I have a property with a mid late 70's oil fired CH system (approx
120,000btu's), configured in what I think is the S plan, which uses 2 2 port valves and one pump acting on an indirect cylinder to provide pumped CH and gravity HW.
Running it in CH or CH & HW it works fine, though in HW only mode (hence gravity flow only) its very slow to heat the cylinder- always has been.
In HW operation, the boiler quickly fires up to the thermostat temperature (80deg C, about 2 mins from stone cold) then shuts off and doesnt cycle again for at least 20mins. Recovering the cylinder to temp takes about
1.5hrs. (!)
The solution to me to improve the HW recovery time would appear to be to add another pump in the HW circuit, switched through the HW valve, converting the HW from gravity to pumped.
Lunacy or not?
I can attempt to describe the actual water circuits if it helps..!
S plan is two 2-port valves and a pump and both the CH circuit and the HW coil are pumped.
Have a look at
formatting link
you have a gravity circuit, it may have C plan. This is a technique to shut down the boiler and to prevent gravity circulation when the HW thermostat is satisfied. However, there is not then normally a second valve for the CH. Normally in that case, the room thermostat causes the pump to run and the boiler to be fired.
Another way to check for a gravity system is that you typically have
28mm pipes to the cylinder whereas a pumped one is 22mm normally.
Please take a look at the hookup. I suspect, though, that you do have an S plan and what is happening is that the heat transfer to the cylinder is poor. There is a clue in that from your comment about the boiler cycling but there can be several factors:
- Cylinders of that era are not that good at heat transfer anyway because they have a small coil area.
- It may be that the coil is covered in hard water scale and that will further inhibit heat transfer.
- It's possible that the thermostat on the boiler has a large hysteresis (temperature difference to switch on to off and back) and that it's taking a long time to switch back on as a result of poor heat transfer. This could be because of the boiler design or a failing boiler thermostat.
With a 35kW boiler, and a poor cylinder, it is not that surprising to have a 10:1 off/on cycle - it suggests only 3kW going into the cylinder. You don't give the cylinder size, but this all fits since you have performance comparable to an immersion heater.
The system works better when the CH is on because the heat load is closer to the boiler production rate.
If the system is indeed gravity to the cylinder, then making it S-plan or even Y-plan *may* help.
However, there is a limit to that if the cylinder is that old and perhaps scaled. If you also make the control arrangements HW priority, it will mean that whenever the HW demands heat, the boiler serves the cylinder exclusively and the CH receives no heat. If the cylinder is poor, you will make the situation worse because the house will cool into the bargain.
A better solution would be to replace the cylinder with a fast recovery type as well. These will absorb heat at a far greater rate and heat the water much faster as well as not cycling. Then you can have a HW priority arrangement where the boiler output will go exclusively to the cylinder when it needs it and should re-heat from cold in 20 mins or so - less for partial use. Then the boiler is returned to CH use. This is a better arrangement because it reduces cycling of the boiler.
Hi andy, and thanks for the already bountiful info. Ok more specifics:
The Boiler is Sectional unit with one water path through it (i.e. just a feed and return).
The HW gravity circuit is in 22mm copper and the cylinder is an 'eco' green foamed single coil indirect situated on the 1st floor of about 5yrs old, some 7 meters from the boiler on the ground floor. Around 5 meters of the 8 comprises ~horizontal pipework running along a floor joist.
The CW circuit at the boiler is in 28mm copper with the pump on the return side.
Both the HW and CH Returns 'Y' into the base of the boiler through about
1/2m of 1.5inch iron pipework.
The Flow from the boiler exits at the top of it through a 1.5inch Iron T piece with the HW in 22mm continuing to rise vertically upto ceiling height. The flow for the CH exits right at the T piece in 28mm copper.
There are 2 port valves on the returns from both water circuits, allowing the progammer to control each.
