Aleph-X builder's thread.

Official Court Jester
Joined 2003
Paid Member
I'm not saying anything for Donut and voltages ...... just saying what's better for you

with IRFP , one per quadrant , you hardly need proper matching ...... while with IRF you must double them , if you wish nice sleep

also , servo is everything but complicated

for input , if you don't have 2SJ , try finding J271 (or smd iteration of ) , buy them dozen or two and you'll find two pairs

aha , emiter follower between LTP and outputs is well worth trouble .....
 
Big thank you...

To Zen Mod...

I used the DC servo built from 2 x BC639 and injected the voltage back to the junction R24/R26.

I have not yet change the Macmillan resistors - can now be higher in value.

Absolute DC changed from 2.2v cold to less than 0.2v. Relative DC is 7mv.

Very simple change and for me it was well worth doing.

Thanks again!
 
I made a small board (20 x 25mm) that just plugs into the R24 position with a short piece of ribbon cable to pick up the voltage rails and speaker outputs. I swapped out the Macmillan resistors (from 4k7 to 100k) and reduced the LTP bias a little.

Absolute DC offset is now +/- 5mv. I think I’ll get some PCB’s made for this next time I put my next PCB order through, should be a neat little modification.
 
Hello to all of you Aleph X experts
After several years in the waiting in the “to be built area of my workshop”, I am finally completing construction of my Aleph X.

I have configured the layout using 12 Fets (IRF140) per channel – I happened to have about 50 matched pairs of these!!.

For the most mundane of reasons, I happen to have a pair of 1200VA toroid’s, with dual 28V secondary’s. This will equate to a DC voltage of about 37 volts (after rectification / filtering/ losses)

By my calculations, and assuming a bias of 6 amps, this equates to a dissipation of about 37 watts per fet, and a total dissipation of about 440 watts.

Power out will be:
8 ohms - 144 watts
4 ohms - 72 watts.

I don’t want to increase the bias current any further, as the dissipation gets out of hand.
I have no issue with heatsinking – have some VERY big brutes on hand.

My question is, has anyone built this amp using a higher supply voltage? and am I asking for trouble. Is this workable solution?

Or should I stop being a cheap skate and go buy the proper transformers.

I look forward to your comments

Kind Regards
George.
 
At first glance, the MPSA18 devices and the 9610's can both handle the rail voltage that you are proposing. I don't recall anyone building an Aleph-X with rail voltage that high, though. As there are different parts lists for a low-voltage (15v) and high-voltage (22v) version of this amp, I suspect some of the resistor values will need to be changed for going as high as 37v rails. I don't have a good starting place for you - perhaps others can make a suggestion.

The power output that you are looking at will generate some serious heat. My Aleph-X monoblocks bake off about 400w each. You can find a link in my signature. If you are looking to drive 8-ohm speakers, higher rail voltage and lower current is to your advantage. If you are looking to drive 4-ohm speakers, lower rail voltage and higher bias current will work better.
 
Last edited: