Install the app
How to install the app on iOS

Follow along with the video below to see how to install our site as a web app on your home screen.

Note: This feature may not be available in some browsers.

Welcome to the forum. Although you can post in any forum, the USA forum is here in case of local regs or laws

Nov 8, 2018
24
9
3
Member Type
DIY or Homeowner
Hi everyone,

I am seeking some advice. We moved in last year to a new build flat which has 6 TBOE radiators with TRV's at the top. The new build flat uses a heat exchange unit as it is connected to district heating system. All pipework is behind the plasterboards and run up in the ceiling.

I have noticed that the return is actually at the top of the radiators and the feed is from the bottom. The radiators do get hot but take around 30 to 40 minutes to get hot. I have also issues with lots of air in the system.

I know this is incorrect so logged this as a defect. The engineers who fitted the system stipulated that it did not matter which way around the pipework was fitted as the TRV was bi directional. So the developer will not charge the pipework around as the radiators get hot.

I have even asked NHBC and they have stated that as long as the radiators get hot it does not matter whether the pipework was fitted incorrectly.

Can anyone clarify why the feed should not be at the bottom and what problems can it cause. Would my heating work more efficiently if radiators were plumbed in correctly i.e fed from the top?

Also can anyone suggest how I could some how get the developer to change their minds. Maybe possibly by getting independent report.

Many Thanks in advance for any advice provided.
 
As Shaun says, it does matter. But it will still get hot. If it is taking that long, ther is a different problem .
 
Basically assume it's inefficient. Engineer used thermal camera and you can clearly see the temperature difference between incoming and return being on 3 degrees. Incoming was 53 and return was 50. Bottom of radiator was 40 degrees. It makes more sense to return the cooler water than the hotter water.

Assume this also will cost me more money in my heating bills?
 
If you only have a 3°C difference from flow to return then there is a chance that your boiler isnt condensing properly either
 
Is the issue just the time that the radiator takes to heat up or is the room taking a long time to warm up as well?
 
Basically assume it's inefficient. Engineer used thermal camera and you can clearly see the temperature difference between incoming and return being on 3 degrees. Incoming was 53 and return was 50. Bottom of radiator was 40 degrees. It makes more sense to return the cooler water than the hotter water.

Assume this also will cost me more money in my heating bills?

The flow should be from top to bottom, the bi directional TRV can be at either end.
You might have a very cold house/room as rads now a days are rated as "50 deg" rads which is (flowtemp+returntemp)/2 - 20. your rads are "32 deg" rads, which will only emit 56% of their rated output except that they are over sized to take this factor into account.
 
Is the issue just the time that the radiator takes to heat up or is the room taking a long time to warm up as well?

Rooms take a long time to warm up. I believe the developer needs to resolve and should get it working as specified according to their design.
Also this should in turn reduce the heating bills.

Thanks for all your responses. I have arranged another meeting with the developers to discuss.
 
Are there other flats in the development, have you spoken to the occupants about thier heating if so?
 
There is nothing wrong with TBOE. However, the flow must be the top. BBOE then yes doesn’t matter with by-directional TRVs (Although I always fit TRVs on flow).
It could be the rads are deliberately oversized to reduce return temp and assist with longer periods in condensing mode.
No just noticed it’s a district heating system, so disregard that.
As John.g says BS EN 442 give a delta temp of 50
so there is a design issue here.
 
Surely it would be an easy job to swap flow and return around as the pipes enter the property or even before? Is the heat exchanger in a cupboard or something, could they cross the pipes there
 
Surely it would be an easy job to swap flow and return around as the pipes enter the property or even before? Is the heat exchanger in a cupboard or something, could they cross the pipes there

Heat Exchanger unit in store. Yes solution is simple, swap the pipes. The issue I am having is to convince the developer to get the work done.
They are currently saying it does not matter when it clearly does matter.
 
