RPi Home Automation Software

[gadgetbazza (Barry Tresadern)]

@mjr

That’s one of the heating controls. A combination of fairly low precision on the manufacturer’s control, how widely the energy prices vary at the moment (thanks to the UK Energy Price Guarantee) and that the last hours before sunset are almost always the warmest part of a day (so best “fuel” for the heat pump) means that 12 timer periods are accurate enough, easier to adjust than a formula and more efficient (mainly in development time) than having a automation/script/whatever recalculate that input to the target temperature hundreds of times a day.

It sounds like what you are achieving here is “control”, not “automation”. Whilst you haven’t shared the details, I suspect that using time as a parameter for your formula on this device is not ideal, except with perhaps the exception of defining a work / home / sleep pattern. The other factors are probably best served with external temperature monitoring and a feed of energy prices if they vary that much throughout the day. Using time to predict when energy rates are high (assuming you are on a tariff that changes by the hour) and also that it will be warmer in an afternoon will be flawed. As for having to adjust a formula, you should only need to do this once maybe with some fine tuning until you get it right. The point being, you are probably having to adjust your schedule approach regularly based upon seasonal change, energy price pattern changes, an automation formula would have this built in and could be left alone to take care of the variations throughout the days / weather / seasons.

In short, using your approach and time scheduling is based upon how “you” want to “control” the system, not defining an “automation” for the “system” to work out how you control devices to meet your parameters. With the automation, it really doesn’t matter if the system has to re-evaluate the heating control every time there is an energy price change or external temperature change, that’s exactly what we have computers for.

@mjr

The HA github specifically says not to open feature requests and tells people to go post on the forum. There are already several forum threads about it which describe what’s needed, probably better than I would. Most attract few votes before being strangled by a moderator or developer pointing to some add-on not easily installed from the HA front-end. For a particularly comprehensive early strangulation example, see Input Schedule integration - Feature Requests - Home Assistant Community - so I don’t feel that I’d stand much chance on that route.

I didn’t say on GitHub? But in any case, I think suggestions / requests are supposed to go in a forum called “What the Heck” or something, a bit un-orthodox, but it’s their way of allowing the HA community to dump their ideas and have voting. I think this month is in fact the month that they review… Not entirely sure.

@mjr

@gadgetbazza

Have you used HA sufficiently long enough and gone through enough broken upgrades to determine the likely impact for yourself?

Well, obviously no, because I installed HA, it doesn’t seem to do the tasks I need and didn’t look like it would within the time I had available.

@gadgetbazza

We’re still not on the same page regarding this… I didn’t have any yaml experience prior to using HA, I’m still not educated in it by a long shot, and I don’t have any python experience, yet doing the simple things you’ve list so far would require neither, so I’m still at a loss as to your challenges other than a simple scheduling tool.

Well, it sounds like you’ve only one pretty simple automation and HA isn’t mission-critical to you.

@mjr

And don’t underestimate the importance of easy simple scheduling to me. I use a lot of scheduling, including for pausing automations for a while. Seems like a bit of a missed opportunity if I can’t quickly tell the home automation to do stuff later and free me from having to remember it then.

Yeah, just one automation!

With regards to underestimating scheduling I refer to my earlier comment about “control” vs “automation”. Yes there is still a place for scheduling, but no more than anything else. In fact, for my own requirements, I can only think of needing a schedule for my economy 7 electricity pricing (it’s specifically driven by time) and a pattern for being home / away / sleep (which I don’t actually have a use for currently).

I get mixed messages from reading your posts, you have advocated that as a platform HA is regularly broken and unstable, yet you’ve not experienced this for yourself, or actually spent any significant time using it.

It seems you’re intimating that you have a mission-critical requirement (by way of you inferring that mine isn’t) but expect to have this served by something that is both free and built to your purpose without bugs or much user effort. To this end, you either need to buy into something significant and spend a lot of money having someone customise it to do exactly what you want (completely against your “free” requirements) or work at it yourself (against your “built to purpose” requirement) IMO.

