Page 1 of 3 123 LastLast
Results 1 to 10 of 25

Thread: EvoHome Hot water temperature incorrect

  1. #1
    Automated Home Sr Member
    Join Date
    Dec 2018
    Posts
    50

    Default EvoHome Hot water temperature incorrect

    Just a quick question really before I go speaking to my plumber...

    Iíve noticed since the install (about 6 weeks ago) that the water temperature indicated on the controller never gets to the 60 degree target and usually is to be found around the 45 degree point, dropping down to 15 or so when weíve used a fair bit of water from the tank (210l unvented).

    I was starting to get a little concerned and looked into it some more today. The water temperature at the tap is 60 degrees (handheld IR thermometer). So thatís a good thing. But the controller is reporting 40 degrees and has been doing so for the past hour, despite the HW demand being 100%.

    My only thought is the sensor must not be contacting the tank correctly, but are there other possibilities? And is the sensor easy enough to have a look at / adjust without calling out the installer? Iím not familiar with unvented tanks so itís all a bit new to me

  2. #2
    Automated Home Sr Member
    Join Date
    Dec 2018
    Posts
    50

    Default

    Hmm. And all of a sudden it’s jumped from 40 to 51. No comms faults. Water temp at the tap is now 62. How frequently should Hot water readings be updated on the display? Seems a huge jump after resting at 40 for an hour.

  3. #3
    Automated Home Sr Member
    Join Date
    Dec 2018
    Posts
    50

    Default

    And 90 minutes later still stuck at 52. Hot water relay is on, no improvement in temperature. I’m a bit stumped really.

  4. #4
    Automated Home Legend
    Join Date
    Sep 2014
    Location
    Scotland
    Posts
    1,828

    Default

    How far up the cylinder is the insertion sensor installed ? Normally it should be about 1/3rd of the way up.

    It sounds like the sensor is not making good thermal contact in the sensor pocket in the cylinder causing it to read too low.

    The sudden jump from 40 to 51 is to be expected if you have the hot water set to 60 degrees - the CS92A sensor only sends temperature updates very infrequently if the cylinder temperature is more than 10 degrees below the set point and is changing slowly, but will send much more frequent updates when the temperature approaches 10 degrees from the set point. The real issue is that it is reading lower than it should.

    I would also add that so long as the Evohome is showing the temperature being too low it will be trying to heat the hot water...

    Presumably you have the original thermostat and/or safety cutout still wired in series ? If the original thermostat is still fitted but only set to 60 degrees what is probably happening is that the Evohome is always calling for the hot water to be heated (never reaching it's target) but the original cylinder thermostat is cutting out at 60C. If your boiler fires via the hot water zone valve switch this will prevent gas being wasted however if you have a separate boiler control relay the boiler will be running trying to heat hot water when it shouldn't.

    If you have another thermostat in series it would need to be set at least 5 degrees higher than the hot water temperature configured in the Evohome to ensure that it's the Evohome that shuts off the hot water heating first and thus has control of it, with the other thermostat being there as a safety cutout in case something goes wrong with the evohome's control of the temperature. (A safety cutout is mandatory with an unvented cylinder)

    So step one is to check the thermal bonding between the thermostat and the pocket and ensure it's in the correct pocket.
    Last edited by DBMandrake; 8th January 2019 at 09:27 PM.

  5. #5
    Automated Home Sr Member
    Join Date
    Dec 2018
    Posts
    50

    Default

    Quote Originally Posted by DBMandrake View Post
    How far up the cylinder is the insertion sensor installed ? Normally it should be about 1/3rd of the way up.

    It sounds like the sensor is not making good thermal contact in the sensor pocket in the cylinder causing it to read too low.
    It's a CenterStore 210l unvented cylinder https://www.wolseley.co.uk/product/c...inder-210-ltr/

    The sensor is located about a third of the way up under the wiring panel, looks like it's installed correctly.

    Quote Originally Posted by DBMandrake View Post
    The sudden jump from 40 to 51 is to be expected if you have the hot water set to 60 degrees - the CS92A sensor only sends temperature updates very infrequently if the cylinder temperature is more than 10 degrees below the set point and is changing slowly, but will send much more frequent updates when the temperature approaches 10 degrees from the set point. The real issue is that it is reading lower than it should.

    I would also add that so long as the Evohome is showing the temperature being too low it will be trying to heat the hot water...
    I'd read about the slower refresh somewhere before, didn't think it was quite that slow but good to know - I thought I'd seen graphs people had produced online showing the tank temperature readings refreshing every few minutes so wasn't expecting a 30m (or thereabouts) refresh period.

