7 zone 2441TH Thermostat setup

[50-50]

Okay, need your expert advice on feasibility.

 

My house has 7 zones for heat, two for cooling (2 of the 7 control two A/C units as well - north half/south half).  Existing Tstats are of the "independent and dumb" variety.

 

The heating season is coming and I need to save fuel.  Propane.  $3+/gallon. 1,700 gallons consumed last year.  New house (2006), well insulated, but the boiler ran almost continuously because each zone is completely independent... and they are rarely, if ever, in sync.

 

4 zones are radiant floor heating (hot water) + 3 zones are hot water baseboard

 

I have an ISY and 40+ switches/iolincs/motion/etc.  ISY is my preferred method for Tstats.  Willing to consider other options, but they need to be easier or cheaper than the ISY setup.

 

The zone controller brings all 7 zones into 3 circulator pumps that drive manifolds.  The manifolds have zone valves like this:

M1:  Radiant floor:  z1: Kitchen & dining -- the whole manifold is one zone == (1 Tstat)

M2:  Radiant floor:  z2 - Living Room (+A/C); z3 - Main Bath; z4 - Master Bath == (3 Tstats) 

M3:  Hydro baseboard: z5 - master Bed (+ A/C); z6 - Second bed; z7 - Third bed == (3 Tstats)

There is an 8th zone (4th manifold) for domestic hot water.

 

Got it?  Simple enough? 

 

Due to the complexity of the setup, when just one zone calls for heat, a lot of pipe needs to heat up.  What I really want to do is intercept the call for heat and modify it in the ISY...  by creating logical primary zones and secondary zones to make this more efficient (using thermal mass / restricting run time of the boiler).

 

Example 1:  weekend daytime:  Living room and Kitchen are the primary; bed3, main bath, master bed, master bath, bed2 are secondary (in that order).

Example 2:  weekday daytime:  Living room, Bed3, Kitchen are the primary; main bath, master bed, master bath, bed2 are secondary (in that order).

Example 3:  nighttime:  Master bedroom and MBath are primary, then kitchen, living, foyer are secondary (in that order), bed2 and bed3 are off unless guest is in house.

 

->  No zone can turn on the boiler except current logical primary zone(s), they trigger at 2 degrees under setpoint

->  When a primary calls for heat, and a secondary zone on same manifold is under its setpoint, heat both (eg. living room and main bath) until each satisfies

->  Attempt to only circulate to 1 manifold at a time (except M4, which is independent)

->  When the requesting primary is satisfied, the other primary is checked for delta to setpoint, if >= 1 degree difference, heat that (follow same rule for 2ndary zones on same manifold)

->  After all primarys are satisfied, follow a pecking order for secondary zones based on weekend/weekday/guest in house (bedrooms 2 and 3 are guest bedroom and home office, respectively)

->  Emergency override if any zone falls below 55 -- heat immediately

->  (future**) If domestic hot water sensor is "hot", check zones in order and "heat to satisfy" if current temp <= 2 degrees below setpoint

->  (future**) force boiler off, circulate to last unsatisfied zone for <time> or until domestic hot water sensor is "cold"

->  (future**) indicate the dist. infra. is hot on a KPL -- time to run washer/dishwasher/take a shower.

 

**  Note the domestic hot water (M4) has a flow valve with a temp sensor (Taco I believe) in it already, I need to figure out if I can tap into this with an IOLinc.  The sensor will go hot when a heating zone is calling for heat and the distribution infrastructure heats up, even if the domestic hot water is not being used.

 

So.. am I nuts?  Can I do this in an ISY?  Is it smart... what if the ISY freezes up when no one is home (its been really, really stable)!  Can I "make it safe" on the condition of a failure of the ISY?

 

To keep domestic tranquility, I'd like the Tstat to show its current setpoint and the actual temp on its display, and if the Tstat is manipulated, for the ISY to recognize it and continue on... (current temp vs set point).  Possible?

 

Thank you!

50/50

[Teken]

Hello 50/50,

 

I have an alternative solution for you that I hope gives you something else to consider. I would suggest if the zones where the temperature needs to be measured can be accessed. Deploying a 1 wire temperature network would provide you the same capability with less hardware invested.

