Could you describe the ‘clunky-ness’ you experiences with the HACS version of the Refoss integration? I’m thinking of getting a Refoss while doing a panel upgrade, and I’d like to understand this a little better. The official documentation from Refoss appears to point only towards the HACS version, so I’d love to know more about your experience,
Sorry but I can’t remember the details. If I recall correctly it wasn’t adding all the properties in the same way. It may differ between the em16 vs em16p.
After fighting with HomeAssistant and YAML everything works as I expect now. I kind of want a better energy meter, so I may create something, but I’m tapped out for homeassistant hacking for the time being.
The energy monitor and integration themselves have been brilliant, zero dropouts.
In case if anybody is considering purchasing one of these units from Aliexpress this is a heads up.
I placed an order yesterday for the EM16P. Last night I received the following email:
Dear customer,
we regret to inform you that the Refoss Energy Monitor EM16P which you purchased on AliExpress, is out of stock. We are contacting you to request that you cancel the order and the platform will automatically refund you.
We do not have the right to cancel the order, and due to inventory being locked during the promotion period, we are unable to modify the inventory, which has caused this misunderstanding. If you can cancel the order, we would greatly appreciate it. Additionally, once EM16P is in stock, we will notify you.
If you do not cancel the order within 48 hours, we will send you an empty package. It may take longer for you to apply for a refund, so we suggest you cancel the order. Thank you!
Ship me an empty box? WTF
Trying to cancel this order on Aliexpress is a robo-nightmare. The online system won’t let me cancel the order or request a refund because the vendor has listed it as “shipped”. The Aliexpress website is set up to be a circular virtual AI help system with infinite pop-up windows meant to make it near impossible to reach support. I suspect the “Live” help agent I ended up dealing with is simply an AI chatbot with lots of
“Thank you so much for letting me know♥”
Hopefully the order will now get cancelled and my money refunded.
Dealing with vendors like this is a PITA. I will no longer buy from Aliexpress or Meross/Refoss after this experience as I feel this is the definition of a disreputable vendor…ship me an empty box, good grief.
Aliexpress can be very hit or miss…and often a loss of money. But I’m not sure how this is ReFoss fault… did their store direct you there?
Aliexpress can be very hit or miss…and often a loss of money. But I’m not sure how this is ReFoss fault… did their store direct you there?
Not ReFoss/MeFoss’s fault? They are the scammers shipping me an empty box. Aliexpress is going along with the scam and not allowing the order to get cancelled. I get caught holding the bag paying for an empty box plus an additional cost for duties and brokerage fees.
Buyer beware. I’ve learned an expensive lesson - don’t deal with disreputable vendors, especially vendors and websites based out of China. Never again.
I do believe you that you got scammed.
I do NOT believe that Refoss did it.
Have you tried contacting Refoss directly?
Also, next time try buying directly from the vendor: Refoss Open API Smart Energy Monitor, EM16P – Refoss Official Store
Are you a Nigerian Prince or my new Russian GF?
Sorry to burst your bubble, I got scammed directly by ReFoss/MeRoss. ReFoss/MeRoss emailed me directly to inform me I was the target of their empty box scam.
ReFoss/MeRoss are the scammers deciding to take my money and send an empty box. Aliexpress is the enabler and perpetuator of the scam.
I’m posting this here as a heads up to other prospective MeRoss/ReFoss/Aliexpress buyers using the HA forum as a reference for their purchasing decisions. Hopefully I’ll save someone else from getting conned out of a couple hundred dollars and getting an empty box.
That sucks. I still believe you got scammed. I still don’t believe it was them. I’ve bought multiple directly from them and recommended others buy from them. I’ve dealt with their tech support. I’ve had zero indication of issues with them.
But it’s clear this conversation isn’t going anywhere so this is my last response. ![]()
Just because the store name in Ali is Refoss or whatever, doesn’t mean that it is related to them at all.
Just got my EM16P x2 today and was quite disappointed when it started to just report 0 for everything unless I gave it an internet connection.
After some digging, easy solution is, it reaches out pool.ntp.org ( NTP server ) every little while, if you create your own local DNS record for this and point it to a local NTP server everything then works perfectly, and completely offline
Amazon has it, easy return. I seen it go as low as $170
The unit’s wireless AP is open by default. Don’t forget to disable it.
I’d be very tempted to reply to their email with a request for a direct link to cancel and let them know that if they ship you an empty box you’ll dispute the charge with your credit card company.
So I just got 2 x EM16Ps for my 2x200A service. I just did the hardware install and I am trying to understand 2 things:
- The C clamps. Is there a reality where you can have an evenly distributed number of breakers between the 2 phases of a panel in the US (8 and 8, assuming they are all stacked nicely)? Or is it really 5 and 11 or 11 and 5 which is not great? Can you configure which phase each C clamp is on? If not, how inaccurate would half of the C clamps be in the blue wire is tied to the other phase?
