What's a good material for a heatsink?

Thread Starter

JuliaM07

Joined Feb 16, 2020
5
I'm working on a wireless transmitter and receiver circuit at home, and the transmitter coil keeps overheating. I created a small heatsink out of 1/4 inch aluminum and I have a metal 5 ohm resistor connected just in case, but I can not leave the circuit running for over 15 min or the coil will overheat. What should I do?
 

Attachments

AlbertHall

Joined Jun 4, 2014
12,345
I was repairing a 15W VHF transmitter and I was putting my finger on things to see if anything ws hot. There was a coil, part of the transmitter output filter, and it felt hot. However it turns out that the coil wasn't hot, the RF power was heating my finger!

Are you sure that isn't the case with the coil in your transmitter?
 

jpanhalt

Joined Jan 18, 2008
11,087
If air is not good enough, you can try a fan, a heat sink with fins, a bigger heat sink, better thermal connection between device and heat sink, or even immerse oil/water, but the latter will change other things too.
 

ronsimpson

Joined Oct 7, 2019
2,989
the coil will overheat.
You need a different coil.
What is the coil's part number? Did you make it? What is the wire size?
Are you heat sinking the coil?
What is the output of the PICAXE pin-2? Frequency, duty cycle, on/off time?
 
Last edited:

Janis59

Joined Aug 21, 2017
1,834
One between many options are using the refridgerator capillary copper pipe staying between cooler screen and freezer box. It is ca 2 mm thick with 0,4 mm inner hollow channel - just a miniature copper pipe. Total length in average freezer is ca 1,5...3 meters coiled in rear. Using such wire for Your coil there is no difficult to use a water-cooling. My experience with this tells the few kW-scale thermal load is manageable there. Thus, look for scraped fridge and keep the pliers in the pocket.
 

Janis59

Joined Aug 21, 2017
1,834
RE:""Copper is also a good alternative to aluminum. It has twice the conductivity. Max.""
One way yes, however selectively for case of high and very high frequencies, where Focault swirle deapth (aka skin effect) becomes near to micron scale, there one must keep in mind that aluminum oxide is perfect insulator whilst copper oxide is semiconductor. Thus, the microwave current will immerse freely through aluminium oxide and will run at perfect and clean aluminum surface, whilst in copper the grand part of all power will run through the copper oxide layer having a catastrophic resistance. Thus for hundreds MHz un several GHz region aluminum routineously is giving better rezults of Q-factor. Be different for kHz to several MHz where copper indeed is better.
 
Last edited:

Thread Starter

JuliaM07

Joined Feb 16, 2020
5
You need a different coil.
What is the coil's part number? Did you make it? What is the wire size?
Are you heat sinking the coil?
What is the output of the PICAXE pin-2? Frequency, duty cycle, on/off time?
It's 100kHz, the duty cycle is 50%, and the on/off time is 1200 μs. I am heat sinking the coil using 1/4 inch aluminum, and the coil's part number is 15N15H06-C01-B.
 

Thread Starter

JuliaM07

Joined Feb 16, 2020
5
One between many options are using the refridgerator capillary copper pipe staying between cooler screen and freezer box. It is ca 2 mm thick with 0,4 mm inner hollow channel - just a miniature copper pipe. Total length in average freezer is ca 1,5...3 meters coiled in rear. Using such wire for Your coil there is no difficult to use a water-cooling. My experience with this tells the few kW-scale thermal load is manageable there. Thus, look for scraped fridge and keep the pliers in the pocket.
Thanks
 

DickCappels

Joined Aug 21, 2008
10,152
From the datasheet, the Q can be as low as 5.95, you can expect it to get warm if you are running much power at all. By the way, this is a coil for inductive charging, not an intentional radiator.

How hot is it? Did you measure the temperature of it?
 

Janis59

Joined Aug 21, 2017
1,834
Bit calculus: if 12V 15W means slightly over 1 Amp, then parallel tank have to do CURRENT multiplication of Q-fold. If lower of MHz Q-factor may be significantly over the 100, dont wonder if in the tank coil is running the 100 Amps !! Thus, even the capacitor must have been designed for large amperage and large reactive power, your case 100A*12V >1 kVAR. Normal larger than average pipe-sized capacitors, say diameter 6 mm and length 30-50 mm may stand for 10-fold less reactant thermal flux. They just blow-up. Average "sausage" of 2 mm thick and 10-20 mm long may stand 10, maximum 20 VAR. One of the methods may be soldering a matrix of SMD caps, if they may stand a best way 10 VAR, then matrix of 100 pieces may stand the load. However, my experience, if Q factor grows from good construction, the reactive energy grows even faster. Just caps made inefficiently may stand, but caps made be perfect blows-up immediately.
For what I tell? - when You shall apply the thermal camera to measure the Temp of coil, use it by the way to measure the temp of capacitor as well.
 

Janis59

Joined Aug 21, 2017
1,834
RE: Dick C
Very strange unnatural figure Q=6. Cant see any resistor artifically diminishing the Q. Or the coil is made of nichromium??
 

Janis59

Joined Aug 21, 2017
1,834
AlbertHall: Absolute right. Any massive sheet near the coil is detrimental for Q-factor. If any material, as ferrielectric, ferroelectric, or diamagnetic is used for fine tuning, the demand is to keep it more distant than coil 3x diameter. Ecrane box outside the 3x outhern diameter and inner screw for frequency adjust thinner than 1/3 of inner diameter of bobbin.
 

Janis59

Joined Aug 21, 2017
1,834
RE: ""circuit running for over 15 min or the coil will overheat. What should I do?""
Actually, why not to use JUST any suitable fan instead?
Only, fans be avoided are those having a Hall sensor inside as at higher EMI they may become stupid and stop. However, its not written in sky Your EMI interference is indeed so high. Simply try and shall see.
 
Top