High Voltage Ground Fault Detection circuit

Robbie Rongstad

Other Parts Discussed in Thread: OPA244, ADS1013, AMC1200, INA149

We are developing a product with a 380Vdc high resistance mid-point ground power distribution system. I need a cheap and simple ground fault detection circuit and attached below is what I have come up with so far. I need to isolate the high voltage DC from the control circuitry. A micro-controller will monitor the output of the ADS1013 ADC to detect a ground fault. If there are no shorts from the +/-190Vdc rails to chassis gnd, the output of the OPA244 will be 0V and the differential output of the AMC1200 will be 0V. If the -190V rail shorts to chassis gnd, the OPA244 will go to ~0.25V and the diff output will go to ~2V.

This doesn't have to be ultra precision, I just need a way to detect if one of the high voltage rails have shorted to chassis. I want the digital output so that I can set the "trip" point in software depending on how sensitive we need the detection to be. So the question is: are there better parts that I could use to accomplish this function?

Thanks

over 8 years ago

0 Rabab Itarsiwala over 8 years ago

TI__Expert 4595 points

Hi Robbie,The image didn't show on the browser.

0 Marek Lis over 8 years ago

TI__Guru 60570 points

Robbie,

Please take a look at INA149 which has the input common-mode voltage range of +/-275V. You could possibly use it together with a comparator for your High Voltage Ground Fault Detection circuit. If you need further assistance, please attach your circuit schematic by clicking on Insert/Edit Media button above.

0 Robbie Rongstad over 8 years ago

Prodigy 90 points

0 Robbie Rongstad over 8 years ago in reply to Marek Lis

Prodigy 90 points

I think that I have the schematic attached now. I looked at the INA149 initially, but that seems more suited to measuring small voltages with a large common mode offset. I'm trying to measure a large voltage and scale it down. Maybe I'm going at this all wrong too, but I've gone around and around and this is my best idea so far. Thanks.

Rob

0 Robbie Rongstad over 8 years ago in reply to Robbie Rongstad

Prodigy 90 points

It was suggested that I do the voltage scaling in two stages to keep the impedances higher. Updated schematic attached.

0 Marek Lis over 8 years ago in reply to Robbie Rongstad

TI__Guru 60570 points

Robbie,

The circuit you show will do what you need in terms of Ground Fault detection BUT you must add bypass caps on each supply and across 232k resistors. Also, place the Transient Voltage Suppressors or Zener diodes on each of the supply (+/-5V) and Schottky diode on each I/O pins to prevent damage to IC's – see below.

Alternatively, please take a look at the TI Power Management solutions under link below:

0 Robbie Rongstad over 8 years ago in reply to Marek Lis

Prodigy 90 points

Thanks for the response. The 232k resistors are actually internal to the source power supplies - there will be four in parallel. I agree that EOS protection is a good idea and I can add that once I get the basic functionality down. I will take a look at the link you provided an see if there is anything there I can use.

Rob

Amplifiers

Amplifiers forum

High Voltage Ground Fault Detection circuit