Advertisement
If you have a new account but are having problems posting or verifying your account, please email us on hello@boards.ie for help. Thanks :)
Hello all! Please ensure that you are posting a new thread or question in the appropriate forum. The Feedback forum is overwhelmed with questions that are having to be moved elsewhere. If you need help to verify your account contact hello@boards.ie

Replacing PT1000 sensor on Solar Panel

  • 15-02-2012 2:20pm
    #1
    Registered Users Posts: 350 ✭✭


    Recently I have noticed that the reading on my solar panel is incorrect. On recent nights when it has been cold outside it is reading 12/13degC whereas the air temperature can be no more that 3/4deg. I measured the resistance going into the panel and it corresponded to a temperature of 12deg (1046Ohm) so the panel doesn't seem to be the problem (although it does intermittently read temperatures exceeding 120deg...), There are 2 joints in the cable coming back to the panel and I measured the resistance at each of these points as approx 1046Ohm so it appears that the problem is with the PT1000 sensor itself.

    I am going to order a new PT1000 cable but want to ensure that there would be no issues with what I order. Do all cables have standard connectors? Is it a simple task to replace the sensor? I am going to try to order a 10m long cable if possible to remove at least one of the joints.

    Also. looking at the piping a lot of the insulatation surrounding the solar pipes has pulled away from the system so I need to fix this as well. Are there good online suppliers of this insulation - I presume it needs to be special high temperature insulation?


Comments

  • Registered Users Posts: 153 ✭✭rayh


    fitzie79 wrote: »
    Recently I have noticed that the reading on my solar panel is incorrect. On recent nights when it has been cold outside it is reading 12/13degC whereas the air temperature can be no more that 3/4deg. I measured the resistance going into the panel and it corresponded to a temperature of 12deg (1046Ohm) so the panel doesn't seem to be the problem (although it does intermittently read temperatures exceeding 120deg...), There are 2 joints in the cable coming back to the panel and I measured the resistance at each of these points as approx 1046Ohm so it appears that the problem is with the PT1000 sensor itself.

    I am going to order a new PT1000 cable but want to ensure that there would be no issues with what I order. Do all cables have standard connectors? Is it a simple task to replace the sensor? I am going to try to order a 10m long cable if possible to remove at least one of the joints.

    Also. looking at the piping a lot of the insulatation surrounding the solar pipes has pulled away from the system so I need to fix this as well. Are there good online suppliers of this insulation - I presume it needs to be special high temperature insulation?

    Just 2 comments:
    1. Had this problem myself and found that the problem was with controller. Might be best to check sender against another controller first.
    2. PT1000 can come in various sizes - be careful if you have small receiver and how it seals


  • Registered Users Posts: 350 ✭✭fitzie79


    thanks for the feedback. should the fact that the sensor is measuring an incorrect value not suggest that the sensor has a problem? i think i will end up replacing the controller anyway but don't want to do this unless it is absolutely necessary as it'll cost me over a hundred euro


  • Registered Users Posts: 153 ✭✭rayh


    fitzie79 wrote: »
    thanks for the feedback. should the fact that the sensor is measuring an incorrect value not suggest that the sensor has a problem? i think i will end up replacing the controller anyway but don't want to do this unless it is absolutely necessary as it'll cost me over a hundred euro

    Have a look at http://www.regin.se/dmdocuments/resistance_table_PT1000_sensors_prsh_en.pdf These units dont give much problems


  • Registered Users Posts: 350 ✭✭fitzie79


    looking up the table my reading of 1046 ohm at the pt1000 (with it disconnected from the controller) corresponds to 12degC which is what the controller was reading. however the air temperature was only 3/4deg so this suggests that the pt1000 has the problem?

    sounds like i'll need to get up on the roof to look at the pt1000 sensor to see what size it is and what size the seal is


  • Registered Users Posts: 153 ✭✭rayh


    fitzie79 wrote: »
    looking up the table my reading of 1046 ohm at the pt1000 (with it disconnected from the controller) corresponds to 12degC which is what the controller was reading. however the air temperature was only 3/4deg so this suggests that the pt1000 has the problem?

    sounds like i'll need to get up on the roof to look at the pt1000 sensor to see what size it is and what size the seal is

    While a sensor failure may seem the obvious, I am not convinced how these units can fail and produce a +10 DegC offset.
    Therefore it is my opinion that your controller is faulty and causing heat to passed to the sensor. Infact this fault can cause your panels to become heat sink at night.
    Can I suggest that at dusk when the sun has declined, that you power the unit down for a few hours and then check if your controller is reading the expected outside temps. Also check the temps of the pipes leading to the panels to see you are getting any thermosyphoning in the system - is one pipe hotter than the other or higher than expected.