Last year the header tank (was open vented) was removed and replaced by a sealed system accumulator etc when radiators were added to an attic conversion to allow them to work. (which they do fine when in CH mode or CH
HW mode)
It strikes me that the head-end of the installation is not as good as it might be in general.
It does sound like you have a hybrid system a bit like an 'S' Plan, but with only the heating pumped. I would move the pump so it pumps both CH+HW (i.e. place it between the 'T' and the boiler) and rejig the pump so it goes whenever either valve is opened.
If that doesn't cure it, you'll also need a more modern cylinder.
Also, ensure that the boiler has an overheat cutout. An old 70s boiler would normally not, which would make your system extremely dangerous now that it has been switched to sealed operation. A simple single failure (of the main boiler thermostat) could lead to major damage, or even explosion or fire.
Having a) 22mm pipe, b) a lot of horizontal run is probably making the flow through the cylinder coil quite poor.
I think that a reasonable starting point would be to convert the system to fully pumped so that at least you can get a reasonable flow through the cylinder coil. That needs to be done anyway.
One approach would be to re-organise the pipework so that the pump is able to drive both circuits (i.e. the classical S-plan from the Honeywell diagram).
Another, which may be simpler, if you don't fancy this amount of pipework reorganisation, would be to put a second pump in the HW circuit. The motorised valves have an auxilliary switch (orange connection on the S-plan) which is normally used to turn on boiler and pump.
You can see that in the classical S-plan, the arrangement is that the programmer, together with relevant thermostat opens the appropriate motorised valve when heat is allowed and required. The orange aux. connections then fire up boiler and pump with a logic that either of them will do it.
If you go for a separate pump for each circuit, you may need to add one, possibly two relays to separate the behaviour of each circuit. In this way, the programmer and thermostat would open the relevant valve as now, but the orange connection would drive the pump for the associated circuit plus a relay. You would then wire a live through the normally open contacts of both relays to fire the boiler.
It sounds more complicated than it is, but may well be a faster option to implement.
In either case, I think that it's very important to draw out a complete wiring diagram for what you have so that you can see what's going on; especially because the current system is not a true S-plan.
This should make a big difference, and reasonably easy to implement. For more performance, you would need to look at a fast recovery cylinder, but you would still need to go for fully pumped, so I'd start with that.
It might be easier to just wire the pump up to the zone valve control. Saves the need for a relay and it will only be pushing against the closed valve for a second or two.
Do you mean wire the pump to the motor side of the valve, Christian?
I have done that in the past, but it made the pump pretty noisy while it was opening. OTOH, in this case, one could probably run the pump for the HW on its lowest setting since the circuit flow resistance should not be as much as for the CH.
I can imagine that being the main issue. It probably depends on the pump type, speed and location as to whether it is a problem in practice. It might be worth a go before mucking about with relays, though. If there isn't annoying whistling, it saves on the bother, and the wet side is identical.
You beat me to it! It shouldn't do any harm for the pump to run stalled for a little while until the valve is open. Not quite sure what happens if the valve fails, though!
I presume - since the boiler apparently works on with a gravity circuit - that it doesn't need a pump over-run, so we're not into by-pass circuits etc.
[I'd not heard of having an unvented gravity system - (not even on old cars!) Is this unique?]
The pump would be left running against a closed valve. I suppose that if the aux. contacts are used to run the pump, if the motor fails the pump can't start via that mechanism. Theoretically the aux switch could stick in the on position, I suppose, but that would be an issue with any arrangement. Probably a very unusual failure mechanism.
This is a good point. Gravity circuits without valves obviously don't need a bypass. However, what has been created here is rather like C-plan on the HW side - i.e. the path is closed when cylinder stat is satisfied.
It is possible, that on the existing arrangement that if the boiler does need bypass that there actually is one around the CH circuit and that the CH pump is connected to an over-run thermostat on the boiler rather than directly from the contacts on the motorised valves. That way, if the HW was running and the boiler is shut down in full flight, the pump would run and dump the heat through the bypass.
