I'm using the LM317 and LM337 for regulation of positive and negative high voltage rails. I noticed in the datasheets (LM317, LM337) that they have a thermal cutoff, but I wonder if this could actually lead to device failure.

I use the regulators floating, so they regulate 220V down to 200V using a voltage divider on the output. The output is therefore 200V with respect to ground, and the input is 220V with respect to ground. Let's also say that I connect the output to an op-amp's power rails. If the thermal cutoff operated, would it not lead to the situation where the input was 220V, and the output 0V? The op-amp could in theory drag the power rail to 0V if one of its signal inputs were grounded, leading to 220V potential across the LM317, which far exceeds its 37V rating.

Would this still be ok for the regulator, or lead to some smoke?

  • \$\begingroup\$ Yes, it would. Floating a regulator is a big boys' game. If you do something like that, you have to know what you're doing, understand all the failure modes, and protect it from all data sheet maximums. \$\endgroup\$ – Neil_UK Jan 10 '17 at 17:58
  • \$\begingroup\$ Show your circuit. \$\endgroup\$ – Andy aka Jan 10 '17 at 19:42
  • \$\begingroup\$ @Neil_UK, thanks. I've seen some designs (e.g. this one) that use the LM317 in combination with an HV MOSFET that ensures the voltage across the LM317 is always no more than 18V. That allows it to safely be used as a temperature cut-off. I'll consider adding a MOSFET. \$\endgroup\$ – Sean Jan 11 '17 at 9:36

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.