Increasing radiator output and improving CH performance

[ajrobb]

I only picked them because they looked very similar to mine, but in light of what you have said in terms of wall thickness I guess there's more to this than meets the eye.

Wall thickness (and water volume) only affect the time it takes a radiator to warm up. Once hot, the convection heat transfer will be practically identical.

 

[mysteryman]

If you have TRVs on most of the rads, the system will be largely self-balancing. I would reinstate the rad to the lounge, and put one of an appropriate output in the conservatory. I would not fit electric heating, especially in the part of the house needing most heat!

The heat load will increase greatly as the weather drops to freezing; 7C is relatively mild.

As regards short-cycling, I would fit an add-on weather compensation controller. This reduces the flow temperature to the radiators, reducing losses and short-cycling, but the real saving comes from the fact that it keeps the boiler condensing virtually all the time. Honeywell make a suitable system.

The Luddites and armchair experts - and competent installers who should know better - will probably start knocking my advice, but it is nevertheless good sound advice.

 

[D_Hailsham]

If you have TRVs on most of the rads, the system will be largely self-balancing.

Question: Why do so many people complain about TRVs not working properly?

Answer: Because the system has not been balanced.

Homework: Download and Read Drayton TRV Commissioning Guide

I would fit an add-on weather compensation controller. ... Honeywell make a suitable system.

Which one is that?

 

[Rhondo]

Where do you get the 15% from???

The figure I have seen is a 1% or 2% increase. BS EN442 tests rads using TBSE.

Just noticed your post D Hail.
I got that figure from memory. Something I read years ago somewhere. But you got me by the nu@s as you quoting bs numbers.
Anyways my customers are happy with what I did and they tell me their rads heat up twice as quick so that so good enough for me.

Doi(fu@k)all will probably be along shortly to give me another rollicking.

 

[big-all]

as an aside
a conservatory of a reasonable size with the the outside temperature at below freezing can loose around 6kw off heat at 21 degrees with 3kw doing not much more than keeping the chill off at around 8 degrees

 

[MJN]

Having purchase an IR thermometer I decided to find out how well balanced the system in case this was the cause of premature boiler cycling...

What was immediate obvious was that it wasn't balanced at all! Whether it was never done at original installation or perhaps interfered with by the previous owners I don't know.

Some of the rads, the very small ones, had a virtually no temperature drop (<3C) whereas the larger ones were more like 10C. Obviously these small rads were acting as short circuits and so it's no wonder the boiler was cutting out given the high return temp.

Several iterations of LSV tweaking later, and dropping the pump speed down from 3 to 1, I have achieved a far more even drop across all radiators of between 7C and 10.5C i.e. a 3.5C variation which I'm quite happy with for now. My results are at:

http://www.newtonnet.co.uk/permanent/20101018-Balancing-Results.htm

I'll probably aim for a narrower range once I've fitted a towel radiator and bigger lounge radiator - I don't see the point in putting too much effort into fine tuning now if I'm going to upset the balance shortly anyway.

Thanks for all the help, not only with this post but the forum in general.

As an aside, does anyone have stock of a dozen LSV caps? All mine are fitted with their adjusting knobs and I'd like to remove them to prevent tampering but don't want to leave the bare spindle! Of course I'm happy to purchase them off you for an agreeable price.

Cheers,

Mathew

 

[D_Hailsham]

What effect has balancing had on your original problem?

You seem to have improved the balance considerably, so I won't suggest you read How to balance a CH system.

 

[MJN]

The main problem was cycling prior to the house fully coming upto temperature. My small rads were previously receiving as much flow as the larger ones yet obviously extracting relatively little heat such that they were allowing a large quantity of return flow at almost flow temperatures. The balancing has addressed that by only giving them as much flow as they need thus enabling the system as a whole to extract the correct amount of heat.