 

The basic's would require that you either deploy a home brew 1 wire network using the RPi (Raspberry Pi) and a 1 wire hardware master controller. Or go with a consolidated 1 wire software box from Autelis (Autelis ISY Bridge) and interface it with the ISY Series Controller.

 

You will of course require the Network Module plugin from UDI for both solutions.

 

I have just recently deployed the first phase of my 1 wire temperature / humidity sensor network in my home. It has so far worked out tremendously in all aspects for my intended use case.

 

Here are a few snap shots of my system and random programs that react to the temperature changes in my home. There are currently twenty 1 wire temperature sensors installed in and around my home which provides local temperature / humidity that interacts with the ISY Series Controller and integrates with the Insteon network.

 

By this years end I should have all 40 sensors deployed and installed. Once done this will allow me to control all aspects of the HVAC system and reduce my energy consumption and costs.

 

Since going down this road the journey to save money and be more efficient has paid off in spades. I have appended my last electrical bill which you can see (actual electricity used) is extremely small.

 

 

 

 

[50-50]

Very interesting.  Let me test my thinking.

 

a)  I will leave the existing dumb thermostats in place for my "oh crap" scenario.  Ie. the network router, ISY or the Autelis bridge go belly up and no one is home.  Just set them to the absolute minimum desired temperature... say 55 F.  This will let the system work "as originally built" if a room temp falls below 55.  It can also serve as a "Manual Override" if I am away or unreachable, and the boss needs heat, now!

 

  Install 7 x one-wire temp probes [next to each Tstat], connect to the Autelis ISY bridge, pump temps into ISY.  Add an additional probe to the boiler hot water feed to know if the water distribution system is "pre-heated".

 

c)  Install 7 x IOLincs next to my heating zone controller (pic of one attached), wire them in parallel with the Tstats N/O (Normally Open) on the 24 VDC HVAC circuit.

 

d)  When a Tprobe fits my logic (primary zone, 2 degrees under setpoint, etc), close the corresponding IOLinc until the Tprobe is satisfied, then re-open it.  For secondary zones, use logic to determine the manifold in use, actual temps vs. target for each room on that manifold and its priority, and if the system is already hot before closing the IOLinc, re-open when satisfied.

 

e) get fancy with the ceiling fans, etc.  In fact, I could experiment with heat/cool times with and without fans to find which one really does help more, normalized to outside temps (fans move air, moving air is "cooler" to the body than still air, does that make the boss want the setpoint higher?  Or do the fans make the room temp more "stable" and therefore more comfortable, while cycling the heat less?)

 

f)  figure out cooling later, but use the same Tprobes.

 

g)  investigate if there is a way to do a 7x N/O IOLinc without 7 large boxes, integrated switching transformers, etc...  Maybe Elk?  Note I have no Elk today and not sure how the costs add up.

 

Cost:  $130 for Autelis bridge + maybe $100 for probes + $315 for IOLincs + wire and time... or $550 plus wire... beats $1,050 for 7x2441TH and since I only have two conductor bundles to the Tstats (don't ask.. I really shook my head in disbelief at this!), I need to run new wire anyway.

 

Anyone see any issues or drawbacks with this idea?  I assume I will need to run Cat5/6 for the digital Tprobes ).. and not just regular Tstat wire? (Teken, I will likely do humidity at the same time as you did... just makes sense.)

 

Thanks much!  50/50

[johnnyt]

Instead of IOLincs would suggest http://www.digital-loggers.com/din.html. Search my username and IOLincs to hopefully find (and be able to avoid) the problems I had with IOLincs. I think just due to too many but I also needed to query them to ensure reliability in my hvac. Too many times my IOLincs query resulted in "safeguards" I had to code activating. The querying also killed ISY performance and lead to other problems. Am on Tapatalk and don't have the links easily accessible. Try also a search of "DIN" to find more on the din relay...

 

 

Sent from my iPad using Tapatalk

[Xathros]

Johnny beat me to it!.  The din relay if FAR cheaper than the IOLincs and is rock solid reliable.  You will need the ISY network module to control the DIN relay but the two items combined is still much cheaper than 7 IOLincs.

 

-Xathros

[Teken]

Very interesting.  Let me test my thinking.

