Help choosing a circulator

[kuribo]

I am trying to figure out what sort of circulator would be best....It will circulate from the boiler to storage and back, using a thermostatic valve.

Boiler is about 60 feet from storage. Lines are 1-1/2" pex. I calculate roughly 8 gpm and a ~10 foot head when the return from supply is at 140F (50f delta T) and roughly 20 gpm at a 24 foot head loss when the boiler is coasting down and the storage tank is nearly up to temperature (boiler output 50,000 btu/hr, delta T at 5F).

I understand that I need to choose a pump that will output 20 gpm at 24 feet of head, but what happens when I only need 8 gpm? Do I need a variable speed pump?

Thanks.

 

[EffectaBoilerUser (USA)]

There are basically two options for this scenario:

Purchase a separate thermostatic loading valve (Danfoss or ESBE), circulator, ball valves, unions etc.
or purchase a complete, insulated loading valve such as the Laddomat.

The Laddomat is a very nice unit and comes in a very nice package (see attached photos).

It is normally available in two sizes: 21-60 and 21-100.

Most use the 21-60 for 25 and 35 kw boilers and the 21-100 for 60 and 80 kW boilers.

The circ. pump on these units are 3 speed models and thus you can select the speed that best matches the application.

These units have 3 gages to monitor the 3 water temps and also have a disc valve that allows the boiler to thermosiphon in the event of a power outage our pump failure.

If you compare the costs of both solutions the Laddomat is slightly higher but gives you many features that a separate loading valve scenario cannot.

 

[kuribo]

I have looked into those but the pumps are too small I think. I need about 20gpm at 25feet of head. My storage tanks are 60 feet away from my boiler.

Thanks for the reply.

I think I have found the error in my assumptions that simplifies this...

I was assuming the boiler would output 180f no matter what the supply temp when in fact I should be assuming a 20 delta T through the boiler. If I assume a 20 delta T through the boiler, then at 140F return temps, I need 20 gpm. When the tank temp nearly reaches full temp, the boiler output will be low, so I will still need those same 20gpm. I don't need to vary flow rate.

 

[JP11]

i don't have my figures here.. but sounds like you are very close to what I had for flows and head.

I have a taco 0011

 

[kuribo]

I redid my calcs and it seems 20 gpm at 20 feet of head is more like it....That means I might be able get by with a taco 13 or a grundfos UP26-116F...

 

[henfruit]

How do you read the temps on the one in the pictures?

 

[Blue Tornado]

Hello Chris:

I too am in the design stage and considering pump options. What I have not seen mentioned in the posts I have read is; should we consider a slower flow?

In the EKO 25 which I am installing, the boiler capacity is 20 gallons. At a flow of 8 gpm, the boiler would do a complete replacement in 2-1/2 minutes. Is that enough time for a 85,000/btu/hr stove to sufficiently heat 20 gallons by 40 degrees? 20 gallons times 8.2 pounds times 40 degrees equals 6560/btu every 2-1/2 minutes equals 157440/btu/hr, that is almost twice the capacity at optimum output.

In my thinking, too fast too soon will keep the boiler cool. So we take a new fire and use the boiler loop to get boiler up to say 175. Now we have the storage pump set to go at the 175 and it begins. In 2-1/2 minutes on the EKO 25, the return from storage replaces the 175 degree boiler volume with say 140 degree water. That 140 is probably warmed a bit from the hotter boiler but not up to the 175. Storage pump shuts down and boiler pump begins again. Cycle, cycle again, and again, and again.

My planning at this point is to have two pumps to service storage, one being a lower flow to charge the first 75% and the second to be higher flow and controlled from a sensor at the 75% mark. With the first pump flowing maybe 4-5 gpm and then with the final 25% both pumps together giving maybe 16-18 gpm.

Does this theory hold any water at all?

 

[kuribo]

Talking to the people at Econoburn, they tell me with the 200 to design with a 20F delta T through the boiler. If the boiler is putting out those 200,000 BTU/hr, then with the 20F delta T, I need:

f= 200,000/(500)(20) = 20 gpm

When the storage tank is nearly up to temp and the boiler is nearly out, my assumption is 50,000 BTU/hr output with a 5F delta T:

f= 50,000/(500)(5) = 20 gpm

So, in my case, unless there is something else here I am missing, a 20gpm flow rate should do it. I thought about using two smaller pumps but decided against the added complexity.

