|
|
Hello All,
I just recently received the GEM module and the physical install has been completed. I am currently stuck on the software portion to obtain the data that is being captured by the GEM. I have been able to initiate the IP link to the EtherX tool, along with IA software. The Engine G software has also been installed on the local computer.
So far I am lost on the part / steps to record, track, and view all of the captured data from this device. The end goal is to either have the data pushed to one of the many hosting sites, or at least remote in and view the data at my convenience. Any insight and step by step instructions on getting this device up and running would be greatly appreciated.
As a FYI: Brultech has stated using all of the above software from the ECM-1240 was possible.
Teken . . .
|
|
|
Supposedly discovery might work, but this is what I did: |
|
Hey topherhansen, Nice outline! I can make this a little easir in SEG buy helping with the discovery magic. In SEG, Create a site and work out it's site token. To do this go edit the site or go to: https://smartenergygroups.com/my_things/show_keys 2. It's not necessary to do anything else in SEG yet. 3. Then go and do all your ECM things that need to be done. The important one here is to give the ECM the SEG site token! 4. Get the ECM talking to SEG (try and use the DNS, we change servers etc from time to time ;) ) 5. Once the ECM is sending data to SEG start a discovery for your site at: https://smartenergygroups.com/tools -> Discoveries 6. If SEG is getting data, then it will create your device according to the ECM name that's coming for the node_name - in the above example this is greeeneye, and it will create all your streams! 7. Then when this is finished, it will bring up the Streams page https://smartenergygroups.com/streams Where you can go and label everything in one place! Note, I don't know what the n_ and t_ data is for... Note temperature the following stream_name as (temperature_1 22.4) (temperature_2 12.6) etc This will create celcius unit streams. Let me know if I can help out! Thanks, Sam. |
|
|
Much thanks to both of you. I shall try to configure my GEM with the steps outlined above. Right now I am having some hardware issues that Brutltech needs to address.
topherhansen,
Would you be so kind and let me know the following in your GEM. When you select the live monitoring via the GEM software. Do you notice that the stream of data stops at some point? Because right now my streaming data seems to stop anywhere from 30 seconds to 60 seconds later?
This seems to be related to when I upgraded the firmware of my GEM to the latest version. Which Ben indicates was to fix some odd issues? It appears to me that this new firmware has created another one. So I am looking for you reference to compare to what I see here.
Teken . . .
|
|
|
Hi all. I am new to this forum. I have just set up 2 GEMs in my home along with the GEM server. Currently monitoring about 62 circuits. While it is "working", the Brultech interface is very slow at the moment. Im trying to debug with Ben at Brultech, but in the meantime, I'd like to send the GEM data to the SEG site. The interface looks very good and im interested in getting it working. I've read the posts above, but not sure how to configure the GEM software to send the packets to the SEG group i created. Any further feedback you can provide? thanks! |
|
|
You will follow steps 6 and onward per topherhansen instructions to point the GEM to the SEG site.
|
|
Sam, I'm getting reliable p_ (power) values from the GEM. I have a question about the e_ (energy) streams though; I'm hoping you can help answer it, since it depends in part on how SEG treats them I think. The GEM reports wattseconds for each report that it sends out. In my case that's set to 5 seconds, so the power value (p_ streams) are computed as difference-in-wattseconds / difference-in-seconds. The e_ stream is simply the wattsecond difference / 3600 to end up with Watt Hours. That could be correct, but only if SEG integrates (sums) the e_ values over the course of an hour. I had (remporarily stopped last night due to ongoing rewiring in my house) a fairly constant p_ stream (fluctuating between 48 and 51 W). I compute the e_ stream by dividing by 3600, and the SEG website shows a 22 Wh usage.. I would have expected something like 50 Wh. Could you explain in a little more detail how e_ streams are processed (if at all) and displayed in SEG, or point me to some documentation? Thanks,
|
|
Hey Erik, He streams are processes according to their engineering units (I was a mechanical engineer once ;) ). Power is averaged, Energy is summed. I won't get into the ddimensional analysis here. I think you are getting rounding issues at 5 seconds. I recommend a rate of 45 seconds for a good balance of data, resolution and rounding issues. You may have seen this: http://smartenergygroups.com/forums/11-SEGmeter... Im not sure exactly what's going on, put perhaps try inspecting the API trace in tools and first verify the data being sent is what's being received in SEG. Let me know of something then seems awry in SEG. Sam. @samotage |
|
|
Just to clarify: When you say set it to 45 seconds. Will this include all captured data in the GEM for the 45 second duration? Or will the GEM simply push data every 45 seconds?
Teken . . .
|
|
Gem to push a single set of data every 45 seconds. With SEGmeter the Wh energy sent every update is fully integrated over the whole sample time, while the W power value is the snapshot. Im not sure what the Gem upstream software does, but this is what it 'should' do ;) Think of your car speedo. Watts / Power is the speed travelling. This varies all the time Watt hours / Energy is how far travelled. This includes all speed variations over a longer time base. Sam, @samotage |
|
Sam, I increased the interval from 5 sec (the GEM default) to 30 sec, and the rounding issues appear to be gone. Data looks good! |
|
Awesome! Great to hear :D |
|
An addendum for calculated streams! Please make sure that your calibration details are set properly for Calculated Streams, and the Calibration Factor should be 1 unless it needs tweaking. Check the child streams units and do some maths to work out if it really really needs changing. If this is set to the 0.0002777778 then the calc reslts will likely be zero or incorrect! Thansk, Sam @samotage |
|
|
Hi folks, I had a devil of a time getting discovery to work using the btmon.py script from http://lancet.mit.edu/~mwall/projects/power/. It turns out that btmon.py uses "XXXnnn" as the device name, where "XXX" is a literal "XXX" and "nnn" is the last three digits of the serial number. Discovery wasn't discovering anything, until on an impulse I changed "XXX" to "xxx". I think there is some case folding magic that needs to get fixed.
eric |
|
Hey Eric, I think I found a small case sensitivity in SEG today and it should now be fixed. Sam, @samotage |