The Aleph-X


Those sinks look too small also. The "base" of this heat sink is not thick enough to spread the heat evenly across the entire surface. You want the base to be at least 5mm thick, preferably more. You want each fin to stand 60-80mm off of the base plate to provide greater radiating area.

This is a picture of the bottom of my amp.
An externally hosted image should be here but it was not working when we last tested it.

The aluminum base plate of the amp with all of the holes in it measures 10" by 20".

Here is another picture from the top:
An externally hosted image should be here but it was not working when we last tested it.
.

Sorry that I don't have any better pictures right now. Both of these are from my web page. Someone in the Mechanical Lab is cutting some custom top and front panels for my amp. When these are completed, I'll post some updated pictures.
 
Those look very nice! While I am by no means the last word on heatsinks, I really like HR-245 - a thick plate, lots of long fins, etc. Four of these in 150mm length would provide total cooling capacity of 0.0625c/w for a mono chassis. That would comfortably burn off 400w of power (100w into 8 ohms, 150w into 4 ohms) while keeping your transistors at a safe temperature.
 
Those look very nice! While I am by no means the last word on heatsinks, I really like HR-245 - a thick plate, lots of long fins, etc. Four of these in 150mm length would provide total cooling capacity of 0.0625c/w for a mono chassis. That would comfortably burn off 400w of power (100w into 8 ohms, 150w into 4 ohms) while keeping your transistors at a safe temperature.

Lenght is tall, right?
 
Correct! Using the image of the HR-245, the width = 300mm, the height = 83mm, and the length is measured along the top edge of any fin that sticks up from the base (not shown 2-D rendering in the diagram). Length = Tall if you stand it up so it is oriented in the manner that it would be mounted to your chassis.

It would be interesting to see if they can produce a 200mm length of this extrusion and how it would compare in price to the 150mm length.
 
F.E. here is mine HR-400 (for NP F3 amp):
 

Attachments

  • HR-400.jpg
    HR-400.jpg
    136.2 KB · Views: 841
Absolute DC Offset jumps up as I lower CCS current

This is my first post of diyaudio. I am building an Aleph X and need some help with settling the Absolute DC offset. I have read through most of the post on this subject including the Wiki and the detailed article by Eric without which I wouldn't have had the courage to get started on this amp 🙂


I managed to bring down the absolute DC Offset well below 100 mv using the the V2 pot and it remains reasonably stable, but the voltage across R23/25(390)is 4.95 volts. If I increase V2 towards reducing the current through Q5/Q7, the DC offset goes up to 7 to 8 volts and stays there.

Here are the details:

The Absolute DC Offset is near zero with V2 set such that the voltage across R23 and R25 is about 4.95 volts and I am measuring .63 volts across the .47 ohm source and output resistors. Resulting in 1.34 amps across the 8 IRF 040s that I am using per channel with the rails measuring 14.5 volts.

readings of my prototype aleph-x build.png


I would ideally like to run the amp with much lesser current, but If I bring down the voltage across R23 and R25 down to 4.5 volts using V2, the Absolute DC Offset jumps up to 7 to 8 volts and stays there. Changing V1 and V3 doesn't help in bringing down the offset.

I will be using this amp in an active setup to drive the midrange of my Emerald Physics CS 2.3 open baffle speakers. I don’t need too much power as the drivers are 8 ohms and and 96db. I would appreciate any help I can get with this problem.

I recently built an Aleph 3 to drive the tweeters and an F5 for a friend both of which sound Fab!

P1040230.JPG

P1040228.JPG

_MG_1656.JPG

_MG_1660.JPG
 
Zen Mod - Thanks for the reply.

I am using the Aleph-x Rev 1 PCB. Got it from a friend who bought them back in 2003. Here is a pic of the PCB from Eric's website.

http://www.facstaff.bucknell.edu/esantane/movies/pcb-ext.jpg


For the current source I am using a 9.1v zener with the Q6a mosfet (irf9610)

Didn't mention in my original post - I have one channel setup and working. It sounds pretty good.