Nelson is braver than I am. I still hear the echoes of a professor I had once upon a time who said,"If you can't put your hand on the heatsink and leave it there, it's too hot." Nelson regularly uses his heatsinks for cooking.
I have used Motorola MOSFETs before. I don't know how robust they are compared to IRF parts as I don't run things as hard as Nelson does, but I will say that they were matched much, much more closely than IRF parts. No guarantees as to whether current parts are that tight.
Oh...yeah, 1/4W is fine for signal level resistors.
Grey
I have used Motorola MOSFETs before. I don't know how robust they are compared to IRF parts as I don't run things as hard as Nelson does, but I will say that they were matched much, much more closely than IRF parts. No guarantees as to whether current parts are that tight.
Oh...yeah, 1/4W is fine for signal level resistors.
Grey
I have used the ON Semi NTP60N06 but they run a little hot(TO220). I'm interested in the Infineon SPP77N06S2 for my next try www.infineon.com. Low C, good thermal properties and high gain. Do I need to consider any other important parameters? I'm very surprised I don't see more experimentation with other devices? I have never totally grasped what makes the IRF044s etc. Nelson's favored parts. Must be some parameters I did not mention???
GRollins said:
Hi grey,
Thanks for the input. I'm not sure that I want to run things too hot but I would like to know if I can get by with just the four outputs on the PCB. I do want to run a 4ohm load. I should have +-16v after regulation and 600vA per channel. I would also like to run without fans if possible. Some of you who have finished one of these amps could maybe tell me if that's possible. I have ordered enough output devices to use 12 per channel if necessary.
I'm plannig to use these heatsinks. One or two per channel. Do they look as though they would be sufficient?
Thanks so much for all of the help.
Blessings, Terry
Hi Terry,
I would go with two heatsinks per channel.
Assuming about 5A bias, you'll
be looking at around 160W dissipation.
I've used heatsinks with a similar profile
(but slightly larger at 11" x 11") and they got
plenty warm at 100W.
Cheers,
Dennis
I would go with two heatsinks per channel.
Assuming about 5A bias, you'll
be looking at around 160W dissipation.
I've used heatsinks with a similar profile
(but slightly larger at 11" x 11") and they got
plenty warm at 100W.
Cheers,
Dennis
Dennis Hui said:
Hi Dennis,
Thanks for the responce. I don't think I'll be at 100W with the 16V rails I'll have.
Did you use more than 4 outputs per channel?
Thanks, Terry
still4given said:
Hi Terry,
Actually I've not built an Aleph-X. My comment on the
heatsinks was based on my A30. The dissipation was
about 100W per channel, with a single heatsink
and 6 output devices.
Cheers,
Dennis
Hi,
Terry 16V rail and 5A bias give 160watts dissipation........Even 4A bias (32 watts 4 Ohm) will give 128 watts dissipation.
sorry,
William
Terry 16V rail and 5A bias give 160watts dissipation........Even 4A bias (32 watts 4 Ohm) will give 128 watts dissipation.
sorry,
William
wuffwaff said:
Thanks Guys,
I can't figure out how to tell how much heatsink I need by knowing the dissapation.
These heatsinks are 11" X almost 10" with 2 1/4" deep fins. How much should one of them be able to dissapate?
Thanks, Terry
OK, I punched in the dimention of the heatink I'm buying into the spreadsheet that Rod Elliot developed and it give me a rating of 0.1317 degC/watt.
Unfortunately, I don't have a clue what to do with this information. Is there a way to use this number in the AXE-1 wpread sheet?
Thanks, Terry
Unfortunately, I don't have a clue what to do with this information. Is there a way to use this number in the AXE-1 wpread sheet?
Thanks, Terry
Terry,
Multiply that value with your power requirement and you get the heatsink temp rise from ambient.
Multiply that value with your power requirement and you get the heatsink temp rise from ambient.
Pass DIY Addict
Joined 2000
Paid Member
Terry, I have found it necessary to include a "fudge factor" when calculating heat sink sizes that are necessary.