    Is the difference between tap water (60-62) and measured water (50-51) degrees an expected difference between the top and bottom of the tank? If so, I think what you say next is the most likely scenario...

    Quote Originally Posted by DBMandrake View Post
    Presumably you have the original thermostat and/or safety cutout still wired in series ? If the original thermostat is still fitted but only set to 60 degrees what is probably happening is that the Evohome is always calling for the hot water to be heated (never reaching it's target) but the original cylinder thermostat is cutting out at 60C. If your boiler fires via the hot water zone valve switch this will prevent gas being wasted however if you have a separate boiler control relay the boiler will be running trying to heat hot water when it shouldn't.

    If you have another thermostat in series it would need to be set at least 5 degrees higher than the hot water temperature configured in the Evohome to ensure that it's the Evohome that shuts off the hot water heating first and thus has control of it, with the other thermostat being there as a safety cutout in case something goes wrong with the evohome's control of the temperature. (A safety cutout is mandatory with an unvented cylinder)

    So step one is to check the thermal bonding between the thermostat and the pocket and ensure it's in the correct pocket.
    I'm 99% sure the original thermostat is in series, haven't checked the wiring but recall that's what the installer said. I've found it is adjustable on the tank, it can be set up to a max of 72 degrees so what you say above is most likely correct - it's set too low and is interfering with the EvoHome being able to control the temperature.

    I'm going to try turning it up a bit tonight and see what difference that makes. I don't have a separate boiler control relay so guess there's not much gas being wasted but I'd rather everything was set up correctly. I could probably max-out the built-in thermostat as that will stop interference and still provide a level of protection @ 72 degrees to prevent meltdown

    Thanks for your help as always DBMandrake!

  6. #6
    Automated Home Legend
    Join Date
    Sep 2014
    Location
    Scotland
    Posts
    1,828

    Default

    Quote Originally Posted by Jemster View Post
    I'd read about the slower refresh somewhere before, didn't think it was quite that slow but good to know - I thought I'd seen graphs people had produced online showing the tank temperature readings refreshing every few minutes so wasn't expecting a 30m (or thereabouts) refresh period.
    Refresh period depends on how far the temperature is from the set point and how rapidly its changing. If it's below the set point minus differential (so 60-10=50 if you still have the default 10 degree differential) and is not changing much maybe a couple of degrees it can go for as long as an hour without any updates.

    But if it's within that differential range and the temperature is rising quickly it can update as often as every minute or so. Some people including me have also had issues where updates aren't sent reliably but this usually results in the hot water temperature overshooting. Battery contacts on the CS92A seem to be problematic - not where they touch the batteries, but where they contact the rear of the PCB. I've had my CS92A replaced and also had to retension the contacts where they go behind the PCB to get it reliable.

    Poor battery contact design seems to be an issue with the Evohome system - I've had problems with the battery contacts in most of my HR92's losing tension resulting in intermittent connection, the contacts in the CS92A causing hot water overshoots, the contacts in the DTS92's causing loss of power, and even the contacts in the Evotouch controller itself.
    Is the difference between tap water (60-62) and measured water (50-51) degrees an expected difference between the top and bottom of the tank? If so, I think what you say next is the most likely scenario...
    No it's not expected. You would expect the reading to be lower than actual temperature if you had recently drawn off some hot water, but not enough to trigger a reheat - because cold water going in the bottom will drop the temperature at the sensor pocket below the water at the top.

    However during a reheat cycle and definitely towards the end and at the end of the reheat cycle convection will circulate the water between the bottom and top of the cylinder so you would expect the temperature reading to be higher than the actual temperature at the tap by maybe a couple of degrees due to heat loss through the pipes.

    I have a vented cylinder so I use the strap on sensor strapped against a section where the foam is cut away rather than an insertion sensor, when I first installed it I found it was reading about 5 degrees lower than it should, it turned out it was a very thin, transparent almost unnoticeable layer of adhesive from the foam that I had cut away that was still adhered to the copper surface, this was enough of an insulator to direct conduction of heat from the cylinder to the sensor to throw the reading off by quite a lot. I sanded the copper back lightly to give a clean raw copper surface and it now reads correctly.

    Likewise if there isn't a good thermal bond between your sensor and the pocket it's in (pocket is too big for the sensor for example so it's just kinda lying in the bottom of the tube) you could easily get a reading that is 5-10 degrees lower than the real temperature. In that case you could either try it in a different pocket if there are others in the vicinity that are a better fit or use some sort of thermally conductive compound between the sensor and pocket such as a suitable grease. If the sensor is not a snug fit in the pocket some sort of thermal grease is almost certainly required to get good accuracy.