If the TBOE is on a one pipe system it matters greatly and will need rectifying Should they be piped on a two pipe system purely to make it easier to adjust the TRV without bending down I doubt it will make a significant difference.
 
I think that the OP is focusing on the wrong point at the moment. If the radiator is operating with a deltaT of 3°C it will make a negligble difference to the heat transfer whether the flow is top-bottom or bottom-top.

The first question should be, is the system is designed with radiators sized to operate with a flow temperature of 53? If not, that's what needs looking into first.

IMO, in discussion with the developer, the OP should stick to complaining about the performance of the system. E.g. "It takes 2 hours for the heating to raise the temperature in this room from 17 to 19 degrees." or "When the temperature outside is below 5 degrees, the temperature in this room never gets above 19 with the heating on full.", etc. Don't try and tell them how to fix it, that's their problem.

If and when you reach deadlock then is the time to employ your own heating engineer, and you'll need one familiar with district heating schemes, to investigate and report on the problem. They'll be able to confirm whether the system meets the appropriate standards and advise on possible remedies.
 
Last edited:
Hi, I have started to make a note of how long it takes to heat the living/kitchen dinner where the thermostat is located. This is open plan area 30sq m in L shape layout with large Patio windows. We are on the ground floor as well.

There are two radiators installed in this room.

Today outside temperature is 10 degrees and the room temperature was 17 degree. Heating has been on for 2 hours and the temperature now is 18.5 degrees. Surely that is way too long time and indicates a problem with the system.
 
Hi, I have started to make a note of how long it takes to heat the living/kitchen dinner where the thermostat is located. This is open plan area 30sq m in L shape layout with large Patio windows. We are on the ground floor as well.

There are two radiators installed in this room.

Today outside temperature is 10 degrees and the room temperature was 17 degree. Heating has been on for 2 hours and the temperature now is 18.5 degrees. Surely that is way too long time and indicates a problem with the system.
 
Meeting rearranged, I was stuck in Germany with food poisoning. It has been rescheduled for next Wednesday.

The heating company is looking at the radiator specs to check how they should be plumbed. That does not feel me with confidence if you have to look at the manual to know you should not have the return at the top with TBOE radiator.
 
Engineer came out yesterday as they had to rehang the radiator which was not hanged properly.

Also tightened the bolts which were loose to get rid of the leak in the Heat Exchange Unit.

Regards to the Delta Temperature.
The Exchanger has flow temperature set to 75 degrees. The return temperature using thermal camera is 50 degrees, I assume this is actually much more like 70 degrees as the incoming temperature was read as 53 degrees.

Anyway we still have a lot of air in the system, he said they will powerflush the system.

He is insisting it does not matter if the return is at the top.

Also stated 3 hours to increase the temperature of the room by 2 degrees is due to room size and that we are on ground floor.

After they remove the air from the system I will have to hire someone to get a report done.

I need someone to explain why the return should be at the bottom and what difference will it make to the system.
 
I really don't think it makes much of a difference as after all its a pumped circulation system, its probably no worse than the conventional bottom to bottom almost universally used system which works pretty well to my mild surprise.
 
  • Disagree
Reactions: chris watkins
I need someone to explain why the return should be at the bottom and what difference will it make to the system.
When designing a heating system using rads a heatloss calculation is completed to determine how much heat is going to be required to keep the room at the required temperature (say 21C) when it is, say -3C outside. This is used to select a suitably radiator to do the job. As others have said the outputs the rad manufacturers state is based on being tested in a standard way (MW-air temp = 50) Here is the important thing in your case - they also connect up the rad in a certain way, typically Top & Bottom Same End (TBSE) if you change the way it is connected then you change the radiator ability to heat the room (stated output).
This is fine if you know how the rads are going to be piped up for instants in this country we mainly use Bottom Oppersite Ends (BOC) which is less efficient at proving heat from it, still that is not a problem we just make them bigger to allow for this.
TBOE connections like yours mean that the rads can be around 15% smaller than if they had been piped BOE, why?? because when they are piped with the flow (hotter lighter) water to the top connection & the return (cooler heaver) water out of the bottom you are working with gravity & not against it. God knows what effect on its output piping it up the wrong way is having on the outputs but effectively you have very much undersized rads which will not do the job they were put there to do. Very simply they are not fit for purpose as they are currently installed.
Don't take any more old crap about this from them, ask for the heatloss calculations for your property so you can see what is require & ask them once again to install them as per the designs.
 
