hello all,
i would like use a thermocouple to measure the temperature in a plasma and the metal where thermocouple is placed has negative polarisation(-600V). The problem here is if we get arcs, we see an overvoltage on the thermocouple which is dangerous for the measurement circuit. What kind of protection circuit can i use, to keep the voltage between 0 to 5mV?

 

Can you make the temperature measurement circuit a "bird on a wire?"

Re-phrased: Can you isolate the thermocouple amp & its supply and tie its ground to the -600V?

 

yes i can make it but do you think this is enough for protecting the measurement circuit?

 

Here is a link to a Burr-Brown (now owned by TI) voltage isolation amp - http://focus.ti.com/docs/prod/folders/print/iso122.html. It's good for about 2KV isolation.

 

thank you all for your help.
I hope i understand it true, isolation amp limits the output voltage at -+ 10V and has gain 1. Which means if the input voltage is less than 10V, we see same voltage at the output but if the input voltage for example 500V, output voltage will be limited at 10V?

 

The point of the isolation is to remove the -600 volts from the thermocouple signal. more accurately, it will allow the thermocouple to float on the -600 volt bias but pass the millivolt level signal to the measuring equipment.

 

The point of the isolation is to remove the -600 volts from the thermocouple signal. more accurately, it will allow the thermocouple to float on the -600 volt bias but pass the millivolt level signal to the measuring equipment.

is this circuit making what you meantioned before, If we connect same ground with thermocouple and isolation amp, we make the thermocouple float in the bias and receiving mV at the output of the isolation amp?

 

Can you use a sheathed T/C?, if so connect the sheath to the -600V, if an arc occurs it would carry it back to the source.

Rick