    BTW while you're diagnosing this problem a tip to avoid the sensor update lag is if you remove the batteries from the CS92A then re-insert them it will always send a single temperature reading within a few seconds. (And then go back to its normal scheduling)

    This may help if you're trying to move the sensor around and want to see the result quickly as you can force it to send an update when you want it to.
    Last edited by DBMandrake; 9th January 2019 at 08:34 PM.

  7. #7
    Automated Home Sr Member
    Join Date
    Dec 2018
    Posts
    50

    Default

    Hmmm. Ok. I took off the cover and had a look tonight.

    What i think heís done is used the insertion sensor but tucked it under the insulation and itís only making a small contact with the cylinder.

    30E1B78B-3CDC-4D3B-B30B-CA122AB6668E.jpg

    I guess thatís why the reading is off then!

    I can either work it further under the insulation so it contacts the tank better or take a look at the strap-on sensor, but not sure thereís space for it anywhere.

  8. #8
    Automated Home Legend
    Join Date
    Sep 2014
    Location
    Scotland
    Posts
    1,828

    Default

    I doubt you'd be able to use the strap on sensor in the wiring box in an un-vented cylinder - it's quite large and relies on a curtain wire strap around the entire cylinder to hold it on...

    I don't have any direct experience with un-vented cylinders so perhaps someone who does use one with Evohome can comment on your picture.

  9. #9
    Automated Home Guru
    Join Date
    Dec 2016
    Posts
    123

    Default

    Quote Originally Posted by DBMandrake View Post
    .... perhaps someone who does use one with Evohome can comment on your picture.
    I have a Megaflo 210i. But my cylinder has a purpose-built pocket for a separate temperature sensor, like the one provided by Honeywell's Evohome CS92. A picture of my cylinder is attached here. The Evohome temperature probe is on the thin black wire to the left of the lower grey terminal box. The later installation manual for Megaflo Eco shows this pocket on page 9 (1/2" BSP sensor pocket). I fear that your version of the cylinder is an earlier one which did not have this pocket. I realise that this is not much help to you but maybe this post will help others planning their systems to ensure they get a cylinder with a separate pocket if they want to use a system like Evohome.

    The manual for the cylinder which you quoted from Centerstore is remarkably similar to the installation manual for Megaflo, implying that the Centerstore cylinder is a Wolesley own-brand rebadging of the Megaflo. Note that the Centerstore manual is issue 15 (final page) whereas the Megaflo manual linked above is issue 19, suggesting the sensor pocket is a more recent addition.
    Attached Images Attached Images
    Last edited by Edinburgh2000; 10th January 2019 at 09:16 AM.

  10. #10
    Automated Home Sr Member
    Join Date
    Dec 2018
    Posts
    50

    Default

    Quote Originally Posted by Edinburgh2000 View Post
    I have a Megaflo 210i. But my cylinder has a purpose-built pocket for a separate temperature sensor, like the one provided by Honeywell's Evohome CS92. A picture of my cylinder is attached here. The Evohome temperature probe is on the thin black wire to the left of the lower grey terminal box. The later installation manual for Megaflo Eco shows this pocket on page 9 (1/2" BSP sensor pocket). I fear that your version of the cylinder is an earlier one which did not have this pocket. I realise that this is not much help to you but maybe this post will help others planning their systems to ensure they get a cylinder with a separate pocket if they want to use a system like Evohome.

    The manual for the cylinder which you quoted from Centerstore is remarkably similar to the installation manual for Megaflo, implying that the Centerstore cylinder is a Wolesley own-brand rebadging of the Megaflo. Note that the Centerstore manual is issue 15 (final page) whereas the Megaflo manual linked above is issue 19, suggesting the sensor pocket is a more recent addition.

    Interesting... I've had another look and definitely don't see any pockets like yours, the cylinder does look a little different from your photo but the manuals have striking similarities . It is a brand new cylinder / installation so possibly just another model with the same circuitry.

    I'll contact Wolesley and see what they say, but doubtful they'll come back with much assistance. If I can get this sensor tucked underneath the insulation better, so it contacts the surface of the cylinder along it's entire length, it will read much more accurately I'm sure. May not be clear from the picture but it's kinda sitting diagonally at the moment with only the tip in contact with the metal.

    I'm sure this isn't an insurmountable hurdle, just a small inconvenience until I work out what size hammer I need to fix it

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •