Injection mixing from storage question

[kuribo]

I'm curious as to what benifet injection mixing has over a 3-way valve?

Sent from my moto g(6) play using Tapatalk

The usual answer to that is faster response, better control, more stability at low hot flow, and more efficiency. Also. injection mixing often can be less expensive in upfront costs. Finally, injection mixing is best suited for high mass, low temp heat emitters....

At least that is what I have read and been told.

 

[chew72]

The usual answer to that is faster response, better control, more stability at low hot flow, and more efficiency. Also. injection mixing often can be less expensive in upfront costs. Finally, injection mixing is best suited for high mass, low temp heat emitters....

At least that is what I have read and been told.

(After writing this post I just wanted to preface this with: I'm not being argumentive. I have genuine interest and just want to share my understanding and perhaps be corrected if it's wrong.)

I'll counter that with its in floor. A fast response is not needed. I think a valve can fully swing in about 30s. Is been my experience the control and stability from a mixing valve is 100% ie if your system is designed to run on 110F water it will always be 109-111F so long as your supply water is 110F or higher. However you may have me at more efficient and its the buffer tank that's key.

I have not seen a buffer tank set up. But with a typical system you have your floor loop circ plus a main loop circ unless your plumbing is in the same room as your storage. So the two circ pumps are both needed all the time to add heat to the system. With the buffer tank you could have your injection coming straight from storage and depending on load and temperatures it might only have to run 10% of the time.

I would think however with a buffer tank to gain that effecency you'd have a greater hysteresis (tempeater swing) on your infloor supply say in the case of a 110F setpoint getting 105-115F but it's infloor and I doubt you'd notice the difference.

Also this has really peaked my interest as I'm about to design my system this weekend.

Sent from my moto g(6) play using Tapatalk

 

[kuribo]

(After writing this post I just wanted to preface this with: I'm not being argumentive. I have genuine interest and just want to share my understanding and perhaps be corrected if it's wrong.)

I'll counter that with its in floor. A fast response is not needed. I think a valve can fully swing in about 30s. Is been my experience the control and stability from a mixing valve is 100% ie if your system is designed to run on 110F water it will always be 109-111F so long as your supply water is 110F or higher. However you may have me at more efficient and its the buffer tank that's key.

I have not seen a buffer tank set up. But with a typical system you have your floor loop circ plus a main loop circ unless your plumbing is in the same room as your storage. So the two circ pumps are both needed all the time to add heat to the system. With the buffer tank you could have your injection coming straight from storage and depending on load and temperatures it might only have to run 10% of the time.

I would think however with a buffer tank to gain that effecency you'd have a greater hysteresis (tempeater swing) on your infloor supply say in the case of a 110F setpoint getting 105-115F but it's infloor and I doubt you'd notice the difference.

Also this has really peaked my interest as I'm about to design my system this weekend.

Sent from my moto g(6) play using Tapatalk

For high mass low temp loads, I agree that fast response is not of much benefit. As for stability, at low flow rates, unless the valve is sized exactly right (and a great percentage are not), you get a hunt and seek at high supply temps from storage. When supplying heat from storage, your hot supply flow varies greatly. Valves are probably better if you have a stable temp in your supply of hot water but when heating from storage, the temp of your supply of hot water from storage can vary from maybe 190F down to your lowest usable temp, maybe 100F. Sizing your valve Cv for 100F supply from storage means you might be moving 12-16gpm for a decent sized system. At 190F, you might be moving .5-1gpm. Hard for the valve to maintain authority at those low flow rates and it will tend to hunt and seek. With outdoor reset, your required supply temps and your supply temps from storage will both be varying. With injection mixing on outdoor reset, you are simply adding enough hot water to the floor loops to make up for the heat loss- like cruise control in a car.

 

[TCaldwell]

Storage is a asset for mixing control regardless of valve or circ. My take on this comes from a data trend chart aspect not recognized comfort.
The circ way is less moving parts/ hardware, easier to plumb, I can’t overlook losing the first 20 percent control to the mechanics of a cap start circ, this will cause hunting with low loads combined with a high storage temp. Remember the control algorithm that determines the output is non adjustable in the controller but often also has odr incorporated.
The valve, more parts, more plumbing, poor results if not sized properly, definitely more math to get it correct. Most valve controllers are adjustable to match the stroke time of the valve, most valves have programmable stroke times, allowing some flexibility. You get a reliable 0-100 output through a sized valve, with full flow at all times.
Both methods obviously do work, but from a data collecting view the valve is more precise at added expense.
I will say as far as control methods I was surprised how well the tekmar floating point system with a belimo characterized control valve worked.

 

[kuribo]

Storage is a asset for mixing control regardless of valve or circ. My take on this comes from a data trend chart aspect not recognized comfort.
The circ way is less moving parts/ hardware, easier to plumb, I can’t overlook losing the first 20 percent control to the mechanics of a cap start circ, this will cause hunting with low loads combined with a high storage temp. Remember the control algorithm that determines the output is non adjustable in the controller but often also has odr incorporated.
The valve, more parts, more plumbing, poor results if not sized properly, definitely more math to get it correct. Most valve controllers are adjustable to match the stroke time of the valve, most valves have programmable stroke times, allowing some flexibility. You get a reliable 0-100 output through a sized valve, with full flow at all times.
Both methods obviously do work, but from a data collecting view the valve is more precise at added expense.
I will say as far as control methods I was surprised how well the tekmar floating point system with a belimo characterized control valve worked.

No doubt both methods can work. A flow balancing valve properly set in the injection circuit should allow the pump to work as designed. The devil is always in the details.

 

[woodboilerwonder]

No doubt both methods can work. A flow balancing valve properly set in the injection circuit should allow the pump to work as designed. The devil is always in the details.

You could also use 2 circulation pumps. A medium and a small. First call is small pump, second call is small one goes offend medium goes on or high call is both small and medium.

 

[kuribo]

You could also use 2 circulation pumps. A medium and a small. First call is small pump, second call is small one goes offend medium goes on or high call is both small and medium.

Mr. Siegenthaler proposed such a staged injection pump system but when I asked him how such a system would be controlled, he said there was no commercially available controller and it would have to be entirely home brewed. Though the idea is a good one, it isn't really practical in my situation without something commercially available to control it.