My one panel is set up as follows:
Left side: 2 double pole breakers : A2, B2, A3, B3 and 4 x single pole breakers A4,B4,A5,B5
Right Side: 2 double pole breakers: A6,B6, C1,C2 and 4x single pole breakers: C3,C4,C5,C6
So I have an even distribution of breakers. I connected the La black wire along with the Lc blue wire on L1, and the Lb red wire on L2. But clearly C2,C4 and C6 are on L2 breakers.
Does it mean that the measurements on the second double pole breaker on the right side (using the C1, C2 clamps) and the 2 single pole breakers using C4 and C6 will be all screwed up? Gemini and ChatGPT say that it doesn’t matter but I’m skeptical.
I have no intention to start skipping slots in the panel just to use more of L1 or L2 to make this work.
- Home assistant Should I give the app a try and decide if I like it or return it before doing the setup in HA, or just go straight to HA?
Thank you so much!!!
My understanding… but I’m not an electrician… is that you’ve got those three wire (black/red/blue) and you connect them into something providing the two poles. So you’re either doing Pole1=AC or Pole2=BC but that C can’t be split because it does some sort of offset calculation.
Do yes. It really is 5 and 11 or 11 and 5. They made it to be variable for many different types of power :- for better or worse… but it is annoying for us 2-pole power people. ![]()
Personally, I went right into HA. That’s where I wanted the data and where it’s most useful to me.
Yeah, it’s a little weird that they say it supports 3-Phase, but don’t have a large C1 clamp… Feels a little weird.
I really wish they had just gone w/ a software approach to the whole A/B/C thing, rather than assigning physical connectors to each. On household split-phase, it’s not a huge deal which leg you connect the “C” wire to - more just an accuracy thing.
Power is measured in Watts. Watts = Amps x Volts (that’s simplified - it gets weird with AC, but…). The CT clamps measure Amps. You could then compute Watts = (CT) x 120V, but the Voltage is never exactly 120V, so products like this add a separate Voltage measurement to determine the precise, actual Voltage (and do so on both/all three legs/phases to catch minor differences).
So picking an arbitrary leg to connect the C (voltage measuring) wire to is totally fine. It just means that some of the Watt calculations for the C channels may be slightly off. You could even connect all 3 to the same leg - nothing will explode; it’ll just make Watt readings for the other 2 phases slightly off.
Some products allow you to select which Voltage channel should be applied to each Current/CT channel. I do wish this was one of them, especially because it really can be done in just firmware… Fingers crossed for an update I guess…!
If anyone finds it useful, you can get more frequent (as opposed to the built-in 15s pushes) data by querying the JSON RPC endpoint directly:
curl -skd '{ "jsonrpc": "2.0", "method": "Refoss.Status.Get", "id": 1 }' http://10.10.10.1/rpc
What happens when you use some C clamps for A leads and some of B leads? In this scenario, the blue wire would be connected to an A lead as well. Will this work at all, or give wildly incorrect values for the B leads? Thank you!
It'll totally work. The instructions even say you can do this, they just say to connect the blue wire to the phase that has the most C clamps (though I'd argue "most current through C clamps" would be better, but that requires more work!).
Generally speaking, each phase should be more or less equal in voltage. There's room for some disparity, but not a ton. I'd (a nerd, but not an electrician) say 5% imbalance would be extreme and possibly indicate that you have a problem. Claude says residential is typically sub 2%. Up to you how you apply salt to those statements! ![]()
Thank you for the response -- I do not see that scenario mentioned in the manual, but I'll give it a try and see if I can get useful values ![]()
Edit: Very strange, so I hooked up C5 to my AC-Breaker's B-Phase, and even though I clamped it in the correct position (consumption going from breaker out), it was recording negative values. I had to flip the clamp the "wrong" way around for it to record positive values. I wonder if that is a side effect of "Lc" (the blue lead) technically being on the A-Phase and me now using spare "C"-clamps for B-phase breakers. Thoughts? @matchlighter
Edit2: Three days in, I can report that this works well. So to recap, Lc (blue lead) is on the A-Phase, and most of the "C"-clamps are on A-phase breakers as well. But two of my "C"-clamps are on B-phase breakers. And this works well, but I had to flip those B-phase "C"-clamps around (essentially the wrong way per the arrow) for them to properly report consumption. I would call it a hack, but it works fine, and the untracked energy consumption in HA is about half a percent of the total daily consumption. I call that good.
Oh, and I should mention that I am using the EM16P, not the EM16. And from Meross, not Refoss. Not sure if this makes any difference, but it is my understanding that they are the same (or related) companies.
For anyone who wasn't having luck getting it to work on a network without internet access...I was running in the same problem, and after inspecting the traffic a bit, it looked like it was trying to use NTP to synchronize the time. So I opened up the network to allow outbound UDP traffic on port 123, and the Refoss EM06P was able to work okay on a network that has no other internet access.