The conservatory radiator now also seems to be more effective presumably as it now gets the flow it requires, and its flow temperature has increased by nearly 10 degrees. It's still undersized as a sole source of heat for the winter so I'll still need to address that.

I suppose there are other benefits where I didn't realise I had any issues such as being able to lower the pump speed thus saving some energy and increasing the drop across the boiler to maximise opportunities to condense.

Mathew

P.S. That looks like a useful balancing reference, however I cannot imagine even contemplating doing it with pipe thermometers - I don't know if I'd have enough patience!

 

[D_Hailsham]

The main problem was cycling prior to the house fully coming upto temperature. My small rads were previously receiving as much flow as the larger ones yet obviously extracting relatively little heat such that they were allowing a large quantity of return flow at almost flow temperatures. The balancing has addressed that by only giving them as much flow as they need thus enabling the system as a whole to extract the correct amount of heat.

The conservatory radiator now also seems to be more effective presumably as it now gets the flow it requires, and its flow temperature has increased by nearly 10 degrees. It's still undersized as a sole source of heat for the winter so I'll still need to address that.

I suppose there are other benefits where I didn't realise I had any issues such as being able to lower the pump speed thus saving some energy and increasing the drop across the boiler to maximise opportunities to condense.

All those who think balancing is unnecessary should read what you have written.

That looks like a useful balancing reference, however I cannot imagine even contemplating doing it with pipe thermometers - I don't know if I'd have enough patience!

That's why it also suggests an IR thermometer. Although you have to be careful as the reading can vary considerably according to actual surface being measured; a shiny copper pipe will read much lower than one covered in paint. This can be overcome by wrapping black insulating tape round all the pipes and measuring off the tape.

 

[MJN]

That's why it also suggests an IR thermometer. Although you have to be careful as the reading can vary considerably according to actual surface being measured; a shiny copper pipe will read much lower than one covered in paint. This can be overcome by wrapping black insulating tape round all the pipes and measuring off the tape.

Absolutely - the emissivity of shiny metals can vary significantly, particularly if oxidised, dirty, etc. I used black tape on my valves as you suggest and even then you would see varying readings if you weren't careful. My approach was to keep the IR thermometer taking constant readings and noting the highest reading obtained (as this necessarilly corresponds to the most accurate reading). You soon get the nack of getting the measurement process well tuned so it only takes a couple of seconds max.

Black insulating tape has an emissivity of 0.97 hence close enough to the 0.95 default of my thermometer (and most others it seems). Masking tape is 0.92 so that'd also suffice.

Mathew

 

[ajrobb]

Still learning:

Of course the boiler stops cycling if all the TRVs are open while balancing. Try turning most of them down in mild weather - the boiler load will drop below the minimum modulated load and must cycle. I can see that balancing and pump adjustment can help improve condensing boiler efficiency, by reducing flow rate and lowering return temperature, but it will have little effect on boiler load. If you don't reduce pump speed, an ABV will provide a short circuit and push up boiler return temperatures, lowering condensing efficiency.

Most of the commissioning guides assume an automatic bypass valve adjusted to something like 0.2 bar. If you don't have an ABV and are relying on 25% of radiators without TRVs, shutting all the lock shield valves will push up the pressure differential. Without an ABV, adjusting the LSVs one at a time will be a waste of time.

On my system, I opened all the LSVs a quarter turn then went round the radiators and opened the LSVs additional quarter turns if their returns were not 'fairly hot'. If I didn't have a stupid dual Keston installation with three circulation pumps, I'd go round the radiators again and tweak the return temperatures to get something like 55°C return with 66°C feed at the boiler. (I don't understand why temperature differentials are quoted without reference to feed temperature. As feed temperatures increase, so will the temperature differentials.)

 

[D_Hailsham]

You soon get the knack of getting the measurement process well tuned so it only takes a couple of seconds max.

You also have to take into account the size of the collection "beam" (like a torch) or you will be including the wall behind the pipe. Having the thermometer touching the pipe seems to be best - ignore the "laser" beam, that close it's useless.

