Hello. I have a tpa3221 aliexpress special so we all know of the "quality". To save space the 4 capacitors are very close to the heatsink and I worry that the heat will very readily transfer. Is adding kapton tape on the side of the capacitors next to the heatsink a possible solution?
Thank you.
Thank you.
Just brainstorming over my head:
You could also apply heatshrink on each capacitor, cut to size (try to get the similar/same size diameter for a snug fit)
It may also act as a "heat barrier" (you may apply 2 times if you're really O.C. about it)
You could also apply heatshrink on each capacitor, cut to size (try to get the similar/same size diameter for a snug fit)
It may also act as a "heat barrier" (you may apply 2 times if you're really O.C. about it)
I doubt any plastic film will do much to prevent heating of the capacitors. Maybe a layer of aluminum foil to reflect the heat away from the caps will work.
Tom
Tom
Kapton is made to withstand high temperatures, but I don't think it's isolation properties are anything special.
If you are concerned, replacing the caps with some that are rated for a higher temperature is probably a better option.
If you are concerned, replacing the caps with some that are rated for a higher temperature is probably a better option.
What temperature does heatsink reach at the nearest point(s) to the capacitors? Can you also measure the temp of the capacitors on both sides (nearest and furthest from the heatsink)? If you can find max temp specs for the capacitors, you'll be closer to knowing if they're safe.
Forced air would work better than adding Kapton tape or heat shrink tubing. Assuming you'll run the amp for long periods of time, adding a static layer of anything that doesn't reflect heat will eventually heat the capacitors the same way (or more) as the naked capacitors. Adding layers of material to the capacitor brings the outside boundary closer to the heatsink, increasing the heat transfer.
Forced air would work better than adding Kapton tape or heat shrink tubing. Assuming you'll run the amp for long periods of time, adding a static layer of anything that doesn't reflect heat will eventually heat the capacitors the same way (or more) as the naked capacitors. Adding layers of material to the capacitor brings the outside boundary closer to the heatsink, increasing the heat transfer.
Maybe providing an actual picture can help us see how close the capacitors are to the heatsinks to better assess the distance and/or severity of the situation.
I only assumed that the capacitors are REALLY close to the heatsink (meaning, just millimeters away), that's why I thought of the heatshrink idea is the most practical and fastest way to add another layer of "skin" for protection due to the limited space (also, once shrinked, it only adds up around a millimeter in thickness), and also, it is safe to use since it is non-electrically conductive (accidental short) as well as non-heat conductive
I only assumed that the capacitors are REALLY close to the heatsink (meaning, just millimeters away), that's why I thought of the heatshrink idea is the most practical and fastest way to add another layer of "skin" for protection due to the limited space (also, once shrinked, it only adds up around a millimeter in thickness), and also, it is safe to use since it is non-electrically conductive (accidental short) as well as non-heat conductive
I know that someone did the heatsink temperature test underload but atm I can't find it. Will look a bit more. Any suggestions on what a material that could reflect the heat? Thank you
It will predominantly depend on the ambient air environment. For example, one end of the rainbow is when you put the module in a closed box then all parts on the pcb will reach about the same temp as the heatsink, and all parts will rise to quite a high local ambient inside the box. At the other end of the rainbow you place the module in front a fan, and then pretty much the heatsink and most parts on the pcb will be at the temp of the air being blown by the fan (perhaps what could be called room air).
If you orient the pcb so that the heatsink is 'vertical' (ie. the pcb was mounted to the side wall of an enclosure such that the fins pointed 'up') then free air convection is through the length of the fins and not across the ecaps, and that is a good use of the heatsink. Only a small % of an ecap will receive radiated heat from one side of the heatsink, and if most of the heat removal of heatsink is by convection then an ecap's internals are not likely to rise much above ambient.
If you wedge in some neoprene or similar (eg. silicone mat or mouse mat off-cut) between the heatsink and the caps then you reduce convection cooling in that region, although the neoprene has relatively poor thermal conductivity so not much heat would transfer through the neoprene etc.
If you orient the pcb so that the heatsink is 'vertical' (ie. the pcb was mounted to the side wall of an enclosure such that the fins pointed 'up') then free air convection is through the length of the fins and not across the ecaps, and that is a good use of the heatsink. Only a small % of an ecap will receive radiated heat from one side of the heatsink, and if most of the heat removal of heatsink is by convection then an ecap's internals are not likely to rise much above ambient.
If you wedge in some neoprene or similar (eg. silicone mat or mouse mat off-cut) between the heatsink and the caps then you reduce convection cooling in that region, although the neoprene has relatively poor thermal conductivity so not much heat would transfer through the neoprene etc.
You seem to be confusing thermal insulation with electrical insulation. Two very different properties.
Thank you for that response. This will be in a boombox which will be ported. Seems like the solution is to mount it on the back panel with the fins (edges) up!
I definitely am at this point. I was under the impression that the tape has both types of insulation properties.
Kapton is heat resistant, it is not heat insulating so it is not suitable at all.
ABS is one of the suitable materials with the best thermal insulation coefficient
https://omnexus.specialchem.com/polymer-properties/properties/thermal-insulation
it is easy to cut, it is cheap and it exists in various thicknesses
1/4"
https://www.amazon.com/Duco-Plastic-Sheet-Inch-Thick/dp/B0B7MDWGJR
1/8"
https://www.amazon.com/Duco-Plastic-Sheet-Inch-Thick/dp/B0B79JJ7WL
so in my view it is the best choice.
It will not be difficult to interpose a shaped sheet between the heatsink and caps.
ABS is one of the suitable materials with the best thermal insulation coefficient
https://omnexus.specialchem.com/polymer-properties/properties/thermal-insulation
it is easy to cut, it is cheap and it exists in various thicknesses
1/4"
https://www.amazon.com/Duco-Plastic-Sheet-Inch-Thick/dp/B0B7MDWGJR
1/8"
https://www.amazon.com/Duco-Plastic-Sheet-Inch-Thick/dp/B0B79JJ7WL
so in my view it is the best choice.
It will not be difficult to interpose a shaped sheet between the heatsink and caps.
You can move the capacitors to the opposite side of the PCB. I know, this brings some construction problems.
Or just simply replace those 85° capacitors to 105° rated types.
Or just simply replace those 85° capacitors to 105° rated types.
Thanks for the input you guys. I bought some nichicon caps that might be just slightly skinnier that are rated for 105c. I'll try orienting properly on the back wall wall of the boombox and will have a piece of ABS on hand. It will take a long time to get it installed, caps replaced and so on but I'll try to report what worked.
I'm just wondering why to adjust something that is not (still) broken replacing 4 caps who have not (still) failed and that if genuine are an appreciated brand.
Frankly I would not despise the "quality" of this board which even at a first visual inspection seems well built and I would leave it just as it is.
If you really can't wait to do "something", put the ABS sheet that does not represent any risk of being a cure that my be worse than the disease...