  • Advertisement
  • Registered Users Posts: 1,627 ✭✭✭quentingargan


    I doubt if the PT1000 is faulty. Do you have tubes or flatplate? I would expect tubes to have a temperature higher than ambient for some considerable time. If your cylinder is up to, say, 40C, then the pump will stop when the panel temperature falls to 44C. The temperature will fall slowly, perhaps to 12 degrees after a few hours.

    If you have flatplates, then the probe should be a lot closer to ambient. UNLESS your non-return valve is faulty. OR if your panel has the anti-freeze mode working.

    If you have the anti-freeze mode on, then when the panel falls below 4C, the pump will run for a minute or so, sending water from the bottom of your cylinder up to prevent it from freezing.

    If your non-return valve is faulty, then some heat will escape from your cylinder up to the panel.

    But I have seldom known a PT1000 sensor to simply fail, and never with a slightly inaccurate reading.


  • Registered Users Posts: 350 ✭✭fitzie79


    i have flat plates

    the non-return valve being faulty is an interesting possibility. with the sunshine over the weekend the bottom of the tank reached about 45 to 50 degrees (it would normally be about 23) i noticed that on saturday and sunday night the reading on the collector was 22 to 23 degrees( instead of 12 to 13).

    if the heat from the bottom of the cylinder was escaping then this might explain what i am seeing. how would i determine if this was definitely the problem?


  • Registered Users Posts: 1,627 ✭✭✭quentingargan


    fitzie79 wrote: »
    i have flat plates

    the non-return valve being faulty is an interesting possibility. with the sunshine over the weekend the bottom of the tank reached about 45 to 50 degrees (it would normally be about 23) i noticed that on saturday and sunday night the reading on the collector was 22 to 23 degrees( instead of 12 to 13).

    if the heat from the bottom of the cylinder was escaping then this might explain what i am seeing. how would i determine if this was definitely the problem?

    If there is heat flowing around the system, it will be on the flow side, so it is more difficult as there should be no exposed pipework on that side.

    Any chance that the pump station was put on the flow side? I've seen cylinders marked the wrong way around.

    Is the temperature at the bottom of the cylinder falling off before you have used much hot water??


  • Registered Users Posts: 153 ✭✭rayh


    fitzie79 wrote: »
    looking up the table my reading of 1046 ohm at the pt1000 (with it disconnected from the controller) corresponds to 12degC which is what the controller was reading. however the air temperature was only 3/4deg so this suggests that the pt1000 has the problem?

    sounds like i'll need to get up on the roof to look at the pt1000 sensor to see what size it is and what size the seal is

    Any update on your PT1000 sensors problem?


  • Registered Users Posts: 350 ✭✭fitzie79


    busy week so haven't gotten back to it - hopefully over the weekend....


  • Advertisement
  • Registered Users Posts: 350 ✭✭fitzie79


    If there is heat flowing around the system, it will be on the flow side, so it is more difficult as there should be no exposed pipework on that side.

    Any chance that the pump station was put on the flow side? I've seen cylinders marked the wrong way around.

    Is the temperature at the bottom of the cylinder falling off before you have used much hot water??

    what do you mean by the flow side? i've attached a picture of the piping in the hot press. there are markings for flow and return for the solar coil. should i measure the temp of the pipe coming from the flow or return pipe or should the temperature be measured back by where the pipes go into the panels on top of the garage roof?



    how could i determine if the pump station was put on the flow side


  • Registered Users Posts: 1,627 ✭✭✭quentingargan


    fitzie79 wrote: »
    what do you mean by the flow side? i've attached a picture of the piping in the hot press. there are markings for flow and return for the solar coil. should i measure the temp of the pipe coming from the flow or return pipe or should the temperature be measured back by where the pipes go into the panels on top of the garage roof?



    how could i determine if the pump station was put on the flow side

    Basically, after a hot day, once the pump goes off, the pipework should cool off. If there is heat in either pipe, then something is wrong.

    The flow side should come off the top of the panel (if it is a flatplate with a top and bottom fitting), and heat from the panel flows from here to the top of the cylinder coil. The return should be from the bottom of the cylinder coil, back up to the bottom of the panel, via the pump station.

    There is a one-way valve in the pump-station to prevent natural siphoning of heat. Otherwise, hot water in the cylinder coil would rise through the coil, on up the pipework into the panel, cooling in the panel and falling back down through the pumpstation.

    Quite a few cylinders were sold with the coils marked the wrong way around.

    It would be my best guess that this, and not the PT1000 is the reason your panel is reading high temperatures after the sun has gone down, but it is hard to diagnose these things on the www.

