Radiator air lock

ChrisGER

I hope someone could help me resolve my newly installed heating system using a WHB055N Knight boiler. Currently one of a total of two radiators installed is constantly building up an air lock resulting into the cold radiator.  We conclude that our plumber did not install the system correctly as per Lochinvar User’s Manual, and also the supporting document published by Bell & Gossett in regards to Primary/Secondary pumping.

Technical Manual TEH-775A at https://documentlibrary.xylemappliedwater.com/wp-content/blogs.dir/22/files/2013/02/TEH-775A-1.pdf

Lochinvar User’s Manual and the above link state the same concept as the specification drawing shown below that we sent to the plumber before the work started.  The Lochinvar buffer tank shown below is used for hydraulic separation of the primary and secondary loop.

However our plumber installed (shown below) both of the secondary circulators serving the radiators and the heat exchanger such that they discharge into the buffer tank instead of pumping away from it.  I understand having the system circulators discharge into the buffer tank results in a decrease in secondary circuit static pressure. Consequently, the drop in pressure creates air problems out in the system by having air come out of solution and form bubbles in the radiator. 

In the Bell & Gossett document, it states specifically 

“The common piping can be considered as the compression tank ‘No pressure change point.’ It is consequently generally wrong to pump into the common piping from the secondary circuit because of a decrease in secondary circuit static pressure.”

 Figure 5 – WRONG–SECONDARY CIRCUIT PUMP DISCHARGES INTO COMMON PIPING

I hope someone could confirm our current setup is incorrect, hence our plumber is willing to change it as per our spec.  Our dilemma is our plumber does not trust our understanding because I am not the expert.  

pedmec

you don't have a primary secondary arrangement. lets get that out of the way. basically if you sketched out the piping you would realize that its really piped as a direct return. it looks different because you have returns on the opposite side but it is still a return.
that drawing comes from bell and gossett primary/secondary installation manual. it does apply here. that common piping is called a crossover bridge and is used for large systems when adding additional heat loads to current system without having to add additional load on primary system circulator. wrong application for your job.

pedmec

that primary /secondary drawing DOESN'T apply here, misspelled. you wouldn't take off the supply for your radiant going into the buffer tank if you want hydraulic separation.

my question is why you felt the need to pipe your boiler this way? i'm assuming because you only have (2) radiant panels for your load you didn't want to short cycle your boiler? if that's the case you could have used a low loss header to accomplished the same thing. plus the whb has a 10-1 downturn. you could have lowered the firing rate.

the circulators should be piping away from the point of no pressure change correct.

ChrisGER

Hi, thank you for your reply. I planned to use a 3-pipe buffer tank configuration as described here
https://www.pmengineer.com/articles/94749-john-siegenthaler-combining-heat-pumps-with-buffer-tanks

and https://m.youtube.com/watch?v=OyYv4ApJfCc

This should still give good hydraulic separation. 

The only problem and difference with what is described in the link is that my system pump (green) is discharging into the buffer tank instead of away from it.

GGross

I would agree it is not piped in the way you have drawn. That is a given. I am trying and failing to hold back the obvious questions here about your piping arrangement choice and will instead stick to the issues. If you have air problems, in a system that has been bled of air, then you most likely have a leak. With all of the money spent on buffer tanks and extra pumps, I should hope you sprung for a water meter of some kind on that fill valve.

ChrisGER

Hi GGross, thank you for your reply. I agree that the actual piping shown in the basement picture is different from both schematics.

The basement picture shows a 3-pipe buffer tank configuration. The system pump (green) discharges into the buffer tank when it should actually pump away from it into the system. My understand of the current setup is that discharging into the buffer tank (PONPC) can lead to a reduction in secondary static pressure. This can result in air coming out of solution and air lock in the radiators even when there is no leak. 

The system pump should have been piped such that it pumps away from the buffer tank.

ChrisGER

GGross said:

If you have air problems, in a system that has been bled of air, then you most likely have a leak.

I don’t see water come out of the installation at any point.

GGross

Yeah looking at your set up the only connections not in the boiler room are likely right at the rads. Either way it's not piped correctly, and you can see the air, so I am not sure what the plumbers hold up is on fixing it. Hope he gets over there and gets it taken care of.

ChrisGER

Hi GGros, do you agree that the system pump should pump away from the buffer tank?

Peter_26

Is the boiler circulator pumping into the boiler?

GGross

Responding to OP Yes definitely.

Zman

You are correct that the circs should be "pumping away" from the PONPC. You are also correct that this condition is contributing to your air issues. In the installer's defense, I am sure your design confused the hell out of him as it is a bit unorthodox and confusing (it also will work just fine).
Moving the 2 circs to the supply side does not look too difficult and the installer should be willing to do that.

ChrisGER

Peter_26 said:

Is the boiler circulator pumping into the boiler?

Yes it is and that’s how it should be according to Lochinvar. The system pump is however pumping into the common piping (buffer tank) which seems wrong.

GGross

I still just don't like that you have that nice buffer tank and have the boiler piped direct to load bypassing it. In my opinion that is a bigger mistake than the pumps in this install. But generally speaking those pumps should not be pumping toward the buffer tank. Not sure if the buffer tank qualifies as "common piping" as described in the B&G drawing, as that is referring to primary/secondary piping, which is not what you have here. You have a buffer tank, with boiler piped direct-to-load, using the mass of the buffer tank on the return. I don't think the primary/secondary drawings necessarily apply to this application. But no matter what, that air needs to be removed.

