tryonziess said:
I'd go for a group buy of some Conrad heatsinks.
http://www.conradheatsinks.com/welcome.htm
If I can locate the proper extrusion from the foundry we could buy it in 22 foot lengths and I could cut it to proper length in my shop. Suggestions on dimensions requested. I would not mind doing this. I would also benefit from a much reduced price. I shall start Googling for this.
Andrew has a good idea to help reduced shipping. Tad
Andrew has a good idea to help reduced shipping. Tad
tryonziess said:
No,
the fins would run in the wrong direction.
The extrusion would be about 12inches (300mm) wide and 1.5 to 2.5inches (40 to 60mm) deep with a 3/8inch to half inch (10 to 13mm) backplate, Fins 3/4inch to 2inches high (20mm to 50mm) tapering from 5mm to 2mm at the tip and gaps between fins of 1/4 inch to 5/16inch (6mm to 8mm) or parallel faced fins @ 10% of height.
The cut length would be about 6inches (150mm).
These suggestion would do for both the 3pair and the 5pair at most normal supply voltages. The smaller fins might suit upto 300W output and the biggest fins upto 500Woutput for non PA duty.
The biggest of those recommendations would suit most of the ClassA designs on the Forum. There could be many orders if the price and carriage come in right.
I will need some construction advice. Hopefully by me feeling stupid by asking these questions someone else has a question or two answered.
Should you secure the outptut transistors on the heatsink and then place the board down on them and then solder?
That seems like the only way to make sure everything is square and there is no stress placed on the transistor leads.
Probably a dumb question but if you solder them on the board first how do you know if they're all square and in the same plane?
Should you secure the outptut transistors on the heatsink and then place the board down on them and then solder?
That seems like the only way to make sure everything is square and there is no stress placed on the transistor leads.
Probably a dumb question but if you solder them on the board first how do you know if they're all square and in the same plane?
AndrewT said:
Hi Andrew
Thanks for the warning - the correct link for this seller is:
http://search.ebay.com/_W0QQsassZbarrredboss
By the way, it is not Dud, but Duda (my nick name).
Since I live in Washington DC, I use dudaindc as my user name. Not very creative, but does the job.
Cheers!
Hi,hags said:
stand the PCB on edge and adjust a three legged device to the height you want.
Attach a clip on heatsink to one leg on the component side and solder the other side.
Remove the clip and move on the the next device. This gives each device time to cool down and thus ensure less risk of overheating.
When all devices are one leg secure, turn the board component side up and adjust each till it is in the correct position/alignment.
Turn the PCB over to solder side up and again attach a sink to one leg and solder and move on through all the devices. Return and sink and solder the remaining leg. Finally sink and resolder the first leg to help remove any stresses due to your re-aligning. That's four times each device has been soldered and why the sink is good insurance.
jacco vermeulen said:
There you go... Good one Jacco - I should have thought of that.
It seems that some of us are waiting on the JRL clone boards to arrive without anything else to do...
-------------------------------------------------------------------------
Tad - I sent payment for 12 boards. Thanks!
-------------------------------------------------------------------------
Have a great weekend everybody!
Cheers!
AndrewT
A few posts back you made some comments regarding the thermal characteristics, or perhaps suggested dimensions, for the heatsink on this project. Jens original heatsink was quite thick--83mm. If the fin count was increased would you think a thinner sink, 50mm, would approximate his calculations. Three inch aluminum is not cheap. I do not know ANYTHING about thermal properties of aluminum, and I am probably too old to learn -- aluminium in British Isles. Thanks Tad I really appreciate the technical input on your part.
A few posts back you made some comments regarding the thermal characteristics, or perhaps suggested dimensions, for the heatsink on this project. Jens original heatsink was quite thick--83mm. If the fin count was increased would you think a thinner sink, 50mm, would approximate his calculations. Three inch aluminum is not cheap. I do not know ANYTHING about thermal properties of aluminum, and I am probably too old to learn -- aluminium in British Isles. Thanks Tad I really appreciate the technical input on your part.
