having hard time with lm317/337

[elic]

Hi all,
I am using the attached design but the lm317 blows out without any reason I can identify - I already blew 6 of them. And yesterday I blew lm337
Input is +38V 0 -38V and output is adjusted to +16V 0 -16V. It is used to power a few opmaps so current is low. The capacitors are 100uf and the zener dioes are 8.2V and I put them to protect against high voltage at the output.
Any idea what's the problem?
Thanks,
Eli

[jitter]

What are the values of the resistors?

[elic]

R1 is 100ohm 1w. R3 is 2K, R2 150ohm.

[6L6]

the lm317 blows out without any reason I can identify

Are you absolutely, completely, totally and without doubt sure that you are wiring them correctly?


?
?
?



If they are in properly, then take the zeners out and see if that makes a difference.

[elic]

I made a PCB for them and I know there were times that I saw both supplies working and adjustable. BTW, when the LM317 blows out, R1 gets hot since which is strange since I would expect R3 to get most of the voltage. Could be a short in my home made PCB?

[googlyone]

Are you sure the input voltage is not exceeding 38 volts?

There are a couple of cases where the absolute input voltage matters a LOT.

-1- on startup. Depending on your reference pin bypass cap, this will take a while to charge, the length of time determined by the resistors you have used from output to reference pin.

During this time the LM317 sees pretty much the full input voltage.


-2- if you short or overtemp the lm317, not unlikely if you are playing with a circuit.

In this case the lm317 will go into protective shutdown - and the delta V between in and out is the full input voltage.


I don't have the datasheet with me, but if I recall the
Max input - output delta V is 37 volts.

Check this, as these devices are usually pretty tough.


Oh and a third subtle failure $ode (though not too likely in your case) is where the positive and negative outputs get shorted. In this case the first regulator to go into over current shutdown has its output pulled to THE OPPOSITE rail. Putting reverse diodes across the output protects against this. Your zeners will likely save you in this case.

[Mooly]

Think the above post covers most things.

Have the zeners ever failed or has one failed open circuit and you don't know it? I would replace those with just a single reversed diode on each output, say a 1n4004

Are there any electrolytics connected rail to rail on the circuit it supplies rather then rail to ground ?

[GLooP]

What is the current the PS need to handle? Having to lower the voltage with more than 20V is not quite optimal. LM317 can handle up to about 20W of dissipation and you will need some hefty heatsink for that. Are your chips insulated from the heatsink?

[sofaspud]

zener dioes are 8.2V and I put them to protect against high voltage at the output.

Huh? What high voltage?
You've made history in my book, blowing a half dozen of these chips.
Are you sure you are blowing the chips? They have overload protection and shut themselves down. Thinking about it... I'd bet you aren't blowing them, you just think you are. Might that be correct?

edit: I don't think your resistor values are correct.
Vout = 1+(R3/R1)+0.1 = 1+(2k/150)+0.1
You'll have to show me how you got 16 volts from that, or show me where I'm screwing up.
PS: and let's dispense with that 0.1 volt error term from the 50uA adjust pin

[elic]

sofaspud: you forgot to multiply the ration in 1.2V to get around the required values.
as for your assumption I did not actually blow them -- could be. I ruled that out since after that happened, I let the circuit rest for some time before connecting power again and got the same results but I could be wrong.