Delay timer for heating pump

[GrahamPye]

Results of some experiments... My preference would be to use some sort of thermostat rather than a timer as I thought that would be "self adjusting". First discovery was that the pump didn't seem to object to circulating through the by-pass with both the motorised valves shut.

So I tried measuring the pipe temperature on the output of the boiler while running the pump, and also looking at the boiler body temperature as before. What happened was that after a couple of minutes, the circulating water reached a fairly stable temperature (about 55C) and didn't get any cooler. This had obviously taken the boiler body down as well, but once I turned the pump off, it gradually started rising again.

Which gives me a problem that there isn't a pipe temperature that I could monitor to turn the pump off again. Perhaps I need a thermostat with a remote sensor that I can put in one of the boiler pockets alongside the other stats and use that to power the pump?

Maybe a time switch is a better idea! But what signal do you use to trigger the time starting - is the boiler being switched off enough?

Or is there some way that I can get one of the motorised valves to open without it bringing the boiler back on again - I can't easily think of how to do this without relays or a stat with at least two pairs of contacts.

Sorry - I'm thinking "aloud" here, or maybe I'm rambling aloud 🙂

Graham

 

[SJB060685]

Results of some experiments... My preference would be to use some sort of thermostat rather than a timer as I thought that would be "self adjusting". First discovery was that the pump didn't seem to object to circulating through the by-pass with both the motorised valves shut.

So I tried measuring the pipe temperature on the output of the boiler while running the pump, and also looking at the boiler body temperature as before. What happened was that after a couple of minutes, the circulating water reached a fairly stable temperature (about 55C) and didn't get any cooler. This had obviously taken the boiler body down as well, but once I turned the pump off, it gradually started rising again.

Which gives me a problem that there isn't a pipe temperature that I could monitor to turn the pump off again. Perhaps I need a thermostat with a remote sensor that I can put in one of the boiler pockets alongside the other stats and use that to power the pump?

Maybe a time switch is a better idea! But what signal do you use to trigger the time starting - is the boiler being switched off enough?

Or is there some way that I can get one of the motorised valves to open without it bringing the boiler back on again - I can't easily think of how to do this without relays or a stat with at least two pairs of contacts.

Sorry - I'm thinking "aloud" here, or maybe I'm rambling aloud 🙂

Graham

You probably would be best to copy and paste your post and start a new thread.

The problem you're experiencing is something I've come across on oil boilers a lot. You're not going mad, the overheat stat is doing its job. There's a lot of residual heat left in oil boilers when burner and pump shut down and as you say creeps quite rapidly on shutdown. There's several ways you could go about this, each would be a somewhat custom design.
A time delay relay on pump and possible heat sink circuit is one way. There are other possible approaches though.

 

[John.g]

Why not consider installing a normally open type motorized valve like the Honeywell V4043B1257 installed between the flow & return as far away as possible from the boiler and wired so that when there is no demand from all zones it will (spring) open and will close with demand from any zone, this may not even need pump overrun due to gravity circulation around the boiler and pipework but if overrun is required a simple pipe stat set to say 85C will start/stop the pump as/if required with its normal 5C to 8C hysteresis.

 

[GrahamPye]

Why not consider installing a normally open type motorized valve like the Honeywell V4043B1257 installed between the flow & return as far away as possible from the boiler and wired so that when there is no demand from all zones it will (spring) open and will close with demand from any zone, this may not even need pump overrun due to gravity circulation around the boiler and pipework but if overrun is required a simple pipe stat set to say 85C will start/stop the pump as/if required with its normal 5C to 8C hysteresis.

John,

Thanks, that's quite a neat idea. I presume I'd wire that into the boiler feed so it closes whenever the boiler is firing. The downside is that it's (for me!) quite an expensive part, and in my house there isn't going to be an easy place a long way from the boiler that I can get at the pipework and run a wire to!

Having rejected the idea of a pipe stat as that doesn't change enough, I've now ordered a stat with a capillary tube, and I'm going to put the capsule in one of the boiler pockets and use that to bring the pump on - I really only need it to trigger at about 90C (or more) so that it will take enough heat from the boiler to stop the 110C overheat stat triggering.

If it turns out that there isn't enough pipework to dissipate the heat, I'll come back to your idea and see if I can work out somewhere to put the valve.

OR.... What if I replaced the existing MoMo DHW valve with the one you suggest and wired it so it would close when the Heating zone was calling for heat, but at all other times it would be open so the pump overrun would circulate through the cylinder. I'd have to work out how the DHW stat/timer would bring the boiler and pump on though. This might also mean that I couldn't have Heat and DHW at the same time - I think I need to draw a diagram!

Graham

 

[SJB060685]

As John said above is another approach. How about a time delay relay to pump, set at say 5 minutes for pump overrun and an additional motorised valve feeding a heat sink radiator or towel rail for example. Wire it in so when there is a demand on boiler live from other motorised valves the new valve closes shut and the boiler and pump energise. When the room stat or cylinder is satisfied power to the boiler live will be cut, the new valve springs open and the pump overruns for the set time around the heat sink circuit. I really do think you're going to need some mass to help prevent overheating, im not convinced a dead short from flow to return is going to be enough, although I could be wrong.

 

[GrahamPye]

As John said above is another approach. How about a time delay relay to pump, set at say 5 minutes for pump overrun and an additional motorised valve feeding a heat sink radiator or towel rail for example. Wire it in so when there is a demand on boiler live from other motorised valves the new valve closes shut and the boiler and pump energise. When the room stat or cylinder is satisfied power to the boiler live will be cut, the new valve springs open and the pump overruns for the set time around the heat sink circuit. I really do think you're going to need some mass to help prevent overheating, im not convinced a dead short from flow to return is going to be enough, although I could be wrong.

Since I'm more of a sparks than a heating engineer, I was hoping to figure out an electrical solution that didn't involve any new valves or pipes 🙂 The problem with changing any pipework not right next to the boiler / cylinder in this house is that all the plumbing is under a suspended floor which is either tiled or covered in solid oak flooring that I can't lift, and where I can get under the floor it's too low to access the other areas.

But I think you may be right that the short loop round the pipework local to the boiler controlled by the thermostat actually on the boiler itself probably won't be enough so I'll have to do something else. I'm still favouring some sort of thermostat to control this rather than a timer - am I going down the wrong course here?

So, I'm going to try my new boiler stat first, when it arrives on Monday, while continuing to think through how I might wire things up to either get one of the existing two port valves open during the pump over-run without everything entering a perpetual loop...

Thanks for your continued suggestions, I'm learning a lot!

 

[SJB060685]

You could use a clamp on pipe stat wired in to bring the pump on at a certain temperature but we still have the problem of where we're pumping this water.
Is there a bypass on the system and if so where with respect to boiler location?

 

[John.g]

You mentioned "the automatic bypass valve" and also that the boiler temperature was stable at 55C so where is this ABV (if installed) and what is the difference in a pipe stat/ capillary stat operation as they both work on a set hysteresis?. So a heat dump may not be required as the stat will cut in/out.

 

[GrahamPye]

You could use a clamp on pipe stat wired in to bring the pump on at a certain temperature but we still have the problem of where we're pumping this water.
Is there a bypass on the system and if so where with respect to boiler location?

I tried a pipe stat but what happened was that after a while the circulating water didn't get any colder so the stat would never have cut out. That's probably going to happen with my capillary connected stat on the boiler casing too, but I thought it was better to measure the problem at source and if I set that stat to about 90C it will only actually run the pump when there's about to be a chance of the overheat stat tripping. Well that's my thinking!!!

 

[SJB060685]

I tried a pipe stat but what happened was that after a while the circulating water didn't get any colder so the stat would never have cut out. That's probably going to happen with my capillary connected stat on the boiler casing too, but I thought it was better to measure the problem at source and if I set that stat to about 90C it will only actually run the pump when there's about to be a chance of the overheat stat tripping. Well that's my thinking!!!

I suspect that happened because there is nowhere to displace the residual heat, which kind of reinforces my point about needing something to help.

 

[John.g]

That implies to me that the stat was set too low, if it was still made at 55C then its setting must have been ~ 65C max even allowing for a (unlikely) 10C hysteresis or else incorrectly wired ie made at falling temperature?.
I use EPH pipe stats which have a hysteresis of 6/8C and actually use a capillary in contact with the pipe.

 

[SJB060685]

A pipe stat would be wired in to make on rise as you say John. I you certain you wired it correctly?

 

[GrahamPye]

A pipe stat would be wired in to make on rise as you say John. I you certain you wired it correctly?

I didn't actually get as far as wiring the pipe stat in, I put another thermometer on the pipe, temporarily wired the pump to come on and then watched the thermometer on the pipe and the one I already had on the boiler. Having watched the temperature readings I then put the pipe stat back in its box as I realised it wasn't going to do the job 🙂 Maybe it would do if there was somewhere for the heat to go to...

So given that I'm sure you're both right and I'm going to need to find somewhere to lose the residual heat to, I thought I'd be better monitoring the "problem" at source, which is why I've ordered the stat with a capillary bulb so I'm looking directly at the boiler rise. I told you was a sparks not a heating engineer!!!

Once I've proved to myself that I can't lose enough heat just circulating round the pipework I'll add something more complicated. It maybe that I can lose just enough heat to keep the boiler sufficiently below 110C that the overheat won't trip which will be OK for now. I'm sure you're going to tell me that this isn't going to do the boiler any good in the long term though!

To answer an earlier question, there's about 7.5metres of uninsulated 22mm copper in the loop from the boiler through the ABV and pump and back to the boiler. I know it should be insulated, but SWMBO likes her airing cupboard to be nice and warm...

Thanks both, Graham

 

[SJB060685]

Yes take that approach first and see if that short loop is enough to dissipate the residual heat. The pipes not being lagged will help in this case.
We're not getting funny at you, was just checking a couple things.
If it turns out that that approach isn't enough we can discuss other options.
Being an electrical engineer you know as well as I do there are units that can be used to get the job done.
The boilers are quite solid and are exposed to some extreme temperatures during operation. As long as the overheat stat is functioning properly then it will cut power to the burner control box when needed.

The one thing I should add is if the water was at and over 100°c and the system was to suddenly lose pressure, be it a PRV or bad joint etc then the potential for the water to flash to steam is very real and can cause catastrophic damage.

 

[GrahamPye]

Yes take that approach first and see if that short loop is enough to dissipate the residual heat. The pipes not being lagged will help in this case.
We're not getting funny at you, was just checking a couple things.
If it turns out that that approach isn't enough we can discuss other options.
Being an electrical engineer you know as well as I do there are units that can be used to get the job done.
The boilers are quite solid and are exposed to some extreme temperatures during operation. As long as the overheat stat is functioning properly then it will cut power to the burner control box when needed.

The one thing I should add is if the water was at and over 100°c and the system was to suddenly lose pressure, be it a PRV or bad joint etc then the potential for the water to flash to steam is very real and can cause catastrophic damage.

Good point about the boiling water, and of course with the slightly increased pressure in a sealed system, it will as you say instantly boil if there was a leak. Hmm, scary!

I have wanted an excuse to deploy a PLC or even something like a Raspberry Pie to control the heating for some time, this might push me over the edge, although I don't suppose the management will approve 🙂

I'm glad of your advice and you're quite entitled to ask a heating amateur obvious questions!

 

[SJB060685]

Good point about the boiling water, and of course with the slightly increased pressure in a sealed system, it will as you say instantly boil if there was a leak. Hmm, scary!

I have wanted an excuse to deploy a PLC or even something like a Raspberry Pie to control the heating for some time, this might push me over the edge, although I don't suppose the management will approve 🙂

I'm glad of your advice and you're quite entitled to ask a heating amateur obvious questions!