The system is slightly odd anyway, which is why it's important for the OP to begin by looking at the wiring and sketching out a plan.
As far as the wiring is concerned, operating the system it appears that the boiler is swtiched through the contacts in the motorised valves, as it will not fire until which ever one the programmer is calling for has opened. The existing CH pump is wired in parallel with the boiler through a relay connected to the CH valve.
Thus the pump only has power when CH or CH & HW are selected.
There are both cylinder stats acting on the valve circuit, and room stat acting on the CH->boiler circuit.
If the room stat is satisfyed the CH valve shuts and the system continues to heat the HW via gravity (or not very well) and if the cylinder stat is satisfyed the HW valve shuts, and the system continnues to heat the CH via the pump.
Hope this provides the missing link..
I am prepared to totally rewire the logistics of the system should it be needed if it results in hot water!!
How easily can you move the pump from the CH branch into the common bit of pipe so that both circuits are pumped? That would require very little change to the control logic - simply put the pump fully in parallel with the boiler and get rid of the relay. It should then exactly match the Honeywell S-Plan diagram.
If - as I think you are saying - the hot water gets hotter quicker when the CH is on than when it isn't, it suggests that pumping water round the radiators in some way helps to speed up the gravity circulation in the other circuit. So positively pumping it will almost certainly solve your problem without requiring a new HW cylinder with larger heat exchanger.
Bizarely I had not considered this!! I could conjure some way to move the pump into the common return just before the boiler yes, by some deft adaperment of the 1.5" iron pipework.
However, would this not take circulation away from the CH circuit which poses the greater 'work' than the easily circulated HW circuit??
The CH cirucit is fairly lengthy (as it would be in a 5+2 bedroom, 3 story farm house) and making an easier path for the water via a single pump might lower the CH performance. (??)
As I said, though, do take the time to draw out a detailed diagram of what you have, if necessary taking off individual wires and using a continuity tester or a meter.
It's all too easy to start disconnecting stuff, getting lost and ending up getting very confused.
If you want to run the CH and HW at the same time, you may well need some sort of restrictor in the HW circuit (e.g. an adjustable gate valve) to achieve some sort of balance and make sure that the CH still gets adequate flow.
However, another possibility is to have independent timing on the two circuits and to contrive - under normal circumstances - not to have them both on together. This is easily achieved with a programmable room stat. You set the timer on your main programmer to go on and off when you want the hot water to be heated. You set CH to "constant". You then programme your programmable room stat for the times when you *actually* want the CH on.
I have done exactly that, and it works like a dream. Even though my HW and CH times overlap a bit, they're rarely both on at the same time because the water is hot and the cyl stat has operated before the CH comes on.
If you *really* wanted to go to town you could by-pass the HW gate valve with another zone valve, and rig up some logic so that it opened when the HW was on but the CH wasn't - but that's not really necessary in my view.
This would be creating a conventional arrangement from the plumbing perspective.
However, there are some solutions to your concern:
1) Run both circuits so that if both thermostats demand heat, both valves are open.
It's hard to guess how the water will flow because on the CH you have a large pipe but several radiators in parallel on it. The HW circuit is in a smaller diameter pipe. You could balance between the two with a valve on the HW circuit if the HW circuit seemed to be getting all the flow and the boiler was cycling a lot.
Unfortunately, it's a bit hit and miss, because if you have TRVs on the radiators, you will get a variation because of that.
2) you can rearrange things to give DHW priority.
This means that if both circuit thermostats are indicating heat demand, only the cylinder is fed. Thus all the boiler output goes to the cylinder, which should then heat quickly, and on completion, the flow is returned to the CH.
Normally this is done with a diverter valve rather than two 2port valves (See other Honeywell plans). You can do it with separate valves, but then the wiring becomes more complicated.
HomeOwnersHub website is not affiliated with any of the manufacturers or service providers discussed here.
All logos and trade names are the property of their respective owners.