Search the forum,

Discuss Mixing Pressure Problems in the Central Heating Forum area at PlumbersForums.net

Messages
30
I have a rather large wet UFH system. Water is flowing around the system, but I don't get any (or minimal) hot water injecting from the primary boiler flow into the ufh system. this means the mix temp never gets above 20 degrees.

I think that the ufh system's pressure might be too high which is causing it to backflow into the t junction where the mix occurs and then this water gets backfilled into the system, backing up the hot water from the boiler.

I've tried fiddling with the diverter valve, but it's so fiddly that half a degree and primary loop goes from cold after the divert flow to red hot!

Pic below if it helps.

I think that I need to reduce the pressure of UFH system but other than that, no idea.
Photo from John Savage.jpg
 
Last edited:
Will try and assist.
I have only seen one UFH system but assume they are all similar, ie the primary pump (one on the left?) takes hot water from the boiler, supplies this to the thermostatic mixer, the mixed water is supplied at the required UFH temperature to the manifold pump (one on the right?) which distributes it to the UFH coils and some of this cooled water returns to the boiler with the remainder mixing with the hot water to give the required mixed water temp.
If you are correct in your understanding of the problem then suggest the following, can you see what the total flow is through the UFH manifold by the flow meters, if this is very low you may have a problem with the manifold pump, either a defect or a wrong setting, could also be the supply pump or the thermostatic mixer.
Can you confirm that the left hand pump is the hot water supply (boiler) pump and the right hand one is the manifold circulating pump and the settings on both pumps and the exact model(s) information.
 
Thanks John

No the left-hand pump is purely to circulate the water around the UHF system, the right-hand side one is the injection pump which pulls hot water in from the primary loop into the ufh system. Flow gages on manifolds are all showing flow around the system. This system works slightly differently, it's an injection system and there isn't a thermostatic mixer. What happens is that when the system calls for heat hot water from the high temp loop is injected into the low temp loop, with a back flow coming through a diverter valve.
[automerge]1606593908[/automerge]
Thanks John

No the left-hand pump is purely to circulate the water around the UHF system, the right-hand side one is the injection pump which pulls hot water in from the primary loop into the ufh system. Flow gages on manifolds are all showing flow around the system. This system works slightly differently, it's an injection system and there isn't a thermostatic mixer. What happens is that when the system calls for heat hot water from the high temp loop is injected into the low temp loop, with a back flow coming through a diverter valve.
The Injector Pump is a Grundfos Alpha 2L and it is set on lowest constant pressure (I've tried it on highest). The other circulating is Grunfos UPS 25-80 180 and was on setting 1 (out of the 3).
 
Last edited:
Hi we we are ufh installers really need to see a wider angle picture that shows manifold as well as pumps . Also has system ever worked correctly?.
 
Hi we we are ufh installers really need to see a wider angle picture that shows manifold as well as pumps . Also has system ever worked correctly?.
There are two sets of manifolds, and they are in different rooms (downstairs and upstairs). We inherited the system when we moved in to the house about 7 years ago and I don't think that it's ever worked properly. There is definitely flow going to the manifolds though as I 've checked the flow gauges to make sure the flow is balanced.
 
Last edited:
Thanks John

No the left-hand pump is purely to circulate the water around the UHF system, the right-hand side one is the injection pump which pulls hot water in from the primary loop into the ufh system. Flow gages on manifolds are all showing flow around the system. This system works slightly differently, it's an injection system and there isn't a thermostatic mixer. What happens is that when the system calls for heat hot water from the high temp loop is injected into the low temp loop, with a back flow coming through a diverter valve.
[automerge]1606593908[/automerge]

The Injector Pump is a Grundfos Alpha 2L and it is set on lowest constant pressure (I've tried it on highest). The other circulating is Grunfos UPS 25-80 180 and was on setting 1 (out of the 3).

Had a quick look at the basics of UFH injector systems, one seems to inject hot water through the injection valve and into the circulating pump suction, the other seems to use the injection valve as a diverter to control the cold water return to the boiler. Your system looks a bit like the former? but afraid I can't make much sense out of it, Is that vertical pipe, the injector/diverter or what Injector system.jpg??. and are the pumps pumping as indicated, or can you indicate. Thanks.
 
As John G has said difficult to make much sense of what you have! The valve you have circled looks more like a manual bypass than anything else .
Any chance or a picture of manifolds please?
 
Morning,
The pumps should be pumping as you have indicated John, and the circled valve is indeed the manual bypass. The problem as I understand it is that if the manual bypass will push the same volume of water back into the primary loop as the pump injects out of it. I've fiddled with this and I can get hot water down to the injector pump but only when the flow back to the primary loop is massively restricted. If I open up the bypass essentially the cold water pushes back past the t junction where the injector pump branches off, and all that happens is that the cold water gets re-injected into the UFH. I've found some old schematics, and the picture of my manifold is there (actuators taken off as I wanted to make sure none of those were causing flow to stop).
IMG_20201129_100948.jpg
IMG_20201129_100840.jpg
IMG_20201129_100716.jpg
 
I'd say you would need a magician to get that going in its present form, there is no injected water control or return to the primary system to keep the system(s) balanced.
You might be able to install a thermostatic mixing valve and utilise the Alpha2 as the manifold pump, you might then be able to use the existing boiler circ pump to provide the hot water and remove the 8M grundfos. Depending on the hot (boiler) flow temp, the mixed flow temp and the manifolds return temps the boiler circ pump may be able to do this or upgrade it, if you had 70C/40C/30C respectively then you only need the boiler circ pump to provide 25% of the mixed flow, for example you may have say a total UFH heating demand of 20kw which will require a manifold circulation flow of 29 LPM but will only require 7 LPM from the boiler circ pump.
I'd say exedon2 would point you in the right direction.
The attachment is a good depiction of the above and shows that whatever hot water flows into the mixer is returned to the boiler to maintain the balance.
 

Attachments

  • UFH Wunda Manifold and Pump.pdf
    1.4 MB · Views: 12
Last edited:
I think the Kanmor box is supposed to control the pump that injects water. The bypass valve I believe is supposed to be passive as its job is essentially just to allow a flow back equal to the amount of water injected and that's how the water returns back to the primary system.

I thought that I might be able to play about with the pump settings and valve and coax the hot water round but there are too many variables for me to play with.

Cheers for the help though, I've often been told that the system I have is a bit different!
 
As John ,g has said never going to work!
I would just go for sorting it properly by fitting a pump/ blender unit to each manifold each feed by a dedicated zone valve. Get rid of current underfloor pump if the one shown is only bypass on system change it for a auto bypass .
On the picture of the manifold all the actuator heads have been removed I take it they do work?
 
There's a bit of a delay as my posts seem to take an age to get through moderation (probably because I'm new). The actuators have been taken off to make sure that I was getting flow everywhere. The Alpha 2L injector is supposed to be controlled by the Kanmor 361 controller box. I've fiddled with it again and managed to persuade a little hot water to come down through the pipe to the injector, but I was hoping that swapping the bypass valve to an auto bypass might have been the answer, looks like it might be a bit more costly!!
 
Yes I'm afraid some cost involved but it's not much use as it is!
Correctly working ufh is lovely to live with .
Make sure you get a heating engineer who is familiar with ufh both pipework and its correct wireing.
 
No it's not much use as it is at all. The problem is that the bloke who built the house used non standard parts for everything, from the boiler to the manifolds. I've just had a good read of the pack and need to change some settings on the mixer controls... You never know, it can't do any harm!
[automerge]1606689787[/automerge]
I think I've got it sussed. Checked the system and there are 2, 2 zone part valves going to the DHW and primary loop. When the system calls for DHW the boiler loop loses pressure and the cold water from the UFH pushes back into the primary loop via the bypass. When the valve opens to supply hot water back to the primary loop, the hot water struggles to overcome the pressure of the cold water coming back through the bypass and can't reach the injector pump. If I play with restricting the manual valve I can usually resume the flow of hot water to the injector but it's very sensitive. I think If I replace the manual valve with an auto valve and then rebalance it might solve the problem..... Well here's hoping (and I hope that all makes sense as I'm a science teacher and not a heating engineer and before Monday knew nothing at all about UFH systems!).
 
Last edited:
Despite our reservations, it is certainly a very interesting challenge. logically, one would think that it should work in some fashion or other, if not perfectly, some design guru went to the trouble of putting this package together with the Kanmor 361 and the injection pump etc, if you have any links to its setting up you might post them.
Re the injection pump control which is one of the key components, the speed/frequency is controlled by the Kanmor but this is only designed for the older type of circ pumps (see quote) some of the newer types can be externally controlled, but only by a PCM (pulse width modulation) signal which is plugged into the pump motor. I wonder was your injection pump changed out sometime for the present Alpha 2L which cannot be speed controlled IMO by the Kenmor.

Quote:
HVAC Systems
The Mixing Control 361 is designed to operate a mixing pump using Outdoor Temperature Reset. It can be used in applications ranging from in-floor radiant to commercial baseboard systems for Boiler Return Protection with full reset. This control regulates a single heating water temperature through Mixing Outdoor Reset. It provides mixing by speeding up or slowing down a single permanent capacitor (standard wet rotor) circulator, and provides control of the mix system pump. Unquote.
 
Despite our reservations, it is certainly a very interesting challenge. logically, one would think that it should work in some fashion or other, if not perfectly, some design guru went to the trouble of putting this package together with the Kanmor 361 and the injection pump etc, if you have any links to its setting up you might post them.
Re the injection pump control which is one of the key components, the speed/frequency is controlled by the Kanmor but this is only designed for the older type of circ pumps (see quote) some of the newer types can be externally controlled, but only by a PCM (pulse width modulation) signal which is plugged into the pump motor. I wonder was your injection pump changed out sometime for the present Alpha 2L which cannot be speed controlled IMO by the Kenmor.
Indeed it was...... I'm off to work, but I'll post settings when I get home.
[automerge]1606731305[/automerge]
Despite our reservations, it is certainly a very interesting challenge. logically, one would think that it should work in some fashion or other, if not perfectly, some design guru went to the trouble of putting this package together with the Kanmor 361 and the injection pump etc, if you have any links to its setting up you might post them.
Re the injection pump control which is one of the key components, the speed/frequency is controlled by the Kanmor but this is only designed for the older type of circ pumps (see quote) some of the newer types can be externally controlled, but only by a PCM (pulse width modulation) signal which is plugged into the pump motor. I wonder was your injection pump changed out sometime for the present Alpha 2L which cannot be speed controlled IMO by the Kenmor.

Quote:
HVAC Systems
The Mixing Control 361 is designed to operate a mixing pump using Outdoor Temperature Reset. It can be used in applications ranging from in-floor radiant to commercial baseboard systems for Boiler Return Protection with full reset. This control regulates a single heating water temperature through Mixing Outdoor Reset. It provides mixing by speeding up or slowing down a single permanent capacitor (standard wet rotor) circulator, and provides control of the mix system pump. Unquote.
Add
Indeed it was.... I'm off to work, but I'll post settings when I get home.
I've got the full plumber/electrician pack with all the settings. We might get this thing figured out eventually. John you seem to know a lot about the pumps (of which I know nothing), do you think that maybe changing the Alpha's setting might help in the short term?
 
Last edited:
It may but last night I was looking at the system schematic and can't figure out why the injection pump inlet is taken from the boiler return (NRV in boiler flow), this means that if no demand for DHW or radiators that the injection pump, is fighting the boiler circ pump (which is on the return), very UNlikely designed that way, have another good look at the arrow on the injection pump which may be pointing upwards, the pump would then be returning cold water to the boiler but then there would need to be (probably controlled ) a hot water supply to the UFH main manifold somewhere, if you look at the schematic which includes the DHW cylinder you will see some form of controller that looks a bit like a pump but is probably a controller as it seems to be wired into the kanmor? high up in the drawing, can you identify this as well this evening.
 

Attachments

  • UFH Pump Injected System near Cylinder.jpg
    UFH Pump Injected System near Cylinder.jpg
    54.5 KB · Views: 7
  • UFH Pump Injected System.jpg
    UFH Pump Injected System.jpg
    65.7 KB · Views: 8
Last edited:
Right then John what do you want?
IMG_20201130_165133.jpg

[automerge]1606760051[/automerge]
The only thing missing from the schematic is the outdoor sensor which I think that the unit uses to calculate the flow need to keep the house at a certain temperature. I've also changed the alpha pump settings to lowest constant flow (I) and that seems to be letting the hot water flow through from the primary circuit.

As for the flow, here's my hypothesis:

The primary circuit has two junctions, the first is the intake and the second the return flow. I'm assuming that if the pressure of the return flow is greater than that of the primary the hot water will take the path of least resistance and flow downwards to the injector pump. This seems to be what's happening in practise, as if the return flow is too fast the hot water never gets to the t junction. If it's too slow, the injector doesn't allow any water to be injected which could be due to the incompressible nature of the ufh circuit, or just because the path of least resistance is forwards rather than down.... (I need a degree in fluid dynamics for this one!)
 
Last edited:
Arrow is definitely pointing down. Pages as requested, but I don't know what I've done but the mix temperature is going up. It's at 27 now and rising and it's never been above 21. Basically I put the alpha on (I) waited until primary was sending warm back to the boiler, changed to the pump up to highest constant and it seems like it's working 🤯
[automerge]1606763805[/automerge]
Just checked again and it's back to 24 but progress.... I suppose it's proof of concept at least.
 

Attachments

  • IMG_20201130_185607.jpg
    IMG_20201130_185607.jpg
    55.8 KB · Views: 3
  • IMG_20201130_190047.jpg
    IMG_20201130_190047.jpg
    85.1 KB · Views: 2
  • IMG_20201130_190032.jpg
    IMG_20201130_190032.jpg
    101 KB · Views: 3
  • IMG_20201130_190018.jpg
    IMG_20201130_190018.jpg
    99.3 KB · Views: 2
  • IMG_20201130_185814.jpg
    IMG_20201130_185814.jpg
    58.4 KB · Views: 2
Last edited:
Thanks but can't read the attachments.

What do you mean by
" waited until primary was sending warm back to the boiler "

Is/was the CH or/and the HW on while taking these tests?

Can you see what that item is in the picture showing the cylinder, above. (post 21)

I honestly cannot see that system working as is with the NRV installed where it is and without the CH or HW on as well, at best then IMO, the injector pump will take some of the cooled water returning from the rads/cylinder coil and inject it into the manifold but that temperature is completely dependent on that return temperature which may be even lower than the required UFH temperature, also the "bypass" is stationed far too close to the injector pump to give any real temperature or contra flows separation.
Can you have a quick read through those pages and see what commissioning problems they had and a explanation of how the system is supposed to work.

Taking the injected water supply from between the NRV and the boiler would definitely make some bit of sense as it is then hot injected water at a set (boiler) temperature.
 
Last edited:
I've tried to annotate the picture to label each item.

Hope this makes a bit of sense.

Water.jpg

[automerge]1606770384[/automerge]
Thanks but can't read the attachments.

What do you mean by
" waited until primary was sending warm back to the boiler "

I just meant that I restricted water flow through bypass so that the return to the boiler heated up.

Is/was the CH or/and the HW on while taking these tests?

I hadn't switched anything off

Can you see what that item is in the picture showing the cylinder, above. (post 21)

Wasn't sure what you meant so I annotated the original photo

I honestly cannot see that system working as is with the NRV installed where it is and without the CH or HW on as well, at best then IMO, the injector pump will take some of the cooled water returning from the rads/cylinder coil

No rads and cylinder coil has its own NRV as shown

and inject it into the manifold but that temperature is completely dependent on that return temperature which may be even lower than the required UFH temperature, also the "bypass" is stationed far too close to the injector pump to give any real temperature or contra flows separation.

The contra flow separation proximity is what I thought the problem was.

Can you have a quick read through those pages and see what commissioning problems they had and a explanation of how the system is supposed to work.

Taking the injected water supply from between the NRV and the boiler would definitely make some bit of sense as it is then hot injected water at a set (boiler) temperature.
[automerge]1606770875[/automerge]
Here's the scanned pages.
 

Attachments

  • TapScanner 11-30-2020-21.08.pdf
    757.8 KB · Views: 3
Last edited:
Can you take a snapshot on the left around where I have continued the red line showing the pipework layout, in particular any NRV (the ones you showed are motorized zoning valves)

Read the Tap scanned pages, the most important item now is to see where (if) this NRV is.


UFH Pump Injected System near Cylinder rev1.jpg
 
Last edited:
I think that's about as clear as you can get it, there isn't a NRV between the T off for the HW cylinder coil and the motorised UFH zoning valve so that means that the hot water is supplied to the UFH once its opened despite the original drawing showing one, I wonder why?.
So, as you said the proximity of the injection pump and by pass vale is the main problem although I find it inexplicable that a system would be designed this way... it may work if mods made as in attachment, I don't see any reason to have the hot & cold looped as this only creates a full fledged by pass once the UFH zone valve is opened, also the original injection pump was a standard pump which can be speed controlled.UFH Baxi rev1.jpg
 
Last edited:
Thanks again John,

I've restricted the flow on the bypass valve and turned the Alpha down to constant setting (II). I'll see if that's enough flow to get at least a little bit of heat into the system. I was also wondering if an auto bypass valve might help.

As you know a bit about pumps do you think this one would be suitable as the documentation says that the original was a grundfos 15-50.


I just can't find out if it's controllable.

I'll speak to my heating engineer and see if he can make the relevant changes (although he's the one who put the alpha on!).

Thanks once again for your help with this enigma! One day I will get it to function correctly
 
No, I don't think that pump is compatible with the controller, I wouldn't be too worried about the actual speed control just now, if you and your engineer thinks the mods may work then just do them first. It may be beneficial to install a ABV in the system.
When did he fit this Alpha??,
 
I moved in to this house about 7 years ago. The UFH has never really worked since we moved in.

I think that as it wasn't working he swapped the injection pump about 5 years ago when he changed the boiler.

The system had a dutch boiler and Danish manifolds. Essentially each year we just swap the pieces bit by bit for things that are actually recognised in the UK.

To be fair to our hearing engineer, everything was pretty much non standard and so trying to pin down the problem was a nightmare.

My initial thoughts were that he had over pressurised the UFH when he replaced my downstairs manifold so a week ago I started doing my research. Then I thought it was the bypass valve, but I didn't know enough to be sure. I think my wife thinks I'm having an affair with the boiler as I've spent so much time in that room "fiddeling". Turns out I actually might have been right lol. Thanks once again.
 
Just re thinking suggested mods, suggest just relocating by pass and isol valve but do not remove the return loop to the boiler as high risk of too low circulation flow through the boiler if UFH only on.
[automerge]1606859215[/automerge]
Back for more punishment, I think that rather than a PRV in the by pass that a two port Tapstat might give far better control, you can attach the sensor to the return (or the mixed flow pipe as desired) and it will/should then control to the requiredtemperature. If you install another isol valve in the by pass you can then isolate it and install/change to which ever device gives the best results.
 
Last edited:
Evening John, It's the enigma that's more puzzling than a Rubik's cube.

I think that whatever you put in there, it's always going to be too close to the injection feed. I've been playing around with it today again and as long as I keep the Grundfos Alpha on it's lowest setting I do get hot water to flow just not enough. Problem is that I'll never get enough if I'm that close to the injector because more water will always equal more pressure which will inevitably mean feeding back into the injector loop. (Managed to get the mix feed up to 31 degrees today a new record, although it only stayed at that for half an hour).

Next enigma is trying to get a heating engineer to come and sort it out before the snow hits, and it's a bit of a trek from the south of Ireland to the North of England...

I'll keep you in the loop (Oh the puns!)
 
Any of the drawings refer to the bypass as just that..... a bypass and it would appear to be just that, ie, a bypass for the injector pump. If there were no injector pump fitted then now that we know there is no NRV fitted, that bypass IMO would supply almost ample flow from the boiler circ pump bearing in mind that it only has to supply only ~ 25/35% of the mixed flow. NO where is there any mention of cold water return to the boiler which is just as important as any injector pump or whatever so was something forgotten in the whole design.
Maybe at the end of the day thermostatic mixing is the real answer?. So, yes, heating engineer or magician needed.
 

Reply to Mixing Pressure Problems in the Central Heating Forum area at PlumbersForums.net

Creating content since 2001. Untold Media.

OFFICIAL SPONSORS

OFFICIAL SPONSORS

Back
Top
AdBlock Detected

We get it, advertisements are annoying!

Sure, ad-blocking software does a great job at blocking ads, but it also blocks useful features of our website. For the best site experience please disable your AdBlocker.

I've Disabled AdBlock