Hi,
a passive heatsink requires space between the fins for air flow.
The normal recommended gap is about 6mm to 10mm.
It is possible to go down to 5mm but at that gap I think the fin efficiency will have fallen off.
But, that's not the end of the world.
There are other gains.
If one adopts 10mm back plate and 50mm fins giving a total thickness of 60mm we are in the ball park.
Now those 50mm fins only need to be 5mm thick at the root where they meet the backplate. The fins can taper, a common thickness at the tip is around 2mm to resist damage. Using these values we could space the fins at 10mm centres leaving a slightly small gap of 5mm at the root but increasing to 8mm at the tip. Most of the gap is now in the preferred passive range and we have used less aluminium getting there.
Now 300mm wide with 10mm centres is 28 middle fins and two more robust (thicker) fins at the ends. Thirty off 50mm fins each 150mm long is a lot of surface area and should give an adequate passive sink for a 5pair amp producing a lot of power.
I have a 75mm tall sink of similar proportions: 270mm wide, 10mm backplate, 27off 30mm fins at 10mm centres tapering from 5mm to 2mm and Rth s-a is about 0.65C/W.
The bigger one I described earlier for the Jens' height PCB would be about 0.35 to 0.4C/W.
a passive heatsink requires space between the fins for air flow.
The normal recommended gap is about 6mm to 10mm.
It is possible to go down to 5mm but at that gap I think the fin efficiency will have fallen off.
But, that's not the end of the world.
There are other gains.
If one adopts 10mm back plate and 50mm fins giving a total thickness of 60mm we are in the ball park.
Now those 50mm fins only need to be 5mm thick at the root where they meet the backplate. The fins can taper, a common thickness at the tip is around 2mm to resist damage. Using these values we could space the fins at 10mm centres leaving a slightly small gap of 5mm at the root but increasing to 8mm at the tip. Most of the gap is now in the preferred passive range and we have used less aluminium getting there.
Now 300mm wide with 10mm centres is 28 middle fins and two more robust (thicker) fins at the ends. Thirty off 50mm fins each 150mm long is a lot of surface area and should give an adequate passive sink for a 5pair amp producing a lot of power.
I have a 75mm tall sink of similar proportions: 270mm wide, 10mm backplate, 27off 30mm fins at 10mm centres tapering from 5mm to 2mm and Rth s-a is about 0.65C/W.
The bigger one I described earlier for the Jens' height PCB would be about 0.35 to 0.4C/W.
AndrewT
The dimensions you have so graciously provided will be easy to duplicate in my home workshop. I have found a wholesale source for 50mm thick x 150mm x whatever length I choose to cut. At 31.00 U.S. per lineal foot (approx. 300mm). This also leaves me more room for finished chassis design with sculpted front corners. This is not a cheap amp design and I do not want to make it look cheap with a tin can chassis. Something nice like a Pass, JRDG billet case. I can live with these prices and protect the bare aluminum with water based dye. This should allow adequate protection from oxidation and still provide good access to the metals grain structure for cooling.
DIY runs fairly deep in my part of the world and I have not given up on home brewed anodizing with a battery charger and some DILUTE acid. It does not appear to be a terribly complex procedure. Thanks again for your effort. TAD
The dimensions you have so graciously provided will be easy to duplicate in my home workshop. I have found a wholesale source for 50mm thick x 150mm x whatever length I choose to cut. At 31.00 U.S. per lineal foot (approx. 300mm). This also leaves me more room for finished chassis design with sculpted front corners. This is not a cheap amp design and I do not want to make it look cheap with a tin can chassis. Something nice like a Pass, JRDG billet case. I can live with these prices and protect the bare aluminum with water based dye. This should allow adequate protection from oxidation and still provide good access to the metals grain structure for cooling.
DIY runs fairly deep in my part of the world and I have not given up on home brewed anodizing with a battery charger and some DILUTE acid. It does not appear to be a terribly complex procedure. Thanks again for your effort. TAD