What Does A Modulating CH Pump Do?

[CareerChangeChancer]

What are the pros and cons of a modulating pump, eg Grundfos Alpha?

Are they suitable for all systems?

Should everyone with vented or sealed systems have one?

Many thanks

CCC

 

[Onetap]

What are the pros and cons of a modulating pump, eg Grundfos Alpha?

Expensive, compared to a 3-speed job.

Variable speed, variable flow rate. Good with TRVs, where the 2-port TRVs reduce the system flow rate and, with a constant speed pump, increase the pressure differential generated by the pump and the system noise. They can also be used with injection mixing systems; unlike 3-port mixing valves they do not have a gland, the usual point of failure.

 

[mickyg]

they have their uses but unfortunately are not compatible with many modern domestic boilers, such as those with the gianonni heat exchanger, as the boilers require quite a high min flow rate, which the modulating pumps may not constantly provide. I think they're useful in large domestic and commercial situations but I'm yet to come across a small/medium domestic property that would get any great benefit from one.

 

[SimonH2]

they have their uses but unfortunately are not compatible with many modern domestic boilers, such as those with the gianonni heat exchanger, as the boilers require quite a high min flow rate, which the modulating pumps may not constantly provide.

I hit that problem when I got my flat - and put TRVs in. The boiler didn't have an internal bypass and it created some interesting noises as I tried to get some semblance of balance into the system. Not helped of course by the fact that it's a combi, and hence sized so that it's minimum output is about 5 times the average load - hence probably at least 10 times the load for much of the time.

I think they're useful in large domestic and commercial situations but I'm yet to come across a small/medium domestic property that would get any great benefit from one.

With a directly heated system you are probably right. With a thermal store, they work very nicely - and you don't need to keep the boiler in both non-condensing mode and short cycling when running the heating.
Compared to my parents house where the valves are often very noisy when it's warm and pressure goes up, the TRVs in the flat are virtually silent.

 

[GeorgeBainbridge]

With a thermal store, they work very nicely - and you don't need to keep the boiler in both non-condensing mode and short cycling when running the heating.

I assume you mean the auto bypass has opened short circuiting the return to the flow. These just drop efficiency like a stone. Most are not set properly and open too early.

 

[ianniann]

With or without a bypass, automatic or otherwise, a boiler will cycle more and more quickly as TRVs shut off. 100' of copper pipe just doesn't shed enough heat, the water in the boiler gets too hot, and then the burners go off. Nothing to do with whether you have a modulating pump, of course. In fact a modulating pump is likely to make it cycle even more quickly by reducing the pump power as TRVs close off. A fixed power pump will just flail away for all it's worth at the few remaining radiators. The result is lots of hissing and all semblance of a delta T gone.

Modulating pumps do reduce noise but I wouldn't have thought it was a major selling point. Maybe that's because I have a modulating pump! Probably the main reason you want one is simply to save money. Boilers and heating controls are increasingly designed to run the system at low temperatures for long periods, so the pump is becoming a more important component of the cost.

The only real downside (apart from costing a little more to buy) is that some boilers just don't get along with them. If you have a system that is designed to work OK with a modulating pump then they're great.

 

[SimonH2]

With a thermal store, they work very nicely - and you don't need to keep the boiler in both non-condensing mode and short cycling when running the heating.

I assume you mean the auto bypass has opened short circuiting the return to the flow. These just drop efficiency like a stone. Most are not set properly and open too early.

I do indeed - but then with a large boiler (such as the grossly oversized ones needed for a combi installation), and a properly setup system with TRVs on most of the rads, you will need the bypass active most of the time in order to keep the boiler running.

I suspect very few replacement boilers spend much time condensing.

 

[GeorgeBainbridge]

With a thermal store, they work very nicely - and you don't need to keep the boiler in both non-condensing mode and short cycling when running the heating.

I assume you mean the auto bypass has opened short circuiting the return to the flow. These just drop efficiency like a stone. Most are not set properly and open too early.

I do indeed - but then with a large boiler (such as the grossly oversized ones needed for a combi installation), and a properly setup system with TRVs on most of the rads, you will need the bypass active most of the time in order to keep the boiler running.

I suspect very few replacement boilers spend much time condensing.

