THZ Tecalor (LWZ Stiebel Eltron) module support and code improvement.

[immi]

Hi Immi, short Info...it works !

happy you made it
>>reference UserReading
try with the config from willy
https://forum.fhem.de/index.php/topic,33452.0.html

[willybauss]

Hi immi,

in the "other" thread there's currently a discussion related to fan speed, power etc. readings inside sGlobal. Can you remember the reason why you used the wording "power" (outputVentilatorPower / inputVentilatorPower)? This wording caused the assumption these would be electrical power values, but me and a few other members don't believe so.

BR Willy

[willybauss]

Hi immi,

as far as I remember you also own a 303 model, right? I have a problem with my 303-SOL in sDisplay readings. The defrost value doesn't get updated correctly. It always stays at 0. Even in defrost cycles it does not jump to 1.

Do you observe same? Or am I doing wrong in any way?

I'm using

THZ module VERSION  0.176
model n.a._04.19HW73SW2.08

Thanks and best regards
Willy

[immi]

Hi Willi
Sorry different model 403
sFirmware version: 04.39
sFirmware-Id HW: 74 SW: 2.08 Date: MAR 9 2011
VERSIONm0.181
modelm4.39t04.39HW74SW2.08

I normally never log sDisplay therefore I cannot tell you if defrost is read properly.
I do not remember who reverse-enginnered sDisplay, maybe you find old post in the forum.

I also do not know what is defrostEvaporator: in sControl (which I also do not log)
immi

[willybauss]

Hi immi,

403-SOL and 303-SOL are identical models, just the compressor is a bit bigger in your 404. Could you do me a favour and get sDisplay during next defrost cycle? I normally would expect the defrost flag to jump from 0 to 1 during defrosting - what it doesn't do in my case.

Thanks for the hint with defrostEvaporator: in sControl. I'll double check if this one delivers the needed defrost information.

What I want to do:
I want to track length of defrost cycles, since in some cases it looks like efficiency decreases after a defrost cycle. So my assumption is that the allowed defrost time (set to 20 minutes by default) might be too short. Since the manual explicitly says "do not change" I try to be careful and get a maximum of data before changing it anyway.

[immi]

Hi willy
logging sDisplay since 1 week and no defrost in logfile.
This doesn' t mean I had for sure no defrost.
Can I trigger a defrost manually directly on the heatpump? are there sideeffects?
immi

[willybauss]

Hi immi,

Thanks a lot for testing! I'm pretty sure you can't force a defrost cycle manually. But your result confirms my observation. As far as I know you're living pretty close to me (around Boeblingen/Tuebingen). So you had definitely some defrost cycles a few days ago, before it warmed up a bit.

Looks like the defrost flag isn't interpreted correctly.

Thanks again
Willy

[willybauss]

Of course you can see the defrost cycles in the plots, when flow temp jumps down to extremely negative values for a few minutes.

[timtom2000]

Hello,
2-3 days ago I could see the defrost cycles several times a day in sDisplay[17] (and in the neg. peaks of flow temp of course). Actually sDisplay is logged every 60s, because my defrost cycles usually take not more than a few minutes (LWZ 304).

Another question:
I get a lot of read errors during a day. My physical connection to the heatpump is somewhat complicated (20m with a CAT6 LAN extension cable; such as LogiLink UA0021D; it should work up to 50-60m).

Therefore I already set these code lines:
line 1521: select(undef, undef, undef, 0.25);
line 1573: my $buf = DevIo_SimpleReadWithTimeout($hash, 0.5);
line 1574: $buf = DevIo_SimpleReadWithTimeout($hash, 0.5) if(!defined($buf)) ;

Is there a way to make the communication even more robust? Increasing 0.25/0.5? What are the drawbacks?

Code Auswählen Erweitern


2019.01.28 07:38:26 3: THZ_Get: Error msg2:  THZ_ReadAnswer: InterfaceNotRespondig. Maybe too slow THZ_Get_Com: error found at step0  -- 0A0928 ->
2019.01.28 07:38:28 3: THZ_Get: Error msg2:  THZ_ReadAnswer: InterfaceNotRespondig. Maybe too slow THZ_Get_Com: error found at step0  -- 0A1781 ->
2019.01.28 07:38:29 3: THZ_Get: Error msg2:  THZ_ReadAnswer: InterfaceNotRespondig. Maybe too slow THZ_Get_Com: error found at step0  -- 0C1531 ->


[toggle]

I have a problem with my 303-SOL in sDisplay readings. The defrost value doesn't get updated correctly. It always stays at 0. Even in defrost cycles it does not jump to 1.

There is another indicator available - evaporatorIceMonitor in sGlobal.

[willybauss]

sDisplay[17]

17?

My sDisplay looks like

2019-01-21_20:58:36 Mythz sDisplay:  switchingProg: 1 compressor: 1 heatingHC: 1 heatingDHW: 0 boosterHC: 0 filterBoth: 0 ventStage: 0 pumpHC: 1 defrost: 0 filterUp: 0 filterDown: 0 cooling: 0 service: 0

So it should be sDisplay[21].

[willybauss]

There is another indicator available - evaporatorIceMonitor in sGlobal.

Thanks for the hint. This works. Unfortunately I log sGlobal just every 5 minutes. This is not enough for checking length of defrost cycles. My logfiles would explode by changing interval to e.g. 1 minute. Additionally the time needed to build plots would increase a lot.

[immi]

Hi timtom2000

Code Auswählen Erweitern

line 1521: select(undef, undef, undef, 0.25);
useless, you just slow down fhem; you just wait before starting talking to the thz; if you think your tecalor is overloaded, reduce the frequency at which you poll data (increase interval)

Code Auswählen Erweitern

line 1573: my $buf = DevIo_SimpleReadWithTimeout($hash, 0.5);
line 1574: $buf = DevIo_SimpleReadWithTimeout($hash, 0.5) if(!defined($buf)) ;
useless; you just slow down fhem;
The longer cable will not influence the response speed of the tecalor (maybe less than 1ms; you increase 0.5s).
Longer cables could make the signal noisy and maybe reset the comunication (UA0021D does not make the signal better)

To my opinion your tecalor is busy: the comunication is resetted too often or your intervals are too small

If you have a nice ethernet cable laying around, why don't you use it for real ethernet?
tecalor - usb - raspi(ser2net) - ethernet cable - fhem_server
tecalor - usb - Freetz(ser2net) - ethernet cable - fhem_server

this config is pretty stable for me
in addition to the ethernet cable, I also used 600km of internet connection and vpn in the past
tecalor - usb - Freetz(ser2net) - internet - fhem_server


p.s. I really like that you are able to ready my spaghetti-code
immi

[timtom2000]

Not to disappoint you...I don't understand the details of the code, I just tried a hint you gave in another forum 

I will check reducing the overload, this probably is a good hint as well, thanks.

@Willi: Display [17] in terms of defining the UserReading in fhem.cfg, Display[21] in terms of defining the gplot-config (there is a offset in counting). So we are talking about the same, I think. But if I reduce the overload of my thz by increasing the intervals I will not see the defrost anymore, probably...

[Stonemuc]

I got a new software update last week. Therefore the service technican installed a extra new sensor...looks like a NTC. Does someone knows something about it?