I’m upgrading my service at home from a 100A 20 circuit panel to a 200A 40/80 circuit panel (US), so this is the time for me to add some monitoring. I’ve got a few questions before I make the purchase.
With the 14 circuit limit on the IoTaWatt, is it possible to slave a 2nd unit for future expansion? As it is, I’ll be running both the double range and oven 230v circuits through one CT to save inputs. It’d be nice to have one for each branch receptacle circuit too.
What’s the advantage of combining multiple circuits using a headphone splitter instead of running them through the CT? Is it when not enough room (or physically unable)? Or accuracy? I’ll miss out on some branch circuits that I figured would get derived from subtracting the difference between mains and all the rest of the CT’s.
With the panel apart, is the SplitCore 200A CT ( ECS24200)more accurate than the clamp-on? I’d do a solid core if I had the option…
And the loaded question - What’s the advantage of the emonTx vs IoTaWatt? Seems emonTx can slave multiple units, and there is temperature/pulse input. Cost would be slightly more at 4 inputs per unit… I’d really like an ethernet option instead of WiFi (would go right to a business class switch instead of my consumer grade WiFi, and no worries about someone hacking in via Ethernet). IoTaWatt has more inputs and seems like a cleaner package, and I’d probably start with one to see how it goes…
Thanks guys, I’m excited to do this! Hope to learn more coding and play with the data at some point too…
I have two IoTaWatt devices. There are different ways to aggregate the data. It really depends on what you are trying to achieve. I have been monitoring my power utilization for many years. Most people would find it pretty boring and I don’t really look at the data that often. But I still believe it has value and can be interesting.
I can see how much power my refrigerator uses and will probably be able to tell when it is not doing so well. I got it to do some demand side load management when on the generator. I have three water heaters and if they all went on at the same time on laundry day it would be too much for my generator. Now, the odds of that happening are pretty rare and I know this because of the monitoring that I have been doing.
Low power loads that are always on are a place where you can save some money. I have an old stereo. It draws ~50W, no matter the volume, within reason. That costs about $50/year if it is on 24hours/day. I have some old VCR/DVD player/recorders. Each draws about 20W when turned “off”. I no longer leave them plugged in.
Having more channels is nice, but you probably don’t really need more than the 14. I got the second one, because I really wanted to see how often the split-phase voltages differed. @overeasy will tell you, “not much” and I can confirm it . Math is easier when you have your mains on the same device.
I looked at the emonTX a long time ago. It is fine, but not as easy to use as IoTaWatt is. I have lots of other devices that I use for temperature and other things. I use Tasmota which supports MQTT and works quite well. I have some Sonoff devices, but mostly I use NodeMCU or Wemos D1 minis. They are all esp8266 based and quite affordable.
You can certainly install multiple IoTaWatt, and a single AC reference can be split to feed all of them. However each unit will only save its own data locally. To aggregate the data, you would need to export to influxDB or Emoncms. Both can be run locally on a RPi, on a private server, or you can use the Emoncms.org service which is hosted and managed for a nominal fee.
There are advantages to both methods. Of course the common advantage is reduced input requirements and ability to combine circuits at the input level.
The splitter method makes it easier to combine circuits that are not close together, or that have conductors that are too fat to fit together into one CT.
Passing multiple conductors through one CT requires fewer CTs.
They have the same turns ratio, @nd both give excellent results. The split-core ECS24200 has a lower and more consistent phase shift, which can increase accuracy slightly in some circumstances. I don’t have a solid core in this current range, but I can’t see one performing much better than the split-core. Maybe the only advantage would be lower cost.
IMO these are two very different things. The TX is a node in their system and must send it’s data to a hub or other agent to either store or forward to a database. Typically they work with an Emonpi for a local database, or with one of their base products to forward the data to Emoncms. That may be an oversimplification, so you should check with them to determine the configuration that would be best for what you want to do and then compare that in capability and price to the IoTaWatt.
Two of the key differences are IoTaWatt’s ability to easily configure a mix of different CTs, and the local storage of all data in IoTaWatt with historical data recovery from server outages or communications failures.
Final question (just about ready to buy…). I’m adding four single zone mini split air source heat pumps to replace my oil furnace (guys at www.greenbuildingadvisor.com forum have been fantastic!). Max current is 11A… I’d prefer to get solid core 15A or 20A CT’s for those circuits vs a 50A.
Should I stop worrying and get the 50A solid core’s from the website dropdown? Or get the custom 20A CT’s…
More and more of my house is going LED with light loads, which I’d like to measure too, so I figured having a few extra CT’s on hand that can measure the low end on my 15A branch circuits would be handy.
What part do I need, and how do I order it (was planning on getting this with the North American bundle). I was hoping to add 5 of the 15 or 20A CT’s to my order, but couldn’t find a phone number to call…
You should stop worrying. I know that another site consistently harps on matching CTs to their full range, and disavows accuracy at the low end. That has not been my experience with IoTaWatt.
I just grabbed a couple of LED bulbs labelled 10W and screwed them into a test jig with an ECOL09, HWCT004, ECS1050 and the ubiquitous SCT013-000.
With one bulb, they all say 10W and occasionally one will go to 11W briefly.
With 2 bulbs they all are a solid 21W.
The CTs have been the goat for poor low end performance of some other monitors. I would maintain that it’s more likely their 10bit ADCs and low sample rates. In any event, no problem with any decent 50A CT at the low end with IoTaWatt.
IMO the YHDC SCT006 that is sold elsewhere is junk, and the only advantage is low price. I don’t carry them. Echun has a 20A split-core CT and it’s just as bad. I don’t carry that either. You can’t go wrong with the ECOL09 or the ECS1050.
Just got it yesterday (thanks OverEasy!)and hooked up the mains. Interesting already - It looks like the house idles at 250W… At 20 cents/KW (go Connecticut!), it’s $.05/hr, $1.20/day, or $438/year. I’ll start flipping breakers to narrow it down, and will get the branch circuit CT’s installed when we do the 200A service & panel upgrade in a few weeks.
Welcome to the fun of data and I also have the wife eye roll.
I’m up in your neck of the woods and we are .22 a kWh, and sit between 600-800w when the house it idle. I must be on the higher side when compared with you. We also hope that solar helps offset the usage but can always try to limit big loads. A great community and supported product, so any questions (which will come) feel free to ask.
I am continually amazed at the disparity in electricity cost between the public utilities and municipal electric companies. In your area there is Hudson Municipal Electric serving Hudson and Stow. Their electric rate is below 10 cents.
My town of Wolfeboro, NH is one of four in the state. Until a few decades ago, the town actually ran their own diesel electric generation. Now they buy it from the grid on long term contracts pegged to the cost of natural gas. Our rate is 13.5 cents.
I don’t have the insight to know why this is the case. It could be that the Munis are just getting a great deal on infrastructure cost, but as I look at the Eversource rates, it appears that distribution alone is higher than the total rate in the munis. Surely the Munis are covering their operating costs. I know in my town they do, manage regular upgrades, keep trees trimmed back from lines, and provide nearly instant service. Been trying to get the town to also go into the internet business, we own the poles…
I agree, one is often locked into the specific provider like you are with cable (for most smaller US towns). Our town is National Grid. We are .22 a kWh and $5.50 a month just for a connection charge. That connection monthly charge is for infrastructure, trees, upgrades, etc. That was one of the driving forces for us to get solar and battery. A friend in a few towns over pays .13 kWh and it will take him double to get any return on solar with a smaller system than ours. Amazing how it can be so different.
One thing to consider w.r.t. CTs is their physical size: typical 50A CTs can be a real chore to fit inside a crowded panel, whereas the 20A SCT-006 I use in my panel slip in pretty much anywhere…
Interesting as I have space issues and though SCT-006 could be a good alternative until I read the comment from @overeasy that he doesn’t recommend it, I’m confused?!
@franck102 , so far did you have any issues? And do you still recommend it since you bought them>
@overeasy I would prefer to use your 50A as you recommend but space maybe an issue. Do you mind explain deeper your thought on why the SCT-006 is bad?
I don’t recall saying they are bad, I simply don’t recommend them. I can’t carry them in the store because they are incompatible with the IoTaWatt ETL safety listing. That said, they do carry a CE marking.
They are supported in the IoTaWatt configuration tables and you can get reasonable results with them up to about 15A. Phase correction is problematic because of the generally high shift and variability from one sample to the next, but that only becomes a problem with lower power factors.
. Math is easier when you have your mains on the same device.