You could fit an oversized plate heat exchanger from the boiler flow to the return. The boiler then just pumps in and out of the plate. On the other side of the plate the CH circuit runs through with a Smart pump on it and TRVs all around. Then full flow through the boiler. If the boiler has integrated weather compensation then it will be efficient enough and the boiler drops the flow temp to suit. The larger efficient plate will allow the CH circuit to extract enough heat to make the return cool enough. There is no direct short cut as the heat in the plate is being transferred.

It also stops sludge entering the boiler, keeping efficiency over time, which normally drops off.

 

[D_Hailsham]

It won't be long before Alpha-type circulators are the only thing available. This is because of EU Directive 2005/32/EC on the ecodesign requirements for glandless standalone circulators and glandless circulators integrated in products (e.g system and combi boilers). This has resulted in the issue of Commission Regulation No 641/2009 in July 2009, which says:

1. From 1 January 2013, glandless standalone circulators ... shall have an energy efficiency index (EEI) of not more than 0.27.

2. From 1 August 2015, glandless standalone circulators and glandless circulators integrated in products shall have an energy efficiency index (EEI) of not more than 0.23.

The method if calculating the EEI is fairly complicated and when the Regulation was issued it stated:

At the time of the adoption of this Regulation, the benchmark for the best available technology on the market for circulators is EEI ≤ 0.20.

Grundfos Alpha Pumps have an EEI of 0.20.

 

[D_Hailsham]

not compatible with many modern domestic boilers, such as those with the gianonni heat exchanger, as the boilers require quite a high min flow rate, which the modulating pumps may not constantly provide.

The flow rate has nothing to do with the type of heat exchanger. A 20kw boiler with a 20C differential will always have a flow rate of 0.25 litres/sec.

It's the head loss throught the heat exchanger which is the problem. My 15kW Apollo Fanfare has a head loss of 0.7 metres; the 15kw Vaillant has a head loss of 1.4m, double that of the Fanfare. This needs to be taken into account when installing a new boiler as the existing pump may not be up to the job.

Alpha pumps are very sophisticated with pressure and temperature transducers incorporated.

 

[SimonH2]

You could fit an oversized plate heat exchanger from the boiler flow to the return. The boiler then just pumps in and out of the plate. On the other side of the plate the CH circuit runs through with a Smart pump on it and TRVs all around. Then full flow through the boiler. If the boiler has integrated weather compensation then it will be efficient enough and the boiler drops the flow temp to suit. The larger efficient plate will allow the CH circuit to extract enough heat to make the return cool enough. There is no direct short cut as the heat in the plate is being transferred.

Congratulations - your idea significantly exacerbates the efficiency problem. The issue is that the boiler itself has a maximum return temperature above which it won't be condensing. By sticking the whacking great HE in there, you've allowed a lot of flow through the boiler but at very low delta-T - thus ensuring that the return temperature into the boiler will almost always be above that required for condensing to occur.

It also stops sludge entering the boiler, keeping efficiency over time, which normally drops off.

You've got a think about sludge haven't you ?

not compatible with many modern domestic boilers, such as those with the gianonni heat exchanger, as the boilers require quite a high min flow rate, which the modulating pumps may not constantly provide.

The flow rate has nothing to do with the type of heat exchanger. A 20kw boiler with a 20C differential will always have a flow rate of 0.25 litres/sec.

It's the head loss throught the heat exchanger which is the problem. My 15kW Apollo Fanfare has a head loss of 0.7 metres; the 15kw Vaillant has a head loss of 1.4m, double that of the Fanfare. This needs to be taken into account when installing a new boiler as the existing pump may not be up to the job.

Actually the problem is that whatever the boiler, it will have a minimum flow rate for it to work properly/at all. When heating, and especially with TRVs on most of the system, the flow rate round the heating circuit may well be below the limit the boiler will tolerate and so you need the bypass. Once you have the bypass open, you are mixing hot flow with whatever comes back from the rads and so are likely to get the return temp into the boiler above the limit for condensing operation.

A modulating pump will make the situation worse by further reducing the flow rate round the rad circuit.