In my view, non professional home automation is still in it’s infancy, it’s firmly in the “Gadget” arena, no one “needs” their lights to turn on as soon as they walk into a room, we’ve always had switches by the doors for that reason, but with the continual shift in our lives to be technology driven, more people are intrigued by the possibilities of them just turning on when you walk in if it’s dark enough to warrant it (as an example).

That includes me, my C4 solution works, but it’s slow, limited and costs money every time I want to do anything with it. Thus far, HA is the only platform that’s showing promise in being able to address those issues. With it, comes a huge expansion in what I can control versus C4 for an acceptable amount of input from me (to avoid paying) and once I get over the multi-room AV distribution issues, I will replace C4 entirely with it instead of using it side-by-side with it as we do today.

I think it’s fair to say we have different opinions on this, so I’ll wish you good luck in a hunt for your solution, but if you want to tease out anything further on HA I’d be happy to try and help, although I’m by no means an expert on it!

[borpin (Brian Orpin)]

@gadgetbazza

With regards to underestimating scheduling I refer to my earlier comment about “control” vs “automation”. Yes there is still a place for scheduling, but no more than anything else.

That is a really interesting distinction.

I combine scheduling and automation; so if my daughter is home (automated via a sensor), I schedule the target temperature in her room and thus the heating. A combination of automation and control

@mjr

the last hours before sunset are almost always the warmest part of a day (so best “fuel” for the heat pump) means that 12 timer periods are accurate enough,

What you need I’d suggest, is an automation, that takes sunset and adjusts when the heating and therefore the HP come on. That cannot be done simply by a timer, but is simple enough in HA.

[gadgetbazza (Barry Tresadern)]

@borpin

@gadgetbazza

With regards to underestimating scheduling I refer to my earlier comment about “control” vs “automation”. Yes there is still a place for scheduling, but no more than anything else.

That is a really interesting distinction.

I combine scheduling and automation; so if my daughter is home (automated via a sensor), I schedule the target temperature in her room and thus the heating. A combination of automation and control

I too started out by thinking that a smart home / automation was about getting control of devices electronically through an app of some form. But in essence, this is “control”, personally I see scheduling as a “control” also, but control of time.

Automation I see as a bringing together of control elements to systemically, in a smart fashion, perform functions that you would otherwise need to do manually. Scheduling can be part of this, but where possible I prefer to use something with more flexibility as scheduling is typically more fixed rather than smart.

@borpin

@mjr

the last hours before sunset are almost always the warmest part of a day (so best “fuel” for the heat pump) means that 12 timer periods are accurate enough,

What you need I’d suggest, is an automation, that takes sunset and adjusts when the heating and therefore the HP come on. That cannot be done simply by a timer, but is simple enough in HA.

Agreed, or better still, since it could be a really cold day but sunset hasn’t yet occurred, add an external ambient temperature monitor and use this, thus capitalising on any time period where it is warmer outside regardless of time of day. Obviously requires additional hardware as the sunset sensor is free.

[gadgetbazza (Barry Tresadern)]

Just on the scheduling front…

Did you see the new scheduling feature added in 2022.9?

2022.9: Home Assistant Birthday Release!

Not sure if it’s sufficient for your needs or not.

[borpin (Brian Orpin)]

@gadgetbazza

Did you see the new scheduling feature added in 2022.9?

If aimed at me, then yes. The reception has been a bit lukewarm. I doubt it will replace my use of Schedy as that can be based on so many different scenarios.

[mjr]

@gadgetbazza

Did you see the new scheduling feature added in 2022.9?

I had, I tried it on a test system, and referred to it above: it’s more limited than contrib schedulers (or other apps) and still only feeds into automations.

It also currently has a practical problem for my application, where it would run on the same Rpi as emoncms, not my test system: recent HA requires a later python (3.9?) than on emonSD (3.7), but upgrading would bring a later php (8.1 I think) than emoncms is fully tested on. Not insoluble, but still another problem.

[mjr]

@gadgetbazza

Agreed, or better still, since it could be a really cold day but sunset hasn’t yet occurred, add an external ambient temperature monitor and use this

An external sensor reading is another input to the target temperature calculation. The timer-ed input is the part only depending on time (including sunset) that can be precooked to simplify the scripted calculation.