Black insulating tape has an emissivity of 0.97 hence close enough to the 0.95 default of my thermometer (and most others it seems). Masking tape is 0.92 so that'd also suffice.

Consistency is more important than the absolute value.

 

[D_Hailsham]

Still learning:

Who, you or us?

I can see that balancing and pump adjustment can help improve condensing boiler efficiency, by reducing flow rate and lowering return temperature, but it will have little effect on boiler load.

The purpose of balancing is not reduce the flow rate and lower the return temp but to ensure that the correct flows rate and return temperatures are obtained. Measuring the flow and return temperature of a radiator is a very crude way of measuring the flow rate as it makes many assumptions. In a large commercial installation flow meters are used.

If you don't reduce pump speed, an ABV will provide a short circuit and push up boiler return temperatures, lowering condensing efficiency.

That shouldn't happen on a properly designed system.

Most of the commissioning guides assume an automatic bypass valve adjusted to something like 0.2 bar. If you don't have an ABV and are relying on 25% of radiators without TRVs, shutting all the lock shield valves will push up the pressure differential. Without an ABV, adjusting the LSVs one at a time will be a waste of time.

0.2 bar is just what most ABVs are set to out of the box. They must be adjusted to meet the specific conditions.

On my system, I opened all the LSVs a quarter turn then went round the radiators and opened the LSVs additional quarter turns if their returns were not 'fairly hot'.

Quarter turns is a large amount.

I don't understand why temperature differentials are quoted without reference to feed temperature. As feed temperatures increase, so will the temperature differentials.

That will only be true if the flow rate decreases as the feed temperature increases.

 

[MJN]

Still learning:

Of course the boiler stops cycling if all the TRVs are open while balancing.

Eh? My achieved goal of reduced/minimal cycling is now that my TRVs are set to their desired temperatures. Of course there'd be minimal cycling cycling whilst balancing with the TRVs wide open. (Or were you not referring to my situation? Apologies if I've misinterpreted what you're saying)

I can see that balancing and pump adjustment can help improve condensing boiler efficiency, by reducing flow rate and lowering return temperature, but it will have little effect on boiler load.

My predicted energy loss in the context of lowering the pump speed was referring to the lower electricity consumption of the slower pump and the increased drop across the boiler (hence more chance of condensing).

I'd go round the radiators again and tweak the return temperatures to get something like 55°C return with 66°C feed at the boiler.

If a non-combi make sure there's enough headroom to adequately heat your HW cylinder up to its target temperature. If this is 60°C you may find a 66°C flow will be very slow to heat it (by virture of energy transfer rate being proportional to temperature differential). Mid-winter CH performance might also be affected.

I don't understand why temperature differentials are quoted without reference to feed temperature. As feed temperatures increase, so will the temperature differentials.

The target temperature differential (the often quoted 11°C) is nothing to do with the radiators but rather the temperature gradient that the boiler has been designed to produce when running flat out at a specified flow rate. This differential remains the same regardless of target temperature.

Mathew

 

[MJN]

You also have to take into account the size of the collection "beam" (like a torch) or you will be including the wall behind the pipe.

Indeed, particularly so with my thermometer as it has a distant to spot ratio of 1:1 so it is pretty much essential to have it flat against the face. Given this ratio mine doesn't have a laser beam, and I knew from the intended application that it would've been no use anyway.

I went for the ETI TN1:

http://thermometer.co.uk/519-tn1-infrared-thermometer.html

For only £15 delivered (from Amazon) it's ideal for this task.

Consistency is more important than the absolute value.

Yes. For all but the most expensive IR thermometers their absolute accuracy is only +/-2°C however their consistency is, as you say, what counts and is more than adequate for making relative comparisons in this application. Repeated measurements on the same spot confirm that in this regard they are extremely good (i.e. within 1/10ths of a degree).

Mathew