 

a)  I will leave the existing dumb thermostats in place for my "oh crap" scenario.  Ie. the network router, ISY or the Autelis bridge go belly up and no one is home.  Just set them to the absolute minimum desired temperature... say 55 F.  This will let the system work "as originally built" if a room temp falls below 55.  It can also serve as a "Manual Override" if I am away or unreachable, and the boss needs heat, now!

 

  Install 7 x one-wire temp probes [next to each Tstat], connect to the Autelis ISY bridge, pump temps into ISY.  Add an additional probe to the boiler hot water feed to know if the water distribution system is "pre-heated".

 

c)  Install 7 x IOLincs next to my heating zone controller (pic of one attached), wire them in parallel with the Tstats N/O (Normally Open) on the 24 VDC HVAC circuit.

 

d)  When a Tprobe fits my logic (primary zone, 2 degrees under setpoint, etc), close the corresponding IOLinc until the Tprobe is satisfied, then re-open it.  For secondary zones, use logic to determine the manifold in use, actual temps vs. target for each room on that manifold and its priority, and if the system is already hot before closing the IOLinc, re-open when satisfied.

 

e) get fancy with the ceiling fans, etc.  In fact, I could experiment with heat/cool times with and without fans to find which one really does help more, normalized to outside temps (fans move air, moving air is "cooler" to the body than still air, does that make the boss want the setpoint higher?  Or do the fans make the room temp more "stable" and therefore more comfortable, while cycling the heat less?)

 

f)  figure out cooling later, but use the same Tprobes.

 

g)  investigate if there is a way to do a 7x N/O IOLinc without 7 large boxes, integrated switching transformers, etc...  Maybe Elk?  Note I have no Elk today and not sure how the costs add up.

 

Cost:  $130 for Autelis bridge + maybe $100 for probes + $315 for IOLincs + wire and time... or $550 plus wire... beats $1,050 for 7x2441TH and since I only have two conductor bundles to the Tstats (don't ask.. I really shook my head in disbelief at this!), I need to run new wire anyway.

 

Anyone see any issues or drawbacks with this idea?  I assume I will need to run Cat5/6 for the digital Tprobes ).. and not just regular Tstat wire? (Teken, I will likely do humidity at the same time as you did... just makes sense.)

 

Thanks much!  50/50

 

A: I believe this is the best balance of fail over, costs, and reliability.

 

C: I agree with others that your money is better spent in the DIN module opposed to the Insteon I/O Linc.

 

E: In my home the ceiling fans, remote fans, over head range, and central exhaust all play a major role in maintaining temperature and overall air quality. Installing ceiling fans are something you can take advantage anytime but allows great flexibility with out the use of the HVAC system which reduces over all costs.

 

D: All you need to do is either change the ISY program for the seasons in question or simply copy the same programs and change their values to reflect the seasons in which they operate.

 

Again, this was just an alternate view of doing the same thing but on a much cheaper budget and with better value / performance. There are many other ways to accomplish this goal but wanted to offer you my use case and how I was able to accomplish it with the devices I listed and lots of help from a good friend and master ISY programmer Xathros and others. 

 

P.S.

 

This is the vendor I have been purchasing my 1 wire sensors from. He has been a stellar seller and the sensors are first rate. You simply can not get 10 sensors for $16.74 USD cheaper anywhere in the world, none!

 

http://www.ebay.com/itm/130702483183?_trksid=p2059210.m2749.l2649&ssPageName=STRK%3AMEBIDX%3AIT

[larryllix]

Do you have outdoor reset on your system? Most will if they run such a huge thermal mass system as hydronic in slab heating to control overshoot and undershoot.

 

I was thinking you could leave all your stats at a high temperature setting in series with your ISY control outputs.

 

Now with an outdoor temperature, wind, humidity and occupancy sensing and logic  you could run a PWM scheme for all your zones and if the things ever locked-on your old thermostats would limit the amount of heat.

 

if the series contacts failed you would need some crude backup scheme to keep the house at a minimum maybe without zones involved.

 

My thermostats have built in "scenes" to accommodate visitors and special occasions etc. I only use two. "everyday" and "away". With the away I have to predict the number of days before I return. When we don't predict properly and come home early it takes about 18-24 hours to warm up the house in the winter from 17c to 22c with the basement slab heating. My complex thermostats run about $2.5k or more for 5 zones so a separate ISY may be worth your while as a dedicated control. Your "play" ISY could be the backup or watchdog or vice versa in the off-heating seasons.