Last edited:
Did or do they make BOC rads with a baffle plate in the bottom header to force the water upwards through "1/2" the columns and downwards through the remainder?.
 
Engineer came out yesterday as they had to rehang the radiator which was not hanged properly.

Also tightened the bolts which were loose to get rid of the leak in the Heat Exchange Unit.

Regards to the Delta Temperature.
The Exchanger has flow temperature set to 75 degrees. The return temperature using thermal camera is 50 degrees, I assume this is actually much more like 70 degrees as the incoming temperature was read as 53 degrees.

Anyway we still have a lot of air in the system, he said they will powerflush the system.

He is insisting it does not matter if the return is at the top.

Also stated 3 hours to increase the temperature of the room by 2 degrees is due to room size and that we are on ground floor.

After they remove the air from the system I will have to hire someone to get a report done.

I need someone to explain why the return should be at the bottom and what difference will it make to the system.

OK then I ran a 1400 x 400 mm double rad BTOE for approx 40 minutes and I can confirm that the performance is certainly affected, unfortunately I didn't have my temperature scanner but the last ~ 1/4 of the rad was much more noticeably cooler than the remainder apart from the top. I would estimate that the performance loss at ~ 20/30%, this would possibly reduce over a longer time period but to quote Chris Watkins..."they are not fit for purpose as they are currently installed."
 
  • Like
Reactions: chris watkins
Thanks. You guys are brilliant offering great advice. Also you have common sense unlike the cowboys I am dealing with. As we are in our defects period I will fight to get this resolved. I will keep you guys updated but this could take some time.

I asked them how long would it take to swap the pipes and they said 2 hours. So do not understand why they simply won't just swap the feeds around.
 
  • Like
Reactions: chris watkins
Regards to the Delta Temperature.
The Exchanger has flow temperature set to 75 degrees. The return temperature using thermal camera is 50 degrees, I assume this is actually much more like 70 degrees as the incoming temperature was read as 53 degrees.

Don't assume things that are central to the diagnosis of a problem.

The fact that the flow temperature was 'set to 75 degrees' does not mean the flow temperature *is* 75 degrees, particularly if your measurements say it's 53 degrees.

If you don't trust the temperature measurement, get another one made using a proper pipe thermometer. Alternatively, press your thumb onto the pipe and count in seconds how long before it's painful enough to make you pull it off. If you only get to one it's 75°C but if you get to more than ten it's 50°C.
 
Thanks. You guys are brilliant offering great advice. Also you have common sense unlike the cowboys I am dealing with. As we are in our defects period I will fight to get this resolved. I will keep you guys updated but this could take some time.

I asked them how long would it take to swap the pipes and they said 2 hours. So do not understand why they simply won't just swap the feeds around.

Got a few temperatures last night with "old" temp scanner but good enough I think for comparison purposes.

BTOE & BOE Rad Temps.jpg
 
Anything in the interest of science.
All temps taken after 40/45 mins, the BOE gave the fastest warm up from cold.

View attachment 37621

View attachment 37622
I don't supposed you took the flow & return water temps of each one do you John? only what would have told us the heat out put. Interesting to see the heat distribution across them & the mean water temp 51.53 top, 60.93 middle & 59.2 bottom
 
Yes, Probably not thermodynamically very accurate but I just averaged all the temps and the BTOE was 51.5C="31.5 deg rad" or 55% output of a 50 deg rad, the TBOE was 59.C="39.1deg" rad or 73% & the best (marginally, apart from the quickest heat up) was the the BOE, 60.9C="40.9deg" rad or 77%.
 