 

[Blue Tornado]

f= 200,000/(500)(20) = 20 gpm

Chris, I don't know the formula you are using (I am not a heat guy) and am wondering what the 500 is about.

 

[ewdudley]

...am wondering what the 500 is about.

(8.333333 lb per gallon ) * (1 btu per lb H2O raised 1 degF) * (60 minutes per hour)

 

[Blue Tornado]

And the 500 is? Gallons? Pounds? Turkey legs? Btu's? My apology for being so dense.

 

[ewdudley]

And the 500 is? Gallons? Pounds? Turkey legs? Btu's? My apology for being so dense.

We have:

(8.333333 lb / gallon ) * (1 btu / 1 lb H2O degF) * (60 minutes / hour)

Put all items on top together and likewise all items on bottom together:

(8.333333 lb 1 btu 60 minutes) / (gallon 1 lb H20 degF hour)

Combine constants, cancel out the lb's:

((8.33333 * 1 * 60) btu minutes) / (gallon H20 degF hour)

Which leaves:

(500 btu minutes) / (gallon H2O degF hour)

Or if you prefer:

500 btu-minutes per gallon-H2O-degF-hour

Also known as:

4180.744 joules per liter-degC-H2O

 

[Blue Tornado]

Thank you for the follow-up.

Using this formula, Chris has shown that a single speed pump is all a system needs to get the btu's to the storage by reducing the output of the boiler.

Case closed?

 

[mikefrommaine]

I'll add that I have my storage about 60' away, round trip of 120' piped in 1.5" black iron. I calculated 20gpm at 9.5' for my system. I went with a grundfos 26-64 witch matched up just right according to the pump curves. And it moves just enough water to keep the boiler from idleing. A case of real world agreeing with the numbers.

 

[kuribo]

I will have my boiler about 50' from the house, and another 20' from the tanks. I will be using 1-1/2" pex and I have about 16 elbows, 8 ball valves, thermostatic valve, air separator, etc., in my circuit. You must have a pretty straight shot to have such a low head loss....

I have been looking at the taco 2400-20. Seems they make a model of this pump specifically for outdoor wood boilers so it is set up for high flow at high head.

 

[ewdudley]

A big pump shouldn't be required. IIRC your system is all in-floor radiant, which means your return from storage will be nice and low. An 80 degF deltaT should be easily attained.

This means your flow to storage would be about 5 gpm, assuming the boiler actually puts out 200000 btu per hour on a good testing day.

The trick is to avoid the n-laps-through-storage problem where you heat storage from 100 to 160 at 6.66 gpm to storage on the first lap and then go from 160 to 180 at 20 gpm to storage on the second lap, and then try to get another 10 degrees by limiting boiler output.

If you go from 100 to 190 on the first lap all you need is a maximum of 4.44 gpm to storage on the first and only lap.

Consider using a small 15-58, 007, Alpha, or Stratos for a boiler recirculation pump, which would give you 18 gpm or so through the boiler, then set up a variable speed ECM pump to inject into that loop with a control that only sends water to storage heated to the exact temperature you desire. The injection pump would only need to pump 4 or 5 gpm to storage to keep up with the boiler. If you plumb it right the flow from storage will add to the total boiler flow giving you in excess of 20 gpm for the 20 degF deltaT through the boiler that EconoBurn seems to want.

 

[kuribo]

The folks at Econoburn tell me the boiler is best configured for a 20F delta T. At full output, that means 20gpm from 140F up to say 170F, then as the boiler output is falling, the delta T is falling, and the flow rate should be able to stay around 20 gpm as the tank coasts to full temp. At least that is how I understand it.

It sounds a bit complicated to me and more expensive with the multiple pumps. Can you point me to a schematic?

Not understanding what a "lap" is exactly???

 

[ewdudley]

The folks at Econoburn tell me the boiler is best configured for a 20F delta T. At full output, that means 20gpm from 140F up to say 170F, then as the boiler output is falling, the delta T is falling, and the flow rate should be able to stay around 20 gpm as the tank coasts to full temp. At least that is how I understand it.

It sounds a bit complicated to me and more expensive with the multiple pumps. Can you point me to a schematic?

Not understanding what a "lap" is exactly???

Like lap around a track, 1600 meter is 4 laps around a 400 meter track. If all the water in storage goes once around through the boiler I'm calling that one lap.