[50-50]

Lots to think about -- thanks Teken, johnnyt, Xanthos & larryllix!

 

I am more worried about underrun of heat, rather than overrun...  Yes, an overrun would cost propane, but that cost is limited compared to frozen and broken pipes, etc.  I also can get an email alert and remotely shut down the "unstuck" zones, with the theory the stuck-on zone will keep the house sufficiently warm (unfrozen) until I can get home and get it unstuck.  Even trigger on/off the stuck zones to try and unstick it... etc... remotely.

 

Unfortunately the radiant floor heat is the "cheaper" version, all the technology and complexity but no slab -- it is nestled in between floor sheathing (again, don't ask -- the previous owners did this).  No outdoor reset.  How does one work.. perhaps its something I can add?

 

I like the idea of a dedicated ISY for this -- not to bog it down with other functions.  Anyone have any luck (I know its not supported) in passing alerts/variables/messages between two ISYs?

 

Has anyone painted the one-wire temperature sensors?  Could I give them a thin coat of rustoleum to get close to matching the wall color without impacting their accuracy?  Also, has anyone found a similar design for humidistats?  Google searches found some pretty large *and expensive* ones, just want to check here?

 

@larryllix - why don't you enable mobilinc to tell the house when you will return -- 24 to 48 hours out?  Is there a reason you opted to not use the "remote" feature?  I don't use the subscription, rather I configured it without their subscription gateway.

 

Thanks again!

50/50

[50-50]

one more question -- electric code related -- I need larger gang boxes to capture all the wire.  Does anyone know of a box that can be "hidden" behind the wall .. ie. a 5 gang-sized box with 3 gangs accessible for switches -- and one or two ganges extending behind the drywall to the left or right  -- perhaps this has to be metal to protect a dummy from injecting a screw or nail next to a switch?  Also... old work would be great   Since a lot are on exterior walls, I don't want to go with deeper boxes as I loose R-value.

 

As far as volume vs. live wires, I am okay (I removed a bunch of 3 and 4 and yes, 5 way switches and replaced with Insteon..  net is I have a ton of dead wire capped in these boxes making it very hard to fit the actual switches in the box (even though they are within the volume for the number of live wires).

 

Thanks,

50/50

[Teken]

The link I provided you for the 1 wire sensors will be fine if painted. If your intent is to measure (physical) devices I would humbly suggest you do one of two things to ensure as much contact and accuracy.

 

You can take any computer related heat sink material (arctic silver) and apply it to the area and wrap it in some sort of insulation.

 

For my outside water pipes I simply taped ( head of sensor ) with aluminum foil tape and it conducts the heat / cold superbly and allows the transfer of temp to stabilize without outside air influences.

 

Wire ties and staples to hold it to the rafters to secure the cable run.

 

Keep us all in the loop as to your next steps and final solution.

 

 

Encrypted By: Phoenix Security Solutions

[larryllix]

@larryllix - why don't you enable mobilinc to tell the house when you will return -- 24 to 48 hours out?  Is there a reason you opted to not use the "remote" feature?  I don't use the subscription, rather I configured it without their subscription gateway.

 

Thanks again!

50/50

You must have a different Mobilinc than I do.  My Mobilinc doesn't do anything automatically. If there is a "remote" feature it would be the first I have heard about anything like that.

 

Any hints where I would find this or is this in the iOS app?

 

 

My thermostats are not Insteon compatible and the RS485 protocol but they talk is not likely to ever talk to any other gadgetry. For the cost of attempting this I may as well leave the heat on full 24/7. I don't see Insteon thermostats that can vary the speed of an injection mixing  pump  through a communications bus to a mixer panel, with outdoor reset, while cycling an HRV unit on a percentage basis of the time period. 

[larryllix]

Lots to think about -- thanks Teken, johnnyt, Xanthos & larryllix!

 

.....

Unfortunately the radiant floor heat is the "cheaper" version, all the technology and complexity but no slab -- it is nestled in between floor sheathing (again, don't ask -- the previous owners did this).  No outdoor reset.  How does one work.. perhaps its something I can add?