ChrisGER

Zman said:

You are correct that the circs should be "pumping away" from the PONPC. You are also correct that this condition is contributing to your air issues. In the installer's defense, I am sure your design confused the hell out of him as it is a bit unorthodox and confusing (it also will work just fine). Moving the 2 circs to the supply side does not look too difficult and the installer should be willing to do that.

The design is from John Siegenthaler’s hydronic piping and buffer tank course here

https://youtu.be/1DMC0w5GscA

I am not sure what you mean with “it will also work just fine” because one radiator get air locked constantly.

ChrisGER

GGross said:

I still just don't like that you have that nice buffer tank and have the boiler piped direct to load bypassing it. In my opinion that is a bigger mistake than the pumps in this install. But generally speaking those pumps should not be pumping toward the buffer tank. Not sure if the buffer tank qualifies as "common piping" as described in the B&G drawing, as that is referring to primary/secondary piping, which is not what you have here. You have a buffer tank, with boiler piped direct-to-load, using the mass of the buffer tank on the return. I don't think the primary/secondary drawings necessarily apply to this application. But no matter what, that air needs to be removed.

I see what you mean but check this video by John Siegenthaler at 48:04 to see the common piping/hydraulic separation. The tee at the buffer tank in short and fat.

https://youtu.be/1DMC0w5GscA

GGross

Thank you for sharing the timestamp. I had not considered that this would still provide separation. Can I ask what the "4th pipe" is doing in your arrangement? and what purpose the heat exchanger is serving in your application?

ChrisGER

GGross said:

Thank you for sharing the timestamp. I had not considered that this would still provide separation. Can I ask what the "4th pipe" is doing in your arrangement? and what purpose the heat exchanger is serving in your application?

In the same YouTube video at 50:35 there is an explanation of adding a DHW detail. 

In my basement picture, the top left connection of the buffer tank is used to connect to a heat exchanger. It is basically an additional zone. A tankless water heater is not installed currently until the issue with the system circulators is fixed.

GGross

That makes sense.

Peter_26

In all of the examples that John Siegenthaler has in his "3-pipe buffer tank system" presentation you can see that he has the pumps piped directly off the "fat" header before the loads so the hottest water can be supplied. I don't see where the idea for the pumps on the return came from, can you elaborate on that?

In the examples he also illustrates that the expansion tank should be at the "PONPC", which is closest to the tank on either side of the tank on the return side.

If you reverse the pumps you are not going to be providing your system with the hottest water possible and the stratification ability of the tank will be gone. I don't know if having them there can also contribute to air problems.

Do you have TRVs installed on your radiators?

@hot_rod has done a lot of work with buffer tanks and can definitely help with this.

GGross

Peter_26 said:

I don't see where the idea for the pumps on the return came from, can you elaborate on that?

I believe that is exactly the issue that OP has asked about. They were concerned because the pumps should be pumping away from the buffer tank, and their installer put the pumps pumping toward it

ChrisGER

Peter_26 said:

In all of the examples that John Siegenthaler has in his "3-pipe buffer tank system" presentation you can see that he has the pumps piped directly off the "fat" header before the loads so the hottest water can be supplied. I don't see where the idea for the pumps on the return came from, can you elaborate on that? In the examples he also illustrates that the expansion tank should be at the "PONPC", which is closest to the tank on either side of the tank on the return side. If you reverse the pumps you are not going to be providing your system with the hottest water possible and the stratification ability of the tank will be gone. I don't know if having them there can also contribute to air problems. Do you have TRVs installed on your radiators? @hot_rod has done a lot of work with buffer tanks and can definitely help with this.

I appreciate your feedback and I agree with you.

I wanted to have it piped as John Siegenthaler specified it. Unfortunately, the plumber who installed it decided to change the design because he thought the location of the system circulators does not matter. I disagree with that. 

I do have TRV on the panel radiators.

Zman

ChrisGER said:

Zman said:

You are correct that the circs should be "pumping away" from the PONPC. You are also correct that this condition is contributing to your air issues. In the installer's defense, I am sure your design confused the hell out of him as it is a bit unorthodox and confusing (it also will work just fine).
Moving the 2 circs to the supply side does not look too difficult and the installer should be willing to do that.

The design is from John Siegenthaler’s hydronic piping and buffer tank course here

https://youtu.be/1DMC0w5GscA

I am not sure what you mean with “it will also work just fine” because one radiator get air locked constantly.

I was referring to the original design "working just fine"...

ChrisGER

 Zman said:

Zman said:

You are correct that the circs should be "pumping away" from the PONPC. You are also correct that this condition is contributing to your air issues. In the installer's defense, I am sure your design confused the hell out of him as it is a bit unorthodox and confusing (it also will work just fine). Moving the 2 circs to the supply side does not look too difficult and the installer should be willing to do that.

The design is from John Siegenthaler’s hydronic piping and buffer tank course here

https://youtu.be/1DMC0w5GscA

I am not sure what you mean with “it will also work just fine” because one radiator get air locked constantly.

I was referring to the original design "working just fine"...

I see. I like John’s design very much. I am from Germany and we have the panel radiator  in every house and apartment. They are the best. Can’t wait for it to finally work after switching the circulator location.