Several Qs regarding GEM

Ask questions about the GEM here.
Post Reply
wci68
Posts: 43
Joined: Thu Dec 17, 2015 4:31 pm

Several Qs regarding GEM

Post by wci68 » Mon Feb 08, 2016 8:12 pm

Yes, I do have quite a few questions. Has anyone considered setting up a wiki to help document the items discussed in the forums?

On to my questions:
  1. I noticed that both my GEMs will blink the left (system/status) LED red each time a realtime packet is sent. Is this expected?
  2. For both volt (potential) and amp (current) values output by the GEM, is there averaging performed or are these instantaneous values from a single sample? If the former, what is the sample size? I already note that the watt-hour readings are cumulative.
  3. I noted in the GreenEye Spec Sheet: Frequency must deviate more than .2% for the selected system frequency. Should this be "must not deviate more than .2%"?
  4. On the same topic: Does 'deviate' mean that the absolute frequency must be with .2% of 50 or 60 Hz, or that it must not vary more than .2% over a short period of time? For electric supplies with a distorted sine wave (fairly common), do variations in this distortion have an impact?
  5. I noticed the realtime clock on both my GEMs tend to drift. Do these use their own crystal or do they use the power system frequency as a reference?
Regarding my questions about frequency, deviation, and distortion I have a particular interest. I have noticed for quite some time, for instance, that the frequency of my supply as measured by digital devices will vary between 59.9 and 60.1 Hz. After watching the waveform on a scope it would appear that this is partly due to variations in distortion that ever so slightly shift the peaks back and forth. Occasionally these variations will be larger, going as low as 59.8 or as high as 60.2 Hz, which would be outside of the range in the GEM specs.

What would be the impacts of a system frequency that varies/deviates beyond spec?
wci68
Posts: 43
Joined: Thu Dec 17, 2015 4:31 pm

Re: Several Qs regarding GEM

Post by wci68 » Tue Feb 16, 2016 9:42 am

Any thoughts on these?

Regarding the time on the GEMs: Are there any implications for changing the time after the GEM has been set up? For instance, if the clock runs a little fast does resetting the time backwards affect the watt-hour accumulations?

I would think that the realtime clock is not involved at all in the measurements or calculations, with the sample period itself being used for energy accumulations, but that would be purely guessing.
ben
Site Admin
Posts: 4271
Joined: Fri Jun 04, 2010 9:39 am

Re: Several Qs regarding GEM

Post by ben » Tue Feb 16, 2016 10:18 am

1. The Sys LED flashes with every packet, yeah. It can also flash if your PT is unplugged, but that would be every second.
2. They're instantaneous values.
3. It should be must not deviate.
4. It means that the accuracy in general should stay within that range.
5. It's a software implementation and not really too important. It was put in for some online services that require a timestamp.
Ben
Brultech Research Inc.
E: ben(at)brultech.com
Post Reply