I made a heat sink out of some thick aluminum sheet and used it for a negative voltage regulator (a 7915). The heat sink is a bit big and long, and is attached to the metal plate of the regulator which happens to be connected to the input pin. Because of this, I got worried that the heat sink might act as an antenna and attract EMI/RFI which would introduce noise to the input voltage. So I decided to insulate the heat sink by wrapping it in several layers of vinyl electrical tape. But now I'm wondering if doing this has reduced and badly impacted the heat transfer performance of the heat sink. Was this not good to do? What do you think?
Thanks!
Thanks!
Wrapping the sink in tape will render it almost useless!
Instead, use a proper insulating washer and bush designed for the device. These are available in a number of materials, but any will do for your application.
i'm still trying to work out why the regulator is connected to the (ungrounded) input terminal...what's the circuit?
Instead, use a proper insulating washer and bush designed for the device. These are available in a number of materials, but any will do for your application.
i'm still trying to work out why the regulator is connected to the (ungrounded) input terminal...what's the circuit?
first of all, any insulating material will not shield from EMI/RFI.
second, you don't need to worry. add a good bypass cap (100nF monolithic ceramic cap) across Vin and gnd will take care of EMI entering the regulator.
dnsey,
the 7915 pinout is pin1 = gnd, pin2 = Vin, pin3 = Vout. pin2 is also connected to the tab in a TO-220 case.
second, you don't need to worry. add a good bypass cap (100nF monolithic ceramic cap) across Vin and gnd will take care of EMI entering the regulator.
dnsey,
the 7915 pinout is pin1 = gnd, pin2 = Vin, pin3 = Vout. pin2 is also connected to the tab in a TO-220 case.
dnsey said:
To be clear, the heat sink is attached naked (metal to metal). Only the non-contacting part is covered with tape. I was just wondering if covering that part with tape reduces the performance of the heat sink in any way. Does it?
Thanks for the washer and bush suggestion! I didn't know such things are used for this, the newb that I am.
What kind of material are they usually made of?
to sum up:
EMI RFI won't be stopped at all by wrapping the heatsink with tape so there is no reason to have the tape to start with..
The heatsink needs to transfer heat to the air as efficiently as possible
Tape all over the sink (even though it isn't between the output device and the sink) is not going to help this process..
Get rid of the tape..
EMI RFI won't be stopped at all by wrapping the heatsink with tape so there is no reason to have the tape to start with..
The heatsink needs to transfer heat to the air as efficiently as possible
Tape all over the sink (even though it isn't between the output device and the sink) is not going to help this process..
Get rid of the tape..
Thanks, I get it now. Mica is a good thermal conductor and at the same time it is an electrical insulator. Therefore, it will allow heat transfer from regulator to heat sink but will not conduct electrical noise currents from heat sink to regulator. Beautiful! Just what the doctor ordered.
Now, I have a similar heat sink on the positive voltage regulator. Do I also need a mica washer here? I guess not, since in this case the tab is connected to ground, and any noise picked up will simply go to ground. Nevertheless, I would like to ask this question just in case there's something I don't know.
Now, I have a similar heat sink on the positive voltage regulator. Do I also need a mica washer here? I guess not, since in this case the tab is connected to ground, and any noise picked up will simply go to ground. Nevertheless, I would like to ask this question just in case there's something I don't know.
Mica insulators need to be smeared on each side with a very light coat of thermal grease to move the heat along. This is similar to that used on as used on microprocessors in computers but is a lot cheaper. The generic microprocessor grease will work..
The heat sinks of your project are presumably attached to your case and the case presumably attached to the house ground, so any RFI goes from the heatsinks right to ground.
The heat sinks of your project are presumably attached to your case and the case presumably attached to the house ground, so any RFI goes from the heatsinks right to ground.
Well if his heatsink is connected to case, it would be a disaster if one of the regulators tabs was not at ground potential, and attached without any insulator.
mtl777 I think you are worrying too much about RFI on heatsinks. The sole reason for insulators is to prevent electrical disasters as outlined above.
If your case is metal and earthed, there will be a vast reduction in any RFI getting into your circuit, and if it did, further grounding the heatsinks is unlikely to do anything as it's already into the circuit. Also, within circuits be careful of shunting RFI and other hash to ground as it can then make your ground poor quality. You need a dedicated noisy ground and audio ground in this kind of situation. However, your situation is not one of those, it's just a simple easy circuit.
mtl777 I think you are worrying too much about RFI on heatsinks. The sole reason for insulators is to prevent electrical disasters as outlined above.
If your case is metal and earthed, there will be a vast reduction in any RFI getting into your circuit, and if it did, further grounding the heatsinks is unlikely to do anything as it's already into the circuit. Also, within circuits be careful of shunting RFI and other hash to ground as it can then make your ground poor quality. You need a dedicated noisy ground and audio ground in this kind of situation. However, your situation is not one of those, it's just a simple easy circuit.
If shorting or RFI is not a concern, can I get better heat transfer performance by installing the heat sink directly without mica insulation? How good a thermal conductor is mica compared to aluminum? (If not as good then maybe it would cause a little degradation of heat transfer performance if used.) Would you still recommend using mica insulation even if shorting or RFI is totally not a concern?
This stuff can be calculated, but for one-off work, most people use rule-of-thumb.
If the heatsink is a reasonalble size, and you're not running the regulators near their limits, you'll be fine with insulators. As a check, as long as you can keep your finger on the tab once the device has been running for a few minutes, it's OK.
If the heatsink is a reasonalble size, and you're not running the regulators near their limits, you'll be fine with insulators. As a check, as long as you can keep your finger on the tab once the device has been running for a few minutes, it's OK.
My chipamp -ve supply voltage is connected to the chipamp's heatsink which directly bolted to an ext. aluminium heatsink. I don't use any insulation I opt for better/maximun heat transfer, all my heatsinks are isolated, ie stand alone no contact with chassis and adjacent heatsink.
I experience no EMI or other problems. My chipamp runs fairly cool except when I play the speakers loud. I sometimes turn on the fan(*) that attached to the heatsink, I use CPU heatsink I bought from computer swap meet, cheap, $5 EACH.
cheers.
* I will later design and install an auto sensor that will turn the fan when it exceeds certain temp.
I experience no EMI or other problems. My chipamp runs fairly cool except when I play the speakers loud. I sometimes turn on the fan(*) that attached to the heatsink, I use CPU heatsink I bought from computer swap meet, cheap, $5 EACH.
cheers.
* I will later design and install an auto sensor that will turn the fan when it exceeds certain temp.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.