Incorrect flow in secondary

radcos

Hi - I had a contractor come out to convert my hydronic piping near the boiler to a “true” primary/secondary system using closely spaced tees for hot and return water to/from the secondaries. I have a 200K BTU non-condensing cast-iron boiler with a 1.25 inch pipe as the primary loop – the primary circ is a Grundfos 15-58 set on high speed.
I have two secondaries that are also run with 1.25 inch pipe – one supplies 3 zones of tubing embedded in concrete and the other supplies 5 staple-up zones. Motorized three way mixing valves were installed for both. The zones are pumped by Grundfos Alpha2 15-55 circulators set on the middle constant pressure curve.
I am having an issue with flow reversing in one of the mixing valves (for the concrete zone). It appears that when the valve is set to mix a certain amount of hot water in with the returning cold water (more than about 20%) the hot water is flowing out of the valve where the cold water should be entering and flowing back to the primary loop along with returning cold water. I can tell this is happening both because the temperature of the secondary spikes and because the piping into the cold side becomes hot.
When I contacted the contractor he blamed it (without coming out) on the Danfoss valve. (Which I picked out but that’s another story.) I don’t believe anything is wrong with the brand new valve and I’ve triple checked to see that the valve stem is oriented correctly. What I see that may be the issue is that he did not make any allowance for a run of straight pipe after the closely-spaced tee for the zone that has an issue. The piping prior to the tees may also be a problem.
About 2” prior to the first closely spaced tee there is an air separator and expansion tank mounted in a long run of straight pipe that begins at a 45 elbow near the boiler. After the second closely spaced tee there is a 90 turn with 1” of straight pipe before the elbow turns (so about 1” of pipe is exposed between the tee and the elbow fittings).

My questions are: Is there anything else to look at as a cause other than the lack of proper amounts of straight pipe prior to and after the tees? This would be 8 diameters (10”) prior to the first tee and 4 diameters (5”) of straight pipe after the second tee. (Got this from the Siegenthaler book.) Is there any chance that only installing the 5” of straight after the second tee will resolve the issue or will I also have to move the expansion tank/air separator that is teed-in towards the end of the long (>10”) run of straight pipe prior to the first tee?
I’ve tried altering the speeds on the primary and secondary circulator as a work-around. That didn’t work. Keeping the valve positioned to admit little hot water to the secondary (less than about 20%) seems to work but I don’t want to leave it there permanently.
The contractor was irritated with the job in the first place because he didn’t take care to install one of the valves in the correct orientation and now he’s unhappy that I am suggesting that it is his piping at fault (and actually says that it doesn’t make any sense) so looking for some advice.

Thanks in advance.

Tinman

A sketch/piping diagram of what you have there would paint a much clearer picture.

GroundUp

Is the circ on that zone pulling water through the mixing valve or is it pumping into the valve?

hot_rod

Is it a motorized or thermostatic mix valve

radcos

GroundUp - the circ is pulling water through the mixing valve.

hot rod - It is a motorized mixing valve.

I'll work on a diagram to show the entire system.

radcos

Here's a diagram of the system. I left out the drain/fill parts. I know that the DHW coming back into the return with the concrete zones is not optimal but it runs on priority so they do not both flow at once. The other sort of odd thing is that for the staple up zones the cold return from the zones tees in before the hot supply to the mixing valve.

hot_rod

I haven't seen an indirect tie in like that before, sort of a parallel loop. Does the boiler circuit shut down on indirect call? Looks like the indirect return could flow into the 3 way?

I wonder why the indirect isn't either a P/S like then other loads, or a true parallel loop? Seems odd to have a circ and zone valve.

Primary pumps seem to be pumping at the PONPC.

The left 3 way seems to be crossed on the closely spaced tees, hot set should be on the left tee, return on right. Do those 3 way close off completely?

Tinman

This looks like a classic pipe as you go job. If the installer starts with a proper schematic based on proven methods, any errors will likely be identified before a tool is in hand.

radcos

When I moved in the system was a real mess without a primary loop at all. It was sending almost whatever the boiler was making through all zones (even though it had thermostatic valves - hard to explain here) so getting it to something "proven" has been a work in process. The indirect was just another zone when I moved in but the guy that fixed that lack of a primary loop put it on its on circ. It could flow into the 3 way but since the controller gives DHW priority and shuts down the rest of the system I don't think it matters (much) - the zone valves for everything else will be closed.

The left 3-way is operating correctly - I probably should have drawn that differently to avoid confusion since it is mounted the opposite way from the right three way. I'm not at all concerned about that secondary (the staple up) - it works great.

The 3-way valves do close all of the way.

Also, I noticed today, that the flow of hot water out of the cold inlet side of the concrete zones valve does not always happen at a given valve setting (> about 20%) - it did not happen when multiple zones were calling and I let the controller move the 3-way valve.

radcos

And, the expansion tank/air separator was moved to the hot side of the boiler by the most recent contractor.

hot_rod

So one mixer works properly, then other doesn't? What brand 3 way, got a pic of how it operates, ports and vane inside? Is it like the cutaway I posted above?

radcos

Hi Bob - The mixing valves are Danfoss/ESBE 130 series.

The valve is installed as shown in the 2b illustration so essentially operates the same as the one you posted.

What is happening is that under certain circumstances (zone 2 calling and the mixing valve set for more than about 20% hot water) the hot water that enters the port marked with a triangle is flowing out of the cold inlet port (marked with a circle). I know this is happening because the pipe going to the cold inlet port gets really hot really quickly and the temperature coming out of the mix port (square) spikes.

I have double checked that the valve is installed in the correct orientation with the "rotating spool" (vane?) in the correct place and that the knob is installed correctly.

hot_rod

Athens circulator is pullimn away from the mixed port square label? Sounds like the valve is operating backwards, did you observe the cane position when you installed? does that vane rotate 360. Sounds like the cold port is opening when the hot should be

radcos

Yeah - I know it sounds like it is operating backwards but I have triple-checked it and I'm as sure as I can be that it isn't. Here's why:
1) There is a flat spot on the stem that indicates where the vane is. I've made sure that it is in the correct position.
2) The vane position is the same as on the other mixer where I don't have a problem.
3) I've turned it all the way hot and cold and it works in those positions even when it doesn't work at the 20% hot position.
4) The problem only happens sometimes. I have seen it when zone 2 is calling and the control is in the (approx) 20% hot position but not when zone 3 is calling or when both are calling. When the problem is not occurring the motorized valve works like a charm and keeps the temp right at the target - when the problem happens the temp in the secondary bounces around like crazy. (Secondary gets too hot then the valve shuts all the way and it gets cold. Then valve opens a bit and gets way hot again.)

I appreciate you trying to rule out the simple causes but I can't find anything obvious so I think it is some type of flow problem - that's why I am looking at the piping. I can't find anything else that would cause it so that's why I am wondering if the lack of straight piping after the tees is causing a problem. I have read that is is for "hydraulic separation" so that flow in the primary doesn't cause (induce) flow in the secondary. So could the flow in the primary be pulling the hot water back through the second tee (along with the cold water from the zone)?

hot_rod

A check would be a way to “cheat” it away😳

radcos

Help me out - what do you mean by "cheat" it away?

hot_rod

You have what seem to be some weird hydraulics going on. I wonder about the boiler circ pumping at the expansion tank, lowering the pressure upstream causing some reverse flow when a combination of mixing flows are going on. Moving the expansion tank connection to the inlet of the boiler circ may change things, fairly simple to do.

Assuming the mixing valve is assembled and working correctly?

Adding a check valve would eliminate the reverse flow, but treat the symptom, not fix the actual cause.

Sometimes you need to just observe, measure and get inside the piping in your minds eye to sort these out.

Attached is an example of how expansion tank placement, PONPC changes the hydraulics and pressures in a system.

radcos

Thanks, Bob. I'll take a look at what you sent. I'll play with it a bit and close the expansion tank valve and see if anything changes.