Thinking about this, there is probably a case for a small cylinder (like a thermal store but without the stress on storage and with no DHW facility) as a buffer between the boiler and the heating - much the same as I've used the thermal store for. It needn't be very big, but would allow the decoupling of boiler flow rate and CH flow rate - allowing fully TRVed rads and a modulating pump on the CH while still allowing sufficient flow rate through the boiler when it has to fire up to reheat the buffer tank.

 

[GeorgeBainbridge]

You could fit an oversized plate heat exchanger from the boiler flow to the return. The boiler then just pumps in and out of the plate. On the other side of the plate the CH circuit runs through with a Smart pump on it and TRVs all around. Then full flow through the boiler. If the boiler has integrated weather compensation then it will be efficient enough and the boiler drops the flow temp to suit. The larger efficient plate will allow the CH circuit to extract enough heat to make the return cool enough. There is no direct short cut as the heat in the plate is being transferred.

Congratulations - your idea significantly exacerbates the efficiency problem. The issue is that the boiler itself has a maximum return temperature above which it won't be condensing. By sticking the whacking great HE in there, you've allowed a lot of flow through the boiler but at very low delta-T - thus ensuring that the return temperature into the boiler will almost always be above that required for condensing to occur.

Note I said oversized plate. Having full flow promotes heat exchanger longevity.

It also stops sludge entering the boiler, keeping efficiency over time, which normally drops off.

You've got a think about sludge haven't you ?

I doubt you have seen inside a 10 year old small tube boiler.

not compatible with many modern domestic boilers, such as those with the gianonni heat exchanger, as the boilers require quite a high min flow rate, which the modulating pumps may not constantly provide.

The flow rate has nothing to do with the type of heat exchanger. A 20kw boiler with a 20C differential will always have a flow rate of 0.25 litres/sec.

It's the head loss throught the heat exchanger which is the problem. My 15kW Apollo Fanfare has a head loss of 0.7 metres; the 15kw Vaillant has a head loss of 1.4m, double that of the Fanfare. This needs to be taken into account when installing a new boiler as the existing pump may not be up to the job.

Actually the problem is that whatever the boiler, it will have a minimum flow rate for it to work properly/at all. When heating, and especially with TRVs on most of the system, the flow rate round the heating circuit may well be below the limit the boiler will tolerate and so you need the bypass. Once you have the bypass open, you are mixing hot flow with whatever comes back from the rads and so are likely to get the return temp into the boiler above the limit for condensing operation.

A modulating pump will make the situation worse by further reducing the flow rate round the rad circuit.

Thinking about this, there is probably a case for a small cylinder (like a thermal store but without the stress on storage and with no DHW facility) as a buffer between the boiler and the heating - much the same as I've used the thermal store for. It needn't be very big, but would allow the decoupling of boiler flow rate and CH flow rate - allowing fully TRVed rads and a modulating pump on the CH while still allowing sufficient flow rate through the boiler when it has to fire up to reheat the buffer tank.

It is called a CH buffer. Thermal stores do it for free, which is great bonus the plumbers here can't understand. They think a store is the equiv to an unvented cylinder or whatnot. You can have a similar buffer by having an oversized plate heat X which is tiny - as I explained to you. A pipe stat can be on the CH side which cuts out the burner/pump when the return pipe would indicate the return to the boiler would be above 54C.When the rad circuit cools enough the boiler cuts in.

 

[Norcon]

A buffer or thermal store between the boiler and heating circuit won't gurantee low return temperatures.
What happens is, the boiler heats the store.
The heating comes on and starts to empty the store. The cold water from the heating system goes to the bottom of the store and is immediately passed through the boiler.

Which turns the boiler on. Or, the boiler is controlled via a thermostat on the bottom of the store, which will signal heat is needed as soon as the cold water from the heating system gets to it and the returning water from the heating system is immediately cycled through the boiler.

Whilst both are running, cold/cool water returning from the heating system will go straight to the boiler. It won't see the buffer tank at all.
So, there's little point having one.

If the boiler is making more heat than the heating system can use, the temperature in the store starts to climb. The return to the boiler becomes a blend of the cool water returning from the heating system and the warmer water in the tank. So the boiler works at an increasingly high temperature. Which is bad for its efficiency.

If the water from the heating system goes straight to the boiler unmoderated, it will always be the coolest possible water that could be returned to the boiler, which will work most efficiently.

 

[SimonH2]

A buffer or thermal store between the boiler and heating circuit won't gurantee low return temperatures.

It will if you have the right controls fitted.

What happens is, the boiler heats the store.
The heating comes on and starts to empty the store. The cold water from the heating system goes to the bottom of the store and is immediately passed through the boiler.

Which turns the boiler on. Or, the boiler is controlled via a thermostat on the bottom of the store, which will signal heat is needed as soon as the cold water from the heating system gets to it and the returning water from the heating system is immediately cycled through the boiler.

Whilst both are running, cold/cool water returning from the heating system will go straight to the boiler. It won't see the buffer tank at all.
So, there's little point having one.

Actually, the flow rates will be different in boiler and heating loops - so the water in the buffer will be rising or falling slowly depending on the flow rate differentials. Thus, you can have a high flow rate round the boiler loop without having a high flow rate round the CH - and you can have a flow rate round the CH without needing any flow rate at all round the boiler.

If the boiler is making more heat than the heating system can use, the temperature in the store starts to climb. The return to the boiler becomes a blend of the cool water returning from the heating system and the warmer water in the tank. So the boiler works at an increasingly high temperature.

No, you shut down the boiler AND it's pump when the temperature <somewhere> in the buffer starts to rise. In practice, you need two stats and a relay so that the cycle goes :
1) Boiler comes on and heats the buffer top-down - you arrange the system so that the boiler never outputs 'half-warm' water as you need a stratified buffer tank with hot water at the top. On a small tank, you may want a diffuser to prevent the flow creating turbulence and mixing the layers. So the boiler is drawing in cool water and outputting hot water. The boiler does not switch off on rising temp at the upper stat because the relay hold it on ...
2) As the hot layer approaches the bottom, the lower stat triggers and turns off the boiler.
3) As the hot water is used in the CH loop - which can be a 100% TRV setup with modulating pump and no room stat - the hot water is drawn off from the top and cool water is returned to the bottom. The interface between hot and cool water rises up in the tank.
4) When the top stat on the tank triggers on falling temperature, the boiler is fired up to reheat it. Goto step 1, rinse and repeat.

The key is that your CH loop does not have a bypass so it only ever returns water to the bottom of the tank that's been through the rads. The flow through the rads is controlled by the TRVs, and so under moderate load the delta-T is high and you get a low flow rate of cool water rather than a higher rate of warmer water that would have been the norm on an old fashioned system setup for a specific delta-T across the rads and turned on-off by a room stat. Under part load (and especially if you oversize the rads) you can get return temps down into the 20s or 30s - plenty cool enough to make the boiler condense.

In practice, unless your demand is high and the boiler sizing is marginal, the buffer will cool slowly and reheat quickly. The worst case is a high demand, so high CH flow rates and correspondingly high CH return temps - but then you are no worse off with the buffer tank, but it's not gaining much either.

It's possible to use a single stat part way up the tank - but then you get frequent short cycles. Using two stats allows you to fire the boiler and run it for as long as it takes to reheat almost the whole of the tank before shutting down completely for a period. The bigger the tank the less frequently the boiler needs to fire - but when it does it will fire for longer.

If the water from the heating system goes straight to the boiler unmoderated, it will always be the coolest possible water that could be returned to the boiler, which will work most efficiently.

You weren't reading were you ? The return temperature into the boiler will NOT be at the return temp from the CH loop if you have a CH loop with TRVs and a modulating pump. The flow rate round the CH loop will be insufficient for the boiler to work and so it will just shut down (or overheat) - hence you need a bypass to keep the flow rate through the boiler up. By the time you've mixed a load of hot water with the return from the CH loop, the return to the boiler will be hot.

 

[SimonH2]

It is called a CH buffer. Thermal stores do it for free, which is great bonus the plumbers here can't understand. They think a store is the equiv to an unvented cylinder or whatnot. You can have a similar buffer by having an oversized plate heat X which is tiny - as I explained to you. A pipe stat can be on the CH side which cuts out the burner/pump when the return pipe would indicate the return to the boiler would be above 54C.When the rad circuit cools enough the boiler cuts in.

I'll just say that I'm sceptical that you could get enough thermal mass for it to be effective. It would be an enormous heat exchanger or the boiler would be short cycling as badly as it would be on a conventional setup.