Body of diode is melting

Thread Starter

pradeeba

Joined Feb 27, 2008
104
Hi all,
In a batch of 25 SMPS kept in the oven with full load for 60 DegC. In that in 5 board's o/p recitifier diode starts melting like icecream after 1 hour. what may be the reason for this?
 

mik3

Joined Feb 4, 2008
4,843
Is the diode capable of working at this temperature ? Check the datasheet

Or maybe the oven is not keeping the temperature at 60 degrees but brings it at higher temperatures, broken thermostat ......
 

mik3

Joined Feb 4, 2008
4,843
Thats intersting, i never tried to put a circuit in the oven to check it at the limits. Try to put a diode alone in the oven and see what happens. Maybe the diode's temperature rises from the current through it too and in combination with the oven it melts.
 

John Luciani

Joined Apr 3, 2007
477
Hi all,
In a batch of 25 SMPS kept in the oven with full load for 60 DegC. In that in 5 board's o/p recitifier diode starts melting like icecream after 1 hour. what may be the reason for this?
How much power is being dissipated in the diode? What is the thermal resistance of
the diode? Does the diode have a heatsink?

Since you have an ambient temperature is 60DegC the diode can tolerate a maximum rise
of 115DegC (175-60). The temperature of the diode junction is

Tj = P * (Rjc + Rcs + Rsa)

where Rjc is the thermal resistance of the diode junction to the diode case
Rcs is the thermal resistance of the diode case to heatsink and
Rsa is the thermal resistance of the diode sink to air.

If there is no heatsink then replace (Rjc + Rcs + Rcs) with the junction to ambient
thermal resistance from the manufacturer's datasheet.

(* jcl *)

www.wiblocks.com
 

Thread Starter

pradeeba

Joined Feb 27, 2008
104
I had kept 25 boards only 5 board's diodes are getting melted. moreover am drawing only 200mA current. so heatsink is not necessary.
 

Wendy

Joined Mar 24, 2008
22,165
How much air flow is in this oven? Sounds like hot spots to me.

I once had a problem with a TEC (peilter cooler) under a table regulating a laser diode temperature (the diode was being tested). Every Monday the cooler had run away, what the problem was the table had a deep lip, trapping hot air. With no movement over the weekend the stagnent air overheated the hot side of the cooler and...

The answer was a fan that was plugged in for several minutes, after that people walking and working around the station did the rest.
 

John Luciani

Joined Apr 3, 2007
477
I had kept 25 boards only 5 board's diodes are getting melted. moreover am drawing only 200mA current. so heatsink is not necessary.
How do you know the current is only 200mA? Have you checked to see if there
is a ripple current that is much higher? The die area in that device is small. It would
not take a long duration current to exceed the maximum junction temperature.

Are there any airflow requirements for the supply?

(* jcl *)
 

Wendy

Joined Mar 24, 2008
22,165
I think I asked this, but is the air being moved in this oven? It matters, if it isn't then the only place that is regulated is the thermocouple/sensor.
 

Wendy

Joined Mar 24, 2008
22,165
OK, you probably are fighting hot spots, and some pretty extreme ones. Bet the problem is mostly at the top of the oven. In a well insulated oven you can have as much as 100 degree C between the top and bottom. I have spent my share of time servicing burn in ovens. You might try some measurements with a thermocouple, and think about how to circulate the air. It doesn't need to be vigorous, just enough to mix the air up, moving the air from the top to the bottom. Remember, the parts are generating heat also, and it all adds up.
 

Thread Starter

pradeeba

Joined Feb 27, 2008
104
Am sorry. The oven is Hot air oven only. but all boards were kept in bottom shelf and all the diodes are not melted only few are melted
 

Wendy

Joined Mar 24, 2008
22,165
Still, they didn't melt themselves. Have you done local area readings with a thermocouple? I'm willing to bet you have a huge variation in temperatures throughout the oven, which is why most burn in ovens circulate their air.

I've been known to take thermocouple readings through the doors of vacuum ovens, running a wire through the door seal isn't hard. In the case of the vacuum ovens, I used teflon tape (as in plumbing tape) to form a temporary seal.

Seriously, I've had ovens that overtemped just from the product. The heaters, other than warming the chamber up after it was turned on, didn't have to work a bit. This is also why most burn in ovens have vent holes, to allow for some cooling if needed.
 
Top