erratic temperatures inner probe / outdoor OKOFEN

[Dams38]

Hello after several years of reading, I take the plunge to post my turn.

To set the context, we live near Grenoble in a well insulated and largely glazed house from 140m2 to 600m altitude. Okfofen Pellematic 15K flanged to 13Kw and standard steel radiator.

Heating curve set, and comfort mode temperature of 21.5C stable without problems. On the other hand impossible to have a satisfactory reduction without anticipating enormously the recovery ... therefore in practice no profit.

So I decided to add a room sensor to boost the times and cut the heating when the large solar gains heat the living room.

Unfortunately, a problem that I had already found on the external probe (with no great consequence luckily) reproduces: the okofen probes return false and especially erratic temperatures, which prevents the calibration from being effective.
At 11h00 you have to compensate 1C
At 15h00 you have to compensate 2C
At 18h00 you have to compensate more.
short ...
Except that on the room sensor, if the temperature is too wrong, the heating stops ... real case yesterday ... 18.5 in the room seen by the probe 22.5 therefore prevents any use.

I read a lot about the problem, but no one seems to have found the answer.

To summarize the false tracks:

* No problem of draft on the probe

* It is new so we can hope it works ... and then even problem on the outdoor sensor which is a CTP 2K very simple.

* shielded cable between the boiler and the probe (about 25m)

* Terminal block tight

Here are my findings:
The principle of remote control first of all according to my analyzes and measurements:

An 1K CTP probe coupled in series with a resistance set 500ohm and 1k on a switch. A potentiometer 47K which allows to modify the resistance seen by the regulation of + - 34ohm (must still I analyze to see how it works).

Basically according to the mode of the command: stop, reduced, prog or comfort the resistance seen by the boiler varies around a fixed value

Ex in prog mode the resistance seen by the boiler is ~ 2500ohm (1500 made with 1K + 500ohm fixed) plus 1K CTP resistance.

When receiving the probe, after connecting the probe, make sure I put the shield to the ground I am delighted to see that the temperature displayed correlates wonderfully with an 19.6 electronic thermostat for 19.7 without even making a setting of probe. Unfortunately after a few minutes the temperature rises slowly to 20.7 and seems to stabilize. I compensate for -1C, it seems to hold until I go to bed because several verification gives a correlation a + - 0.2C.
Wake up unfortunately, 18C in the house, heating cut ... ambient temperature seen by the boiler almost 23!

So I'm looking for people who have ideas, or even who would have solved the problem. On my side I am afraid that this is a structural problem to the measuring system of the probes.

I think that the measurement of the temperature is done by injection of a constant current (of the genus 1mA) and measurement of the voltage at the level of the regulation ... except that the least noise of feeding or parasite will induce modifications of voltage of the same order (qq mV) that a variation of 1C ... seen how the terminal block of the regulation is packed with 220V son of the motors just next it seems alas very possible.

As this hypothesis depresses me and leave me without solution ... I hope that my analysis is wrong and that the problem is much simpler ... Otherwise remains more than to buy a digital remote control or to build it myself sniffing the RS485 bus ..

[Dams38]

Good bin..I have made good progress ...

It is necessary to activate the analog CAD by search auto and not by probe environment !!! ... to read the doc diagonally and to be based on the forums ... we make dumplings.

Now, I have the boiler that reacts to the movement of the selector ...

The measured temperature seems more stable (well I wait a few days to confirm) ... I would update the post in resolves if it is confirmed in the duration

[Did67]

Ah! Finally an "electronics engineer" who unpacks this ...

Christina regularly complained about her probe and erratic data ... Changed without success ... Etc ...

I never had a problem.

CAD = ????

Can you just give the references? or photo of the probe he is questioning ...

And summarize all this (if it turns out to be conclusive, of course), for the good series of Boeotians that we are ??? It would be great!

[Dams38]

Good evening Did67,

The Boeotian on the Okofen is me. ..I peeled so much of your post before and after buying my bug.

This is actually the standard Okofen probe CAD = Order
analog remote.

First problem solved related to my stupidity ... bad installation ... it was necessary to install the probe by "automatic search" and not via "the room probe menu" From this moment I had more likely measurements and a slight compensation to do 1C.

Before that ... the probe was giving really horrible results ... it jumped from 18 to 23C ... A check with Christina ... it's such an enormity on my part that I do not really believe ... but hey we never know ... I saw on another forum an unfortunate man who had this problem and he gave up ... when the heating engineers who leave the compensation to 0 ...

To be sure, just see if the boiler mode changes when you modify the selector. On the new regulation, we can clearly see "reduced" comfort, which changes in the heating menu.

I'm doing the temperature measurements right now, but I have to go back and forth in the boiler room from the outside ... At this moment it quails and it's raining ... it limits the number of measuring points possible. .

