I had a reply from Eurotronic around the warning in the manual about not using rechargeable batteries.
Here is their response. It is translated from German, by them. I can paste the German text they included if that helps.
Thank you for your email.
Our thermostats are designed to be powered by 2 x 1.5V AA batteries require the cumulative voltage of these
two batteries = 3V.
Unfortunately, with 2 reachable batteries you get 2.4V and it is too low.
Furthermore, it is the general problem with reachable batteries, that the internal resistance of the batteries at the still part of the 2.4 V voltage drops at higher current decreases.
For this reason, we advise against using normal batteries, as we can not guarantee that the thermostats will work fully.
Interesting... I wonder if that is all rechargeable batteries? It also might explain the issues I have seen on my wireless texecom alarm sensor that I use rechargeable batteries on.
Either way, not being able to use rechargeables is (assuming they last a year or more as some people suggest) a small price for devices that connect and stay connected reliably, and respond immediately.
I am still going to talk to Heatmiser about their thermal activator and using that as a TRV with a relay, but I have now ordered 4 Spirits for the bedrooms and keen to test out the new plugin. First though I need to work out how to wire the temp sensors to rj45 socket and get them into each room and then hooked up my pibeacon Pi. Time for some soldering!
siclark wrote:Interesting... I wonder if that is all rechargeable batteries? It also might explain the issues I have seen on my wireless texecom alarm sensor that I use rechargeable batteries on.
As I recall, every rechargeable I've ever tested with my meter has lower initial voltage than a non-rechargeable alkaline.
I was going to respond to them around Eneloop batteries, as I was making the assumption that rechargeable batteries drop voltage from 1.5 (each) to 1.2 fairly quickly, as suggested by their (Eurotronic's) response, and maybe Eneloop didn't have that problem.
This led down a rabbit warren, but it turns out that AA rechargeable batteries do, indeed, top out at 1.2v each..
Mostly.
This is due to unbreakable rules of physics and chemistry, it seems, but there have been some breakthroughs with Lithium and (rechargeable) Alkalines. Except Lithium batteries will explode unless charged in a very specific way and 'Alkaline' rechargeable batteries seems to be a grey area. as various sites mention them for sale, but never available. Basically I didn't find a nice easy, recharable, 1.5v AA answer.
However, normal batteries start off as 1.5, then drop off fairly rapidly to 1.2 and on to 1.0.
It puts a different perspective on the Eurotronic 'WARNING: Do not use rechargeable batteries'. 'Warning' sounds like a danger, but I think they mean they will not be responsible for malfunction with less than overall 3.0v. As normal batteries are going to start at 3.0 (for two), then drop anyway, whereas good rechargeables will start at 2.4 and stay there until final drop off, they are basically dropping out of responsibility when less than 3.0v, no mater what the power.
That doesn't mean they will not actually work though, and there are other options with AA battery eliminators.
Or maybe have 24v power to all of them, then step down to 3.0v locally and into fake batteries.
I really don't want to be changing batteries all the time. It is already a pain for me now, before even adding TRVs.
Indeed, I would much rather there be a whole host of z-wvave devices that were powered by wire, 12v, 24v or USB 5v (preferable). And there already are, various motion/temp sensors are now USB optional power, and of course pattress box installs are AC240v.
As a manufacturer, this is something you are going to know.
It seems that AAs will fall somewhere between 1.5v and 1.0v. Most vendors seem to have decided that their devices will stop working once the voltage drops below Nx1.0v, where 'N' is number of batteries.
Which is fine, but I would expect Eurotronic Sprits to work from 3.0v down to 2.0v, 2.4v if I'm being generous. Not be fragile around being precisely 3.0v.
Again, I'm not clear if they really are that fragile, or it is caution with respect to warranty to work.
i mentioned (above, from the manuals) that StellaZs and Honeywell HR92s do not seem to have this problem, indeed Honeywell even suggest rechargeables. Maybe it is something to do with FLiRS?
autolog wrote:... I am hoping that I will have something that can be released for alpha testing at the end of next week.
I should take a lesson from Matt and Jay and not quote target dates!
I have encountered a few problems which I am steadily resolving and so the alpha testing version won't be available for at least another week - apologies for raising expectations.
The issues have entered around determine where the request for heat came from and how to handle it.
For example in my setup, a Spirit TRV can be controlled by:
Temperature input from a Remote Thermostat
Setpoint adjustment from a Remote Thermostat
Setpoint adjustment from the Spirit Thermostat (by pressing the buttons on the physical device)
Setpoint adjustment of the Remote Thermostat device from the Indigo UI / Touch UI
Setpoint adjustment of the Spirit Thermostat device from the Indigo UI / Touch UI
Setpoint adjustment of the TRV Controller device from the Indigo UI / Touch UI
Setpoint adjustment of the TRV Controller device from the plugin's scheduler
I am testing the system by controlling my heating and I am keen to get this into a workable state as soon as possible.
If it helps for testing, I am happy just controlling the temp from indigo. I have a remote temp sensor, not a remote thermostat, and no one is going to try and control the TRV themselves..
I am actually playing with the new iMessage plugin and looking at using that as well. Guests, where most of the variability in schedule is in the guest rooms, can message in to boost or extend heat, in response to an initial welcome message (assuming they log onto the guest network and I have them setup in unifi!)
siclark wrote:... If it helps for testing, I am happy just controlling the temp from indigo. ....
That's the bit I am testing at the moment and currently isn't working.
The plugin went into a Z-Wave loop and locked out Indigo which I eventually managed to get into to be able to stop and restart the Indigo server.
I thought I had done the most complicated bits first; handling manual control and scheduling - apparently not
Don't worry, as soon as I have got something that will more-or-less works, I'll let you try it out.
Ooops.. ok I'll leave you alone to keep playing. I have to find time to get all the TRVs connected and included into the network, then the not so small task of working out how to solder the temp sensors to cable that is wired to a rj45 plug, then find other end of that cable in the mass of cables I didnt bother wiring into my patch panel, then connect that end up the rPi! So no rush.
And learn to solder as well, as it appears I am not as good as I was 25 years ago, the last time I soldered!
siclark wrote:Ha. I thought the same, seems excessively deep, but if it means a decent readable screen rather than the danfoss/Popp ones I'm all for it.
Yep, totally agree. The size, of course, diminishes somewhat when it is actually on a radiator. All looking good so far.
