Adding a secondary pump to my radiant heating system

Gordy

I think

That he means 180 boiler temp, and 140 mixed temp supplied to radiant.

Gordy

Tiles

I see you have tile. Is this on all the radiant floors?



I think you have a lot of lose ends to get in order before you worry about buying a bigger circ.



You say there may be air in the system this can effect flow rates. You also state all the tubing is coming down to be reorganized? To what extent.



Lots of people come here wanting to install a bigger more expensive to buy, and operate circulator thinking its going to fix a flow problem. A lot of the time it's not the circulator selection.





Really have to do the math to see where your at..



What's the heat loss to the space heated by the radiant?



This tells you tubing size needed centers spacing, flow rate, and water temps.



You are already bound by the 3/8 pex which of I understand correctly is a plated under floor install?



So now unless you redo the emitter end you have to design around that, and make it work as best as you can.



You state that it is working fine, but the complaint is that it takes to long to get up to temp.



Unless you follow SWEI's advice ODR with constant circulation you can't do much better than what you have. A big pump won't fix it. All it will do is give diminished returns. The amount you gain will be far less, and cost more where ODR will give you improved comfort, and save money.



I run a B&G series 100 HV on my system which has a steeper curve than the 100. My system is all radiant ceilings and floors from the 50's. 1/2" and 3/8 copper tubing 200' loops 6",8" and 12" on center 26 loops piped parallel. The system head is 11.5 feet at 15gpm. I run an average water temp of 107 degrees on a design day which is -10. Each loop averages just under .6gpm.



That's what tube spacing allows you to do lower water temps, and flow rates. The wider the spacing the higher the flow rates, and the higher the temp needed. The tubing has its flow, and temp limitations to design around.



Hope this helps you understand a little more.

SWEI

Confused by the PDF above

titled "Radiant Sketch" which shows a supply manifold with ports marked

140 140 180 190 170 170

and a return manifold with ports marked

 .7    .7    .6    .6    .7    .7



I figured the numbers on the supply ports were temperatures, but can't imagine how those could result from the piping diagram shown (unless they represent cheap clip-on thermometers which have not been calibrated.)

Gordy

SWEI

I think those are both sets of flow rate numbers. Unless I'm looking at the wrong sketch. First link right hand sketch.

SilverCheetah

Follow Up

PLEASE SEE IN BOLD. APOLOGIZE FOR BEING BOLD BUT AT LEAST THE ANSWERS WILL BE THERE.

 see you have tile. Is this on all the radiant floors?

YES, ONLY THE TILES HAVE RADIANT FLOORS. THE FAMILY ROOM HAS 8" SPACING AND THE OTHER AREAS HAVE 16" SPACING.



You say there may be air in the system this can effect flow rates.

I HAVE A FLOW CONTROL VALVE AND SLIGHT LEAK AT COPPER TUBING. AT ONE POINT THE FLOW WAS ABOUT 4.5 GPM AND NOW AT ABOUT 4 GPM

You also state all the tubing is coming down to be reorganized? To what extent.

NOTHING MAJOR. JUST ORGANIZE THEM AT THE MANIFOLD.



Lots of people come here wanting to install a bigger more expensive to buy, and operate circulator thinking its going to fix a flow problem. A lot of the time it's not the circulator selection.





Really have to do the math to see where your at..



What's the heat loss to the space heated by the radiant?

I CALCULATED THE HEAT LOSS BASED ON HIGH CEILINGS, 9 FOOT CEILINGS, ETC. AND BY EACH ROOM AND THIS IS WHAT I CAME UP WITH.



LAUNDRY ROOM 1,620

KITCHEN810

FAMILY ROOM & FOYER 36,000

SMALL BATH 405

Total = 38,835



This tells you tubing size needed centers spacing, flow rate, and water temps.



You are already bound by the 3/8 pex which of I understand correctly is a plated under floor install? YES, I AM ALREADY BOUND BY THE TUBING. I WILL NOT TAKE THIS TUBING DOWN. IT IS JUST WAY TOO MUCH WORK AND $$$ AT THIS POINT. THE WORK BEING MORE THEN THE $$. I CAN EASILY MOVE OVER THE OLDER TRANSFER PLATES AND ADD ANOTHER 4 RUNS, ADD ON TO THE MANIFOLD AND THEN MOVE ON TO THE PUMP. BUT BECAUSE OF 3/8", I DO NOT WANT TO TAKE THE RISK OF THE SYTEM NOT WORKING THE WAY I WANT IT TO.



