A question about three speed pumps

  • Active since 1995, Hearth.com is THE place on the internet for free information and advice about wood stoves, pellet stoves and other energy saving equipment.

    We strive to provide opinions, articles, discussions and history related to Hearth Products and in a more general sense, energy issues.

    We promote the EFFICIENT, RESPONSIBLE, CLEAN and SAFE use of all fuels, whether renewable or fossil.
Status
Not open for further replies.
rvtgr8

rvtgr8

New Member
Hearth Supporter
Jan 6, 2008
155
Elbert Colorado alt. 7300
abe21.net
Am I correct in thinking that if I set my three speed Grundfos pump that circulates hot water through my HX on the Garn side of my system on its slowest setting and I place the Grundfos pump on my house side of my system on its fastest setting that the result will be lower temps on the house side of the exchanger? I hope this makes sense. :red:
 
Correct, However it may not change the temps that much.
 
My first cut on this is that it is more complicated than that. I think the answer relates to 1) how much, if any, your plate hx is oversized; 2) pump head on both sides of the hx at the various speed settings; and 3) the load. The easy and inexpensive way to get an answer is to put a thermometer on the house side of the hx and see what it reads at the two speeds. A meat probe thermometer will work good enough; cable tie it to the pipe and wrap insulation around the probe.

If you want to reduce and aim for a more consistent temperature on the house side, might you consider a mixing valve?
 
The thing is... The deltaT on the garn side will go up at slower speeds on the emitter side and the deltaT on the emitter side will go down but the output temps MAY
not change much.... esp if the exchanger is oversized.

I would run it at the lowest speed that heats the house.... unless you need specific temp for in floor.... then use a mixing valve.

Kris
 
Some random thoughts:

I'll go with 'It won't make as much difference as you'd think".

Mostly you'll affect delta T on each side. If you run the Garn side slower, then the water will be in the HX longer and will exit closer to the house return temp. That should give you more stratification in the Garn (if I understand Garn plumbing).

If you run the house side faster the water will be in the HX for less time and will pick up less heat. It will also spend less time in the house and will lose less heat - smaller delta T both ways. That MIGHT be desirable if you have a long loop with many feet of baseboards and you want to get hotter water to the last baseboard.

Of course, higher circ speeds use more electricity and at some point start to create noise from water turbulence.
 
I'm plumbing my GARN and have a couple of related questions:

1. I plan to use the FlatPlate (GEA) GBE 5x12 80 plate. This is considerably oversized, but I'm hoping to get temps very close to the house side as the GARN. I can't seem to find info on the head loss in the exchanger. Is there a rule of thumb for this? I need to know the expected head loss to size the pump, or select the speed.

2. 2" black pipe will leave the GARN. I belive I read the UPS15-58FC manual indicates it should have 9' of water above it in an open system. I can't do this for my GARN, can get close to about 5'. Am I likely to have a problem with the suction boiling as a result? I will plumb the pump vertically, with the shaft horizontal, but the flow will be going down...not ideal for caviation bubbles to rise.

3. Are the pump flanges for a Grundfos pump the same as a Taco? Or are they different? It looks like an 005/007/008 are 6.5" in "length", and the super brute is 6.375". Is this difference taken up with the gasketing, or can one not switch? I see Grundos advertises the super brute as a replacement for these Taco's, so I assume this 1/8" can be taken up by the gasket?

4. The HX has 1.25" ports. At what point do I switch the piping over to this size? After the pump on the supply, I plan to switch over to copper to the HX. This will give me 5' of black 2" pipe before switching over to copper. On the return, I plan to have 5' or so of black pipe before switching to copper. Copper will mate to the HX. Is this all OK, or am I missing something?

Sorry for all the questions!

Bruce
 
Thanks everybody! It never fails to amaze me just how quick and willing all of you are to help your fellow Hearth members. The ultimate solution I suppose is the mixing valve. I am currently following the advice of so many here in the Boiler Room...just get the system up and run it for a season and then do the tweaking when you have gotten your feet wet. Pun intended. My concern is that I do not want to significantly raise the temperature going into my floor zones. My heat exchanger is a GEA 50 plate. To the best of my knowledge this is about the correct size for my system. But if my initial figures are correct, I am sending water that is about 15 °F warmer into my floor zones. When I built the house, I admit that I was not as aware of hydronics as I should have been and was dependent on my plumber who had designed systems in Alaska using in floor radiant. He just wanted the temperature on the propane boiler to dump 130 °F into the high mass of my concrete slab. It worked great and the house is wonderful during winter months. The house is a slab on grade and the zones are fairly long. But it concerns me that the water is so much hotter coming off of the the HX from the Garn. I guess that I could burn the Garn at lower temperatures and fire more often, but that would not be my first choice. I am hoping to cool the water on the house side by controlling the speed of the circulation through the exchanger. Unfortunately it sounds as though I won't be able to affect the temperature other than by a few degrees. Does anybody have any other ideas for me other than the mixing valve. I just can't see myself cutting into the system one more time this season. I am going to try and follow some of the advice above and get some actual numbers to go with my thoughts and will post later.
 
Variable speed or motorized mixing valves are another choice. With either you could connect a weather responsive controller. In the spring and fall shoulder seasons the slab will perform with lower supply temperatures than at design days. A warm slab, or excessively warmed slab, is like a freight train. It's hard to rev it down as the conditions warm.

With an outdoor reset control properly installed and setup you will get virtually constant circulations and the radiant adjusts to actual conditions based on indoor and outdoor temperature conditions. With a setpoint thermostatic mixer it is accurate for one condition. The day the system needs the temperature supply that it is set to.