Yes, Probably not thermodynamically very accurate.

If the flow rate was the same in each case, it would more accurate to infer the emitted power from the fact it's proportional to deltaT. If this is correct, there is only a 10% advantage for TBOE over BTOE at deltaT = 9 °C. The heating engineers the OP is arguing with are claiming TBOE vs BTOE makes an insignificant difference and this result supports them; if the OP's radiators are running at deltaT = 3 °C rather than 10 °C the difference will be even smaller.
 
If the flow rate was the same in each case, it would more accurate to infer the emitted power from the fact it's proportional to deltaT. If this is correct, there is only a 10% advantage for TBOE over BTOE at deltaT = 9 °C. The heating engineers the OP is arguing with are claiming TBOE vs BTOE makes an insignificant difference and this result supports them; if the OP's radiators are running at deltaT = 3 °C rather than 10 °C the difference will be even smaller.
I wouldn't mind betting the 3 °C was not as per the design. With these communal systems the heating pant is invariable N.gas powered & capable of condensing so the design should be very much that of the domestic gas boilers system.
 
  • Agree
Reactions: Chuck
It must be remembered that normally the mean rad temperature is conveniently taken as (flowtemp+returntemp)/2, this in turn infers that there is a gradual fall in temperature between the flow and return which can be seen quite clearly in the sketch in both the BOE & the TBOE cases but not in the BTOE case where several temperatures are only in the 40s. To put it another way, when I average the 15 readings in both the BOE & the TBOE and compare them with their mean temps based on (flow+return)/2, one gets (BOE) 60.9 vs 61 & (TBOE) 59.1 vs 59, very close in both cases.
If one compares these numbers in (BTOE) one gets 51.5 vs 60.5, a huge difference (obviously because of the non linear fall in the actual rad temps).
If one accepts this reasoning, then the BOE is the best performing rad and is a (61-20) "41 deg rad"(partly because of the low flow temp of 65C), the next best is the TBOE, a (59-20) 39 deg rad, almost identical (apart from slower warm up) than the BOE.
The BTOE rad is only a (51.5-20) 31.5 deg rad so its performance using the BOE as the bench mark is only outputting (31.5/41)^1.3, ~ 71% of the BOE output?.
 
I wouldn't mind betting the 3 °C was not as per the design.

I agree. The deltaT = 3, which I inferred from post #20 seems far too low. The flow temperature, if it is 53 not 75 is also too low and is where I'd look. Anyway, I'm repeating myself so I'll shut up now.
 
  • Like
Reactions: chris watkins
I have just cleaned up the sketch, it seems pretty clear that convection just won't work properly in any system with the return on the top. In the BTOE there are 8 readings between 45C and 38C, the flow just seems to flow upwards to the top of the rad and then partially short circuits out the other end whereas in the other two it flows along the top of the rad and then drops down in parallel flow paths and out the return, (makes sense I suppose).
BTOE, BOE & TBOE Rad Temp.jpg
 
John.g how did you arrive at the figure of 71% output in comparison with the BOE? I understand everything but your calculation of (31.5/41)^1.3 ~71%. I’m not an installer, I work with oil fired appliances but I’m keen to fully understand this.
 
Don’t worry I was reading it slightly wrong. I know your numbers are estimates but the output is 77% of 41 not 71%?
 
