or it could just be quiescent dissipation.
5 chipamps need 5 times the heatsinking of one chipamp.
typical 3886 heatsink for Ta=25°C can be from 3.0C/W to 7.1C/W
For normal domestic type use I recommend double the heatsinking (cf. NS fig34) for cool operation and to delay the early onset of "Spike" protection as the chipamp heats up.
5off 3886 would require 3C/W/5 to 7.1C/W/5 i.e. 0.6C/W to 1.4C/W when the heatsink is in 25°C ambient air.
I would use 0.3C/W to 0.7C/W for domestic use. (But, I don't build multi-channel amplifiers, two is a many as I ever reach).
An interanl heatsink with Ta=50°C requires 0.42C/W when 8ohms loading with ±35Vdc is applied and double that would be 0.21C/W to match my cool operation recommendation.
5 chipamps need 5 times the heatsinking of one chipamp.
typical 3886 heatsink for Ta=25°C can be from 3.0C/W to 7.1C/W
For normal domestic type use I recommend double the heatsinking (cf. NS fig34) for cool operation and to delay the early onset of "Spike" protection as the chipamp heats up.
5off 3886 would require 3C/W/5 to 7.1C/W/5 i.e. 0.6C/W to 1.4C/W when the heatsink is in 25°C ambient air.
I would use 0.3C/W to 0.7C/W for domestic use. (But, I don't build multi-channel amplifiers, two is a many as I ever reach).
An interanl heatsink with Ta=50°C requires 0.42C/W when 8ohms loading with ±35Vdc is applied and double that would be 0.21C/W to match my cool operation recommendation.
Last edited:
OP said the heat sinks were 0.9 K/W (assuming natural air flow and delta-T of 70 ºC most likely). The idle dissipation of five LM3886es running on a, say, ±30 V supply is 3 W each, so 15 W total. It could be as much as 25 W, as the worst case idle current of an LM3886 is 85 mA.
15 * 0.9 = 13.5 degrees of temperature rise ... assuming air flow. In the chassis shown, there's hardly any air flow - even with the fan. So yeah. It's reasonable that the heat sink will get hot to the touch.
The solution is to punch some holes below and above the heat sink. An alternative solution would be to punch the bottom only and run a duct from the fan to the heat sink to direct the airflow along the fins of the heat sink.
Tom
15 * 0.9 = 13.5 degrees of temperature rise ... assuming air flow. In the chassis shown, there's hardly any air flow - even with the fan. So yeah. It's reasonable that the heat sink will get hot to the touch.
The solution is to punch some holes below and above the heat sink. An alternative solution would be to punch the bottom only and run a duct from the fan to the heat sink to direct the airflow along the fins of the heat sink.
Tom
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.