Yes it is more cost and complexity, but does increase comfort and reduce fuel consumption, it's a nice feature especially on high mass radiant. Think of it s a cruise control for your heating system.

hr
 
Everything what was said in this thread is certainly true, but:

1. If you restrict the flow in the Garn loop, the house supply temperature WILL drop, eventually. If low pump speed will not do it, I would try partially closing a valve on the discharge side of the Garn pump (to avoid pump cavitation). You will have to fiddle with this setting all the time though. Also do a search, valves do not like this mode of operation.

2. Really dumb solution, but it just might work:
Put an aquastat on a house supply side, in series with Garn pump. Set it to e.g. 130. Once the house supply temp will reach 130, Garn pump will stop. Once the house supply cools, it will start again. This will create process oscillations, I am guessing in a minutes range. There will be some under and over shoots, but only by experimentation can be determined how much. The pump may not love this mode of operation, but it is cheap...
 
My question that is raised here is how will closing back on the valve on the discharge side of the Garn loop impact the pump? I understand why you would want it on the discharge side because of potential cavitation, but won't it create back pressure on the pump that could probably damage it? :red:
 
Partially closing the valve on the discharge side simply increases the head - to the pump there is not difference between this and adding more pipe... As long as you stay withing reason (look at the pump curves to get an idea of what to expect). E.g. UPS 15-58 on low should do as little as 2 GPM at head 10 ft.
 
I have absolutely zero experience with radiant floor systems. That doesn't stop me from having opinions, though ;-)

I've always wondered about the wisdom of low flow rates on radiant systems, especially those with really long runs. Seems to me like you'd get a huge temperature drop and some parts of the slab would be way cooler than other parts. I suppose careful routing might reduce the effect, but I'd think you'd want a mixing valve that gives you the temp that you want (maybe automatically adjusted via outdoor reset) and a relatively high flow rate within the loop to minimize delta T between slab supply and slab return.

Should I just go back to topics where I actually know something?
 
Just to be clear: we are talking here about limiting the flow on the Garn side of the exchanger. The secondary side (house side) should remain on high, precisely to limit the delta-T...
 
I would certainly think you'd want a mixing control solution, either a valve or the injection pump. I use inject pumps. Do you heat DHW with the Garn as well? If not, then you have a HUGE tank of water that can heat you for a very long time...as you can likely go down to 110 or lower before having to fire. If you heat DHW, then I'd be concerned about making hot water when the temp is that low.

I have a Garn 1500 newly arrived, and I have lots of inslab radiant. But I also heat DHW.

If you look at simple controllers to do your mixing, like those from Tekmar, Taco, etc. they are quite easy to setup. If you only have one zone to mix water for, the Taco RMB-1 is a one-unit solution...with the mixing pump and zone circulator, and control, all integrated into one unit. There's a thread here from last week with short discussion about it.

I'd be weary of putting too warm of water into the slab. My Tekmar controls are set to a max of 110 I think, and they rarely ever reach this. But with the mixing controls, it is the absolute most constant type of heat I have ever experienced....and you should be able to maximize the time between burns with the Garn.
 
I forget your plumbing layout... If you heat DHW from the house side of the exchanger than you really need to keep the temps up....
If not then in theory you COULD slow down the flow on the garn side to control the radiant temps.
 
Keep the delta t low on residential radiant floors 10-15degrees. Commercial jobs that are not as comfort sensitive can run as high a 30 degree delta t. and higher supply temperatures... no barefeet!

Ideally the surface temperature would match your skin temperature low to mid 80's so you don't even notice where the heat is coming from.

Tight tube spacing and counterflow piping helps even out the temperature spread in slab radiant, especially bare concrete.

hr
 
...the wisdom of low flow rates on radiant systems, especially those with really long runs. Seems to me like you’d get a huge temperature drop and some parts of the slab would be way cooler than other parts.
Click to expand...

I've wondered about this also. Some of it must relate to the high pump head created in these lines, which results in low flow rates to keep the head down. For example, I think 300' is the max run for 1/2" pex. A flow of 1 gpm produces head of about 11-12 feet, and at 1.5 gpm the head is up to 25 feet, while 3/4 gpm the head is a little over 7 feet, and 1/2 gpm the head is 3.5 feet.

Another factor is the density/spacing of the runs of pex; 12" is common. I don't recall where I picked up the following info, but it might help:

The RPA rule of thumb is .6 gpm for 1/2 pex. With 4 loops you need to move 2.4 gpm. At .6 GPM flow in 1/2” pex the pressure drop through 270 feet is about 3.5 ft., not including any other pressure drop in that circuit.

So chose a pump that moves 2.4 gpm at 3.5 feet of head, ideally in the mid range of the pump curve. The smallest 3 speed wet rotor pump you can find, maybe a Grundfos 15-58 on speed 1.

If the tube is 12” on center in concrete with no floor covering expect around 23 btu/ sq ft of output.
Click to expand...
 
Status
Not open for further replies.

Similar threads

M
Pump speed when it gets cold
Replies
16
Views
2K
The Boiler Room - Wood Boilers and Furnaces
duramaxman05
D
DerrickH
Wood Furnace ducted into with Heat Pump and unsure of dampers
Replies
9
Views
1K
The Boiler Room - Wood Boilers and Furnaces
brenndatomu
brenndatomu
J
Taco 009-F5 oversized for my application?
Replies
13
Views
847
The Boiler Room - Wood Boilers and Furnaces
Joacchim
J
RowCropRenegade
Pump got fried.
Replies
0
Views
433
The Boiler Room - Wood Boilers and Furnaces
RowCropRenegade
RowCropRenegade
P
Push water towards boiler or pull from it ?
Replies
9
Views
2K
The Boiler Room - Wood Boilers and Furnaces
chew72
C
Top Bottom
Back