Rads are now based on a "50 degree" basis (formerly 60C), this is the mean rad temperature minus a nominal room temperature of 20C, For example if your rad had a flow temperature of 75c & a return of 65C then you would have a "50 deg rad", ((75+65)/2)-20. If you had a flow temp of 75C & a return of 45C ( to get full condensing benefit) then you would have a "40 deg rad, ((75+45/2)-20, you would then have to install a bigger rad(s) to compensate for this and all the rad manufacturers have correction tables, I don't bother looking at them because I figured out (from the trend line on a spreadsheet) sometime that they are using (generally) a correction factor of ^1.3, so the correction factor for your rad is,(50/40)^1.3, 1.34 which means that the rad would have to be 34% "bigger" or if you installed the rad based on 50 degrees then it would output only, (40/50)^1.3, 75%.
The reason I used 41 degrees as the base is because the boiler had its set point at 65C which resulted in my rad(s) being only a "41deg"one so I just compared the output of the BTOE rad based on this. I could have and maybe should have corrected both of them to the 50 deg standard, this would equate to outputs of 77% & 55% respectively to the BOE & the BTOE rads so the BTOE rad would still output (55/77), 71% of the BOE one as stated above.
 
Last edited:
  • Like
Reactions: SJB060685
"^" is to the power of, you have heard of "squared", 2 squared is 2x2 or 4 but if you were to do this calculation in a spread sheet, you would enter = 2^2, and so on,
 
  • Informative
Reactions: chris watkins
Based on your calculations I’m arriving at the same answers, just doing it differently lol. Like I said I’m not an installer, I service, repair and commission oil fired appliances. I’ll leave the plumbing to someone who wants it lol
 
On the N.gas designs (with DHW cylinder) I use (70+50/2)-21 (or 22 bathrooms) 39 (38) against the rad manufacturers stated outputs of 50.
Do you think your results would differ much with the higher temps ?
 
  • Like
Reactions: SJB060685
Im keen to see Johns answer to your question Chris. Like I said I don’t install, just service oil appliances. However I always went by the formula of flow temp is say 70, return is say 50, so average rad temp is halfway between. Then minus room temperature of 21 gives the final figure. Similar to how you guys have said you do it above.
 
Yes, this is quite correct in the case of a BOE or a TBOE plumbed rad but is not (IMO)) the true mean temp in a BTOE (wrongly??) plumbed rad.
BOE Two Outputs.jpg


Re Chris's query above, I can,t really get higher than 70C flow temp as I have a oil fired boiler which has a max set point of 75C and a hysteresis of 10/12C so hard to get a steady flow temp for readings but they are quite reasonably correct.
His "39 deg rad" should emit 72.3% of a "50 deg rad" and the "38 one "will emit 69.9%
 
Can anyone recommend a cheap infrared thermometer that I could use to measure the temps. I might go into another flat and compare the temps of the RADS.
 
I agree, I will focus on the fact that the flow temperature is 53 degrees where it should be much higher.
Then they can rectify the fact that they have piped it incorrectly.
They also need to get rid of the air in the system.
When engineer was on-site he showed me the heat exchange unit had temp of 80 degrees but I believe that was the incoming hot water from district plant.

Anyway I will order a thermometer to take readings of all my rads.
Will then see if neighbor will assist to provide theirs.
 
One more thing I can recollect in my mind is when the first visited the flat last year, they did mention they moved something to the highest setting. It could have been the pump or it possibly could have been the thermostat for the heating in the heat exchange. The thermither range is between 35 and 75 degrees. Maybe there could be a possibility that the thermostat setting is still stuck on the original setting or maybe it's not functioning correctly. I can't turn it as it seems to be stuck in its current position which is at the highest setting.
 
If the rad inlet temperature is 75C then you won't be able to keep your hand/finger on this for very long, if its = or < 53C then you will be able to.
 
  • Agree
Reactions: Best
Ok I have a response.
They contacted the manufacturer stelrad and they told them they need to swap the flow and return.
I have also asked them to resolve the incorrect flow temperature as it should be much higher.
So hopefully all will be resolved, it is a wonder this into web thing.
 
Ok I have a response.
They contacted the manufacturer stelrad and they told them they need to swap the flow and return.

I have also asked them to resolve the incorrect flow temperature as it should be much higher.

Good result, do hate how these developers take the pish so badly, you really shouldn't have to fight that hard for them to provide a pre agreed upon service/result!
 