So now unless you redo the emitter end you have to design around that, and make it work as best as you can. I WOULD ADD WALL RADIATORS IN THE KITCHEN AND FOYER PRIOR TO DESIGNING AROUND THE SYSTEM. THERE IS AN EXTRA PORT IN THE BASEMENT FOR THE FIRST AND SECOND FLOOR PUMPS (OUTSIDE THE RADIANT HEAT), MAKING THIS 2 EXTRA COPPER SUPPLY AND RETURN PORTS (CAPPED OFF). THE COST WOULD BE ABOUT THE SAME AS THE RADIANT, ALTHOUGH THERE WOULD BE 3 OR 4 LARGER RADIATORS IN THE AREAS THAT WOULD NEED THE HEAT.



You state that it is working fine, but the complaint is that it takes to long to get up to temp. THE SYSTEM WORKS FINE, BUT RUNS A LONG TIME BEFORE TEMPERATURES CAN BE REACHED. IN THE MORNING TEMPERATURE IS AT 67 DEGREES AND IT WILL TAKE ABOUT 2 HOURS BEFORE REACHING 69-70 DEGREES (ONCE I TURN THE SYSTEM ON) ONCE SYSTEM SHUTS DOWN THE HEAT WILL NOT TURN ON FOR AT LEAST 6-8 HOURS IF NOT LONGER.

ALTHOUGH THE HOUSE HAS 2 QUITE LARGE HIGH CEILING ROOMS AND ABOUT 800 SQ. FT LARGER THEN MY NEIGHBORS HOME WITHOUT THE HIGH CEILINGS. MY GAS BILLS ARE ONLY $30-$50 MORE PER MONTH THEN HIS. HE KEEPS HIS HEAT AT 62 (NIGHT TIME) AND 68 DURING THE DAY. MINE CONSTANTLY RUNS BETWEEN 68-70 ALL DAY LONG AND NIGHT. SO SOMETHING IS WORKING RIGHT.



Unless you follow SWEI's advice ODR with constant circulation you can't do much better than what you have. A big pump won't fix it. All it will do is give diminished returns. The amount you gain will be far less, and cost more where ODR will give you improved comfort, and save money. MY AQUASTAT MINIMUM IS AT 180 DEGREES.  I WILL LOOK INTO THE ODR TO DETERMINE IF THIS WILL WORK.



I run a B&G series 100 HV on my system which has a steeper curve than the 100. My system is all radiant ceilings and floors from the 50's. 1/2" and 3/8 copper tubing 200' loops 6",8" and 12" on center 26 loops piped parallel. The system head is 11.5 feet at 15gpm. I run an average water temp of 107 degrees on a design day which is -10. Each loop averages just under .6gpm.



That's what tube spacing allows you to do lower water temps, and flow rates. The wider the spacing the higher the flow rates, and the higher the temp needed. The tubing has its flow, and temp limitations to design around. I DO UNDERSTAND THIS A LITTLE MORE. WHAT I DO NOT UNDERSTAND IS IF THE 3/8" TUBING IS SNAPPED INTO ALUMINUM TRANSFER PLATES (WHERE I HAVE FELT THE PLATES) REAL HOT, IMO THE TUBING IS NOT COOLING DOWN IN THE BEGINNING BUT AT THE END OF THE LOOP. I GUESS WHAT I AM TRYING TO DO IS JUST CREATE FOR EVEN HEAT WITH WHAT I HAVE. I DO NOT KNOW WHY THE INEXPERIENCED PLUMBER LAYED DOWN A RADIANT HEATING SYTEM WITH 16" APART INSTEAD OF 8". THIS IS WHY I THOUGHT INCREASING THE FLOW RATE FROM 4-4.5 GPM ACROSS THE BOARD TO ABOUT 6 GPM MIGHT DO THE TRICK. THAT IS ONLY ABOUT .3 GPM PER LOOP. BUT IT SOUNDS LIKE I WILL HAVE DIMINISHING RETURNS.  



Hope this helps you understand a little more. IT DOES. APPRECIATE ALL YOUR HELP. TO ADD TO THE CONFUSION, I GUESS. I HAVE 6 LOOPS. IF I SHUT DOWN THE FIRST 2 LOOPS 8" ON CENTER (140 FOOT RUNS EACH) TO THE MANIFOLD, HERE IS WHAT HAPPENS.

THE FAMILY ROOM IS NOT WAS WARM AND FEELS KIND OF COLD INSIDE THE CENTER OF THE ROOM. ALL THE RADIATORS IN THE FAMILY ROOM ARE ON THE OUTSIDE (A LOT OF RADIATORS). THE TILES ARE NOT WARM. THIS ROOM HAS BOTH. I ADDED THIS HEAT 5 YEARS AGO. NOW WE ARE LEFT WITH THE KITCHEN, SMALL HALLWAY AND FOYER ALL WITH 16" SPACING AND RUNS OF 180, 180, 170, 170 TOTALING 700 FEET INSTEAD OF ABOUT 1,000 FEET. MY DELTA CHANGES AND THE DIFFERENCE IS ABOUT 20 DEGREES, BUT THE KITCHEN HEATS UP FASTER.

I HAVE EXPERIMENTED MANY DIFFERENT WAYS TRYING TO FIGURE THIS OUT.

SilverCheetah

Reply to SWEI

Confused by the PDF above

titled "Radiant Sketch" which shows a supply manifold with ports marked

140 140 180 190 170 170

THIS IS THE FEET OF LOOPS PER RUN SNAPPED INTO THE 3/8" JOIST TRAK TRANSFER PLATES.



and a return manifold with ports marked

 .7    .7    .6    .6    .7    .7

THIS IS THE FLOW AT THE PEAK OF THE SYSTEM WITH NO AIR IN THE SYSTEM. THERE IS AIR IN THE SYTEM AT THIS POINT.



I figured the numbers on the supply ports were temperatures, but can't imagine how those could result from the piping diagram shown (unless they represent cheap clip-on thermometers which have not been calibrated.) THE CLIP ON ARE WINTER THERMOMETERS AND THEY ARE ACCURATE. I HAVE TESTED THEM AT ROOM TEMPERATURES, ETC.

Gordy

Okay

I did not see those numbers on the first radiant sketch PDF .

Gordy

Odd

Silver cheetah



Are you sure that the circ is a series 100 and not a 100 hv?. IEEE you said it's 1/12 hp which is the 100.



The reason I ask is something is odd. If your flow meters are accurate, and the tube lengths are accurate, 190' of 3/8 pex with .6 gpm is 11' of head according to my rad pad. Are you sure it's 3/8 pex?



The series 100's shut off is 8'

SilverCheetah

I am 200% Positive

Silver cheetah



Are you sure that the circ is a series 100 and not a 100 hv?. IEEE you said it's 1/12 hp which is the 100.



Yes, 200% positive. #'s match from website # 106189.



http://www.pexsupply.com/Bell-Gossett-106189-1-12-HP-Series-100-NFI-Circulator-Pump-3540000-p?gclid=CJGat6nYlbYCFYSK4AodbFIASA



The reason I ask is something is odd. If your flow meters are accurate, and the tube lengths are accurate, 190' of 3/8 pex with .6 gpm is 11' of head according to my rad pad. Are you sure it's 3/8 pex?



200% positive it is 3/8" pex. I have some left and have original receipts and have 3/8" fittings that fit the tubing.



The series 100's shut off is 8'

That is what I am always understanding that the 100 is 8 feet of head and pretty flat from "0" to "10" GPM.  

I could be off slightly on the lengths but not that much. I have been through this a hundred times in 3 months.

I have over 85 (90 degree bends in the pex tubing) about 15-20 (90 degree bends in the pex tubing) along with the manifolds, mixing valve, copper tubing, flow control valve, etc. Many pictures if you need to see them. This could be causing the wrong data to be calculated.

Based on your calculations, what should the longest run or average of all 6 be on the 3/8" pex tubing?

 

Lastly, I tested out loops 2-4 (16" spacing) and today the Delta T was between 15-5 as the boiler shut off and turned on. The temperature varies as the supply side becomes hotter so does the return side. After boiler shuts off the supply side cools down about 10 degrees and return side cools down about 15-20 before boiler turns back on.  

Gordy

Okay

Your reading the flow meters from the top should be reading the bottom of the plunger.



So you are from .4 to .5



That changes head a lot to 5.5 -7.5'



See how head goes up to the square of the flow rate.

SilverCheetah

That is what I thought

That is what I thought as well. They start at .40 at the top without any flow. So I found this odd as well. I will call Pex Supply or the manufacturer for this part.

But, when you go to the website for Pex Supply, here is what they write?

1. TruFLOW Visual Flow Meter, 0.15 to 0.8 gpm
   
   Q: how do you read the flow meter? From the bottom of the yellow, or at the top where the spring makes a line across the yellow?Asked on 1/16/2012 by hop from McKellar, Ontario
   
   Know the answer? [u][color=#0066cc]Answer this question[/color][/u]
   
   1 answer
2. 
   
   CUSTOMER CARE
   
   
   
   A:You read the meter at the top of the yellow line, where the spring meets the yellow indicator.
   
   Answered on 1/31/2012 by PexSupply Staff from NY

SWEI

PEX head

I'm sure my Rad Pad is here someplace...



I get 3.79 ft/100' (at 180F) or 4.06 ft/100' (at 120F) from the Uponor CDAM for 3/8" PEX at 0.6 GPM (2 ft/sec.)  This would imply somewhere between 7.2 and 7.7 ft depending on temperature for the loop.



I'd start with an iSeries-R and see how things improved.  Perhaps a Bumble Bee for the radiant loops later depending on what resulted.

SilverCheetah

Please explain the comment below?

I'd start with an iSeries-R and see how things improved.  Perhaps a Bumble Bee for the radiant loops later depending on what resulted

What is iSeries-R and a Bumble Bee? Please accept my apologies as I am not familiar with those terms?

SilverCheetah

Considering we know more facts

Considering I am only getting only 3 GPM (reading from bottom of the flow meter), would the 0015 pump @ speed # 3 would produce 16 feet of head at 5 GPM  and 15 feet of head at 6 GPM. Would this pump not overcome the resistance needed to produce slightly more flow or are we still on the same assumption? The cost associated with the pump to test out the waters is not that much?

SWEI

iSeries-R

Is the ODR-controlled Taco mixing valve I linked to above http://flopro.taco-hvac.com/products/index.html?category=188



The Bumble Bee is a variable-speed ECM (electronically commutated motor) ∆T circulator from Taco.  Its highest curve is roughly equivalent to a 008, but it will constantly adjust to maintain the set ∆T.

SilverCheetah

Is this the PUMP you are referring to?

http://www.pexsupply.com/Taco-0013-VDTF3-0013-Variable-Speed-Delta-T-Cast-Iron-Circulator-Pump-1-6-HP

If so, TACO talked me out of this to purchase the 0015 pump?

The NRF-36 is similar but has a manual 3 way switch (I am more comfortable with this) and cost is less?

Gordy

Pumps

You have to design around a needed flow rate for the loops.



So lets say you want .8 gpm through the 190' loop the head would be 20.9' so now go to the 0015 curve chart 21' of head on speed 3 will only produce .5 gpm. It ain't going to happen. The 009 is a better match but still is over kill.



Stick with what swei is telling you. Your wanting to waste about 800 dollars in pumps that won't do what you want. Swei is proposing something that will increase comfort, and save fuel for about the same cost or less.

Gordy

Taco

I don't think they had the proper info to spec the 015.



http://www.blueridgecompany.com/radiant/hydronic/747/taco-bumblebee-high-efficiency-variable-speed-pump



This is the circ swei is describing



It might settle in at .6 gpm on speed 4

SilverCheetah

Assuming the loop is 190 feet

Gordy:

Maybe I am missing something. I am looking at the bottom of the chart. Are you looking at the top of the chart? How are you figuring out 20.9' of head? I trust your figures. Is there any 3 speed pumps out there that will produce the .8 GPM which may not be needed. Thanks! I do not remember these pumps being $800 but more like $300.

Gordy

Wrong to me

If the customer care is correct then how do you read say .15 flow. They may have confused it with another flow meter.

Gordy

Charts

I guess I should ask what chart you are reading.



Mr. Pex chart is .11 feet of head per foot for .8 gpm that's 20.9' for 190' of loop.

If youmtrymto push 1 gpm that's .18 feet of head per foot or 34.2 feet of head for 190' loop.



Rad pad is close to the same.



Now plot the head of 21 feet on the 0015 curve. Come down to gpm your less than .5 gpm on your longest loop.







As far as pumps go 275 for the series 100 you won't use, and 375 for the 015 plus some one installing them unless you are doing it..

SilverCheetah

Some of the verdict is in

Well I bit the bullet and purchased the B&G 3 speed pump NRF-36. Wow what a pump. Price was very reasonable as Pex Supply matched the lowest price on line. On $249 + free shipping.

So far here is what has happened. I ran the pump @ speed # 2 in which the flow meters passed the .8 GPM (which is the max). Except # 4 which I believe the flow meter may be off (not sure) as this is the longest loop. I have new flow meters up to 2 GPM but will not install them until another 2 weeks or so (after the heating season). And within 10-15 minutes  from a cold start, the Delta was between 10-15 degrees. This will provide me even temperatures through the aluminum transfer plates, which was not the case before. This was with the mixing valve open all the way @ 140 degrees. Again, I do not know where I am at the chart because of the flow meters. Tomorrow I will run the heat to determine how long it will take to go from 67-68 until 70 degrees. Usually the time is about 2 hours with the B&G Pump. It will be cold tonight and temperature drops to 32 degrees. Chances are the heat will not even turn on at all.

Changing out the pump was a piece of cake. I needed a fold out chair and my knees to hold the pump. The actual size of the NRF-36 length is exactly the same as the B&G 100 as nothing had to move. Installation was about 30 minutes, if not less. Replaced the washer/O-Rings with the new ones and used the old flanges that were already part of the system (Unfortunately they sent me the 3/4 NPT fittings and I need the 1").  The manifolds did not move. This was done by myself. Isolated all the shut off valves. Made sure I followed the instructions and filled the system with water first before I ran or turned on the pump. Pump is very quiet compared to the B&G 100 pump.

With the Series 100, the total flow was 2.9 GPM and I always had a 30-40 difference between supply and return temperature. With the NRF-36, on the lowest setting, speed # 1, I am at 4.1 GPM. Based on the chart @ 4 GPM, that is about 19" of Head Pressure.  I did not have time to determine what the Delta T will be on this setting. Tomorrow I will be testing the system at speed # 1 and # 2 with different settings on the mixing valve. I will use the 110, 125 and 140 degrees at both speeds to determine the temperature from the supply and return.  Speed # 2 pushes me past the .8 GPM (max on flow meters) so this is hard to tell but I am assuming that I will be at a minimum of 1 GPM on each loop. Even on the highest speed and medium speed I can adjust the balancing valves on the return side and there was no noise or banging around in the pipes. My goal is 1 GPM, and I doubt that any more flow will be needed. One thing I noticed is the pump gets hotter then the Series 100 as it is a smaller pump.



If anyone can answer this question I would appreciate it:



I have a FLOW CONTROL VALVE 1" that is right after the boiler pipe on the supply side that is leaking. The plumber will be changing this out for a new one in about 2-3 weeks. Because there is so much resistance in the 3/8" pex tubing, is the valve really necessary? Does it serve a purpose? Is this Code? It is supposed to act as a check valve (whatever that means) and also to prevents hot water from migrating into a zone that is not calling for heat. 

SilverCheetah

FINAL ANALYSIS

The reason why I purchased the pump is that eventually I will add 2 more loops in the kitchen only, either 3/8" or 1/2" which will give spacing every 8 " either all with 3/8" pex or 3/8" & 1/2" pex. I do not know if this combination will be over kill as I may now have uneven heating because of the bigger tubing. Nothing to worry about now.