.....

 

Thanks again!

50/50

Outdoor reset is typically used when you have milder weather and you have a huge thermal mass involved in your heating radiator, like a large concrete slab. The outdoor temperature is sensed and the temperature of the water is varied to make the heating more gentle and also to not shock the radiator material.  In a "bang-bang" style heating system if you turned your thermostat up suddenly, from a cold house, the hot water may crack the slab or in your case "cook" the flooring, perhaps cracking ceramics and/or creating some hotspots and overshooting the setpoint, causing discomfort and wasting energy when you have to open your windows.

 

The water temperature response is controlled by various methods including PWM (pulse width modulation) of the injection pump, variable speed injection pump, or proportional mixing valve to get the temperature of water desired.

[50-50]

@larryllix -- no not automatic, rather human intervention via MobiLinc (remotely, ie, no one home) after an automatic email or SMS alert that the interior temp is outside a defined range, say 60F to 78F in heating mode and 70F to 90F in cooling mode.

 

@All -- I am about to pull the trigger on the DIN/Autelis ISY Bridge/DS18B20 temo sensors.  Will keep posting as to status.

 

I'd still like to do humidistats at the same time, came across DS1923 iButton cells, but they don't seem to be an exact fit - more of a track/log/dump use case.  Anyone have a suggestion for a one-wire humidistat?

 

Appreicate the help!

50/50

[50-50]

Outdoor reset is typically used when you have milder weather and you have a huge thermal mass involved in your heating radiator, like a large concrete slab. The outdoor temperature is sensed and the temperature of the water is varied to make the heating more gentle and also to not shock the radiator material.  In a "bang-bang" style heating system if you turned your thermostat up suddenly, from a cold house, the hot water may crack the slab or in your case "cook" the flooring, perhaps cracking ceramics and/or creating some hotspots and overshooting the setpoint, causing discomfort and wasting energy when you have to open your windows.

 

The water temperature response is controlled by various methods including PWM (pulse width modulation) of the injection pump, variable speed injection pump, or proportional mixing valve to get the temperature of water desired.

Thanks for this, on my system there is a thermal feedback on the domestic hot water w/ an automatic mixing valve, I presume to keep from scalding.  However, in a related experiment, I'm not entirely sure its working... for another day to diagnose.   

 

Since this house is 80% pex, I like the idea of warming the pipes more slowly, will have to see if there is an easy way to do this.  Perhaps since it takes so long to get the big, long boiler loop hot, and the zone circulators engage in parallel, I have this without extra controls.... overshoot has not been an issue I have observed  

 

On the positive side, the system as-is has been going for 9 heating seasons and I've not detected any leaks (yet).

[larryllix]

Thanks for this, on my system there is a thermal feedback on the domestic hot water w/ an automatic mixing valve, I presume to keep from scalding.  However, in a related experiment, I'm not entirely sure its working... for another day to diagnose.   

 

Anti-scalding and, lest we forget those that gave their lives, anti-legionella bacteria.

[larryllix]

....

 

Since this house is 80% pex, I like the idea of warming the pipes more slowly, will have to see if there is an easy way to do this.  Perhaps since it takes so long to get the big, long boiler loop hot, and the zone circulators engage in parallel, I have this without extra controls.... overshoot has not been an issue I have observed

 

On the positive side, the system as-is has been going for 9 heating seasons and I've not detected any leaks (yet).

I believe thermal shock can separate the PEX from the concrete slab and then thermal contact becomes a problem for spotty heat zones. With PEX embedded in wood it shouldn't be as much of a problem.

 

Are you aware of the de-zincification of brass PEX fittings? I believe the class-action suit is still ongoing for cashbacks. Looks for white powder on your fittings.

[50-50]

Wow, no, I was not aware of the de-zincification issues.  Does it look like the pic attached?  I thought this was salt from the water softener?

 

I also attached a pic of the whole heating system, for those intrigued.  The supply/return are the two copper pipes to the right of the expansion tank.

[larryllix]

Wow, no, I was not aware of the de-zincification issues.  Does it look like the pic attached?  I thought this was salt from the water softener?