    If there is heat in the pipework after about 8.00pm on a sunny day this week, either plumbing is wrong, or your non-return valve has failed (or possibly you are pumping hot water up to the panel because the holiday mode in your controller is switched on.


  • Registered Users Posts: 1,379 ✭✭✭dathi


    hi i have seen this on a few systems that are not working as expected. if you look at the picture of the cylinder where the two pipes for the solar coil leave the cylinder, one pipe is going straight up. this allows the the hot water in the cylinder to thermosyphon up the pipe as far as the non return valve, this being made of brass heats up and transfers the heat to the other side of the non return valve where it makes its way to the roof. or worse still its on the other side of pipe work and there is no non return valve to stop the heat thermosyphoning to the roof. the solution to the problem is to make the two pipes travel down for 100mm from the cylinder before they start to rise to the roof this stops any water in the cylinder from thermosyphoning


  • Registered Users Posts: 182 ✭✭saibhne


    I agree with Dathi, I have seen this in a number of installations where heat is lost through thermal movement despite the presence of non return valves. Plumbing pipes downwards at the cylinder does work.

    S.


  • Registered Users Posts: 153 ✭✭rayh


    fitzie79 wrote: »
    Recently I have noticed that the reading on my solar panel is incorrect. On recent nights when it has been cold outside it is reading 12/13degC whereas the air temperature can be no more that 3/4deg. I measured the resistance going into the panel and it corresponded to a temperature of 12deg (1046Ohm) so the panel doesn't seem to be the problem (although it does intermittently read temperatures exceeding 120deg...), There are 2 joints in the cable coming back to the panel and I measured the resistance at each of these points as approx 1046Ohm so it appears that the problem is with the PT1000 sensor itself.

    I am going to order a new PT1000 cable but want to ensure that there would be no issues with what I order. Do all cables have standard connectors? Is it a simple task to replace the sensor? I am going to try to order a 10m long cable if possible to remove at least one of the joints.

    Also. looking at the piping a lot of the insulatation surrounding the solar pipes has pulled away from the system so I need to fix this as well. Are there good online suppliers of this insulation - I presume it needs to be special high temperature insulation?

    Whats the update?


  • Registered Users Posts: 350 ✭✭fitzie79


    good timing with the question! i also had an issue with the controller showing extremely high temperatures. On cloudy days the controller would read in excess of 70deg on the panel and start the pump even though the temperature was actually only 20 to 30 deg. Turning on/off the controller would "reset" the temperature. As I had this issue and felt that perhaps this was happening at night and effectively taking hot water from the tanks I decided to replace the controller. I did this last night but it hasn't resolved the issue I have with the spike in the solar panel temperature reading (this morning at 8am I allegedly had 50deg at the solar panel despite the deluge of rain outside)

    I need to resolve this issue and the non-return valve issue as well. I am looking to get in expert advice. Does anyone have any suggestions for someone in the cork region who may be able to look at the system and fix these issues?


  • Registered Users Posts: 1,627 ✭✭✭quentingargan


    That is odd. I would have suspected moisture reducing the resistance of the probe, but that would show a lower temperature.

    The first point though is to see if it is a false reading, or if you are getting heat leaking up to the panel from the cylinder. Is there any way you

    1) At night, physically disconnect the pump
    2) Early following morning, see if the panel is reading hot. If it is, feel the pipes in the attic in close proximity to the panel (cut a bit of insulation if you need to get at this).

    If there is any warmth up there, it is an issue with either how your non-return valve is working, or the pump station is on back to front.

    There were a lot of cylinder sold which had the flow and return marked the wrong way around on the solar coil. The flow should be going to the top of the coil, and the return comes from the bottom, and goes to the pump station. If that is the wrong way around, it may have the effect you are seeing, as the one-way valve is wrong way around (eh, I think:confused:)


  • Registered Users Posts: 350 ✭✭fitzie79


    Yesterday I disconnected the temperature sensor from the cable that brings the signal from the garage (where the solar panels are mounted) to the solar controller in the utility room and instead connected a PT1000 sensor that was sitting in the garage to this point. The reading has been solid all at ambient (about 15deg yesterday) since. This to me shows that the issue is not with the cable running from the garage to the controller.

    I then measured the resistance on the cable coming from the solar panels on the roof last night and it was fluctuating wildly. I manually turned on the pump and after 5 minutes of running the resistance measurement settled down (to a temperature corresponding to approx 20deg - I think it was this high as it took some of the heat from the bottom of the cylinder). Is it possible that there is air in the system that is causing this issue?

    Any recommendations on a good installer in the Cork region who could fix this and look at the non-return valve?


