Vaillant 438 with 3 zones, frequent S53

[fezster]

I think your boiler is probably oversized because:-

1. You don't know if, or if so what figure, an allowance was made for water heating was used if the heat loss calculation. The recommendation is that just 2 kW is used ( but many installers use far higher figures! )

True, I dont know what the original installer did. But I have run through the idhee calc myself (came to 34kw) and we know the megaflo cylinder is a 250l indirect, so therefore if 2kw is an appropriate number to add, we come to a total of 36kw.

2. 40 kw rad output would be very high for about 20 rads, most average about 1.0 - 1.5 kW but obviously I have never seen your house!

Many rads are large (eg. 600mm by 1800mm). Ive taken the kw output from the screwfix website. All are kudox, nearly all are double panel, one is double convector. The total comes to over 40kw.

3. You have already said that its currently set at just 26 kW. If that's adequate for heating your house then a 38 kW boiler is grossly oversized when a 28 kW boiler would suffice!

This point does confuse me. Im looking to you guys for an explanation here. The rads do indeed add up to 40kw, the cylinder is as stated, all trv's are open. Yet the boiler set to 26kw is able to heat up all of the radiators (the hw may or may not have heated up at the same time).

There is NO advantage oversizing boilers and it just reduces efficiency and causes problematic short cycling!
Tony

Yes, totally understood. By the way, Im not trying to argue with you above, but just trying to understand your reasoning and, hopefully, ultimately come to some conclusions as to how my heating system currently works and how it can be made to work better.

 

[Agile]

Oversizing rads does no harm and is indeed beneficial as it enables the required heat output to be obtained with a lower flow temperature.

This results in improved boiler efficiency as its co0ndensing more.

The maximum boiler output is only required when the outside temperature is at -1 C. For the rest of the year its only a small proportion of the maximum.

Tony

 

[fezster]

Ok thanks, that makes perfect sense. And indeed, I asked the original installer to oversize the radiators, as per advice read from past posts on these forums (perhaps by you).

I'm contemplating whether to drop another £300 - £400 putting a bigger pump in. If I knew for sure it would fix the flow problem on u/s, Id be happy to. Any suggestions on how to proceed ?

Also, the original installer did not do a very good job of balancing the system. Ive pretty much done this myself. Whilst all radiators do get hot now, Im wondering whether it's possible to reduce the flow/return temperature differential by adjusting the balancing ? If so, I'll invest in an IR thermometer and do the balancing more scientifically, rather than just by feel.

 

[Agile]

An IR thermometer is not much good for balancing! You need a contact thermometer.

If you spend a little while "calibrating" your hands they can be surprisingly good for balancing!

Its only anyway an approximation as it varies as TRVs adjust themselves!

More important to allow the system to stabilise after each adjustment, that's usually about 15 minutes! That's why balancing ( properly ) is so time consuming!

Tony

 

[fezster]

Do you know if balancing can reduce the temperature differential across the boiler ? At the moment, the temperature differential across the boiler with u/s only is over 30 degrees. Can I reduce this through balancing ?

If so, am I looking to increase flow or decrease flow in one, some or all radiators to achieve this ?

One thing I did try was fully opening the first radiator on the u/s circuit, believing that would boost the return temperature since a larger proportion of water would be passing through this radiator and returning to the boiler (at least, that was the theory). However, this had absolutely no effect whatsoever on the return temperature, which left be baffled as to how this works.

 

[Agile]

The whole point of balancing is to adjust the diff temp across each rad and by doing this ensure the design heat output is achieved.

A high diff at the boiler indicates an inadequate flow rate and I am surprised you boiler does work fairly well as those models do not like much more than about 23 C. In most cases its difficult to achieve more than about 15 C diff!

I would almost wonder if a fixed ( or auto ) bypass on that zone might be a bodge solution to reduce the diff at the boiler. But that would not be a normal thing to do.

It would also reduce the pump pressure across that zone and might reduce the heat output from the rads too much. It could also have an effect of the system when other zones are on.

