I have salvaged a couple of skived heatsinks from old Carver 900 receivers. They used to hold 6 x TO-3P (2SA1264 & 2SC3181) and 2 x TO-126 (2SC1846) transistors.
If I wanted to estimate their R(theta), I was thinking to run a hefty transistor in a single-ended Class A operation, like a MoFo., and then measure the voltage & current across D-S (to get power), and the temperature at the gate pin.
Thoughts?
Kind regards,
Drew
If I wanted to estimate their R(theta), I was thinking to run a hefty transistor in a single-ended Class A operation, like a MoFo., and then measure the voltage & current across D-S (to get power), and the temperature at the gate pin.
Thoughts?
Kind regards,
Drew
Remember Carver used voltage tracking power supplies in his amps didn’t he? If so those heat sinks are not designed for the kind of class A load you’d be putting on them. The whole Carver thing is waste as little power as possible while class A wastes as much power as possible!
Why do you want to run Class A at all?
Waste of power.
Use a Class D amp instead, or AB if you want to keep the heat sinks.
Carver used some sort of triac / similar to control the power, so the heat sinks were smaller in comparison.
And go through heat sink catalogs to get a decent idea of the power rating of the ones you got in scrap.
Waste of power.
Use a Class D amp instead, or AB if you want to keep the heat sinks.
Carver used some sort of triac / similar to control the power, so the heat sinks were smaller in comparison.
And go through heat sink catalogs to get a decent idea of the power rating of the ones you got in scrap.
Sounds good, but you should make the temp measurement on the sink itself, not the heat generator, otherwise your measurement will be corrupted by additional thermal resistances.
You will need to run the setup for half an hour or more, until the final equilibrium is reached
Basically correct. But why the effort with a SS device when a simple resistor, attached to the sink, will also do?
Best regards!
Best regards!
He just wants to use the class A to heat the thing so he can measure thermal data. He didn't say he wants to use a class A for audio.
Can't people read anymore??
jan
Can't people read anymore??
jan
Suggest you attach a known power dissipating device to heatsink, your transistor is fine, also a power resistor the kind which comes embedded in its own aluminum heatsink, bolted to the one under test, then drill a small hole in the HS and place a thermocouple inside or at least a glass thermometer bulb.
Previously dab a drop of thermal paste inside the hole to improve thermal contact.
You will get Rth, then determine whether it´s enough for your amp or not.
I guess it will, even if original one was designed for an "efficient" Class AB amp, it was a BIG , hundreds of watts one; those HS must be enough for any reasonable Class A amp.
Are we talking 25W RMS or less?
Mmmhhh, I guess you are in the wrong Forum.
Try "Efficient Amps Forum" instead.
A most extraordinary statement coming from you, who continuously suggest salvaging scrapyard car speakers as proper replacement for powerful PA speakers or high quality Hi Fi ones.
As of using a known dissipation proper transistor, an aluminum heatsinked resistor is fine, I suggested it as an alternative myself, but in general diyers already have some power transistor, mica and grease right there on the bench, so why not?
Last edited:
AFAIK, heatsinks have no knowledge of whether they came from a dumpster receiver or new off the shelf. If I've ever bought a new one, I can't remember when and with the current prices of aluminum new ones probably aren't cheap. Anyway, I'd do the measurement with a big aluminum housed power resistor, well greased to the sink.
Reading the first post again, it seems the OP wants to do measurements of transistor D-S, and case temperature, in which case my remarks were way off.
The heat sink will be a temporary support, or similar.
Please suggest the best setup for a test rig / jig for doing this .
What the poster will do with the data is his decision.
The heat sink will be a temporary support, or similar.
Please suggest the best setup for a test rig / jig for doing this .
What the poster will do with the data is his decision.
I would think someone who posts about how inexpensive everything is would appreciate trying to save money. Heatsinks are more expensive than tubes - that's why I've only built one SS amp.
A Heatsink case for a class A amp is almost 400$ shipped to Canada from China. Those Carver heatsinks would cost almost 50$ as surplus here. Best if you can find a broken amp in the garbage and salvage 🙂
The number most people care about is temperature rise in degrees C per watt. Not too difficult to measure with a known input power to a resistor.
How is that done properly, Conrad?
I have an amp with two MOSFETs that each dissipate 50W, and there's a resistor and a diode. Total power about 105W for the heatsink... Temperature is almost 100°C (and I don't care because it works).
