EEVblog Electronics Community Forum
Electronics => Projects, Designs, and Technical Stuff => Topic started by: max_torque on December 09, 2016, 06:21:13 pm
-
I'm think of using some water cooled resistors as a load, and they will have a max of around 500Vdc driving current through them.
Although the claimed resistance between the water channel and the actual resistance (wire wound) element is huge (>10E8 ohms) i'm wondering if a failure case exists from an overheated resistor getting shorted to the water channel, and hence the cooling water loop itself?
Of course, we can use silicone pipes to interlink the resistors to the water distribution manifold, but the actual water itself will be conductive (it's around 50:50 water/glycol).
Without buying anoff the shelf Isometer, anyone got any clever ideas to monitor leakage currents or measure the impedance between coolant and HV ?
(guess a CT around the DC link in-out wires a-la RDC kinda device could work??)
-
If you are afraid of such failure mode, you could consider an RCD. However, care must be taken that you buy one suitable for DC.
You can diy an megger using an isolated voltage source and (u)current meter.
Eg: backlight inverter.
But, that is far from safe compared to a real megger. Which should be in your toolbox if you work with >100V regularly.
They don't have to be expensive. http://nl.rs-online.com/web/p/insulation-testers/8937922/ (http://nl.rs-online.com/web/p/insulation-testers/8937922/)
-
Yes, we have all the usual test equipment like Meggers etc.
I'm looking for a permanent installed solution that will open a safety contactor if excessive ground leakage is determined. The Resistor bank is controlled by a micro controller anyway, so adding leakage detection should be easy.....
It certainly needs to work with DC, so no simple "common mode" style passive current transformers will work, but perhaps using a low range LEM hall effect CT might ?
Ie DC bus + and - wires both through CT, amplify output, and trip safety contactor if it goes over some set point etc.
Can probably arrange some sort of auto-zero type affair where the output of the CT is read whilst the contactor is open at power on, and that value becomes the "zero" leakage value?
-
So basically you need a DC GFCI?
You might be able to find such a thing by looking at the solar industry. For instance, here's one that's rated to 300V. Might be able to find a higher voltage one somewhere. I doubt it will be cheap.
http://www.midnitesolar.com/productPhoto.php?product_ID=252&productCatName=Breakers&productCat_ID=16&sortOrder=4&act=p (http://www.midnitesolar.com/productPhoto.php?product_ID=252&productCatName=Breakers&productCat_ID=16&sortOrder=4&act=p)
-
I'm looking for a permanent installed solution that will open a safety contactor if excessive ground leakage is determined. The Resistor bank is controlled by a micro controller anyway, so adding leakage detection should be easy.....
It certainly needs to work with DC, so no simple "common mode" style passive current transformers will work, but perhaps using a low range LEM hall effect CT might ?
The CTSR 0.3-P seems like a good candidate. Vacuumschmelze have some similar ones on the market as well, but they are more expensive.