Radiant Heat Loop Not Working #2

Ralphxyz

At 27# that is 62.28':


So it does appear the circulator is too big!

Ralph

HomerJSmith

I've been thinking, a first for me.

I don't know what the CV of the RTI thermostatic mixing valve is. I have tried to find out. I like Honeywell thermostatic mixing valves as they have a high CV, 3.2 . Higher is better.

The flow thru a infloor radiant sys should be in the range of 1-2 gal/min. I'm thinking PEX here. Perhaps, a balancing valve on the radiant return to the boiler from the pump set to 1-2 gal/min. The 15-42 at 2 gal/min is outputting 13' of head energy, WOW! Is it possible that you are shutting off the spool in the RTI mixing valve, preventing a mixed flow?

AGAIN, are you sure the radiant pump is working? and how do you know? And how does the radiant pump turn on and off?

At 27# that is 62.28'--what that all about. The pump curve at the rated flow tell you the head for that pump. There is static pressure and dynamic pressure. Don't add the two together. The pump curve is the rated dynamic pressure at a given flow.

"It is appearing that if I just add zone valves and isolate the loops would be the simplest thing to do plus move the tee if I can find my pipe wrenches." Listen-- I would first add a balancing valve and set it to 2 gal/min.

Look, for you to have flow, you must have a higher pressure and a lower pressure. If the pressures are equal, guess what--NO FLOW.
If the 15-42 is turning, as a pump creates a low pressure at the input and a higher pressure at the output, how come you have no flow.
Flow is the conveyor belt that move heat energy around the sys.

HomerJSmith

The fact that you don't have any flow (an assumption) and the pump is turning indicates to me that the RTI mixing valve is the problem. If the pump is turning, you have higher pressure at the input to the RTI on the cold side than the pressure at the RTI hot side. What that means, I don't know. Perhaps that's why the valve isn't regulating. When the boiler pump turns on, it increases the pressure to the hot side of the RTI and and lowers the pressure on the cold side of the RTI and the valve functions.
Maybe the pump is moving water, maybe just cold water thru the RTI.

Jolly Bodger

I think the flow control is holding you up. Try cranking it open and see if you get hot water to the supply side of the mixing valve.

HomerJSmith

If I was going to install a check valve, I would put a spring check on the output of the boiler pump and a swing check between the radiant return and the boiler pump return, on the horizontal pipe connecting the two.
The way it is set up now the boiler pump is going to push water thru the boiler and thru the radiant return pipe up to the RTI and to the radiant loop. If both pumps were working at the same time you would have both pumps pushing against each other. The one with the greatest head energy would win out in terms of flow. You could put the swing check anywhere on the return line as long as it is a horizontal pipe. Just a thought.

HomerJSmith

Boiler pumps that came with cast iron boilers were usually a Taco 007, occasionally a 010.

GroundUp

The 15-42 is a non-issue here. Also the static pressure plays absolutely no role in calculating the feet of head in the system. A 250 foot loop of 1/2" pex will flow approximately .6-.7 GPM at ~2-3 feet of head. So with 2 loops, you are doing roughly 1.3 GPM at 2-3 feet of head through the floor. The additional piping and equipment in the system will create additional head loss, maybe a total of 8-10 feet with a low cV mixing valve, but the 15-42 is of adequate size to be just fine with that. Honestly it's pretty tough to overpump a radiant slab, before ECM circs the 15-42 was one of the smallest you could get and there are thousands of them out there doing 2 loops like yours just fine. Given the missing wrenches issue, perhaps the other option of a 3 way zone valve and a bypass would be more fulfilling, albeit at a higher cost. This would mean (my recommendation) a 3 way zone valve after the existing 1-1/4" x 3/4" tee, to either bypass the radiation during a radiant-only heat call so the primary circ just sends heat to the tee for the secondary circ to grab, as it does now, and then return it to the boiler instead of sending it through the radiation OR if both zones call simultaneously, the bypass would close and direct flow to the radiation as it does now while the radiant can still pull as it does now. Does that make sense?

HomerJSmith

The head loss of 1/2" pex 250' long at 100F with pure water is 9'hd @ 1 g/m. The 15-42 at 1g/m is 14'hd. The same parameters @ .5 g/m, the head loss is 2.5' hd. The boiler hd loss is .5' hd. thru the HX. Piping with the exception of the flow check, negligible, I think. The only way to guarantee flow thru the pex is with a balancing valve, but the pump is close as it is @ 1 g/m. I only see one pex circuit.

Maybe the pump doesn't have enough head to push thru the flow check, I don't know. Do as Jolly Bodger says and lift the weight on the flow check and see what happens.

GroundUp

Homer, he has some photos earlier in the thread showing his manifold, with two loops. Piping, elbows, mixing valve, flow check, all will have some resistance but primarily the mixing valve. Regardless, 3ft or 13ft, the circ is a non-issue here. He can easily verify flow by checking temp on the return side of the loops, a balancing valve would be useless.

Ralphxyz

Now I am really disappointed:
From weird to weirder.

I have shut off any flow from the return to the mixing valve.
And guess what: No Heat!



I put globe valves ( Stop and Waste) on the mixing valve return feed. With the globe valves closed there still is no hot feed.

FACT:

With the house circulator on and the radiant loop circulator on
the loop heats as expected!

Possible the tee location "might" have a effect.

If I isolate the two loops, house and radiant, then it "should" work.

HomerJSmith, that is correct "Boiler pumps that came with cast iron boilers were usually a Taco 007, occasionally a 010."
However we are talking about the radiant loop not the boiler.

Also the two pumps do not push against each other they work in conjunction and flow together.

??Pushing through the Flow Check?? Where does that come in?
Interesting I can try adjusting the flow check.

Yes GroundUp, that makes sense, I do not see the need for a three way zone valve.

Thanks again, we will work this out!!

Ralph

hot_rod

Just to be clear on pressure readings. A gauge on a circulator discharge, supply side for example.

Turn off all circulators, what does the gauge read?

That number is static pressure, the pressure you put into the system when you filld it. Say 15 psi.

That pressure is to lift the water to the highest point in the building. About 1/2 psi to lift water 1 foot (.433 actually)
so if you want to lift 20' to the highest point X .5 = 10 psi fill pressure. 12- 15 psi is plenty for a two story building..

Now turn on the pump and read the same gauge. That reading is the dynamic pressure, the additional mechanical energy added by the spinning, centrifugal circulator.

Let's just say that same gauge jumps to 20 psi. If so you have added 5 psi. Plug 5 into your calculator to convert to feet of head.
5 X 2.31= 11.5' of head.

In a perfect hydronic world you would calculate all the piping in the circuit the circulator attaches to, and determine feet of head. Then use a pump curve to select a circulator. GPM and feet of head.

Jolly Bodger

@Ralphxyz With the mixing valve return line valve closed what is the gauge pressure on the radiant circulator suction? Only radiant circulator running.

With the mixing valve return line valve open what is the gauge pressure on the radiant circulator suction? Only radiant circulator running.

HomerJSmith

Yes, I do see two circuits on the manifold, GroundUp. I was looking at his drawing. How long are the circuits, each one, and it is 1/2" pex? The reason I ask is to kinda figure out the Hd loss in the sys. Now I'm kinda thinking that the sys loss is greater than the capability of the 15-42 Hd energy and when the boiler pump turns on it acts as a booster pump.
The reason for the RTI mixing valve is to limit the supply water to a temp compatible with the pex circuits. 160F down to 100F. The RTI may not work because there may not be enough pressure on the hot side to get flow.
Re-piping the supply of the RTI to the air scoop would certainly eliminate the Flo-Chek.
You are pumping into an expansion tank, you realize. I would move the Expansiion Tank to the input of the boiler pump.
I know that Ralphxyz wants a simple fix. I can't see why, if the pump has enough Head energy to overcome the pressure loss in the circuit at a given flow rate that it wouldn't work.
I agree that there isn't any hydrolic separation and the boiler pump is going to interact with the radiant pump and it does because the radiant works when the boiler pump works.
I would want to know the pressure loss in the radiant circuit from supply to the return, the whole circuit thru the boiler.
I would also entertain using a Taco i-Series set point mixing valve,controlled by the thermostat, instead of the RTI . It has a high CV, which would be a benefit.
How about a picture of the whole sys at about 8' away.

HomerJSmith

If the sys pressure loss is greater than the 15-42 can overcome, then a Grundfos 15-58 3speed might do the job. I would still ditch the RTI in favor of the i-Series 3way. More bucks, tho.

hot_rod

For the size of his system the thermostatic is plenty of valve. I suspect it is around a 3 Cv valve, and he has two loops, plenty of flow. I think he replaced or rebuilt it already?

Now if the 1/2 loops are 500 or 1000", that's a different problem

HomerJSmith

More like 1.9 CV, I think If it is anything like Danfoss's ESBE. Then switch to the 15-58 to get the flow. We already know it works with the boiler pump on. Tells me that the 15-42 ain't cutting it. The 15-58 on speed 3 will give him 19'hd @ 1g/m. One only knows when one adds up the pressure loss in the circuit what size the radiant pump needs to be.
I would still like a check valve on the output of the boiler pump. When the radiant pump is on and the boiler pump is off, some of the radiant pump's head energy is wasted on a reverse flow thru the boiler pump, I would think.
"For the size of his system the thermostatic is plenty of valve." I was thinking of the pressure loss thru the RTI. The i-Series has a CV of 13 or greater and it has a sensor to set the mixed flow that can change the mixed temp as the supply temp changes. I think the i-Series can maintain the mixed temp better than the RTI with less pressure loss.

hot_rod

Never seen a 1.9 Cv mixer? In a 3/4" size? What would be the application?
Possible a small 3/8 compression under the sink ASSE 1070 valve would have a low Cv.

Even if his radiant needs 3 GPM that is a small 1 psi drop in a valve like that.

Ralphxyz

I replaced the mixing valve with a ESBE VTA 3/4" FNPT

Static Pressure:


Pressure Radiant Loop Circulator NO HEAT:


Pressure Radiant Loop and House Circulator HEAT:


That <2 psi makes the difference!!

I am going to take the time to find my two 24" pipe wrenches moving the tee is starting to make more sense, I could also bring the radiant loop lead out 3' with 1 1/4" pipe to minimize any pressure loss.

Ralph

Jolly Bodger

I question where the pressure is making the difference. If the radiant loop is not circulating because of high head then I am not convinced the re-piping will correct it.

If the radiant loop is circulating but just can't draw in hot water then the re-piping should make all the difference.

It would be real easy to pop in a bigger pump and see. What size is on the boiler?

HomerJSmith

What happens when the cold water return on the RTI is at a higher pressure than the hot water supply to the RTI?
Ralph said,--- "It is strange the first time when everything was cold when the thermostat called for heat and the circulator kicked on with the mixing valve turned all the way to the right the pipes did not get hot on the supply side but when I turned the knob all the way to the left the supply got hot and the Mix got hot." This is the reverse of how the RTI should work.
Then he says, " Now with the circulator running I barely feel the Mix side heat up and that appears to be coming from the return as that is warmer than the supply. The warmth in the return is residue from running all morning, it had gotten warm (not hot). But at 10:00 a.m. I could not get any new heat from the supply it remained cooler than the return. No matter where I turned the mixing valve knob I never get the supply to heat up." I would think that the RTI would close off the cold return to the RTI to maintain the mixed temp setting. I don't think this may be happening.
My theory.---The radiant pump is running. He says,"...the supply it remained cooler than the return." Then he says, "...No matter where I turned the mixing valve knob I never get the supply to heat up." We all know, I hope, that water flows in the path of least resistance. Can we take that as a givin'? I don't think the RTI is closing the cold return spool and that flow is still occurring thru it. The path of least resistance from the radiant pump is thru the cold return of the RTI. The path of greatest resistance is thru the boiler circuit. The pressure from the radiant pump is greater on the cold return to the RTI than the pressure on the hot supply side of the RTI because of the resistance in the boiler circuit. That would prevent any hot supply water from getting thru the RTI. The radiant pump would be circulating hot water that gradually cooled and then continues to circulate that water. Very little or no circulation goes thru the boiler circuit.
This is why I said that the 15-42 was too big a pump and that it might be overriding the hot supply of the RTI, but it's not too big for the boiler circuit, in fact, it may be too small for the pressure loss in the boiler circuit.
This why I said that you need a balancing valve on the input to the RTI cold return to force more flow thru the boiler circuit. How many gal of flow is need, only enough to balance the hot and cold flow thru the RTI. You can use a Caleffi QuickSetter Balancing Valve w/ Flow Meter or if you're cheap like me a globe valve properly installed on the input to the cold return to the RTI and then turn it down until hot water flows thru the mixed outlet. Globe valves have a directional flow pattern.
It may necessitate a bigger radiant pump which is why a pressure loss calc need to be done on the whole circuit thru the boiler. Or you can just guess and install a Grundfos 15-58 on speed 3.
What do you think?

HomerJSmith

Hot Rod, I was reading from Danfoss and that's where I got the 1.9CV (Flow vs. Pressure Drop for VTA Series of Valves). Perhaps I got that wrong.

HomerJSmith

Ralph, before I would start grabbing pipe wrenches and changing tees I would install a globe valve on the input to the RTI. Do the least invasive thing first.
I saw the pressure readings. It appears the boiler pump acts as a booster pump to the RTI creating flow to the hot supply. I believe my theory is correct, pending further review.
2 psi=5'hd

GroundUp

Homer, he already added a valve to the cold side of the mixer. That is not the problem, nor is the circ. It is piped wrong, it is piped wrong, and it is piped wrong. Nothing more, nothing less. This has been made to be a month long debacle that was as simple as a 1 hour fix from the start. KISS

Jolly Bodger

@HomerJSmith , @Ralphxyz already put the globe valve in and tried to force the return to the boiler. It did not work.

I do not see a terrible amount of restriction in the return to the boiler. But we can't see everything and sometimes a rag gets left in the pipe. All I can see as a restriction in the pictures is the flow-control valve. Have you tried opening it up?

HomerJSmith

I see your globe valve W/Drain in your closeup pic. I have no idea where in the circuit they are located.
GroundUp, again I say, Explain to me why Ralph has heating in the pex when the boiler pump is operating with the radiant pump and no heat when the boiler pump is off. That's the crux of the whole problem.
You say, "Then run that 3/4" line all the way back to the boiler and cut in a new tee immediately after the air scoop, as I said in my last post. This will allow hot water to be drawn directly from the boiler, using the radiant circ." You don't think that the current connection draws water directly from the boiler? It seem to me that running a direct line, all that does is eliminate a lot of piping and pressure loss. Why does it matter if you pull water from the tee on the air scoop or further down the line except for the pressure loss. A circuit is a circuit is a circuit. I'm sorry, I just don't get it. I'm open to your idea, I just don't understand it.
I do think the 15-42 is possibly a big deal. The whole point is to run the radiant circuit thru the boiler without the boiler pump.

DZoro

Guys this works when the mixing valve can see the hot water....

Just needs to be properly piped and turn the boiler on when there is a call for heat.

Really not all that hard, but have to say it is very entertaining
D

Ralphxyz

Jolley Bodger, there are no flow control valves, only the flow check which is in the correct position to stop gravity feed possible turning the knob counterclockwise might do something.

Yes closing the valves on the mixing valve return feed did nothing!

The Grundfos variable speed circulators look interesting.

Changing the valve will be more than a hours work!

Ralph

HomerJSmith

Yes, "Guys this works when the mixing valve can see the hot water...." How do you get hot water, flow, Flow, FLOW! Gad! That's the problem, NO FLOW! Do I have to spell it out for you? N-O-F-L-O-W
I guess it is entertaining.

Jolly Bodger

I call it a Flow-Control valve because that is what B&G calls it. We are talking about the same thing. What harm does it do to open it up and see? If it works then problem solved. If by some chance you get gravity feed in your other zone then either pipe the radiant in before the flow-whatever. Or throw a spring check in between the house pump and return to the boiler. Possible a bigger pump will overcome whatever the problem is without any piping changes.

HomerJSmith

Jolly Bodger--If you do open the flo-check and the radiant works without the boiler pump, it just prove that the current radiant pump doesn't have enough head energy to open the flo-check. Solution get a pump with higher head energy suited for the desired flow requirements. Of course, if it doesn't allow flow it doesn't mean the solution isn't appropriate, you still might need a bigger pump.

GroundUp

Homer, I see this all the time. The head between the mixer and circ is less than the head from the mixer back to the boiler- so without the boiler circ running to bring hot water to the tee and flow to the loop, the radiant side simply short cycles or deadheads as there is no hot water to draw from the "primary" loop because it is stagnant due to the flow check. If you care to do the math on a 1.9cV valve and 2 loops at .6 GPM and compare it to the curve of a 15-42, you'll see that the 15-42 can easily handle upwards of 10 loops with the same valve and pipe size. The circ is not an issue.

Ralph, nobody said anything about changing a valve a valve- however you're correct as that would be a 20 minute job. I was referring to the repiping back to the boiler ordeal.

HomerJSmith

"Homer, I see this all the time." Well, I sure don't. What do you mean, "The head between the mixer and circ is less than the head from the mixer back to the boiler..." First, head is head energy which can be converted to pressure. It is not pressure loss. It is not distance. Do you mean this: The pressure loss between the RTI mixed output to the radiant pump is less than the than the pressure loss from the hot input of the RTI thru the boiler to the output of the radiant pump? So, the radiant pump stagnates because the pressure loss of the output of the radiant pump to the RTI hot input is too great for flow to occur and the flo-check is the culprit. That seem to me to be a pump issue. That's the issue I raised.
The pressure loss at a given flow is the whole circuit and that is what the radiant pump must overcome, from start to finish.
So, when the radiant pump is on and the boiler pump is on, they are in series (well not exactly in series more parallel) and that increases the flow (pumps in series increases the pressure and pumps in parallel increases the flow) which overcomes the pressure loss of the piping and flo-check creating flow to the hot input of the RTI.

So, why are you repiping from the hot input of the RTI to the air scoop? How does that help the problem of the radiant pump operating only.
I decline, math isn't one of my strong points.

Jolly Bodger

@HomerJSmith . what I suggested is a diagnostic test that would take 30 seconds. Maybe not the whole answer. But it could be tested quickly with out spending a dime. Collect Data then make decisions on that data. Or just start replacing parts and piping randomly. Eventually both methods will end up at the same place.

HomerJSmith

Both methods will end up at the same place, but randomly replacing parts, I end up a lot richer.
Look, I agree and it would tell a lot.

Ralphxyz

HomerJSmith what exactly is your test?

Ralph

HomerJSmith

Ralph, do what Jolly Bodger says in regard to the Flo-Check. OPEN IT.
Asked and answered--If you do open the flo-check and the radiant works without the boiler pump, it just prove that the current radiant pump doesn't have enough head energy to open the flo-check. Solution get a pump with higher head energy suited for the desired flow requirements. I would off hand recommend a Grundfos 15-58. Of course, if it doesn't allow flow it doesn't mean the solution isn't appropriate, you still might need a bigger pump. Look, your pressure gauges show that there is a 2psi increase when the boiler pump turns on that equates to a 5'hd increase. Your 15-42 @ 1.5 g/m is 14'hd. If you need 19'hd @ 1.5 g/m, what do you think will happen? You can not figure out the pressure loss of you radiant sys thru the boiler unless you add up all the pressure robbing elements from the radiant pump output to the radiant pump input. I recommend you do that or you just put in a bigger pump and wing it. I asked you for the pex length of both your pex circuits and I guess I missed it just like I missed the far away pics of your sys.

Ralphxyz

Here are some pressure images.
Static: OFF COLD
Boiler


Radiant Loop Return at Boiler


Mixing Valve Return (Cold) feed


Radiant Loop Circulator Inlet


Radiant Loop Circulator Pressure Outlet


House Circulator: On COLD





House and Radiant Loop Circulators On:
HOT

Jolly Bodger

Flow between the radiant loop and the boiler is the problem. either remove restriction or put a bigger pump in.

Is that a check valve on the suction side of your radiant pump? Pumps like to pump, not suck. Put Check valves on the head (discharge) side.

Ralphxyz

No that is a ball valve.

Ralphxyz

JollyBodger where is the restriction?

Here is the radiant loop, which is 20' from the circulator.