Attached is the spreadsheet of all the calculations that have been done for about 1 hour. Amazingly the boiler turned on for 3 minutes and shut off for 7 minutes. This is with only the radiant running and neither the 1st floor or 2nd floor + hot water holding tank. One day I will test the system with all 3 floors calling for heat. Next years project.

If you look @ the sheet I was able to obtain a difference or lowest delta with the mixing valve allowing the most water to be mixed in with the warm water. This gave me an average temperature of 126 degrees with an average delta of 8. Considering this is in the sub-floor and it takes time for the heat to rise, should I be running the system with the mixing valve all the way opened? 

Mixing valve fully opened with an average temperature of 133 degrees and an average delta of 16.  There is more consistency with the lower temperature.  The boiler did not change its cycle for on/off times. 

Where should I be running the system? At which point. Mixing valve at 1 or at 6? Thank you.

 

Gordy

Did you

Get the longest loop to flow yet?



Yes you should have the flo control valve in place to prevent ghost flows. Is the packing around the stem leaking?



In viewing your spread sheet I have to ask what the outside temps were at those times?

What is the design day temp used for the heatloss.





You have 43 min of on time, and 15 min of off time in a 63 minute period observed. But yet your on times are 7 min.



I will take it that this was one continuous call for heat?



Is boiler set at 180*?

Were all zones in the house calling for heat?

Were only the radiant zones calling?



It seems your bouncing off high limit frequently. Is the boiler correctly sized for the total heatloss of the structure?



With your setup all you can do is set everything for design day supply temp requirements, and as it gets warmer than design day the boiler will cycle more. Or you could go down , and adjust the manual mix valve in relationship to outdoor temps. Or you could get a motorized mixing valve coupled with outdoor reset, and it will all be automatic as SWEI was trying to tell you.



Otherwise it could be set at 120 now when temps are mild, and be reset to 140 next year when the weather gets cold.



Your wrapping yourself up in the delta t of the loops,and forgetting about the supply temps your design needs.



You could have narrowed your delta with the old circ if you would have lowered your supply temps to 120 also.



To really dial things in you need to look at everything as a whole. You said it worked fine, but was it fine as far as satisfying the heat call? Were your floor surface temps even?



Something else I would investigate is the rim joist insulation detail. With an under floor setup poor rim joist sealing, and insulating can really rob btus in the joist bay.



Let us know

SilverCheetah

Answers for Gordy



Did you

Get the longest loop to flow yet?

Yes, I did. The longest loop and I checked the tubing at 180 feet (give or take 5 feet max) and this loop is at exactly .65 GPM which makes no sense. I believe the flow meter is off or something is wrong. I will know better when I swap the flow meters and fix the loops. If I have the same problem then I have to check the loops. Even at speed # 3 which the head pressure is huge does not push the loop past .8 and all the other loops are way past that, so I am assuming something is wrong. Maybe the manifold supply and return is not opened all the way. I will have to examine them to make sure nothing broke inside the manifold.



Yes you should have the flo control valve in place to prevent ghost flows. Is the packing around the stem leaking?

Yes, it is leaking. I tried tightening the nut but no luck. Very little leak. I have the new part and it was only $50. The plumber will be fixing a few minor items for only a few hundred dollars.



In viewing your spread sheet I have to ask what the outside temps were at those times?

The temperature this morning was 32 degrees outside and the water temperature in the manifold were at room temperature at 65 degrees. The water in the boiler I believe was about 130 because of heat in other parts of the home.



What is the design day temp used for the heat loss..

I do not understand the question. Please explain.

What is the design day temp used for the heat loss..

I do not understand the question. Please explain.

You have 43 min of on time, and 15 min of off time in a 63 minute period observed. But yet your on times are 7 min..

What this spreadsheet is showing is that the boiler shut off for 7 minutes while the pump was pumping the water at all times. Then the boiler turned out at around 155-160 degrees (20 degree difference) and it took about 3 minutes to reach 180 degrees before the boiler shut off and then the cycle continued like clockwork.



I will take it that this was one continuous call for heat?

Yes, the temperature was at 68 degrees this morning and the temperature was 32 degrees outside. A continuous call for heat. It took quite a while to bring the temperature up in the kitchen because of the 16" spacing. About 2 hours. Under normal circumstances it would take about 3 hours with the old pump. Once the kitchen reached 70 degrees and thermostat shut the system down, the temperature increased to about 72 degrees, probably because of the heat buildup.



Is boiler set at 180*? Yes, this is the minimum setting on the aquastat. The old aquastat was at 160.

Were all zones in the house calling for heat? No, I had shut down all the other zones which were pretty much at 70 degrees so I can test the system out.

Were only the radiant zones calling? Yes, only the radiant



It seems your bouncing off high limit frequently. Is the boiler correctly sized for the total heat loss of the structure? I honestly believe the boiler is oversized for the home, based on my heat loss calculations.. I do not understand the "bouncing off high limit frequency", please explain. The boiler shutting down for 7 minutes and only taking 3 minutes to reach temperature seems pretty good. I can not adjust the differential in the aquastat.



With your setup all you can do is set everything for design day supply temp requirements, and as it gets warmer than design day the boiler will cycle more. Or you could go down , and adjust the manual mix valve in relationship to outdoor temps. Or you could get a motorized mixing valve coupled with outdoor reset, and it will all be automatic as SWEI was trying to tell you.

Otherwise it could be set at 120 now when temps are mild, and be reset to 140 next year when the weather gets cold.

As long as the 20 Delta is sufficient so I have better spacing heating then I do mind leaving the mixing valve at 6 to keep the temperature up. The cycling times are consistent no matter what setting I used.



Your wrapping yourself up in the delta t of the loops,and forgetting about the supply temps your design needs.

Unfortunately the system was never designed properly, so I can not figure out properly the design needs. The only thing I can do is work with what I have. I could add the extra 2 loops in the kitchen and chances are that within the next year or so that will be done. Taking down the insulation or system will take no time whatsoever. I can have everything down in 2 hours. Putting it together will take time, probably 2 days. I have priced everything out and I can get everything done for under $1,000. Although working on possibly an extra radiator in the kitchen. I do not know if this is the time to do this. Definitely after next heating season.



You could have narrowed your delta with the old circ if you would have lowered your supply temps to 120 also.

I tried lowering the mixing valve and trust me that I have spend endless hours downstairs with the old B&G Pump. The supply temps were going in at the same temperatures as you see on the spreadsheet but the return side was between 30-40 difference. On setting # 6 because it took so long for the water to return the temperatures were even hotter. I have seen temperatures of 160 degrees going in and 110 coming back for a short period of time. There was not enough head pressure to overcome the difference. Now, at speed # 2 on the new pump and after about 15 minutes the return side loops are smoking hot (by hand). Before this only would the return be hot for a short period of time and then within minutes be cool to the touch again. The delta was not narrowing to any extent no matter whether I had the mixing valve either at 1 or at 6 or anywhere in between. The return temperature was at most 100 degrees max. I will know better the real difference in flow per loops when I install the new flow meters. There was too much inconsistent heat in the tiles. The kitchen was split in 2 sections. In each section only 50% of each section the tiles were at 75-80 degrees (surface temperature) and the next few tiles over where 5-10 degrees cooler. Because of the 16" spacing I was not getting even heat at all. At least now the areas that are over the aluminum transfer plates are all even, which is what is helping out.



To really dial things in you need to look at everything as a whole. You said it worked fine, but was it fine as far as satisfying the heat call?

The system worked fine on the coldest days (temperatures from 35 degrees to 25 degrees in the overnight) until the kitchen reached 68 degrees. If you tried to have the kitchen reach a temperature of 70 degrees, the heat and pumps ran for over 3 hours. Not only is the spacing 16", but because in the subfloor it takes time to reach the surface, etc.



Were your floor surface temps even?

The floor temperatures are not even in each tile. Tiles are 12". Only in the areas above the transfer plates. So hypothetically if each plate was center under each tile (which is not the case), that tile would be about 80 degrees and the next tile would be about 70-75 degrees. I will place 3 thermometers on the tiles next to each other after the system has ran to test out the surface temperatures to determine how fart apart the surface temperatures are.



Something else I would investigate is the rim joist insulation detail. With an under floor setup poor rim joist sealing, and insulating can really rob btus in the joist bay.

I have the silver bubble wrap stapled up real nicely. In addition I used special tape and there is insulation tucked in nicely in every area that you can think of. Of course there is possibly some heat loss, although I took my time insulating the area. This must work well because when the system shuts down at 70 degrees from a 68 degree start and 16" spacing, the heat rises another 2 degrees. So I believe there is so much heat buildup underneath the subfloor it continues to rise and reach a peak. Sometimes I come home even hours later and I can feel the heat still coming up and tiles are warm. Kids are always on the floor. A lot of times the winter coats and stuffed animals are still warm 6 hours later after I lift them up off the floor (again this area is the 8" spacing).

　

Question: Given with what I have should I ben running the system at the 135 average temperature or the 125 average temperature. I believe I read somewhere that a 20 degree delta is sufficient for consistent heating. I have to work with what I have.



Let us know



Let us know

Gordy

Design day

Is the design outdoor temperature used to calculate your heat loss for the structure. It is a temperature that you may only see 1% of the heating season.



So it appears you did your own heatloss calculations. What temp did you use for outdoor, and what temp did you use for indoor? And did you look up what outdoor design temp is for your area?

SilverCheetah

I have no idea

I went on line and figured something out. If you can direct me to a website, I can easily design something. I used a website for heat loss based on square footage of each room and height of ceilings plus how well insulated the area is. House is well insulated. Here is the chart that I used. These are calculations. The lowest figures are the most and best insulation, while the highest is literally no insulation at all.

www.bgmsupply.com is the website.

SilverCheetah

Here are the surface temperatures of the Tiles

Here are the surface temperatures. Each tube is snapped in to an aluminum transfer plate. There are 4 foot aluminum transfer plates, insulated with bubble wrap and then regular household insulation.



The surface tiles that have 8" spacing of tubing measured at 12"  apart and with 3 different thermometers were 81 - 82 & 81 - 82 & 81 - 82 degrees. 



The surface tiles that have 16" spacing of tubing measured at 12" apart and with 3 same thermometers were 77 - 78 & 78 - 79 & 78 - 79. This area has a thin sheet metal stapled underneath the sub-floor that covers the whole joist and then the transfer plate in the middle. The sheet metal does absorb and transfer some of the heat across the whole joist, but I am assuming not as well as the aluminum transfer plate. This could be the reason why the temperatures are pretty consistent, although lower.  



At the end of the day, the difference between 8" spacing and 16" spacing is about 3 - 4 degrees.



Temperature outside during the testing was 42 - 44 degrees F.

Gordy

No idea what you have there

Did it ask for your geographic location?



That program is a guess as stated a " back of hand calculation"



You need to perform a more thorough one than that. I think you will find your heat loss is high based on that program you used. With out knowing winter design temps for your location it's a guess.



If you have your house in Alaska the heat loss will be higher than your house being in Illinois. The lower the outside temperature the higher the heat loss will be. Delta t drives this.



So you must do your heat loss based on the design day for your area. This is going to be the coldest on average for a winter. Could be colder, or warmer, but its an average.





Any type of heating, or air conditioning sizing calculations are based on mathematical equations it IS a science. The math works, but you need to plug in accurate numbers, or all is for nothing . Garbage in garbage out if you will.



Was this system installed by you, or existing?





You have provide a lot of information, but some of it has been inaccurate, and that turns into a dog chasing its tail.



It is hard to trouble shoot over the Internet with out being there. There are a lot of variables some as simple as thermostat location to very complex.



So lets kick this off with an accurate heat loss, and go from there that is always the first step even if it is just to replace the boiler. Slant fin provides a free one that is easy to use, and also in iPad app format. Do one with that,and compare the numbers to the back of your hand one. Tell me your location, and I can find the design temp for you. I will bet an accurate one is 40% lower than the one you did.

Gordy

Outdoor design temps

http://www.energystar.gov/ia/partners/bldrs_lenders_raters/downloads/Outdoor_Design_Conditions_508.pdf

SilverCheetah

Appreciate your help

I will review this later this evening.