Tony

 

[fezster]

I made a post a few months ago (I think you replied), where one radiator on the u/s circuit was not warming up. It turned out to be a faulty or incorrectly set autobypass, which was constantly bypassing some (but not all) water.

I wonder if the original installer was trying to achieve what you just described. Suffice to say, replace the ABV drastically affected the heat output from the boiler and all radiators got much hotter, and the furthest rads finally started getting hot too.

Ok, so circling back, is a bigger pump the solution here ? And short of spending £300 buying one and it not working, is there anything I can do to try and prove this ? Eg. If it is a flow problem due to a large head, would turning one or more radiators off reduce the head, and therefore flow should increase ?

 

[Agile]

You need to increase the flow through the rads to reduce the diff temp across the boiler!

If you did not realise that it makes me wonder if they have been totally wrongly balanced because the objective was incorrect.

Tony

 

[fezster]

My balancing was to achieve a 20 degree drop across the radiators. Obviously, by hand, this was slightly guess work, but it had the desired effect of getting all radiators to heat up, whereas previously some were not.

Given the system is "balanced" (perhaps not perfectly, but somewhat), I am now looking to reduce the temperature differential at the boiler. Therefore, I would have assumed that increasing the flow to the first radiator on the circuit would have achieved this. Presently, the lockshield is open about a quarter of a turn. Even opening the lockshield fully (about 8 turns !) I do not get a sudden rise in return temperature at the boiler. Do you agree I should, given what you said in your last post ? And if not, then how would you achieve a greater return temperature ?

 

[Agile]

It would seem that you need to SLIGHTLY increase the flow through each rad. That will be a very small angular rotation of the L/S valve.

But if all the rads have a return temp giving a diff of 20 C then that's what should be presented to the boiler.

This is where a contact thermometer is needed so the temps at the rads and at the boiler can be compared.

Tony

 

[fezster]

Thanks, will invest in some contact thermometers. Anywhere I can pick up some cheap ones ? Otherwise I'll invest in a multimeter with thermometer from maplin for about £30.

Btw, reading this post again from ChrisR (tenth post down):

//www.diynot.com/forums/viewtopic.php?t=19661&start=0


I'm calculating:

26000/(4185*20) = 0.31 l/s

0.31 * 3.6 = 1.11 m^3/hr


And looking at the grundfos technical spec for the UPS 25-55 (which apparently is the equivalent for the UPS 26-50 that I have), I can see this gives ~ 4.5m of head:

http://www.anchorpumps.com/media/downloads/75/Grundfos Light Commercial Circulators.pdf

Not knowing how much head I need, though, kind of makes this information pretty pointless at this stage (!)

 

[Agile]

You do need a thermocouple type. I prefer the bare wire kind as they have a low thermal inertia and quickly pressed against the pipe.

A simple meter with Type K socket or a more elaborate two input type which can be used to show the differential directly.

Since the flow is fairly accurately controlled by the boiler a single measurement on the return is usually adequate. Balancing is anyway always approximate.

I don't think you should be trying to spec a new pump yourself. But if you must then note the pressure the boiler required which you have already been told buy don't seem to have digested.

Tony

 

[fezster]

I don't think you should be trying to spec a new pump yourself

I'm being driven to the point where it seems it may be easier to learn how to do it myself, because there seem to be so few so-called qualified people (at least locally to me) that are able to do it properly.

Thanks for the advice on the thermometer, I've ordered a meter, so will give that a go to see if I can improve things.

 

[ALEC1]

by far the easiest and least costly thing to do is to change the PCB, to one that has been modified to accept low thermal loads.

I have to say I do not know how to ensure that what you buy is the modified one, speak to vaillant would be my advice

 

[fezster]

by far the easiest and least costly thing to do is to change the PCB, to one that has been modified to accept low thermal loads.

I have to say I do not know how to ensure that what you buy is the modified one, speak to vaillant would be my advice

Are vaillant replacing these FOC on request ? If not, how would I go about obtaining the modified PCB ?