where to measure temp on radiator when balancing

[Roger]

The message from Si contains these words:

Not in normal situations. If you subtract the outlet temperature from the inlet you get the average temperature of the radiator. Taking the temperature in the centre merely short circuits that calculation. Temperature decay is dependent on temperature difference so the centre is not quite average but within the limits of the exercise you are carrying out it and the thermometer you are using it matters not a jot. And you get the added bonus that your results are less likely to ruined by measuring the temperature of the wall rather than that of the pipe.

If you have clip-on thermometers you are stuck with measuring the pipe temperatures but have a bit less need to get simultaneous results which is a good thing as using clip-ons may take you 10 times as long.

No. Just fully open the TRVs. Opening the windows will introduce an unwelcome variable into the equation. As I said above speed is of the essence but it takes very little time to go round the house pointing an infra red thermometer at each radiator and recording the result, and as a precaution you can measure the first one again at the end of each circuit.

When all is said and done all you are trying to do is to get approximately the same flow rate through each radiator. The TRV will control the actual room temperature which is why cowboy installers don't waste their time balancing radiators.

You do need to make sure that the radiator without a TRV which you should have in the same room as the room thermostat is not favoured over the other radiators otherwise the remainder of your house might take too long to get up to temperature.

[Si]

In message , Roger writes

The radiator without TRV is in the bathroom. We don't have a room thermostat anywhere. I've always thought in terms of only allowing that bypass rad' to run with as little flow as possible as the bathroom is small and upstairs so it seems a bit of a waste. I also tend to keep the bath mat hung over it to minimise its heat loss. The boiler doesn't seem to complain.

[Si]

Yes, it does.

So experimentation and observation is the order of the day - Ok. :)

(SWMBO is just about to go out for the day so I may get to do this today. Shhhhh, no-one tell her. :)

[fred]

In article , Roger writes

If you subtract the inlet from outlet you get the temperature drop not the average temperature but I imagine that was a slip of the keyboard. I wouldn't recommend guessing where the average temperature point on a radiator was and using that to attempt a balance, it's the drop that's important and you'll do a much better job by taking 2 measurements and subtracting to get the real drop.

[fred]

In article , Roger Mills writes

Hmmn, open a window or take off all the heads, think I'll just open the window. As you say below, max setting on the TRV is 30degC so you're not going to get that at this time of year with a window open.

Also, balancing with the room temp at elevated levels distorts the balance as the rads won't operate as efficiently with the difference between room and radiator temp being reduced.

eg. Condensing boiler, MRT 55degC, room 21 gets 34deg diff, MRT 55degC, room 30 gets 25deg diff which is plenty to distort the balance result.

[Andrew Gabriel]

You beat me to it -- the temperature in the middle of a radiator isn't the average in the general case. I played a bit with this when I installed my heating, in order to understand how radiators really work. Generally, you'll find the top of the radiator is at pretty much the inlet temperature, and the bottom is pretty much at the outlet temperature. The air as it leaves the top of the radiator tends to be close to the top of radiator temperature, which is the inlet temperature regardless of the flow rate. As you increase the water flow, this hotter area extends further down the radiator. It doesn't much affect the temperature of the air leaving the top, but it increases the extent of the chimney effect and increases the quantity of hot air leaving the top of the radiator. There isn't a linear temperature drop down a radiator at any flow rate, as the rate of heat loss is higher from the hotter areas (and isn't just by convection in any case), so the idea that the centre is any kind of average is bogus. I suspect the whole idea of average radiator temperature is bogus.

[<me9]

It takes only about 10 seconds for the two measurements with an IR thermometer anyway, and the inlet temperature may vary, particularly with a non modulating system. It's prolly quicker to do it properly. I find it much easier in the early part of the warmup (before the bioler thermostat cuts in) for that reason, although it may lose accuracy on other counts, like lower flow temperature.

[Roger]

The message from fred contains these words:

Not exactly a slip of the keyboard, more the disassociative effect of a large glass of single malt. :-)

My method was based on the impossibility of me getting a consistant reading on my inlet and outlet pipes and all I can say in my defence is that it seemed to work for me at the time although that time is sufficiently far in the past for me not to remember exactly how I went about it.

[Roger]

The message from [e-mail] (Andrew Gabriel) contains these words:

To start at the end rather than the beginning there is nothing bogus about an average temperature. Whether or not the centre of the radiator is anywhere near the average of the inlet and outlet temperature is a moot point but under steady state conditions there will be a point somewhere which is and likewise the relationship between that and the actual temperature at the centre should be consistent.

Other things being equal if you increase the flow rate you automatically decrease the temperature drop across the radiator.