[Leonardo]

Hello I’m new on the Community. A friend of mine gave me a not used “EMONPI” (the aluminum chassis with display, shield and stuffs) I added a RPI, a power sensor and I’m playing with it.

I’m a Domoticz user, I use it since years ago and I use it for whatever I can in my home automation and I’m trying to use Emonpi for energy meter.
At the moment I’m able to read voltage and power usage (I had to correct scale in emonhub.conf) and it seems it’s ok,
I’m reading data with MQTT and I’m updating Domoticz via curl/json with a crontab script.

It works, but it’s a workaround: reading via MQTT takes seconds, crontab script is not realtime, so I’d like to write these data directly in my Domoticz (using MQTT or json).
I think the way is reading /dev/ttyAMA but… I was not able to understand HOW the data are calculated… I tried to understand the output but… no way, looking for info I just became more confused… is there anyone can give me some suggestion? I’d like to understand protocol to write code directly reading ttyAMA output and not reading it via MQTT.
What I get with (for example minicom) is:
OK 5 145 0 0 0 145 0 193 85 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 (-0)

my “scale” is:
scales = 1.3,1,1.3,0.01,0.1,0.1,0.1,0.1,0.1,0.1,1

but I’m stuck…

thanks a lot for any support, Leo

[Robert.Wall]

Welcome, Leo, to the OEM forum.

I can’t help you with MQTT or Domoticz, but I can tell you how to decode

@Leonardo

OK 5 145 0 0 0 145 0 193 85 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 (-0)

It is a string of bytes with the value given as a decimal number. But you need some more information to know what the bytes represent. Your emonHub configuration file will tell you (look in emonCMS → Setup → Emonhub → Edit config) and find Node 5. It starts “[[5]]”

It might be like mine

[[5]]
    nodename = emonpi
    [[[rx]]]
        names = power1,power2,power1pluspower2,vrms,t1,t2,t3,t4,t5,t6,pulse1count,pulse2count,E1,E2
        datacodes = h, h, h, h, h, h, h, h, h, h, L, L, l, l
        scales = 1,1,1, 0.01, 0.01,0.01,0.01,0.01,0.01,0.01, 1, 1, 1, 1
        units = W,W,W, V, C,C,C,C,C,C, p, p, Wh,Wh

My guess is you won’t have pulse2count, E1 & E2.

All the numbers match 1:1 left to right, so the first element is power1, the datacode is h, the scale factor is 1, and the units are w.
The full list of datacodes is here: emonhub/configuration.md at emon-pi · openenergymonitor/emonhub · GitHub

The part you need to look at is datacodes =
h is a signed int16_t (2 bytes), L is an uint32_t and l is an int32_t (both 4 bytes)

You decode your data like this:
OK - no checksum error
5 - the node ID the data comes from
145 0 - power1, 2 bytes, value = 145 + 0×256, this is then multiplied by the scale factor 1.3 (why? - it should not be necessary, do you have a different c.t. or a.c. adapter to our ‘standard’ ones?)
0 0 - power2, 2 bytes, value = 0 + 0×256, this is then multiplied by the scale factor 1
etc.
The pseudo-code to decode the bytes is

for extracting signed integer

  x= ([1st byte] + ( [2nd byte] * 2^8) ) 
    if x > (2^15)
        x = (x - (2^16)) 
    return x  
or
  x= ([byte 1] + ( [byte 2] * 2^8) + ( [byte 3] * 2^16) + ( [byte 4] * 2^24)  ) 
    if x > (2^31)
        x = (x - (2^32)) 
    return x  

and for an unsigned integer, ignore if x > …

[Leonardo]

Hi Robert and thanks a lot for clearing me up on this!
I didn’t consider TWO bytes for power, now I can understand it

@Robert.Wall

My guess is you won’t have pulse2count, E1 & E2.

for pulsecount you mean an AC/AC adapter connected to the Emonpi? I connected an old 9V AC power supply, maybe is not working well… I have to check it.