It seems that due to a number of thermal transfer inefficiencies you need to incorporate a "fudge factor" to your heat sink calculations. Much of the math behind calculating junction to case, case to heatsink, and heatsink to air conductance rates is based upon ideal, theoretical circumstances - circumstances which are never achieved in practice. This conclusion is based upon two examples: my a40 amp and the Pass Labs Aleph2 monoblock amp.
Example 1: For my a40 amp, the design calls for heatsinking rated at .25 c/w for each channel in order to limit the theoretical temperature rise of the heat sink to 25c while dissipating 100watts for that channel. Using the formula of 25c temp rise / (c/w rating of the heat sink) in this case yeilds 25/0.25 = 100w. Thus, it looks like 0.25c/w of heatsinking per channel hits our expected heatsinking demand exactly: we can dissipate 100w of power while limiting thermal rise to 25c. Multiplying this by 4 (because I used 4 heatsinks per channel) dictates that we need 4 heatsinks, each rated at 1.0c/w - one heatsink per output transistor. However, for my completed amp, I used heatsinks rated at 0.67 for each output transistor and still have a 24 to 25c temperature rise. Thus, theory calls for 1.0c/w per transistor, while practice shows us that 0.67 is really necessary to keep thermal rise below 25c. Thus, we need to derate the capability of the heatsink to approximately 70% of its claimed dissipation.
Example 2: The Pass Labs Aleph2 monoblock delivers 100wpc into an 8 ohm load and dissipates 300 watts of heat all of the time. Nelson indicates that the Aleph2 requires heatsinking of 0.06c/w per monoblock. Using the theoretical calculation dissipating 300 watts of power should require 0.083 c/w worth of heatsinking to limit thermal rise of the heatsinks to 25c above ambient. Derating the theoretical result of 0.083c/w to 75% provides us with a more realistic figure of 0.0625c/w - much closer to the target value of 0.06c/w indicated by Nelson.
So, for the conclusion: Perform your calculation of the power that you need to dissipate, determine the appropriate sized heatsink, then derate its dissipation to 75% of the calculated dissipation rating. This is the real size of a heatsink that you need - unfortunately, you alway need more heatsinking that you think!
Just some of my observations... Please feel free to object
It seems that due to a number of thermal transfer inefficiencies you need to incorporate a "fudge factor" to your heat sink calculations. Much of the math behind calculating junction to case, case to heatsink, and heatsink to air conductance rates is based upon ideal, theoretical circumstances - circumstances which are never achieved in practice. This conclusion is based upon two examples: my a40 amp and the Pass Labs Aleph2 monoblock amp.
Example 1: For my a40 amp, the design calls for heatsinking rated at .25 c/w for each channel in order to limit the theoretical temperature rise of the heat sink to 25c while dissipating 100watts for that channel. Using the formula of 25c temp rise / (c/w rating of the heat sink) in this case yeilds 25/0.25 = 100w. Thus, it looks like 0.25c/w of heatsinking per channel hits our expected heatsinking demand exactly: we can dissipate 100w of power while limiting thermal rise to 25c. Multiplying this by 4 (because I used 4 heatsinks per channel) dictates that we need 4 heatsinks, each rated at 1.0c/w - one heatsink per output transistor. However, for my completed amp, I used heatsinks rated at 0.67 for each output transistor and still have a 24 to 25c temperature rise. Thus, theory calls for 1.0c/w per transistor, while practice shows us that 0.67 is really necessary to keep thermal rise below 25c. Thus, we need to derate the capability of the heatsink to approximately 70% of its claimed dissipation.
Example 2: The Pass Labs Aleph2 monoblock delivers 100wpc into an 8 ohm load and dissipates 300 watts of heat all of the time. Nelson indicates that the Aleph2 requires heatsinking of 0.06c/w per monoblock. Using the theoretical calculation dissipating 300 watts of power should require 0.083 c/w worth of heatsinking to limit thermal rise of the heatsinks to 25c above ambient. Derating the theoretical result of 0.083c/w to 75% provides us with a more realistic figure of 0.0625c/w - much closer to the target value of 0.06c/w indicated by Nelson.
So, for the conclusion: Perform your calculation of the power that you need to dissipate, determine the appropriate sized heatsink, then derate its dissipation to 75% of the calculated dissipation rating. This is the real size of a heatsink that you need - unfortunately, you alway need more heatsinking that you think!
Just some of my observations... Please feel free to object
If you have the sink in hand, by far the best thing to do is attach a few heat sinkable aluminunm power resistors to it and let it cook. Use a variac and measure the power dssapated and let it cook for a few hours, then measure the temp. Be sure to mock up the exterior conditions accurately.... raise it off the floor the correct deminsion and orient the fins the way they'll be in the amp etc. This will give you real numbers you can trust. Also, if you plan to run 2 of them side by side, this will reduce dissapation.
Plug the number into cell B66 - Kelvin degrees are the same as Centigrade, just referenced to absolute zero. (IIRC, 0 C= 273 K)
That number seems optimistic, though. I'd suspect that in reality it will be a little over .2C/W
Don't forget to plug in the K of your isolators in B68 and B69. I just stuck my cheapo DMM's temp probe on the sink next to the device and again on the metal tab, and measured the difference. With pink silpads, I got around 1.4 C/W total, somewhat verifying the comment. With aluminum oxide isolators (the mouser part mentioned earlier) and a thin coat of thermal grease, I measured 1.2C/W. using a higher number is more conservative.
If you will be mounting your output devices directly to the heat sink with just an isolator, B67 should be 0.
I'd use at least two of those sinks per channel - It may be overkill, but 1 probably wouldn't let you get the bias up where it belongs.
That number seems optimistic, though. I'd suspect that in reality it will be a little over .2C/W
Don't forget to plug in the K of your isolators in B68 and B69. I just stuck my cheapo DMM's temp probe on the sink next to the device and again on the metal tab, and measured the difference. With pink silpads, I got around 1.4 C/W total, somewhat verifying the comment. With aluminum oxide isolators (the mouser part mentioned earlier) and a thin coat of thermal grease, I measured 1.2C/W. using a higher number is more conservative.
If you will be mounting your output devices directly to the heat sink with just an isolator, B67 should be 0.
I'd use at least two of those sinks per channel - It may be overkill, but 1 probably wouldn't let you get the bias up where it belongs.
BobEllis said:
I got the heatink calculator from here.
I tried to enter all the right numbers. Maybe I'm off. I'm OK with using two sinks per channel. I have 4 of them. I could probably shorten them some if I use two don't you think? If not I'll leave them like they are. Funny, My first amp was a P101, which is 200wpc and it's the shortest of all of my amps. This amp will be the lowest wpc amp I have and it will be the tallest.
I will probably use 8 outputs per channel. Would you mount all of them remotely or mount the first four on the main PCB and then mount the others off the board? I'm not sure what would be better. I think I could get better distribution on the sinks if I spread them out some.
Thanks, Terry
Rod's heat sink calculator seems to understate the lost efficiency due to height. According to NP, doubling the height of a sink only gets you about 1.4x the heat removal capacity. While Rod's sheet shows that doubling the fin height doesn't quite halve the K, it is nowhere near as pessimistic as NP's rule of thumb.
Welcome to the wonderful world of overbiased single ended class A operation.
Yes, you probably could cut the sinks down a bit - maybe to 8-9", but then you might not be able to take full advantage of your monster transformer 😉
Welcome to the wonderful world of overbiased single ended class A operation.
Yes, you probably could cut the sinks down a bit - maybe to 8-9", but then you might not be able to take full advantage of your monster transformer 😉
BobEllis said:
Thanks Bob,
Now about the outputs, would you mount them all off of the main boards or would you mount the ones that fit on the board and mount the other to the sides?
Thanks, Terry