> Without going down the whole reverse engineering, desgn calculation
> path, does someone have a simple, rule of thumb value for emitter
> resistors to promote current sharing in paralleled bipolar transistors ?
>
> Now that I actually need 5A out of my 5A bench supply (Wonix YW-
> APS305 from Dick Smith for those who may also have one), I find that
> they have 2 2N3055's in parallel (all three leads) and invariably, one of
> them cooks and dies. Pity they put thermal sensing on the other one.


I hope that's not the DSE Q1760, 30v, 5A supply that I just bought :-)
CLAIMS to do 5A "at all selectable output voltages".

A bipolar transistor has ABOUT 0.6v Vbe when driven and more like 0.7V on a
power device under heavy drive.
If you cause 0.1v emitter drop at full current that would start to haven a
significant affect and 0.2v more so.

So if it's 5A/2 = 2.5A each then R = 0.2/2.5 = 0r8 per transistor should be
about right.
Either 0r68 or 1r0 are probably available at DSE.
Power = 2.5 x 0.2 = 0.5w so a 1w would do and 1 5w cheapo wire wound would
run cool. .

1r0 5w    R1602        in stock everywhere
0r47        R1601

No 0r68s


The 3055's dying at 2.5A is pretty unusual.
With proper heatsinking they take vast watts.
At say 30v max in and 2.5A that's 75W and non shared it's more. Even a TO3
with a good heatsink would be pressed at that level.



        Russell McMahon

--
http://www.piclist.com hint: PICList Posts must start with ONE topic:
[PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads