Triangle Tube PTE110, hydronic boiler is short cycling.

murry0529

Hello, I was wondering if anyone has any ideas on what is going on with my boiler. I recently installed a Prestige combo boiler (M# PTE110) in my new construction home. I live in northern WI and although is has been a mild winter so far, I have been struggling to get our new building enclosed enough to fill the boiler/infloor heating system before the entire mass becomes a large ice cube. To give you a little background on the building itself, it is basically a floating slab. The slab's basic construction is: plastic vapor barrier, 2" horizontal insulation, 5/8" pex tubing (max. loop length 250'), approx. 16" of sand and 5" concrete (2" vertical insulation around entire perimeter). I would call it a medium mass heat sink. The thought process for the 16" of sand and shorter lengths of 5/8" pex tubing, was in case I ever decided to interconnect a solar hot water heating system with a D/C circulator to the existing system. I have 2 infloor radiant heat manifolds (2 zones) that are currently connected together just to get my system up and running and to a moderate temperature inside. The boiler is piped primary/secondary. My primary loop is extremley short and piping into the secondary loop with a Webstone 1" purge tee. On my secondary loop, from the supply side of the purge T is a spirovent microbubble and a Taco Bumblebee, set up for 20 f delta T operation. I filled the system this weekend and the slab was cold, very cold. After a very long day of attempting to purge and fill the system with straight water, I fired the boiler. Two days later, my thermostat is still calling for heat (and I expect it to keep calling for quite some yet), I have purged air several times from the system, my secondary circulator (the Taco Bumblebee) is showing approx. 14.0 GPM flow and my infloor (secondary loop) supply temp is approx. 114-120 f.

Here is my problem, my boiler seems to be short cycling and I am not sure why. My outdoor sensor is installed, boiler fires, runs for maybe 1-2 minutes max. and the burner shuts off (CH setpoint reached). The main (primary circulator) continues to run until the temperature inside the boiler drops and the boiler fires again. I seem to be running a maximum of 6 f delta T across the primary loop side of the bolier, most of the time it is less. The 3-speed factory installed (primary loop) Grundfos circulator is currently set at low speed in hopes of increasing the delta T on the primary loop. The delta T on my secondary loop is massive. The last time I looked it was at about 66 f. I guess I would expect the delta T on the secondary side to be large until I am able to bring the large, cold mass up to temperature. It just seems as though my boiler (primary loop) is not sending much heat to the secondary loop. Maybe I am overthinking things due to lack of sleep. Does anyone have any ideas on what is going on? Thank you.

CMadatMe

Pictures. With the flow rate #s you're supplying on the secondary side you should be pulling all the flow from the boiler supply side out and mixing with about 8-9gpm of secondary return water to get 14gpm that you state the Bumble Bee is giving.

That means the boiler return should be seeing all that ice cold water and shouldn't be reaching high limit.

But the Bumblee at 14gpm can only overcome 2' of head so it's not pulling anything out of the primary side. Bumble Bee is the wrong pump is where my money is..

Rich_49

You are not overthinking things . The boiler is seeing high limit reached because of the narrow Delta . Try changing the Delta setting on the Bumble Bee to 10* which is where it should be for a radiant floor anyway . That will increase the flow on the system side twofold and possibly the system side flow will exceed boiler flow and the short cycling will lessen .
Once the slab is up to temp however , it will begin again and exist forevermore . Might wanna think about a buffer at this point if you want any chance to have the boiler operate the way it was supposed to and you assumed it would .

Use a 2 pipe buffer config with the load between the boiler and tank .

Zman

It sounds like you have low flow on the radiant side (the bumble bee may be lying) or the piping on the primary/secondary is not right.
If you put the bumble bee on high speed does anything change? Do you hear water moving?
Unless you have dozens of loops, the bumble bee will never push that much water through 250' of 5/8" pex.
Any chance you are air locked or frozen?
Pictures of piping?

FranklinD

Where in Northern WI are you located? I too am a northern Wisconsinite (enjoying the mild winter myself...we just finally dipped into single digits a few nights ago).

I'm no professional, but I have met a few guys around my location (Twin Ports area) that are. Only one company that I would consider to be radiant experts, though, and they're busy as all heck most of the year. Mostly we have existing gravity conversions and baseboard around here.

Sounds like an interesting system, though...I haven't seen many like that around here. Seems like most new construction in my area is scorched air. You've definitely come to the right place for advice, though!

murry0529

Thanks for all the quick feedback everyone. I am kinda old school, so I am not even sure how to reply, so bare with me. When you post a question do you typically reply to each individual and if so, how do you do that? Sorry, I am not that familair with all of this. Jason

Chris:
In regards to your comments on "the boiler should not be reaching a high limit", I agree. The boiler should be getting slammed with low return water, but it appears to not be mixing/mixing very little. On my infloor design, my total system (building) requirements should be 5.1GPM @ 1.3' (20f delta T) or 10.2 GPM @ 4.3' (10f delta T). As far as the Bumblebee is concerned, my guess is that it is trying to screem the water through the system to achieve the 20f delta T, which it cannot do because of the ice cold slab. According to the Taco pump curve, the Bubblebee should easily be able to handle the design/install (under normal conditions that is).

I will try to get pictures tommorrow. My piping is very minimal and kinda temporary, as this system will eventually serve as my backup system to a 3-ton water to water geo system connected to off-peak, a buffer tank and most likely an HBX control. My primary loop is literally about 16" of 1" straight pipe and two 90's into a 1" webstone purge/fill.

Rich:
My guess is that changing the delta T before the system is up to a normal operating temperature is a waste of time wouldn't you think? I'm running over 60 f right now and not by choice.
A buffer tank is in the near future, but won't be installed for a month or so yet. Coming with my geo system when I am ready for it.

Zman:
Trust me, I have lost a lot of sleep this year so far worrying about speeding up the building process, so I could install and fill this system sooner. Didn't happen though. I attempted to fill my system mid morning on Saturday. I had minimal water flow through most loops and zero flow through a few. My first manifold is 6 loops and my second manifold is 4 loops. I figured that due to the large mass of sand around the tubing and the depth that it is buried, that if I could get 1 loop per manifold with flow, that I could eventually thaw any remaining. As far as being airlocked, I think I have purged and bleed air about 5 or 6 times already. It is pretty hard to force fill and bleed air from a system that is partially frozen. As of this morning (Monday), I am holding consistant pressure (no fastfill or backflow connected yet!!), maintaining a consistant secondary loop supply temperature and very slowing increasing room temperature (45 f). Boiler is still just on and off, on and off.

Zman

The bumble bee is is not pushing 14 gpm. On it's best day it could move about 10 gpm through your system. At a delta t of 66, That would mean 330,000btu/hr being moved. That is obviously not the case so disregard the bumble bee.

I would try putting the bumble bee on high and the boiler at 180 degrees.

Are you able to isolate the loops and purge them one at a time?

I probably should not ask, since you are having trouble sleeping.
Is the pex tubing at the bottom of 5" concrete and 16" of sand?

The way you answered more than one question was the best way to do it.

murry0529

Zman,
This is my first experience with the Bumbulebee circulator, but I am pretty sure I can force it to a high speed and bypass delta T mode.

I can get the boiler to a higher temperature easy enough by choose a different reset curve and yes, I can isolate each loop individually if needed. What are you thinking?

My tubing is attached directly to the top of 2" foam, then 16" sand, then concrete. Not ideal in my opinion, I would have liked to install the tubing part way into the sand bed, but what a nightmare that would have been!

hot_rod

Ah, the old sandbed radiant concept again. Dry sand as you know is a very poor conductor, not the best for moving heat, certainly not quickly.

The sandbed concept was intended to be a solar flywheel where it has months of summer and fall solar input to slowly ramp up for winter heat, then slowly un-wind throughout the winter.

Zman

Assuming the primary/secondary is set up right, You must have a flow issue on the secondary side. It could be frozen or it could be airlocked. In either event, more pressure and higher temps will help the problem. There is no other explanation for a 66 degree delta over a long period of time

Normally when you purge the system you would go one loop at a time until no air comes out. Purging tubes that are in a sub freezing environment is scary. The only safe way is with glycol. You might get away with hot water.

16" of dry sand has a similar r-value to the 2" ridged insulation under the tubes, It is going to be difficult to get the heat out. I don't think that is your problem at the moment.

bmwpowere36m3

How are the primary and secondary loops "tied together"? Closely-spaced tees or a prefab fitting? PEX loops shouldn't be that difficult to bleed… all the loops are on a manifold correct? Can each loop be isolated, i.e., does the manifold have flow setters or shut-off or zone actuator fittings?

Pictures of your system would be helpful.

I'm thinking either the system side is air-bound, with no or little flow OR the boiler side is flowing way more than the system OR the primary/secondary connection isn't right and not allowing proper mixing.

1GPM thru 250' 5/8" PEX is about 3.3' of head… not sure how many loops total you have (for a system total of 5-10 GPM). That's not including manifold/piping/boiler head loss. So I doubt the bumblebee is actually pumping 14 GPM.

Zman

bmwpowere36m3 said:

How are the primary and secondary loops "tied together"? Closely-spaced tees or a prefab fitting? PEX loops shouldn't be that difficult to bleed… all the loops are on a manifold correct? Can each loop be isolated, i.e., does the manifold have flow setters or shut-off or zone actuator fittings?

Pictures of your system would be helpful.

I'm thinking either the system side is air-bound, with no or little flow OR the boiler side is flowing way more than the system OR the primary/secondary connection isn't right and not allowing proper mixing.

1GPM thru 250' 5/8" PEX is about 3.3' of head… not sure how many loops total you have (for a system total of 5-10 GPM). That's not including manifold/piping/boiler head loss. So I doubt the bumblebee is actually pumping 14 GPM.

This has all been covered in the above posts...

bmwpowere36m3

I'm not seeing answers or pictures to those questions…

CMadatMe

Open up all the radiant and run the Bee at Fixed Speed #4 and don't worry about the delta-t right now. That'll give you 10gpm @6' head you may pump 11 but that's about it. Figure out the flow problem 1st.

Now you should pull all the flow rate out of the boiler/primary side into the secondary and that boiler return should see all that nice cold water. Take baby steps..

Rich_49

Rich:
My guess is that changing the delta T before the system is up to a normal operating temperature is a waste of time wouldn't you think? I'm running over 60 f right now and not by choice.
A buffer tank is in the near future, but won't be installed for a month or so yet. Coming with my geo system when I am ready for it.
@murry0529
getting the system side flow greater than the boiler loop flow is not a waste of time . Try closing the valve between the 2 boiler fittings on the purge Tee and run it direct with the Grundfos on high speed and remove the Bumble bee temporarily . Raise the water temp also , this will lower the head of the circuit bit by bit as it runs and the Delta Narrows .

As Hot Rod pointed out , sand is a terrible conductor when dry , wet is another story . The other problem is you are suffering immense downward losses . 2" of foam as someone else also pointed out has a similar R value to 16" of dry sand , problem is that there is another .5 R value on top of that sand making more Resistance above the tubing than below . Is the earth below that insulation wet or even damp ? If so you will never get heat up and out of the slab since the moist soil below will grab those BTUs and move them away . That's why we account for a water table within 6' of our slab installs and also why we insure that we have less R above then below .

maybe trying to get this slab to temp is more a waste of time than lowering the Delta on the Bee .

Gordy

What centers are the tubing laid at?? This will be running for weeks. 21" of cold wet mass. If the sand is dry it gets worse.

murry0529

Pictures of install

murry0529

Spent a little more time looking things over tonight. The manifolds are a Roth manifold. I tried to force flow thru each loop, but my booster pump is not in the greatest of shape. It seems as though I have flow through 6 of the 10 loops. If the flow indicators are somewhat accurate, each loop that seems to be flowing is showing approx. .5 GPM each. As of tonight I am still holding pressure, my secondary loop supply line is warm (I believe it was approx. 110-112f) and I am low 40's for indoor temp. Boiler has cycled 700 times since Saturday night and a total run time of 8 hours. I must have something piped wrong.

Zman

The bumble bee may be able to purge the lines. Try turning all but one off for a few minutes with the bee still on high.See if you get air. I would not isolate the lines for long if freezing is still danger.
The bumble bee should be able to push about 1.75 gpm with one loop at a time open.
Your piping is not going to win awards, but should not be causing the problem.

CMadatMe

.5GPM x 6 = 3GPM or 15,000 Btu/hr on a 10 System Delta-T

Minimum Boiler firing rate is 30,000 Btu/hr

Supply 112 - 10 = 3GPM of 102 Degree System Return Water Temp

Boiler Flow I'm guessing off the curve 5.8gpm
Boiler Supply Temp based off your System Supply is 112 Degrees

5.8 -3 = 2.8GPM of 112 Degree Water Back to Boiler Return

(2.8x 112) + (3*102)/5.8 = 106 Degree Boiler Return Water Temp. You're not pulling anything out of the boiler, thus hitting the curve set point so you short cycle. Load is 50% less the minimum boiler firing rate.

Now I'm making lots of assumptions. I don't know what you have the curve set at nor the pump speed nor your actual system return water temp.

Gordy

All of this is a mute point if lines are not purged, and flow established in each loop.

bmwpowere36m3

Those Roth manifolds look different than mine, but to purge each loop start with the return manifold.

• Close all the loop returns on the manifold, but 1. Those white caps should be threaded (they are on my Roth HK). When you tighten/thread them in they close off each loop by depressing a pin/plunger.
  
• Close the ball valve on the return manifold (the blue one).
  
• Open the drain/bleeder on the other end of the return manifold (which I assume is part of the cross-shaped contraption on the end).
  
• Engage your auto-fill/regulator (quick fill if you have the option) and monitor the pressure so you don't blow the boiler relief valve.
  
• Purge at least 3-4 gallons (what a 250' loop should hold) and till the water runs without air.
  
• Disengage quick-fill and close loop return (white cap) and drain/bleeder on return manifold.
  

Now repeat for all the other loops. Then open all the valves/loops and set the pump to its highest speed and let it run for a bit. Verify on the manifold that there is "some" flow thru each loop based on the flow-meter.

As @Rich mentioned, closing the valve between the Webstone tee, setting the boiler pump to its highest-speed and disabling the bumblebee (disconnect power) is a good way to direct-flow... that way the boiler should see ALL that cold water.

murry0529

Zman,
I am going to see if I can get my hands on a better pump from work to purge air, maybe some antifreeze as well (which I was trying to avoid). I will try to isolate each line with the Bumblebee tonight after work as well per your suggestion. Yes, my piping is not going to win awards for appearance, but everything between the boiler unions to the manifolds is temporary, it just has to function. I am not sure if you can tell by the pictures or not, but the only insulation in my walls is a 1" layer of spray foam. Until the walls are fully insulated (2 weeks out), no more plywood on the walls. Unless I want to tick off my insulator (who is a good friend of mine).

Zman

I see you don't have the system hooked up to the water supply yet. Do you have pressurized water in the house. Purging would be easier with more pressure.
What model purge pump are you using?

Regardless of what some are suggesting, Until you have flow through all the loops, you cannot effectively trouble shoot the rest of this.

murry0529

Zman,
Yes, the system is not connected to water supply yet. My constant pressure well system was just installed and the water quality in my area is known to be fairly acidic. I have filled everthing so far with distilled water using a Liberty 331 pump which is what we have at our shop. It is a pretty old pump that has seen it's better days. I was trying to avoid using the well water, but that is probably not possible at this point. If I am going to use the well water I could connect a hose to my domestic water inlet on my indirect water heater, crank up the mix valve and try to purge the remaining lines with hot water??

Rich_49

Caleffi and Fernox both make a quality unit made just for this operation . Keep the trash water out of the system . Look into investing in the flush cart from either manufacturer .

Use the right tools and personel for the operation to avoid frustration .

Zman

That is a pretty powerful little pump.
It looks like by the picture that you are purging backwards though the system.
Many balancing manifolds don't like reverse flow. If you have and check valves in the system (circulator checks) they wont like it either.
Is there any reason not to purge in the direction of flow?

murry0529

@Rich We might have our Geo flush cart somewhere in the shop, but that might be a litle overkill.

@Zman It's probably hard to tell from the pictures, but I have a Webstone purge/fill valve installed right before the microbubble and have been purging in the direction of flow. I also don't have the check valve installed in the Bumblebee, so I should be able to purge either direction.

Rich_49

Murry , you cannot overkill initial purge . This discussion kinda verifies that point huh ?