  • Registered Users Posts: 1,627 ✭✭✭quentingargan


    fitzie79 wrote: »
    I then measured the resistance on the cable coming from the solar panels on the roof last night and it was fluctuating wildly. I manually turned on the pump and after 5 minutes of running the resistance measurement settled down (to a temperature corresponding to approx 20deg - I think it was this high as it took some of the heat from the bottom of the cylinder). Is it possible that there is air in the system that is causing this issue?

    Any recommendations on a good installer in the Cork region who could fix this and look at the non-return valve?

    I wonder what you mean by fluctuating wildly. Generally heat probes should gradually move up and down with some time delay as the probe and its mounting, as well as the manifold itself all heat up. If it was fluctuating slowly like that, it may be genuinely reflecting the temperatures at the panel, but if it is zooming up and down, it is more likely to be a fault in the cable.

    What sort of cable was used? I have seen training manuals suggesting that you can use bell wire for this. I would always suggest the use of stranded wire which is less vulneable to breaking when bent and twisted around the place.


  • Registered Users Posts: 64 ✭✭soundskin


    Is your sensor PT1000 running parallel to any electrical cables?

    I've seen plenty of heat pump sensors (non PT1000) when ran next to 230v cables in parallel and not shielded to cause sensors reading to vary wildly, ie increase and decrease rapidly over 20+ degrees.

    Danfoss says the following for it's sensors;
    "NOTE! The cables to the outdoor and indoor sensors carry low voltage.
    Interference from surrounding cables may affect these and cause incorrect information to be sent to the control unit. It is therefore vitally
    important that careful consideration be given to positioning of cables.
    Low-voltage cables must be kept well separated from high-voltage
    cables. It is important that low-voltage cables are run parallel to high-
    voltage cables and separated by a distance of at least 25 cm for cable
    lengths up to 100 m. For longer cable lengths, the distance between
    cables should be increased."


  • Advertisement
  • Registered Users Posts: 350 ✭✭fitzie79


    thanks for the response - my sensor cables are directly next to the mains cable in a trench running from the garage. they are using 2 cores in an armoured cable so they have a metal shield and are twisted but....

    i did an experiment previously where i connected a pt1000 sensor that was sitting in the garage instead of the sensors on the roof. the reading from them was rock solid but there the pump from the solar panel was not running. i will repeat this with the pump running to rule out interference...


  • Registered Users Posts: 648 ✭✭✭PeteHeat


    fitzie79 wrote: »
    Yesterday I disconnected the temperature sensor from the cable that brings the signal from the garage (where the solar panels are mounted) to the solar controller in the utility room and instead connected a PT1000 sensor that was sitting in the garage to this point. The reading has been solid all at ambient (about 15deg yesterday) since. This to me shows that the issue is not with the cable running from the garage to the controller.

    I then measured the resistance on the cable coming from the solar panels on the roof last night and it was fluctuating wildly. I manually turned on the pump and after 5 minutes of running the resistance measurement settled down (to a temperature corresponding to approx 20deg - I think it was this high as it took some of the heat from the bottom of the cylinder). Is it possible that there is air in the system that is causing this issue?

    Any recommendations on a good installer in the Cork region who could fix this and look at the non-return valve?

    Hi,

    You could try This Gent he is in Cork and has experience of a lot of different systems.

    .


  • Registered Users Posts: 350 ✭✭fitzie79


    ran my final experiment last night. broke the connection with the sensor in the panel and instead hooked up a pt1000 sensor dangling in the air in its place. got a hair dryer and with a low heat setting raised the temperature to 45deg on the sensor. the controller turned on the pump and ran no problems for 15 minutes. i then increased the heat from the hair dryer so that the sensor temperature was 75deg. again it ran for a period of time with no issues.

    i think this rules out an issue with the cable coming back from the garage and also any noise from mains cables causing interference. the only issue can be the sensor itself or how it is connected.

    hopefully my next post will be to report that all is working normally again!


  • Registered Users Posts: 350 ✭✭fitzie79


    have made some progress with this recently. contacted the person as recommended by peteheat. there was a bit of air in the system but this didn't turn out to be causing the problem with the temperature reading fluctuating. i asked him to replace the pt1000 sensor and even though he had only seen dodgy sensors giving a constant reading (not fluctuations like i had) he replaced it. when he took the old sensor out he noticed that it was a cylinder sensor that was put in and it was the incorrect size. with the new sensor the temperature has been solid for the past two weeks with no fluctuations :D

    i have other issues with the system though as the pump is undersized and there may be heat escaping from the tank as the non return valve is placed out by the pump in the garage and not next to the tank in the house. apparently the guy who did the installation (no longer in business) was a complete cowboy and i am not the first instance that the new guy has come across who has had problems with this installer. at least though i am no getting some hot water and can work on resolving the other issues


Advertisement