@Robert.Wall

this is then multiplied by the scale factor 1.3 (why? - it should not be necessary, do you have a different c.t. or a.c. adapter to our ‘standard’ ones?)

I’m in Italy and the “device” providerd by my energy supplier has a display and I can see power usage (Watts) in realtime. The value on device was always 1.3 bigger so setting the “scale” factor I always have the correct value on Emoncms too.

Maybe with a good AC/AC adapter for pulsecount I don’t need this?

The other strange thing is: the VRMS I can read on serial is 193, but EmonCMS shows the correct value between 220 and 230V…

regards, Leo

[Robert.Wall]

@Leonardo

for pulsecount you mean an AC/AC adapter connected to the Emonpi?

No, I mean the input on the RJ45 connector labelled “RJ45 I/O” that’s NOT the Ethernet one for counting pulses either with S0 contacts or the optical pulse sensor.

@Leonardo

Maybe with a good AC/AC adapter for pulsecount I don’t need this?

The a.c. adapter measures the mains supply voltage. You need that plus current transformers to accurately read real power. If you need to multiply by 1.3, it means either or both are now giving the output we expect. The a.c. adapter and the c.t. always give you the power reading at worst 10 s old. You can only know the average power from a pulse count after you have received a pulse - so it is always history.

@Leonardo

The other strange thing is: the VRMS I can read on serial is 193, but EmonCMS shows the correct value between 220 and 230V…

I think your friend must have changed something inside emonCMS to calibrate it. If you look at the Input processing, you might find they multiplied the number coming from the “emon” part (the numbers you see in the emonhub log) by 220 / 193. The “193” is wrong because the wrong voltage comes from your a.c. adapter - but it does not matter, you calibrate the emonPi to make it correct.

[Leonardo]

@Robert.Wall

No, I mean the input on the RJ45 connector labelled “RJ45 I/O” that’s NOT the Ethernet one for counting pulses either with S0 contacts or the optical pulse sensor.

No, I didn’t connect anything yet… I have to read documentation, I though was not important to me.

@Robert.Wall

If you need to multiply by 1.3, it means either or both are now giving the output we expect.

it’s a very old one, imagine it’s marked U.S. Robotics, it was the power supply of an old 14.400 modem I guess

@Robert.Wall

I think your friend must have changed something inside emonCMS to calibrate it.

I rebuild SD downloading the img file, the Emonpi I got was without RPi, so I put one of mine inside it

@Robert.Wall

The “193” is wrong because the wrong voltage comes from your a.c. adapter - but it does not matter, you calibrate the emonPi to make it correct.

yes but… the EmonCMS shows the voltage accurately, the problem is ONLY on the ttyAMA output… this is very strange for me…

[jchdel (Jean-Charles de Longueville)]

For one i use my EmonPi as sole home automation engine.
I live in an old farm off-grid with only 6 PV panels. I have to drive my power consumption harshly in winter time…
I have 1 EmonTx and 4 EmonTh in the house.
I do write glue scripts in bash as mqtt publishers or suscribers and publish whatever I want to store as mqtt topics to be picked by EmonCMS.
On the Pi, I’m running GitHub - jchdel/weewx: WeeWX code repository to publish weather data.
I use GitHub - jchdel/mpp-solar: Python package to communicate to MPP Solar PIP-4048MS inverters (and similar) to publish inverter data.
I wrote a small 1-wire2mqtt script also for a couple of sensors (should publish in my github).
I have a wifi2mqtt relay.
Plan to introduce zigbee2mqtt next week, with a couple of smart-plugs and Utility meter reader to start with.
The script to pilot my 23 years old dumb fuel heater via the relay is about 60 lines of bash and takes quite a lot of parameters fetched from EmonCMS via it’s API into account.
The whole point to me for home automation is automation. Therefore I do not need an interface to pilot the house. Just some dashboards and a way to visualize timeserie values. That’s perfectly done by the OEM team so I do not have to reinvent the wheel as a geek and I, as a geek I’m, just dump a couple of bash lines on top of an already complete infrastructure!
Thanks guys!