 

[kuribo]

wouldn't it be rather unconventional to task the boiler with a 90F temperature rise?

With 1000gal of storage, I would need roughly one load of wood (110lbs) to raise the temp from 110F to 190F. At 20gpm, that would be about a "lap" an hour, for a total of roughly 3 laps....That is assuming no load is being drawn from the tank while it is being charged, which is probably not realistic...

 

[bioman]

Do I need a variable speed pump?

I would say yes to the variable pump, thats what i use & I have more control over the system. Though every setup is different. Hope this helps.

 

[DaveBP]

A case of real world agreeing with the numbers.

The real world always agrees with the numbers once you have the right numbers.

 

[Coal Reaper]

Somebody here needs to play with the bumble bee before i need to bite on one.

 

[ewdudley]

wouldn't it be rather unconventional to task the boiler with a 90F temperature rise?

With 1000gal of storage, I would need roughly one load of wood (110lbs) to raise the temp from 110F to 190F. At 20gpm, that would be about a "lap" an hour, for a total of roughly 3 laps....That is assuming no load is being drawn from the tank while it is being charged, which is probably not realistic...

For example: 4.7 gpm comes from storage at 100 degF. It mixes with 16 gpm of recirculation flow at 185 degF. Mixed flow of 20.7 gpm enters boiler at 165.7 degF. Rise through boiler is 19.3 degF, approximately the 20 degF required.

4.7 gpm leaves the top of boiler and goes to top of storage with deltaT of 85 degF, having absorbed heat at a rate of 200005 btu per hour. The other 16 gpm goes back to the return port to mix with the 4.7 gpm coming from storage. The boiler does not see the 85 degF deltaT, it sees 19.3 degF deltaT.

I don't see how it would be expensive, it can be done with two $75 conventional circulators, e.g., a 007 or 15-58, plus a $40 PID controller and $12 RIB, and no mixing valve. You'd be drawing about 1.1 amperes on average as opposed to 1.9 amperes with a $255 2400-20. Spare parts inventory would be $75 + $40 + $12 = $127 as opposed to $255. You do plan on a complete critical component spare parts inventory, don't you?

With a correct design and nice low return temperature coming from in-floor radiant, there is no reason for flow from storage to boiler and back to ever to exceed about 5 gpm with a 200000 btu per hour boiler.

 

[kuribo]

With 1-1/2" pex from storage to the boiler, I think I need at least 8gpm...

Low flows at a very high temp versus high flows at a moderate temp.

When the tank is at 100F, one pump is circulating 4.7 from storage and the other is recirculating a total of 20gpm. As the storage temp climbs, I will either need more and flow from storage injected into the loop or more and more flow through the boiler recirc loop. As the storage temp reaches 165F 2/3 of the way into the burn, if I keep the flow from storage constant at 4.7gpm as you suggest, the mix temp would be about 180F and the flow through the boiler recirc loop would have to increase to around 40gpm, assuming at the 2/3 burn stage, the boiler output has dropped to 100,000btu/hr. Unless my numbers are wrong (good chance) a variable speed pump moving 20-40+gpm is pretty costly....

 

[ewdudley]

With 1-1/2" pex from storage to the boiler, I think I need at least 8gpm...

Low flows at a very high temp versus high flows at a moderate temp.

When the tank is at 100F, one pump is circulating 4.7 from storage and the other is recirculating a total of 20gpm. As the storage temp climbs, I will either need more and flow from storage injected into the loop or more and more flow through the boiler recirc loop. As the storage temp reaches 165F 2/3 of the way into the burn, if I keep the flow from storage constant at 4.7gpm as you suggest, the mix temp would be about 180F and the flow through the boiler recirc loop would have to increase to around 40gpm, assuming at the 2/3 burn stage, the boiler output has dropped to 100,000btu/hr. Unless my numbers are wrong (good chance) a variable speed pump moving 20-40+gpm is pretty costly....

If you fill the tank using water that is heated to the desired temperature (185 degF or whatever you choose) on the first and only lap then the will be no 'as the storage temperature reaches 165 degF' phase, at least not for long. When storage is filled with 185 degF water return temperature will rise rapidly to 185 degF and you're done filling storage. Heat all the water from 100 degF (or whatever) to 185 degF (or whatever) once and only once and be done with it.