Injection loop piped backwards

Solid_Fuel_Man

Working at a large church built in 1996 radiant throughout. After replacing the third tekmar first one was in 2004 I noticed at commissioning that the injection Loop was at 100% and not meeting Target temperature even on a mild day. Further study of the piping revealed what appeared to be secondary Loop flowing backwards in relation to the injection point. It's been this way for 20 years am I missing something I've never pipe them that way seems counterproductive. My thinking is that this short circuit could be burning up the injection controllers for working too hard to maintain target.

Taylor

Solid_Fuel_Man

This place is in the early days or radiant slabs, way over controlled and micro zoned to the max. There are 3 low temperature injection loops and 3 fan coils in air handlers, 2 Smith boilers. The mechanical room is a cluster, so this didn't jump out at first.

Paul S_3

I would switch the injection pump....its piped improperly....i think your on the right path

Solid_Fuel_Man

How has this even come close to working for the last 20 years? That's my real question. I'd think this would have a significant effect on the life of the injection controllers at nearly 100% duty most of the time. And all three loops backwards? It had 370 controllers which I replaced with 402's with room feedback thermostats. The 370's never seemed to last more than 10 years.

Taylor

Ironman

Paul S said:

I would switch the injection pump....its piped improperly....i think your on the right path

Paul, I don't think that will fix it.

A simple way to visualize it is to look at Tekmar's drawing and erase the common piping between the Tees on both loops. What you'll see is 3 circs in series in the supply lines and 3 returns in series. Now, do the same to Solid Fuel Man's drawing and you'll see the piping error. It can be confusing, but this makes it clear.

The injection bridges need to crisscross to make his drawing right.

Paul S_3

You are right Bob.....i over looked that

Solid_Fuel_Man

This is actually what I was thinking. I will need to recipe the injection loops. Or poverty the secondary loop circulator to the other side of the secondary loops. But there are many zone valves and actuators on the secondary. Will have to do more investigation if flow direction will effect their operation.

Taylor

Paul48

It doesn't matter how that segment flows. You have hydraulic separation. What goes into a tee, must come out of a tee. What winds up at the emitters is a function of a difference in flow between each segment. You have another problem if it won't come up to temp.

Ironman

Solid_Fuel_Man said:

This is actually what I was thinking. I will need to recipe the injection loops. Or poverty the secondary loop circulator to the other side of the secondary loops. But there are many zone valves and actuators on the secondary. Will have to do more investigation if flow direction will effect their operation.



Taylor

It wil effect it becaus the injection circ is pushing against the loop circ.

This is the easiest fix:

Solid_Fuel_Man

That's what I'm going to propose. I doubt they will go for it "since it's worked fine" for 20 years. I wonder if the original drawings were like that.

Taylot

Paul48

I'm not trying to be argumentative.......Hydraulic separation= It doesn't even know it's there. The best way to fix it, is to eliminate the secondary (tertiary?) circs, at the emitters. Eliminate the closely spaced tees on the emitter end. Run 2 pumps, primary/secondary and you have made your job a whole lot easier.

Ironman

Solid_Fuel_Man said:

That's what I'm going to propose. I doubt they will go for it "since it's worked fine" for 20 years. I wonder if the original drawings were like that.



Taylot

Show them the Tekmar drawing and explain that it's effecting SYSTEM efficiency.

Ironman

Paul48 said:

I'm not trying to be argumentative.......Hydraulic separation= It doesn't even know it's there. The best way to fix it, is to eliminate the secondary (tertiary?) circs, at the emitters. Eliminate the closely spaced tees on the emitter end. Run 2 pumps, primary/secondary and you have made your job a whole lot easier.

That would eliminate ODR control of the the SWT in the secondary circuit and send boiler water temp directly to the floor. Do you really think it's a good idea to send 180* water to a radiant floor?

Or, are you saying let the injection circ pump the floor directly? That would result in flow being greatly reduced to the floor loop as the pump modulates. That would cause the first portion of the floor to be heated, but not the remainder of the loop.

I'm not trying to be argumentative either, but i don't understand what you're proposing and it contradicts Tekmar's drawing and instructions.

Paul48

No.......Let's just stay on the P/S aspect. What you've achieved by criss-crossing the supply and return, gets you back to the same thing. It's a closed loop. What will be supplied to the emitters is just a function of the difference in flow between the 2 circs. There may, or may not be a reversal of flow between the closely spaced tees, based on a difference in flow. That's true of any P/S circuit. From a practical matter, the direction of flow, as drawn shouldn't make any difference, but I'm just an enthusiast, and not a pro. I may be completely confused as to my understanding of P/S.

Ironman

Have you every done variable speed injection mixing?

I think you're only seeing the hydraulics and not the function of the controls that's being effected.

Paul48

No, but I'm always eager to learn. Can you explain how pumping it "backwards" adversely affects the controls or the system, in general?

VDBLU

Wouldn't reversing the boiler pumps correct the problem? That would just be a matter of unbolting the pumps and reinstalling them.

ratio

What's happening is that we're injecting heat into the radiant system & taking a portion back out a few inches later. The loop is effectively getting less heat—not water—than we're injecting into it. Reverse the taps at the radiant side.

Hydraulic separation is there, but follow the BTUs.

VDBLU

ratio said:

What's happening is that we're injecting heat into the radiant system & taking a portion back out a few inches later. The loop is effectively getting less heat—not water—than we're injecting into it. Reverse the taps at the radiant side.

Hydraulic separation is there, but follow the BTUs.

I see that with the OP's diagram. And Ironman's cross-over fixes it but if the flow of the boiler loop was reversed it would fix the BTU flow as well. I guess that would cause the in and out taps on the boiler to be used wrong though. Hadn't thought of that before. How do you keep the second boiler from heating the output of the first?

ratio

I think I'm off somewhere. The original piping will result in higher temp water coming back through the injection–return pipe than if the tees were in the right order—not all of the BTUs injected make the trip through the emitter, some come right back to the primary loop. The primary loop return temp will be higher than it could be. The boiler will hit high limit sooner, & cycle shorter? What are the flow rates, temps, etc?

In any case, I've got a hard time imagining that this is causing your pump controller to fail. It should be able to operate at 100% indefinitely. Have you verified the injection pump meets the specs for that output? Is the controller mounted in a hot hot boiler room? Low cooling airflow available to the controller? The hotter it runs, the shorter it'll last. & of course, sometimes engineers are ...optimistic... about their specs.

Paul48

"Hydraulic separation is there, but follow the BTUs."

OK.......The primary circ is pumping 10 gpm. Each of the secondary circs (between the tees) is pumping 5 gpm. Each of the secondary circs (emitter circs) is pumping 5 gpm. You will move all the btus from the boiler to the emitters regardless of which way the flow is between the tees. It's no different than if there was a buffer tank there, instead of the tees.

Solid_Fuel_Man

Yes, the ONLY fix I can see is replying the injection loop. I can't reverse any circulators do to flow issues from the boiler or on the secondary side from flow issues from zone valves in that system. As far as I'm aware you cannot make an injection loop with any fewer than three circulators. One on the boiler Loop, 1 for injection, and one on the secondary Loop. I was just wondering if anyone else has seen any of these reversed injection loops on early injection systems was it ever spec'd that way?

Taylor

ratio

Oh, yah, IDK which way the flow between the tees is, doesn't matter. Isn't injection pumping a form of mixing? Let's say we meet the heatloss exactly with an emitter AWT of 100°, 20° ∆T, 10 GPM. Boiler AWT of 180. How much do we inject? (I'd attempt the math, but I'd embarrass myself.) What would the return–to–the–boiler temp of the injection connection be with the injection preceding the return? Should be about 110° right, due to our emitter AWT? So we'd be returning that temp to the primary loop at whatever flow we inject at. If the return–to–the–boiler connection precedes the injection port, we must be returning about 90° water to the primary loop, at whatever GPM.

I still don't see how that would effect the emitter loop, as the AWT will be the same regardless (if the injection pump can keep up), since the secondary pump hasn't changed. Unless I've simplified something important out. Maybe if the secondary loop & injection loop are closely matched, but why would anyone do that? Maybe they are/the injection pump is undersized for the amount of BTUs needed for the emitter. That would explain why it's running at 100%.

IDK, in just a washed up electrician.

Paul48

VARIABLE SPEED INJECTION A standard wet rotor circulator is connected to the 356 at the back of the control. The 356 increases or decreases the power output to the circulator when there is a mixing demand. The circulator speed varies to maintain the correct mixed supply water temperature at the mix sensor. For correct sizing and piping of the variable speed injection driven circulator, refer to essay E 021. A visual indication of the current variable speed output is displayed in the LCD in the form of a horizontal bar graph.

Based on the diagram above, said mix sensor should live on the output side of the emitter circ. The numbers I used were just an attempt to show that it is possible to move the btus regardless of the flow direction.

hot_rod

I agree with Paul, the flow in and out of a tee. Tekmar use to show reverse injection and they still show a two pump injection method

http://tekmarcontrols.com/support/tekmarglossary/33-support/glossary/194-on-off-injection-output.html

hot_rod

http://www.manualsdir.com/manuals/594652/tekmar-365-mixing-control.html?page=4

ratio

Thanks, @hot rod. That's just what I wanted to look at.

Ironman

Well, that's interesting.

I finished up a radiant job a couple of months ago where in my haste I had gotten the injection bridge piping reversed like in Taylor's drawing. I talked with a Tekmar engineer about and sent him a diagram. He made it very clear that it had to be re-piped or it would not function properly.

@hot rod
It appears that the drawing you posted is a very old one and I've never seen anything but the drawing which Paul S posted included with any Tekmar controls. Could it be that Tekmar found out long ago that there were issues with the bridge being reversed and abandoned that idea?

hot_rod

It does appear to be a 2000 vintage tech sheet. But as Paul 48 mentioned it is about flow and the mixing of varying flow rates. Same happens in hydroseps and it changes boiler feturn temperatures

Good article about that in HPAC magazine last month

Paul48

@Ironman

For that engineer to tell you anything other than what was printed in the install instructions, would have required him to do the nasty four-letter word........WORK. I know it's not correct, I'm just questioning it being the source of a problem. From what I've learned about P/S, it shouldn't be.
The easiest way to settle it is to re- pipe/pump it to the instructions, and see what happens.

Solid_Fuel_Man

That's what I'm going for. Meeting with the head trustee tomorrow if schedule allows.

Taylor

Ironman

@Paul48,
I'm not disagreeing on the flow part of the equation. What I'm referring to is the other part: namely, the temperature. By short circuiting some of the flow backwards, the injection control would have to increase the pump speed to get more btus injected into the secondary. At the very least some more electricity is used and that MAY effect the life span of the control?

My mind is too tired right now to entertain what effect it would have on the boiler.

Paul48

We'll call it "done", because we can't seem to agree that the flow part of the equation translates into the temperature. There's no sense in "beating a dead horse".