For the probe I dismounted the beast and took pictures of the interior to make the circuit diagram ... If people are interested made me sign but it's not Doisneau ... The content of the probe c is really not much ... precise resistors, a thermistor and a potentiometer ... good the quality of the welds is blah blah ... but nothing that explains problems of variation. If it happens again is that the problem comes from the regulation next to the boiler or parasites on the cable.

For info, the probe
OKOFEN E1229 Reference and Serial Number
E1391 1596373.

[Did67]

The Boeotian on the Okofen is me. : cheesy: ..I have peeled so much of your post has


... when the heating engineers leave the compensation to 0 ...

No, electronics, I do not know anything about it. And we can see that it could serve!

And it is unfortunately also worse at the level of many "heating engineers", who are often only "plumbers" [a very honorable job - I am talking about skills!] And for whom these animals are even more confusing! When I see the configuration nonsense some of which forumThey have reported to me !!! So I'm not talking to open one of these boxes!

I hope christina will go through there. Because it had an erratic probe, replaced by another probe ... just as erratic!

[christina86]

Thank you for informing me

- at the workshop (ITI and solar contributions) the second probe is ok

- at home, replaced too, but still not fair ...
installed by the technician as it should
(Be careful indeed, "room sensor" is only for the optional wall regulator, and as there are three instructions for use, not easy to navigate at the beginning ...)
he calibrated that one, but it held a day or two maxi

So I disabled it - my heating curve is good, it works much better without (stable temp).

I had added it to the house afterwards - less for solar gains than for the opposite (too cold): wind, rain, etc. perceiving that they are not taken into account by the external probe (house not really waterproof ...) and possibly as remote command on occasion.
But at the moment I was not conscious enough how much the house is inertial -
I have also abandoned the night shift, the offset is more than 12h ... *
(You can mount the heater when the walls are already hot, but lowering the night is not so fast - and the boiler cuts instead of sending less hot - and there is more time to go up)
(* I lowered the 19 setpoint to 18 the other day, and it put 24h)

So I gave up asking to replace her again, because I do not know if in the case of my house it would have much effect ... (but I have never been able to experiment with a probe that works)

[Dams38]

Hello everybody

For the moment the biggest difference observed during my "artisanal" measurements ... 0.4C but not constant. Difficult to complain with such a small difference.

Just take off the reference thermometer from the wall with an insulator induces a variation of the same order of magnitude.

But I will continue my surveillance during the week to be sure not to fall in the same case as Christina with a probe that starts to go in java after a few days.

It's still against all logic this story ... we can not do more simple as probe principle ...

[Chaboum]

Diams38: I have long had this problem on all my probes !! and I still have it !!! no one has ever been able to solve it !!! I had posted a long time on the forum , but ... not being an electronics .....:
little reminder: my external probe tells me -7 ° while it is -13 °, then tells me that it is 15 ° when it is 15 ° !!
ambient sensor: I pretty much managed to stall the beast: wheel turned 3e point to the right to tell me about the actual temperature !! and especially I touch it more !!! otherwise disaster !! Imagine, I'm a little cold, I decided to turn right to get hotter !! and well there disaster: the boiler receives for example the information that it is 22 °, therefore useless to send hot water !! result: it was even colder then we wanted to be warmer !!! This room sensor is a puzzle for me, I thought to use it for comfort: the ambient temperature is the one I want to get, so if it is measured by a probe that SHOULD be a plus! And the wheel to operate should allow to go up a little in case of cold .... and it's been years that nobody has solved my problem!

thanks to the new and old of your advice .... dare I hope again ?? thank you

[Did67]

I can only say that my "old" wired probe obeys the finger and the eye since the beginning of 2008 ...

Right now, I spend my time "going down / up" based on the weather forecast; having PCBT, I am very out of place: transition to comfort at 4 am. So I heat up a shot in the morning: if the weather is fine, I heat a minimum - probe to 17 °; the slab heats up a little then the sun does the rest; if it's going to be gray, I put the 20 ° back - I don't put a kopek on the sun.

They are incredible - but true, I do not doubt it! - your erratic probe stories. Christina has already testified.

[christina86]

Good evening Chaboum
my probe is also on the 4e point right to indicate the right temperature, I dropped to claim an 3e replacement since I have disabled it - it does more than indicate the temp int.

But against me at home it reacted at least in the right direction, ie if you turn it completely to the right it indicates colder as it should - at home it's the opposite, as if it was plugged upside down, son reversed, but I do not know how it works, it may not affect (note, to know it would be enough to try)?

Did you were lucky, maybe it's like so often with the old stuff, they worked better than the new ones - weirdly the "progress" rarely goes in the right direction ... the old wireless tel which received much further than the following ones, I am in 3rd since 2000 and each time we lose a little (Al ... el then Si ... ns)