Almost finished my M^3 + Sigma11 build, some questions.
May 29, 2011 at 9:42 AM Thread Starter Post #1 of 8

Meliboeus

1000+ Head-Fier
Joined
Apr 17, 2008
Posts
1,134
Likes
15

 
 
I've almost completed my new M^3 amp, the sigma11 unit has been configured for 36 volt output, using a 2x18v 30VA toroidal transformer, and seems to be working pretty well.
 

 
 
That's a work in progress picture taken while stuffing the M^3 board, i've used good components, sticking to the recommended ones where possible, so i went for Panasonic FCs and Nichicon HE electrolytics, Wima mks, dale resistors and so on; the gain was left as stock as well and i bypassed the bass boost circuitry. I've chosen Opa604 opamps which are rather cheap, maybe i will upgrade to opa627/637 later.
 

 
Here's a picture of the amp in its Hifi2000 chassis, the Sigma11 will be housed in an external enclosure, as you can see i used solid locking XLR connectors as power interconnects between the two boxes, they should do the job.
 

 
Everything went well with the first testing, the opamps supply voltage is +/- 17V, and i managed to adjust the bias as recommended to get about 85mA of quiescent current through the output MosFets. What puzzles me is the Dc offset at output, which is higher than the recommended maximum of 5mV ( according to the official AMB M^3 webpage ). With the 604s i get 1.5 mV on ground, but  around 10mV on both channels, what i'd like to ask you is if the 5mV limit is strictly specific to the M^3 circuit with the recommended 8610 opamps and 24V supply voltage, and higher values are still tolerable for normal use in different configurations, or if it's better not to go beyond that value as a general rule.
 
Swapping the 604s for a pair of Opa134 i got around 3.5mV of DC offset which is ok, so i don't think there's something wrong with my implementation, it's probably due to the opamps; however the amp seems to work perfectly, it's dead silent and sounds great.
 
 
How hot are the mosfets supposed to get ? Right now the amp has been running for half an hour or so and i can barely touch the heatsinks. Still colder than the Little Dot MKII.....
 
May 30, 2011 at 10:22 AM Post #3 of 8
+1. I would not be concerned with 10 mV as long as it doesn't drift much higher as the amp gets hot. 
 
As far as temperature, "too hot to touch" can give you quite a large temperature range, a lot of which is perfectly fine. If your DMM has a temperature probe you should be able to get an accurate reading, otherwise I wouldn't worry about it as long as you used the recommended heatsinks.
 
Also, if you weren't aware amb has his own set of forums on his website for questions/problems. He is excellent about responding, often getting an answer to you within minutes of posting.
 
May 31, 2011 at 8:21 AM Post #5 of 8
yeah i would definitely make a measurement of the current draw, as i recently built a pair for a balanced build and they werent very hot at all with 100ma
 
Jun 3, 2011 at 12:39 PM Post #6 of 8

 
 
The Sigma11 chassis has arrived, it's the same size as the main one, as the two will be stacked together, there's plenty of room left unused but the M^3 board was too large for the narrower Galaxy enclosures.
 

 
Here's the final result...yes, i know the dual chassis solution is probably overkill if not completely useless for an M^3 build, but i liked the idea. I've run the amp for some hours now and everything seems fine, it's dead quiet as you hear no hiss or noise even with the volume knob turned all the way up when driven by my 0404, and sounds great to me with every headphone i have at hand.
 
 
 
Jun 6, 2011 at 9:27 PM Post #7 of 8
Looks a really neat build, congrats! I like a two box solution, bettering your chances for a quieter amp.
 
Jun 8, 2011 at 2:33 PM Post #8 of 8
 
Dual casing definitely helps keeping things simpler and cleaner, cable routing and chassis earthing especially. I've taken a pic with the rest of the system just to show the fancy power button glowing, i'd say it looks nice, if only it didn't cost me as much as the Alps pot....
 
 
 
On the left there's a four channel LM1876 power amp, and it's almost half the size of the M^3,i had to disassemble it today, working with crowded enclosures can be a nightmare....
 
 
 

 

Users who are viewing this thread

Back
Top