0
\$\begingroup\$

I want to add reverse polarity protection on my electronic load which will be capable of sinking 20A and with a max. voltage of 50V (max. power is limited to 250W). Because I want to go with low input voltages for e.g. testing batteries, a typical solution like a PFET at the positive load input terminal is not a solution because of the V_GS_th of the PFET.

Diodes are also not suitable because of the high current and the related power dissipation.

I came across the solution of Spehro Pefhany which is shown in the picture.

enter image description here

I got one question about it: My electronic load input terminals (Vx) are going to be floating (no connection to mains earth because of a transformer). The metal case is going to be connected to mains earth.

Is there a possibility to screw anything up, if some strange connections are made?

E.g. connect the negative load input terminal to the case and attach there the positive terminal of my DUT. Now I connect the negative terminal of my DUT (which is mains earth referenced) to the positive load input terminal. That would short out Q2 and the current would flow through the body diode of Q1. Is this the only case which could cause problems?

What would happen if I connect a mains earth referenced power supply to my load terminals in a normal way (positive to positive and negative to negative)? That would connect the negative floating input terminal to mains earth and would bridge the isolation. Could this be a problem?

\$\endgroup\$
  • \$\begingroup\$ "Diodes are also not suitable because of the high current and the related power dissipation" - so you think that a reverse connected FET will not produce the same power eh? \$\endgroup\$ – Andy aka Dec 1 '16 at 15:10
  • \$\begingroup\$ If you just reverse the polarity of the DUT (Vx), the FET Q2 will block the reverse voltage. I want to know if the cases I mentioned above (connect something mains earth referenced to the load inputs or to connect a floating load input to the case ground (mains earth)) are possible and causing a problem or if I am missing something. \$\endgroup\$ – electricar Dec 1 '16 at 16:01
  • \$\begingroup\$ Your schematic makes no sense at all, so why not gives one that does? \$\endgroup\$ – WhatRoughBeast Dec 1 '16 at 16:13
  • \$\begingroup\$ My schematic is quite big and not finished yet. Just assume that the ground on the picture is isolated of the mains earth ground and a metal enclosure is placed around my circuit which is mains ground referenced. The voltage source Vx would be the power supply you want to test. \$\endgroup\$ – electricar Dec 2 '16 at 7:26
1
\$\begingroup\$

E.g. connect the negative load input terminal to the case...

Don't do that - problem solved!

connect a mains earth referenced power supply to my load terminals in a normal way... would bridge the isolation. Could this be a problem?

If you connect a ground-referenced device it will ground one of your terminals and your circuit will no longer be floating, but it's not 'bridging the isolation'. Your electronic load's internal ground can still be at a different potential, which is the only reason you need to 'isolate' it.

\$\endgroup\$
  • \$\begingroup\$ I won't do that, but I only wanted to know if I am getting this right :D Thank you for the clarification of "bridging" the isolation, it's obvious now! :) \$\endgroup\$ – electricar Dec 2 '16 at 7:23

Your Answer

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

Not the answer you're looking for? Browse other questions tagged or ask your own question.