sometimes 30 seconds or so will allow for a quiet turn off.
I really doubt if the size of the caps is the reason this one thumps. More likely the feedback loop or the chips themselves.

I now use the Evolution design. It has an active servo to handle offset. It always powers down silently.


George
 
#2,716
maxlorenz said:
The only black point is a big turn-off "TUMP"
Click to expand...
I don't think that is an offset problem. A small offset would produce more of a "click" than a "thump". My guess is that the power supply of the 318 starves faster than the 3886 supply does (and the 318 looses control), given the cap sizes and the quiescent current of the devices. One could size up the supply caps of the 318 section, but I would prefer triggering the 3886's mute pin as soon as AC mains ceases (that would require some additional circuitry). This way the 3886 is down immediately and will not dump current into the speakers.

Regards, Klaus
 
#2,719
Thanks guys, for the replies.

I was incomplete in my description, sorry.
*DC offset (cold) is 2mV 😎 I haven't measured it when "hot".
*Yes, I turn off immediatelly without letting the amp "rest".
*I only made one channel, so far.
*The relay part of the circuit is stock, as far as I 'm aware. Power-on relay function is OK.
*I lack the "relay part" of the schematics, though.

Ah! One idea :att'n:
I used single path (two pins) power interruptor, instead of double path (four pins) interruptor. Could this be the cause...i.e. instantaneous asymetric power rails or something.

I remember I used R-C accross the other amp's power interruptor (to prevent arcing I guess). Is it worth it here?

Hi Nordic,
Yes, I guess some of us are sensitive to capacitor influence in sound 🙁

Hi Panelhead,
I'm aware of the Evolution amp. I would love to have it...as I would love to have SKA, Twisted X overture, Symasym...etc
But I'm risking divorce right now, with three DIY systems 😀

Thanks again for your help.
M
 
You must log in or register to reply here.
Top Bottom