  • Agree
Reactions: chris watkins
To be honest, most people wouldnt even check. As long as their rads get hot they would be not realising there is a problem.
Anyway they also have to resolve the fact that there is a lot of air in the system. Not sure how power flush is going to solve it. Assume their must be air ingress from somewhere.
 
Anyway they also have to resolve the fact that there is a lot of air in the system. Not sure how power flush is going to solve it. Assume their must be air ingress from somewhere.
Check the inlet side of the pump & pipe work until you get to the expansion vessel.
What is the pump set to ? (let me guess fixed speed III)
 
That's interesting.
Perhaps the developer looks after the secondary circulation as well? and has a flow meter plus the flow and return temps to calculate the energy usage for each user billing purposes.
 
There is a meter on the pipework in the heat unit. Now I will try to figure out how that actually works.
 
If they are serious about accurate energy consumption then they would have to measure the deltaT as well. Is the bill presented as kWh/month? Or what and purely for comparison purposes you might like (or not) to post one period's usage.
Any circ pump visible there?.
 
They provided pay as you go meter. This is only for hot water and heating. I have standard meter for electricity.

For months where I don't use heating it costs £20 a month. During winter it cost me around £60 a month with the inefficient setup.

Basically this is for a family of 4.
Also I know they take £10 a month in service charges to cover for the gas in the central plant and for maintenance of everything including the unit in the flat.

This seems far more expensive then gas boiler.

I will look at the Meter and check the KW usuage. The rate is at 7.9p per unit used. I believe there is a daily charge of 20p a day. So around 6 to 8 KW a day usuage.

I am quite Fortunate, apparently one bedroom flat was paying more than us to assume something is wrong with his meter.
 
  • Like
Reactions: John.g
You are doing very well, I would have expected at least ~ > 30 kwh/day for HW+CH. If you are/were using say 7 kwh/day for HW only; this equates to a usage of 30 ltrs/person/day @ 60C or 50 ltr/person/day @40C, pretty normal/lowish usage IMO.
 
  • Agree
Reactions: Pickwickpick
Hadn't copped on to the£60/month in winter which would equate to ~ 25 kwh/day for HW+CH, quite good.
 
Plumber is here now. He said 53 degrees flow is fine for the flow. He said 70 degrees would be too hot. That means the radiators must be oversized as the are delta 50 rads.
 
Ok works completed. 58 degrees on the flow and the return is 52 degrees. One thing we noticed, radiators can be bleed of air and the system has no air anymore. Silent central heating for the moment. Radiators are much more hot to feel. Still don't think they are running at full potential and to be honest he should have atleast balanced them as there are temperature variations between flow and return between all rads.

I will leave heating on for atleast an hour a day to see if air keeps returning in the system.

Does anyone think that the flow temps should be higher but don't forget that we have district heating system. Input Temp is around 80 degrees so not sure if heat exchanger can heat the cold water to 70 degrees. Engineer mentioned that the water in the radiators would be higher than what he is seeing on his thermal camera.
 
You have, (58+52)/2)-20, "35 deg" rads which will output 63% of 50 deg rads, the rads may have been over sized to take account of this, who knows?. If you gave dimensions of a room and the dimensions of its rads then someone on here would have a fair idea IMO as to under sizing or not.
I see you have 2 rads in a 30 M2 L shape so maybe you can just post the dimensions of those two rads.
 
Last edited:
My main concern now is my daughters room. I believe bthey have have undersized the radiator as it's only 600 X 600. It's stelrad compact K2 I believe. Output is 3100 BTU. Room size is 14.25 by 11.25 ft.

Room has two external walls and one of these has a French door leading to balcony which is around 2m by 2m in size.

Problem is I don't think they will ever change the size of the radiator.

Open Plan will be ok as we also cook there.
 
My main concern now is my daughters room. I believe bthey have have undersized the radiator as it's only 600 X 600. It's stelrad compact K2 I believe. Output is 3100 BTU. Room size is 14.25 by 11.25 ft.

