I was just wondering if anyone knows how quickly the GEM samples the channels and if it uses some sort of signal averaging?
To put this into context I am building a 'power divertor' to divert excess solar power to water heating. This requires the use of a burst firing thyristor which drops cycles from the AC supply (see image). http://www.practicalcontrol.com.au/imag ... ontrol.png.
So there will be very brief periods where no current flows and periods where current does flow.
My concern is if I put a CT clamp around the supply of the control thyristor there is a risk that the GEM may sample during a period of no flow and report the current draw as 0. Or will the GEM take samples and report the average current being drawn over a 'period' of time.
How often and how quickly does the GEM sample? How many AC cycles are used the calculate the wattage? Perhaps Ben might be able to answer this?
Hope this makes sense.
Thanks, Marty.
Sampling rate or signal averaging for the GEM channels
-
mminehan
- Posts: 33
- Joined: Wed Oct 14, 2015 1:08 am
-
wci68
- Posts: 43
- Joined: Thu Dec 17, 2015 4:31 pm
Re: Sampling rate or signal averaging for the GEM channels
The GEM Hardware Installation Manual has technical specs near the end that include the sample rates.
For a proper sine wave power would be be properly and accurately measured and accumulated. Previously it was noted that the voltage and current readings are "instantaneous". I haven't seen clarity on exactly what that means (IE: RMS over a single full cycle?).
However the method by which the noted control works may cause other issues with the GEM measurements, to which someone from Brultech (Ben?) would need to reply.
For a proper sine wave power would be be properly and accurately measured and accumulated. Previously it was noted that the voltage and current readings are "instantaneous". I haven't seen clarity on exactly what that means (IE: RMS over a single full cycle?).
However the method by which the noted control works may cause other issues with the GEM measurements, to which someone from Brultech (Ben?) would need to reply.
-
mminehan
- Posts: 33
- Joined: Wed Oct 14, 2015 1:08 am
Re: Sampling rate or signal averaging for the GEM channels
Thanks wci68,
I didn't see that spec in the manual but have since found it: Current sampled @ 10.4 kHz.
But I'm not sure how many samples are used to make the calculation. I'm just hoping that it averages it over a few seconds or something like that.
The burst firing thyristor costs about $150. I guess in the end the only way to find out will be to buy it and see what happens. I might have to do dome clever programming and do something like have brief periods where the thyristor is turned off so that the system can estimate the excess power and the turn on again at the right power level for a short period.
Marty.
I didn't see that spec in the manual but have since found it: Current sampled @ 10.4 kHz.
But I'm not sure how many samples are used to make the calculation. I'm just hoping that it averages it over a few seconds or something like that.
The burst firing thyristor costs about $150. I guess in the end the only way to find out will be to buy it and see what happens. I might have to do dome clever programming and do something like have brief periods where the thyristor is turned off so that the system can estimate the excess power and the turn on again at the right power level for a short period.
Marty.