[Amps] diode junction temps and thermal resistance

[Hal W5GHZ]

Thermal compound can be highly conductive if you buy the right kind.
 
Hal

--- On Tue, 10/19/10, Kevin Normoyle <knormoyle at surfnetusa.com> wrote:


From: Kevin Normoyle <knormoyle at surfnetusa.com>
Subject: [Amps] diode junction temps and thermal resistance
To: AMPS at contesting.com
Date: Tuesday, October 19, 2010, 10:34 PM




Carl: "There is an ample front to rear aluminum shield that is more than adequate to dissipate the 8W or so in normal use."

All heat gets dissipated eventually. The question is: what is the junction temperature inside the diode and is it over 
specification as a result of the thermal issues?

The 1N2971A data sheet says 10 deg C/W thermal resistance for junction to stud. That can't change since it's the part 
packaging. The maximum allowed junction temp for the part is 175C (from the spec)

The stud to aluminum thermal resistance is probably at least 2 deg C/W. So that's 8W * (10+2) = 96 deg C rise (for the 
diode's junction)

But we have to add the temp of the aluminum chassis wall (the max temp of the wall relates to it's ability to radiate/be 
cooled by air). I'll guess the aluminum wall around the stud can be measured with an IR thermometer to be 150 deg F. 
(after some hours of 15M RTTY contesting) In the summer, with room temp at 90F, that's easily imaginable?

150F is 66 deg C, so the junction temperature would be 66 + 96 = 162 deg C. (compared to 175C allowed)

A poor path from stud to aluminum wall would be maybe 3 deg C/W. So that would add 8 deg C. to get you to 170C. 
(remember max allowed is 175 C)

I think it's fair to say that there is marginal headroom, if 8W occurs in normal use. Does 8W get dissipated by each 
zener? Since they are right next to each other, they locally heat the aluminum wall that's doing the dissipation. My 150 
deg F guess is just based on probable heating within the tube cavity (this is just memory..may be wrong)

The main point: you can do numbers on all this, and IR measurements..it's not just guessing.


Now: It may be true that none of these common zener failures are due to overtemp. If so, then what is causing them? Is 
RF causing them? or what?

I also noticed that Ameritron put the thermal compound between the case and the solder lug washer..not just between the 
case and the insulator+washer. I Thought the thermal compound was insulating? If so, than that interferes with the 
electrical connection between the diode case and the solder lug/washer.

-kevin
AD6Z





_______________________________________________
Amps mailing list
Amps at contesting.com
http://lists.contesting.com/mailman/listinfo/amps