We're here to offer free, impartial and safe advice.
I'm glad you understand the connection between pressure and boiling point. In case you didn't know water which flashes to steam expands some 1600 times, so if your system consists of 100 litres volume (quite a large system and used as an example) then you would have 160,000 litres of steam contained within the boiler and system pipework, this would obviously be causing extreme pressure and something will rupture in a flash, most likely taking out a large portion of the dwelling, I've seen it on unvented cylinders a couple times.
I can't comment on the PLC or Raspberry Pi control systems as I have zero knowledge or experience with them, although I will have a read up later.
If you do decide on controls like that then make sure its compatible with your system.

 

[John.g]

I didn't actually get as far as wiring the pipe stat in, I put another thermometer on the pipe, temporarily wired the pump to come on and then watched the thermometer on the pipe and the one I already had on the boiler. Having watched the temperature readings I then put the pipe stat back in its box as I realised it wasn't going to do the job 🙂 Maybe it would do if there was somewhere for the heat to go to...

So given that I'm sure you're both right and I'm going to need to find somewhere to lose the residual heat to, I thought I'd be better monitoring the "problem" at source, which is why I've ordered the stat with a capillary bulb so I'm looking directly at the boiler rise. I told you was a sparks not a heating engineer!!!

Once I've proved to myself that I can't lose enough heat just circulating round the pipework I'll add something more complicated. It maybe that I can lose just enough heat to keep the boiler sufficiently below 110C that the overheat won't trip which will be OK for now. I'm sure you're going to tell me that this isn't going to do the boiler any good in the long term though!

To answer an earlier question, there's about 7.5metres of uninsulated 22mm copper in the loop from the boiler through the ABV and pump and back to the boiler. I know it should be insulated, but SWMBO likes her airing cupboard to be nice and warm...

Thanks both, Graham

7.5M of uninsulated 22mm copper will emit ~ 450 watts (0.45kw) at only 75C so I would be (very) surprised if this isn't sufficient to dissipate the heat in a relatively short period and well before the temperature reaches 110C, you don't want to be wasting energy either so if the stat is, as you said, set to ~ 90C then I think (hope) you will be in business, I also think you have already proven this anyway as the temp was only 55C with the pump running in overrun, the stat/pump may come in a few times but that should be it.

 

[GrahamPye]

7.5M of uninsulated 22mm copper will emit ~ 450 watts (0.45kw) at only 75C so I would be (very) surprised if this isn't sufficient to dissipate the heat in a relatively short period and well before the temperature reaches 110C, you don't want to be wasting energy either so if the stat is, as you said, set to ~ 90C then I think (hope) you will be in business, I also think you have already proven this anyway as the temp was only 55C with the pump running in overrun, the stat/pump may come in a few times but that should be it.

Excellent - I'll report back once I've wired the new stat into circuit next week!
Thanks, Graham

 

[John.g]

We're here to offer free, impartial and safe advice.
I'm glad you understand the connection between pressure and boiling point. In case you didn't know water which flashes to steam expands some 1600 times, so if your system consists of 100 litres volume (quite a large system and used as an example) then you would have 160,000 litres of steam contained within the boiler and system pipework, this would obviously be causing extreme pressure and something will rupture in a flash, most likely taking out a large portion of the dwelling, I've seen it on unvented cylinders a couple times.
I can't comment on the PLC or Raspberry Pi control systems as I have zero knowledge or experience with them, although I will have a read up later.
If you do decide on controls like that then make sure its compatible with your system.

If you have 100 litres of water at 110C and a massive leak/rupture occurs then this water will fall to 100C instantly but 100 litres of water will not flash off as steam, ~ 2 litres will, and will expand as stated above, the remaining 98 litres will remain as water at 100C, still not pleasant if it rains down on you.

 

[SJB060685]

If you have 100 litres of water at 110C and a massive leak/rupture occurs then this water will fall to 100C instantly but 100 litres of water will not flash off as steam, ~ 2 litres will, and will expand as stated above, the remaining 98 litres will remain as water at 100C, still not pleasant if it rains down on you.

My understanding is any water at or over 100°c will flash if there is a rupture, although not all the system volume will be at those temperatures. At the moment the pressure drops water will retain that heat but be at or below atmospheric pressure quickly flashing to steam. That's the problem with unvented cylinders wrongly installed and YouTube videos will show the damage possible.
If I'm misunderstanding something then please correct me.