Heat Issues with LM732 Power Supply

[Gristle McThornbody]

I plan to build a power supply based on a LM732 voltage regulator driving a couple of
2N3055's. This supply is targeted at 24 vdc at upwards of 10 amps, but I am making it adjustable.

I am more familiar with small power regulators where any excess voltage in the supply has to be dumped across the regulator (along with the accompanying heat). With a LM732-type design, is this still an issue? If I supply, say, 30 volts to the regulator (max input is 40), and my output voltage is dialed down to 13 volts, will all that power/heat have to be dissipated, or does this type of design have to deal with that problem? TIA

 

[BobK]

This is a linear regulator. Any linear regulator is simply wasting the excess power as heat. The power wasted is (Vin - Vout) * Iout.

If you want a regulator that does not waste power, look into switch mode regulators.

Bob

 

[Gristle McThornbody]

Ok, but since the LM732 cannot dissipate any heat, where does it go? The 3055's?

 

[jackorocko]

Ok, but since the LM732 cannot dissipate any heat, where does it go? The 3055's?

even linear regulators with pass transistors still dissipate power, the heat produced is just limited to something the linear regulator can handle. The rest of the current is routed through the 3055's and these will dissipated a little bit of power as well. You can still use the formula bob provided, worse case would be the max current through the regulator. You will need to provide more detail if you want specific numbers. A schematic would be nice

 

[Rleo6965]

To avoid too much heat or voltage drop on 2N3055 in DC output of 13V or lower.
Design your power transformer to have secondary winding of 18VAC and 32VAC. Place a switch or relay to select 18V winding for +13V DC and the 32V winding for +24V DC output.

 

[Gristle McThornbody]

Thanks everyone. I think I've got it now.

And Rleo, this PS will be used exclusively in a range of 18 to 24 vdc. If I were making it to use from 0 to 24 volts, I would definitely implement your suggestion.

 

[TedA]

What's an LM732?

Are you sure you are not making an FM radio?

( uA723 and uA732 are Fairchild numbers; I don't think National ever made the '732.)

Could you possibly have some digits swapped in the part number?

A linear supply is big, heavy and inefficient, but can be more practical to design and build for your own use. And it tends not to generate a lot of nasty noise.

I bet you will need to provide a driver transistor to get enough base current for your 2N3055s. For best stability, you want to keep the heat out of the regulator IC. And you might want to use more than two 2N3055s, at this power level.

You should attempt to make it so it does not overheat with an overloaded or shorted output. A really big heatsink, and possibly a fan may be in order. Possibly a thermal cutout.

Ted

 

[Gristle McThornbody]

Oops, yes I wrote the wrong part number. I'm working with an LM723 not 732. Sorry for the confusion.

This circuit uses a BD139 driver transistor between the regulator and the four 2N3055's. This will be a linear regulator because I don't want any HF noise issues. And like you said, it's simple enough that I can build it I will be using substantial heat sinks and a fan.

The circuit has short circuit protection, but I would appreciate some thoughts on how to add thermal protection. I haven't worked with that sort of thing before.


And by the way, I really appreciate all the help from everyone. This is a great forum.