I have an amp with two MOSFETs that each dissipate 50W, and there's a resistor and a diode. Total power about 105W for the heatsink... Temperature is almost 100°C (and I don't care because it works).
Here they cost about $4 a kilo, the price of aluminum has gone up, and maybe those being an odd item, the price of some sections will be above average.
So my view is different from yours.
It will be how many kilos?
And sometimes it is a constraint, you have to design the amp with the existing heat sink in mind.
I would not know about the kind of material available in your area, and in a DIY forum creativity is important.
Engine valve covers in cast aluminum might work, for example. Drill and cut to shape.
Even mag wheels.
I have seen Papst LED coolers that look a lot like CPU coolers...another way to cool the hot bits.
So my view is different from yours.
It will be how many kilos?
And sometimes it is a constraint, you have to design the amp with the existing heat sink in mind.
I would not know about the kind of material available in your area, and in a DIY forum creativity is important.
Engine valve covers in cast aluminum might work, for example. Drill and cut to shape.
Even mag wheels.
I have seen Papst LED coolers that look a lot like CPU coolers...another way to cool the hot bits.
Last edited:
I have an auto electrician, he was saying the new LED head lamps have little fans for cooling, and when they pack up, the entire unit has to be replaced. Fans and LEDs are not available separately.
So use a bigger than needed heat sink, not a small one.
The suggestion made above by me to compare the heat sink to a current production one would help decide on the rating for the capability of the sink to dissipate power, and maybe fans would improve the performance.
That will help decide the amplifier output which can use this sink.
Scrap Aluminum would be 25 to 30% of new cost here...$1.20 or so per kilo.
Bear in mind how much air freight would cost to send a heat sink from China. That would be a large portion of the $400 it would cost delivered.
Also, so much fuel used, adding to global warming...but then so does Class A.
So use a bigger than needed heat sink, not a small one.
The suggestion made above by me to compare the heat sink to a current production one would help decide on the rating for the capability of the sink to dissipate power, and maybe fans would improve the performance.
That will help decide the amplifier output which can use this sink.
Scrap Aluminum would be 25 to 30% of new cost here...$1.20 or so per kilo.
Bear in mind how much air freight would cost to send a heat sink from China. That would be a large portion of the $400 it would cost delivered.
Also, so much fuel used, adding to global warming...but then so does Class A.
Last edited:
What ambient? And what is "almost"?
Assuming 95C above a 25C ambient, 70C rise.
70C/105W is 0.67C/W
Thank you to everyone who replied. If nothing more, I hope we are reminded that large geographic differences can have a large impact on our expectations and our perceptions. As can significant differences in experience.
As per my signature, I have a short road behind me in my diyaudio journey; considerably shorter than many who weigh in here. To those, I offer special thanks. As it happens, this isn't my only hobby, and I don't even have a dummy load built yet, nor even the resistors to do so. But I do have a small collection of TO-220 transistors (IRFP250, &ct), hence my question, as @JMFahey alluded.
Mea culpa; I did not mention that the receiver was rated at "90W into 8ohms", but I didn't think that mattered much. I don't have a specific design in mind for them yet; this is part of the reason for the question: Using what I have, what kind of amplifier could I reasonably use it for? I thought that knowing the dissipation would be helpful.
Kind regards,
Drew
As per my signature, I have a short road behind me in my diyaudio journey; considerably shorter than many who weigh in here. To those, I offer special thanks. As it happens, this isn't my only hobby, and I don't even have a dummy load built yet, nor even the resistors to do so. But I do have a small collection of TO-220 transistors (IRFP250, &ct), hence my question, as @JMFahey alluded.
Mea culpa; I did not mention that the receiver was rated at "90W into 8ohms", but I didn't think that mattered much. I don't have a specific design in mind for them yet; this is part of the reason for the question: Using what I have, what kind of amplifier could I reasonably use it for? I thought that knowing the dissipation would be helpful.
Kind regards,
Drew
Just from a very rough "rule of thumb", and on first approximation, IF your receiver was rated 90W RMS heatsinks should be useful to dissipate up to some 25W safely.
I´m talking "thermal" power, not "audio" power output.
If so, I guess you can build an up to 25W RMS Class A amplifier.
Of course, measuring rules.
I´m talking "thermal" power, not "audio" power output.
If so, I guess you can build an up to 25W RMS Class A amplifier.
Of course, measuring rules.