Variable supply using LM317 with 32Vrms transformer rating

[Sukhbinder]

We needed some variable power supplies back at my college where i teach and came across some old multi voltage supplies which were dead but the transformers were fine.  I planned to make a variable supply using the LM317 but the transformer has a winding of 32Vrms that i plane to use but 32Vrms means about 45V max.

Can i use the LM317T for a variable supply, variable from 0V to say 30V. the data sheet specifies the max input-output differential as 40V for the LM317T?

I was planning to use two LM317T's in series. the first can drop the voltage from say 45V to 35V along with a series pass transistor for the current boosting and then the second LM317T can work as the variable regulator. The currents needed are max 1A (using it as a lab supply for basic electronics experiments)

There is the LM317HV that supports 0-60V but that is a bit more expensive and lower availability version here.

Any suggestions for the supply design. I would like to keep the circuitry simple and small.

thanks in advance

 

[audioguru2]

The datasheet for the LM317 shows that it reduces its max output current as the voltage across it increases above 15V. So if the input is 35V and the output voltage is very low then the max output current could be as low as about 200mA.

With 15V across it and 1.5A of current then the little LM317 must have a huge heatsink that carries away 22.5W of heat. If it gets too hot then it will shutdown.

A current-boost transistor does not allow current-limiting unless a separate current-limiting transistor is added.

 

[Hero999]

Yes get the 60V version, you've no chance of doing it with the normal LM317.

Does the 32V winding have any taps? If you add a relay you can reduce the power dissipation at lower voltage settings.

The LM317 can't got below 1.25, without a negative supply and additional components.

The current output can be boosted by adding extra transistors. I've written quite a lot about this on the forum.
http://www.electronics-lab.com/forum/index.php?topic=19090.msg87897#msg87897
http://www.electronics-lab.com/forum/index.php?topic=20224.msg92354#msg92354

 

[Sukhbinder]

Thanks for the inputs, The transformer that was in the supply earlier has taps:
0-9.5v: i'll use for the negative bias for bringing the o/p of LM317 to 0V
0-10.5v: use with a 7805 for 5V supply
0-11.5v: not used till yet (could use for supply for a panel meter)
16-0-16: use with 7815/7915 for +/- 15v
0-32v: planning to use for LM317 variable supply.

I could use a lower winding but then the 0-32v goes un-utilized and there are no taps in it.

thanks again

 

[Hero999]

Then it looks like you're stuck with having to throw away large amounts of power.

Here's  circuit which shows you how to bias the LM317 to 0V with an op-amp. It's better than the circuit shown on the datasheet because the tolerance of the LM317 is taken into account. Unfortunately it does use more components.
http://www.silicontronics.com/index.php?action=ezportal;sa=page;p=19

 

[Sukhbinder]

thanks for the alternate circuit. I seemed to have a 5V DC supply also from the same transformer (the windings are different) and i will use a ICL7660 to generate a -5 for the negative bias of LM317