samsagaz said:
Did you measured Output bypassing protection circuit ?
How much DC Offset?
samsagaz said:
In the Italian forum several people had problems with transistors that doesn't had the right hfe values but were BC546 and BC639... if you had to replace the ones in kits...
These Mouser's order codes (are the ones that are working in my MyRef) are OK:
512-BC546
512-BC639
You could compare these ones parameters with yours
you probably know about some of these sites, but for newbies who are looking for guidance with C13: (in no particular order)ClaveFremen said:
http://www.head-fi.org/forums/f21/orgy-capacitors-cap-thread-284863/
http://www.humblehomemadehifi.com/Cap.html
http://www.ecp.cc/cap-notes.html
http://www.enjoythemusic.com/diy/0708/capacitor1.htm http://www.enjoythemusic.com/diy/1108/capacitor1.htm
Sonicaps and Obbligattos are always great choices, especially for their prices
Re: C9 & C21 Caps comparison
ClaveFremen have you tried the Nichicon Muse KZ for this application???
ClaveFremen said:An externally hosted image should be here but it was not working when we last tested it.
My last post on caps test had no comments but when I start something usually I try to finish it...
I've tested all caps in the photo (C9 and C21 positions) and from last post a lot has changed.
C21 position:
- MKP1837 is not the best cap here (it could be for the value, 10nF), too much highs
- Evox-Rifa MMK is not the best too, soundstage is significantly smaller, and sounds harsh
- AVX BQ (like the ones in kits but 100nF) soundstage is even smaller than MMKs but sound is smoother
- Wima MKS2 sounds good,dynamic and involving, it's smooth and has a wide and deep soundstage
C9 position:
- Rubycon ZL here is rubbish
- Panasonic FM is quite good, dinamyc and involving, but has a bit of harshness and basses are a bit bloating
- Nichicon KW is quite good, less harshness than FMs but also less dynamic and a bit confused
- Elna Silmic II (with 40 Hours burn-in) are smooth, with the widest and deepest soundstage, a bit more dynamic than FMs with a stronger and firmer bass. In comparison with KWs and FMs highs are less but I wouldn't call them veiled, IMHO Silmic's tonal balance is more correct.
I hope that this could be useful to someone.
I'm still asking myself why TP swapped MKS2s with BQs...
ClaveFremen have you tried the Nichicon Muse KZ for this application???
ClaveFremen said:
well, i fixed it. I goto the store and purchased BC546C (instead A because cant find the right value, this have a lot more of HFE but looks like works too), BC639, and 1 LM3886, after replace it the amp works!
Really dont know what happend with the other ones!, i was getting too much Offset at output, and that was the problem, the relay dont make contact for the high offset! The LM3886 was fried i think.
now i have 2 channels working, sound are great!
Russ was kind enough to post all the cad files on the original my ref thread. Uriah and I didn't want to mess with the design for fear of screwing things up. But if someone has the background and knowledge they could do a design iteration. Maybe something for another group buy?AndrewT said:
peter
after listen the amp for many hours im really happy with the sound, but i think that dont have enough bass, i compared it with other lm3886 that i have and have more bass. Why that happend? is possible that the other have more bass because have more capacitors in the PSU? in the other im using 4700uf x 10.
Now i need to goto and purchase 2 more LM3886 for the other 2 monoblicks that dont works too.
Now i need to goto and purchase 2 more LM3886 for the other 2 monoblicks that dont works too.
it's more than just the PSU capacitance.
It is also related to the RC time constants of the input of the opamp and the RC time constant at the input to the power chip and the RC time constant of the feedback network, as well as the peak current ability of the WHOLE PSU to speaker and return route.
It is also related to the RC time constants of the input of the opamp and the RC time constant at the input to the power chip and the RC time constant of the feedback network, as well as the peak current ability of the WHOLE PSU to speaker and return route.
Dear Samsagaz,samsagaz said:
I don't know why your perception of bass is different on the two amps. For starters they may have different frequency responses. The first thing would be to eliminate that as a potential application. For all I know the revC is more neutral and your ears got used to a design that artificially enhances the bass... (could be... I just don't know).
Capacitor size in the PS is a contentious subject. I found (as is so often the case) the writing (and simulation and measurement) of Rod Elliott on this subject very enlightening http://sound.westhost.com/power-supplies.htm.
Note that some of the highly regarded chip amp designs have as little as 2000uF on each rail. (Personally I think that's pushing it, but people report liking them quite a bit.)
I keep coming back to an earlier question/challenge/hypothesis (which, as you read Rod Elliott, also applies here): when people run with higher rated AC toroids they are really comparing with a setup in which the rail voltage is effectively higher (at the same nominal voltage rating of the toroid) and that is the effect they are seeing. A similar argument can be made when increasing the capacitance in the PS (over a certain range).
Anway, long winded ramble in response to your question.
peter
samsagaz said:
Are you still using the 100va trafo?
Mauro suggested 200VA MINIMUM. I am using 160 and it is good but as low as I would go FULL-RANGE or MAIN system.
The Apex.jr trafos are good for high passed units or small system amps.
Very happy that you got it working.
If your chipamp has more bass, bi-amp!.. Best of both worlds, Mauro's on top and chip-amp on bottom!
troystg said:
Yes, still using the 100Watts, just have one of the 150VA trafos when get another will change the case of the amplifier and will start using 150VA ones. The case taht i made right now are for just one heatsink, and come hot after like 10 minutes at high volume (like 80-90%), i measured it and get like 55 C
schro20 said:
Great!
widman said:
Well if maure recommends 220/stereo, i think that 100VA/channel will be not so low