Room has two external walls and one of these has a French door leading to balcony which is around 2m by 2m in size.

Problem is I don't think they will ever change the size of the radiator.

Open Plan will be ok as we also cook there.

From your info you have a 0.91 kw rad ("50deg") heating 14.9 m2, but because of lower running rad temps you only have a 0.57 kw rad ("35 deg") and I doubt very much if this was the design heat loss for this 14.9 m2 (38watts/m2).
You would need a ~ 60% bigger rad to compensate if a 50 deg rad was their standard.
As stated above it would be nice to get the heat loss calcs.
 
Last edited:
Hi, I never responded but thanks for everyone's input, help and advice. Eventually might change the radiator in my daughter's room but will leave it for now. Developer never ever responded with heat loss calculations. I think it is simple enough to upgrade the radiator and be done with it.
 
Yes it matters return should be at the bottom so when the water cools it flows into the return
Not necessary at all .
One of the biggest myths of small bore heating . TBOE doesn’t care what the configuration is and the so called “ traditional wisdom” of putting the return at the bottom or lower pipework is flawed.
Correctly as you state , as the hot flow gives up it’s heat it becomes less buoyant more dense and will sink , but there is no counter convective interaction and although the efficacy is improved with TBOE , it would improve further with Hot flow at bottom return at top . Why ? Because it created counter flow , which is far more efficient in mixing flows and therefore gives greater thermal efficiency and efficacy . Heat from the convective sense naturally wants to rise . It will of course give up heat becoming cooler and want to fall back down . If the hot feed is at bottom, it counter flows and blends far better . The remaining heat wants to rise anyway so it’s sensible to return from the top , not forgetting it’s still relatively hot and the useful heat exchange has taken place in the rad.
All large heat exchangers function with counter flow
 
Not necessary at all .
One of the biggest myths of small bore heating . TBOE doesn’t care what the configuration is and the so called “ traditional wisdom” of putting the return at the bottom or lower pipework is flawed.
Correctly as you state , as the hot flow gives up it’s heat it becomes less buoyant more dense and will sink , but there is no counter convective interaction and although the efficacy is improved with TBOE , it would improve further with Hot flow at bottom return at top . Why ? Because it created counter flow , which is far more efficient in mixing flows and therefore gives greater thermal efficiency and efficacy . Heat from the convective sense naturally wants to rise . It will of course give up heat becoming cooler and want to fall back down . If the hot feed is at bottom, it counter flows and blends far better . The remaining heat wants to rise anyway so it’s sensible to return from the top , not forgetting it’s still relatively hot and the useful heat exchange has taken place in the rad.
All large heat exchangers function with counter flow
Having the return from the Top certainly made a big difference in the few tests I carried out over 3 years ago
It seems pretty clear that convection just won't work properly in any system with the return on the top. In the BTOE there are 8 readings between 45C and 38C, the flow just seems to flow upwards to the top of the rad and then partially short circuits out the other end whereas in the other two it flows along the top of the rad and then drops down in parallel flow paths and out the return, (makes sense I suppose).View attachment 37636
1675794568639.png
 
  • Informative
Reactions: Chuck
I think that what Jonny_boy is saying that the conventional flow-top return-bottom arrangement results in a stratified temperature distribution but putting the flow in at the bottom and return out from the top would disrupt this pattern resulting in a higher average temperature of the radiator surface.

While this might be true for some designs of radiators, if you try it in practice with a typical domestic radiator the flow scoots up the inlet side, along the top and straight out of the return. This is the 'conventional wisdom' and @John.g's measurements seem to confirm it.

The flow speeds inside a radiator are really quite low and I have no difficulty believing that convection forces play a significant, maybe dominant, role even in a pumped system.
 
Last edited:

Official Sponsors of Plumbers Talk

Similar plumbing topics

We recommend City Plumbing Supplies, BES, and Plumbing Superstore for all plumbing supplies.