There have been arguments before on this ng as to how much of the heat output of a radiator is by radiation and how much by convection but the amount of radiation is considerable and is probably over 50% for a plain single radiator. Radiation varies at the 4th power of the absolute temperature so on a typical setup there would be rather more than a 20% difference between the radiation at the warmest and coldest spots on the radiator.

Convection I am led to believe is directly related to the temperature difference between the hot surface and the bulk of the adjacent fluid. ISTM that if Andrew is right about the temperature of the air leaving the radiator then as the temperature gradient in the air adjacent to the radiator is greater than the temperature gradient in the radiator the lower part of the radiator will give out more convection heat than the upper part which would go some way at least to cancel out the imbalance in the radiation output.

[fred]

In article , Roger writes

LOL

Gotcha, I understand the pipe targeting problem, maybe try the flat front surface of the rad directly adjacent to the inlet and outlet points, they should be as close to the pipe temps as makes no difference, that would be my spot to target.

[Si]

Right it's Wednesday & I've been trying to balance my radiators...

Temperatures hadn't settled after 10 (someone else suggested) nor 15 and only seemed to be stabilising after 30 minutes but then suddenly both in & out temperatures started increasing/decreasing.

Thanks.

The temperature differences in the round I've just completed were ("C)

0.4, 1.0, 1.0, 5.7 (bathroom bypass), 2.5, 2.5, 3.0. Before that, 4.4, 3.9, 5.1, 4.2 , 5.3, 2.5, 2.1.

However, measuring the living room again (the first in the list) the difference is 5.2. :(

What am I not doing / doing wrong ?

These differences are much smaller than previously mentioned. Should/could I slow the pump down?

[Roger]

The message from Si contains these words:

I will stick my neck out again and suggest it is catch 22. For you to measure the real temperature difference across the radiator the inlet temperature has to be stable long enough for the return temperature to stabilise. For the inlet temperature to be stable the return temperature at the boiler has to be stable and the boiler has to be firing. If you have a modulating boiler even this might not be enough but I don't know enough about the nitty-gritty of modulation to be sure.

The longer you wait for the system to stabilise the more chance there is that the boiler thermostat will decide the water is hot enough and cease firing. If the room thermostat is still calling for heat (or in the short term the pump overrun is operating) the water will continue to circulate and you might get the situation where the inlet temperature is the same as the outlet temperature.

What I am about to suggest may or may not work as it is only theoretical. Take a single radiator at a time. Make sure the boiler is firing. Monitor the inlet temperature until it ceases to rise, then likewise the outlet. If the inlet is then still at the same temperature then ISTM that you can be sure of your figures. If by then the inlet temperature has started to drop the figures may still be good enough but the bigger the discrepancy the more chance there will be of a significant inaccuracy.

Don't see any reason why not (if you have a slower speed to go to) unless you are then faced with unacceptably long time for the rooms to warm up or in extremis rooms never getting up to temperature. You can always turn the speed up again later in the unlikely event of a really cold spell.

[Ed Sirett]

Try with the boiler on flat out and if possible and under you control try a slower pump speed. You need to get the system to that the boiler is always firing (albeit not at full power) and there is enought difference to measure.

OTOH if all the radiators are nicely hot, then the system can be declared as good enough, set the TRVs and boiler themrostat the required setting and enjoy.

[Andrew Gabriel]

I recently bought myself a slide rule off ebay, no longer having the ones I used at school. One day when there are a few people around my desk and I need to do a calculation, I'll get it of my drawer and use it, and see what reaction I get. I found I still remember how to use it, even though I haven't done so for 30 years.

[geoff]

In message , Andrew Gabriel writes

Hmm, no idea what happened to mine

[Si]

In message , Andrew Gabriel writes

I still have my Thornton's Comprehensive. I think we were the first year to get them in the clearly long-lasting (but otherwise nasty) plastic cases. I don't have the instructions but can still use the main functions. :) Log's in plastic instead of those endless tables - magic!

[Roger Mills]

I think I've still got one or two of mine in a drawer somewhere (a long fairly accurate one and a short - pocket - quick and dirty jobby) but haven't used either of them for - I guess - over 40 years.

Is there a market for them?

[Huge]

A quick look on eBay seems to suggest so, but the prices fall into the "can't be arsed" range. I still have my Faber-Castell from 6th Form in 1971.

(On a similar point, I put my original Sony Walkman on the charity shop pile a few weeks, ago, looked on eBay out of idle curiosity and promptly took it off and put it on the eBay pile.)

[Andrew Gabriel]

Yes, but there's no shortage. I think mine cost me only a couple of quid on ebay, and that was for one with original box and instructions.

Dad has a cylindrical brass slide rule -- the rule is a spiral around the cylinder so its effective length and accuracy are much higher. Those can command higher prices in good condition, but there's no way he'd sell his.

[LSR]

From what I remember of slide-rule calculations, are you sure it's 40 years and not 4.0 or 400 years?