 

I also attached a pic of the whole heating system, for those intrigued.  The supply/return are the two copper pipes to the right of the expansion tank.

The actual events are really hard to track down and there are rumours and guesses everywhere on forums. From what I can make out some cheap brands of fittings had too much zinc in the brass and the zinc leaches out of the brass causing the fitting to collapse eventually. The best report I heard was that the "zinc accumulates in the PEX, causing a blockage and the pressure builds up until the PEX pipe bursts."   I did notice most of the case studies had PEX-AL-PEX crimped with various rings, a complete no-no to crush the aluminum strength layer.

[LeeG]

The other possibility.   House built in 1990 using PEX and brass fittings.   No leaks or other issues with couplings.  

[larryllix]

Mostly in systems built 1995-2007.

 

It started with these guys and has spread to most of the other brands now.

http://www.classactionsnews.com/settlements/zurn-pex-f1807-fittings-litigation-settlement

 

"Problems found to date are with the following products:

IPEX Inc – sold as Kitec, PlumbBetter, IPEX, AQUA, WARMRITE, Kitex XPA, AmbioComfort, XPA, KERR Controls, Plomberie Amelioree

ZURN PEX – sold as QPex (may not be sold in Canada)

Uponer Inc – sold as Plasco Plumb-PEX, P Pex, MB Pex

These companies are all currently involved in lawsuits."

[Xathros]

 

I like the idea of a dedicated ISY for this -- not to bog it down with other functions.  Anyone have any luck (I know its not supported) in passing alerts/variables/messages between two ISYs?

Easy if they are on the same network.  You can pass variables or set devices/scenes on/off all via the /rest interface using the network module on the other ISY.  This is supposed to get even easier with the 5.X firmwares due yearend or early next year.

 

-Xathros

[50-50]

Instead of IOLincs would suggest http://www.digital-loggers.com/din.html. Search my username and IOLincs to hopefully find (and be able to avoid) the problems I had with IOLincs. I think just due to too many but I also needed to query them to ensure reliability in my hvac. Too many times my IOLincs query resulted in "safeguards" I had to code activating. The querying also killed ISY performance and lead to other problems. Am on Tapatalk and don't have the links easily accessible. Try also a search of "DIN" to find more on the din relay...

 

 

Sent from my iPad using Tapatalk

Alright, have all the parts.  Time to build. 

 

I was able to read the temperature using a probe connected to an Autelis Bridge in the ISY with a state variable!  (Teken's posts helped a ton!)

 

Is there a tutorial out there to now get the ISY talking to the Data-Loggers DIN?

 

Best -

50/50

 

EDIT:  a Forum search on "DIN" only yields this thread... somehow I find this odd.

[johnnyt]

50-50, it is strange DIN doesn't return more. I used dinrelay to find a post by Xathros, who put me onto it and has posted his network resources that you can import. Am not near my own ISY and am using Tapatalk so not easy to point you to it or post what I did now but let me know if you don't find it and I'll try to be more helpful when

 

 

Sent from my iPad using Tapatalk

[50-50]

50-50, it is strange DIN doesn't return more. I used dinrelay to find a post by Xathros, who put me onto it and has posted his network resources that you can import. Am not near my own ISY and am using Tapatalk so not easy to point you to it or post what I did now but let me know if you don't find it and I'll try to be more helpful when

 

 

Sent from my iPad using Tapatalk

 

Searching for DINRELAY helped.  I found this post:

http://forum.universal-devices.com/topic/13164-any-recommendations-for-a-relay-interface/?hl=dinrelay&do=findComment&comment=108635

 

The question becomes how do I define the DINRELAY as a Resource in the network module?

 

>> Program: Relay_4_Control

>> If

>> $s.Relay_4 is 1

>> Then

>> Resource 'DINRELAY: Relay 4 On'

>> Else

>> Resource 'DINRELAY: Relay 4 Off'

[50-50]

FYI, Xathros pointed me here:

 

http://forum.universal-devices.com/topic/13909-call-for-network-resources-crowdsourcing-request/

 

Very clever -- and after import -- very clear.  Should get me going just fine.

 

Now to find a slick way to drill through inter-stud blocking 4' up from the floor to run new cable... else I am doing a ton of drywall repair.

 

Thanks again!!

50/50