Aleph J-X Amp Project

To me no cap will ever be worth $275, Ive built amps for less.
IME theres no such thing as a final and ultimate DIY build, and your tastes may change too.
You may hear something one day and say thats what I want and have to do a complete rethink;-)
BTW SLR5000 who pays your gas bill😀 I hear ford will discontinue the Falcon in teh coming years, who will compete with Holden at Bathurst?
 
> just want to let you know that littlediode.com in the uk still has the 2sj109
> they are $6.99 eu each

I just got 11.19 Euros plus VAT, so more like 13.


Patrick
 

Attachments

  • 2SJ109 Little Diode.jpg
    2SJ109 Little Diode.jpg
    49.1 KB · Views: 750
Hi All

I intend to build the aleph J-X version with 12 mosfets per channel. Attached is the heat sink size. I am using Peter's board and I think extenting the mosfet out will be better. The heat sink size is 300H x 130w x 50d. Are these heat sink good enough? These are the best that I can find here.
 

Attachments

  • AJ-X.png
    AJ-X.png
    97.2 KB · Views: 658
  • AJ-X 2.png
    AJ-X 2.png
    66.8 KB · Views: 654
Well, if you are speaking from experience, I stand corrected. 😀

I have a pair of F-4's running and using 4 independent devices instead of a rectifier block, I don't run any sinking and have no heat issues. I know my F-4's aren't likely pulling as much juice, but that seems like a lot of heat. Would a higher efficiency rectifier produce less heat?
 
aleph j-x

Hi, thanks. Would it be better to move the Source resistor to the extension board as well? I have seen many people done this. Can the heatsink be reduce to 250mm height and still able to cope with the heat? Also, I like to incorporate a soft start to the PSU. The attached scheme is stolen from somewhere in the post. Is this soft start OK for the PSU?AJ-X 5.png

AJ-X 6.png

delay.jpg
 

Attachments

  • AJ-X 4.jpg
    AJ-X 4.jpg
    59.9 KB · Views: 733
Last edited:
The location of the source resistor isn't as important as the gate stopper. Place it wherever is convenient, allowing a bit of space for air circulation.

The critical number for heat is the device junction temperature. Shoot for <100C for reliability.

The thermal resistances you need to concern yourself with are:

R(jc) Junction to case
R case to isolator
R isolator to sink
R sink to air

R(jc) is from the device data sheets, the R across the isolator usually ends up somewhere around .7 C/W, and the Sink to air is from the data sheet. Adjust the sink R to account for the length if other than specified (often spec'd at 3", doubling the length usually gives around a 33% reduction in resistance) Adjust for sink temp lower than the test condition specified.

Multiply the thermal resistance sink-air by the power dissipated on the sink. If this rise is under 30C, you are OK and your sinks will stay below the 55C target (won't burn anyone who accidentally touches them)

Add up all the thermal resistances, multiply by the power dissipated on the sink. This will be the junction temperature rise above ambient. Add 25C (nominal room temp) and verify the number is under 100C. If so you are good to go, if not, bigger sinks, lower bias or more output devices are required.

Post your calculations and we can sanity check them.
 
Hi and thanks...
I have found this link : Heat Sink Temperature Calculator

From IRFP240 data:

Maximum Ambient Temperature.................30 (°C)
Maximum Junction Temperature...............150(°C)
Thermal Resistance - Junction to Case.....0.83(°C/Watt)
Thermal Resistance 1............................do not know(°C/Watt)

I do not know the thermal resistance of the heat sink..

Power.....100w(?) for aleph j-x?
Junction Temperature??(°C)

If i do not put in the value for the thermal resistance, I got a junction temperature of 113 °C. Very confusing

thanks
 
Look around suppliers to find an extrusion that matches yours reasonable closely. Based on your sketch, let's use .this profile. M&M doesn't give too much information about how they calculate Rtheta, but at 1.3C/W/3" It's probably safe to assume that at 250mm tall the net Rtheta for the section is .5-.6°C/W.

What bias did you plan to use? If idling at 100W as you mentioned, then you are looking at 50W per section. That will give you 25-30°C rise above ambient. This leaves the sinks at NP approved 50-55°C. So far so good.

With two output devices per sink, you have 25 W per device. Assume your isolator adds .7°C/W and your device is.83°C/W junction to case. Add up the Rs to 1.53 multiply by 25 W and you've got another 38C rise above heat sink temperature, a comfortable 88°C at the junction.

If you were planning 12 output devices and shooting for 100W output, then you probably ought to use the 300mm long sections or monoblocks with 4-6 sections per amp for the outputs.

HTH
 
Last edited: