Wood burner with in-floor heat

gpjazz

Hello all, I am new here, but I've been in the hvac business for 22 years. I am a member at HVAC-Talk, and I've read a lot of stuff from Dan Holohan. So figured I'd post something here to see if I can get some input. This is regarding a system / set-up that I have not done before...
I got a project my customer asked me about and I'm trying to think it through. I was asked to hook up and commission a system where majority of the components are already installed, or on-site to be installed... But I'm having some trouble thinking about the proper way to do this. I've never installed this type of system before. So any advice would be appreciated.
The customer has built a garage with in-floor radiant. He has an outdoor wood burner to be used specifically and only for the garage heat. It's a Central Boiler model CL 6048.. The wood burner is way bigger than necessary for this 50x80 garage, but that's a different story.
All tubing in installed in floor and concrete is complete. Outdoor wood burner is already in place, and supply and return tubing is installed from wood burner to garage. Customer has manifolds and all fittings on-site, he has the necessary circulator's, flanges, and also a mixing valve. He even has a thermostat and a zone control. This is a single zone by the way...
So my issue is, how is this system supposed to be hooked up? To my understanding, the outdoor wood burner is not a sealed hydronic system and no regulated water pressure is to be applied to the wood burner. It just gets filled with water and circulates as needed. I assume the wood burner has a water reservoir of some sort and is open to atmosphere. But the wood burner is installed at an elevation lower than the garage to be heated. So this right here is confusing me to start. How is the water gonna stay in the system when water will find it's own level? Won't all water eventually feed back to wood burner and overflow?
As far as the radiant floor system, is this set up required to have some type of a heat exchanger between garage and wood burner to isolate the systems and have a constant regulated pressure to the radiant? Looking thru the manual of wood burner, this is NOT what they are suggesting. But some of the information from wood burner manual is vague and un-clear. It looks to me that the manual says to pipe wood burner directly into radiant floor system, with the use of mixing valve to maintain temperature to floor, and also a thermostatic valve to maintain temperature of wood burner.
But I'm still confused with this whole idea. Again, the floor of the garage is at an elevation higher than the wood burner (which is open to atmosphere)... So how is the water gonna stay in the system and how is air gonna be eliminated???
Any help or information will be appreciated. Any drawings to reference on piping diagrams?
Thanks for your time

Jamie Hall

You've got a point there, judge. The water will try to come back to the boiler and go away. However, there is a way to solve the problem. First, you pump away from the boiler, and on the outlet to the pump, between it and the radiant, you place a check valve. In fact, I'd use a double check assembly. To keep the water from flowing back through the pump when the pump is off. Then, at the other end of the radiant system, you arrange a high point -- go up the wall, maybe, and then over a bit and down. Then back to the boiler return. On that high point you put a vacuum breaker. Now when the pump shuts off, the water in the return line will flow back to the boiler, but that shouldn't be all that much, but the vacuum breaker will prevent the rest of the water from being sucked out.

I hope everything is corrosion resistant... you'll have a lot of oxygen in there to work its mischief.

I also hope that you have an arrangement for a heat dump of some kind -- wood boilers just don't turn off that fast, and one that big is going to overheat pretty badly pretty regularly.

gpjazz

Thanks for the quick response. In piping it that way, are you suggesting that the radiant floor side of the system does not have to be pressurized and isolated from the wood boiler?
All piping/fittings would be copper, brass, and pex tubing. No black iron anywhere.
And I'm not familiar with the term "heat dump"... What are you referring to with this?

Jamie Hall

No, I don't think it has to be pressurized and separated -- although, of course, it could be. The fittings should be OK. How about the pump?

A heat dump is, simply put, a place -- often controlled by a thermostatic valve -- where the system can get rid of excess heat until the boiler calms down. Could simply be a large tank. One would pipe it so that if the boiler got carried away -- it will -- you could circulate water to and from it to keep the boiler cool (relatively speaking).

hot_rod

It is not unusual to see the heat emitters above the water level of the un-pressurized OWF like yours, think multi story homes.

When the pump turns off that section above the water line in the OWF will be see sub-atmospheric conditions, negative pressure.

At some point, high temperature and negative pressure water will boil below 212F (sea level).

The water will stay in the piping above the OWF as long as no auto air vents are installed.

I'd rather see a plate HX to isolate the non-pressurized system from the radiant. This allows the radiant to have ferrous components, and auto air vents.

Since the OWF is open to atmospheric rust and corrosion never sleep, no sense in pumping that into the floor tubing.

GroundUp

Put the pump right off the supply port at the boiler, pumping toward the garage. Once you establish a closed loop, it will stay that way regardless of elevation difference as long as the vacuum isn't broken. Run that pump 24/7 during the heating season. Many will run an atmospheric floor system as well because they're too cheap to buy an exchanger and parts to pressurize it. Bad idea. The 6048 is mild steel (unless brand new model) and WILL plug the floor loops with crap. Pipe in a heat exchanger. I do a lot of these, and as a matter of fact had one of those where the customer supplied all the parts that I piped just yesterday. I'll try to post a pic for reference

GroundUp

Yes, here is the one I piped yesterday. Customer already had the plate hung and connected to the OWB, that crooked stuff was him lol

GroundUp

Jamie Hall said:

You've got a point there, judge. The water will try to come back to the boiler and go away. However, there is a way to solve the problem. First, you pump away from the boiler, and on the outlet to the pump, between it and the radiant, you place a check valve. In fact, I'd use a double check assembly. To keep the water from flowing back through the pump when the pump is off. Then, at the other end of the radiant system, you arrange a high point -- go up the wall, maybe, and then over a bit and down. Then back to the boiler return. On that high point you put a vacuum breaker. Now when the pump shuts off, the water in the return line will flow back to the boiler, but that shouldn't be all that much, but the vacuum breaker will prevent the rest of the water from being sucked out.

I hope everything is corrosion resistant... you'll have a lot of oxygen in there to work its mischief.

I also hope that you have an arrangement for a heat dump of some kind -- wood boilers just don't turn off that fast, and one that big is going to overheat pretty badly pretty regularly.

While I would typically agree with Jamie, very little of this is accurate. If you let the water drain back to the boiler, it won't find its way back with a circ pump. Absolutely no vacuum breaker in these things. Central Boiler sends water treatment with all new stoves and requires annual samples and treatment to maintain warranty, corrosion is of little concern if they keep sending in their water samples and adding treatment like CB tells them to. No need for a heat dump in a system like this either. These things are natural draft and will never run away unless the damper sticks open, at which point you'd need a 500,000 BTU dump to keep up with the 6048's output.

hot_rod

keep the OWF pump as low as possible on the piping, give it as much head as possible.

Larry Weingarten

Hello, I'm curious about how many feet of what size tube is in the floor. It might be a very long loop!

Yours, Larry

gpjazz

GroundUp said:

Yes, here is the one I piped yesterday. Customer already had the plate hung and connected to the OWB, that crooked stuff was him lol

Looks nice, but where is the supply water inlet for radiant side? Ultimately I think a heat exchanger to isolate the system is the best way to go, but being that I've never done this type of set-up, figured I'd ask some questions...

gpjazz

Larry Weingarten said:

Hello, I'm curious about how many feet of what size tube is in the floor. It might be a very long loop!

Yours, Larry

I don't know the exact total length. But there are 9 loops altogether, 50x80 garage. Customer says they are spaced 12" on center, but I can't confirm or deny that. So a little math can estimate the total amount of tubing in floor...

gpjazz

I really appreciate all the help so far. I respect everyone's thoughts on this...
But it seems to me that the general consensus is that it's best to install a heat exchanger to isolate the floor from the wood boiler? I initially thought this would be best also... Am I correct here?

hot_rod

The heat exchanger isolation is the better way to accomplish this. It does take more components and $$. It would also allow a back up boiler or HP to be easily tied in if needed.

GroundUp

> @gpjazz said:
> Yes, here is the one I piped yesterday. Customer already had the plate hung and connected to the OWB, that crooked stuff was him lol
>
> Looks nice, but where is the supply water inlet for radiant side? Ultimately I think a heat exchanger to isolate the system is the best way to go, but being that I've never done this type of set-up, figured I'd ask some questions...

The supply water inlet for the radiant is the pipe that enters the supply manifold on the bottom. Hot water enters the mixing valve from the bottom and is mixed with return water from the top to create mixed water that is pumped into the floor

GroundUp

Keeping the pump at the lowest point is usually not correct unless the low point is at the boiler. Pump needs to be as close to the boiler as possible to prevent any vacuum/cavitation from the friction loss of the length of pipe previous to the circulator

hot_rod

> @GroundUp said:
> Keeping the pump at the lowest point is usually not correct unless the low point is at the boiler. Pump needs to be as close to the boiler as possible to prevent any vacuum/cavitation from the friction loss of the length of pipe previous to the circulator

correct, low on the open system OWF as shown in the first schematic

gpjazz

Yes, the wood boiler would be the lowest point in the system

hot_rod

gpjazz said:

Yes, the wood boiler would be the lowest point in the system

Good, also the pump mounted as low as possible on the ODF, bear the ground, pumping away from the supply connection

gpjazz

Ok sounds good. This all seems easy enough if the radiant side is isolated with a heat exchanger. I've done these many times.. Just never did a system with outdoor wood boiler.
Thanks for all the replies and information

gpjazz

So I guess I'm gonna move forward with this job and will be using a heat exchanger... What are some recommendations for the best type/brand heat exchanger to use ??

Zman

Any flat plate heat exchanger will do the job. Most manufactures have on line calculators to help size them.
What size and length pex is running from the Boiler to the house?

gpjazz

It's 1" from wood boiler to the garage. I don't know the exact length, but it's probably about 200 feet. Customer installed this...

hot_rod

gpjazz said:

It's 1" from wood boiler to the garage. I don't know the exact length, but it's probably about 200 feet. Customer installed this...

The 1"pex will be the bottleneck. So 400' of 1"? Here is a pressure drop chart, use the 180° column.

Also do you know what the load is, how many btu you need to transfer? Having some gpm or btu data is the best way to size a HX. Probably a 30- 40 plate 5X10".

SWEP is a top brand, Flat Plate another, plenty online sales also.

gpjazz

hot rod_7 said:

gpjazz said:

It's 1" from wood boiler to the garage. I don't know the exact length, but it's probably about 200 feet. Customer installed this...

The 1"pex will be the bottleneck. So 400' of 1"? Here is a pressure drop chart, use the 180° column.

Also do you know what the load is, how many btu you need to transfer? Having some gpm or btu data is the best way to size a HX. Probably a 30- 40 plate 5X10".

SWEP is a top brand, Flat Plate another, plenty online sales also.

I have not done the loads yet, and unfortunately I can only take the customers word on some of the information because everything is closed up already. But it is a 50x80 slab,, customer says he has 2"foam underneath with 5" concrete and 1/2" pex spaced 12"o.c. All metal building, metal finish inside and out with R19 in walls and R30 in attic. 14ft ceiling height. I still gotta check on the garage doors type and size for the load. No windows

hot_rod

So just a wild guess, 4000 sq ft X 22 BTU/ sq ft would be a 88,000 btu/hr load. So you need to move something like 8-9 gpm.

9 GPM thru 400' of 1" pex will be around 40' of head! (9.92' per 100' of tube)
Therein lies the problem with undersizing those OWF pex lines.

A high head circulator like that becomes very prone to cavitation when used in high temperature, atmospheric pressure OWF.

gpjazz

hot rod_7 said:

So just a wild guess, 4000 sq ft X 22 BTU/ sq ft would be a 88,000 btu/hr load. So you need to move something like 8-9 gpm.

9 GPM thru 400' of 1" pex will be around 40' of head! (9.92' per 100' of tube)
Therein lies the problem with undersizing those OWF pex lines.

A high head circulator like that becomes very prone to cavitation when used in high temperature, atmospheric pressure OWF.

Yes,, but like I said before, that part was all on the customer. He has the OWF installed with tubing to garage and he has the circulators purchased already as well. Whatever problems come about on the wood burner side of the system is on him. I agreed to "hook-up" the system using the parts and materials he supplies.
And I said I will handle the radiant floor side of the system as my responsibility. Customer is in agreement with this, so we'll see how things work out.
I'm just at the point now of deciding which plate heat exchanger to use...? What brand?

hot_rod

You will pay more for the name brands like SWEP, a 5X12- 40 plate for example.

Depending on where you like to shop. I've seen imports on e-bay for under 200 bucks.

All the online wood stove OWF sites have plate HX from mild to wild prices.

You have a quandary,
size the HX for what that OWF can actually produce, although I don't see a spec on Centrals website? 200,000 BTU/hr?

Or size it to what you can reasonably expect to shove thru that 1" pex. I'd say 8-10 gum size for the pex you have, could be 20 gpm or more for what the boiler can produce.

Without a load number on the building you will need to guesstimate, may as well go with a larger size, incase they add more buildings or upsize the poet someday So a 40 or 50 plate size is my WAG
Or go with what Central recommends.

One of many sites when you Google plate heat exchangers for OWF.
https://www.hvacbrain.com/swep/

gpjazz

Thanks for the response's

GroundUp

The 6048 is 500,000 BTU output, and the load in that building will suck the life out of those 1" lines regardless of circ. You'll cavitate the pump before you squeeze any more than 7 GPM out of a 400ft loop. If it were mine, I'd go with a 50 plate 5x12 to allow for some inevitable blockage down the road and keep the SWT to the floor as low as possible (like under 100 degrees) to keep the delta low and the load balanced. Are you installing the thermostatic valve that Central Boiler provides for boiler protection or are you skipping that part?

hot_rod

Or let someone else get saddled with a system destined to fail Data so far indicates it will suffer from ongoing and expensive pump replacement is addition to possible under performance

gpjazz

GroundUp said:

The 6048 is 500,000 BTU output, and the load in that building will suck the life out of those 1" lines regardless of circ. You'll cavitate the pump before you squeeze any more than 7 GPM out of a 400ft loop. If it were mine, I'd go with a 50 plate 5x12 to allow for some inevitable blockage down the road and keep the SWT to the floor as low as possible (like under 100 degrees) to keep the delta low and the load balanced. Are you installing the thermostatic valve that Central Boiler provides for boiler protection or are you skipping that part?

Yes, my intentions are to use the thermostatic valve Central Boiler recommends. Having never done a wood boiler before, I figured I'd try to stick with at least some of the mfg recommendations.

Zman

First off, you can tell yourself that what has been purchased by the customer is "on them" if you like. In reality, they will blame you and expect you to fix it.

You can compensate for much of this if you are careful in sizing the HX. First determine what flow rate you can get out of that 1" piping with the circ provided (you might also look at a "plan b" circ). Keep in mind that you likely only need 120 water to the slab so with the right heat exchanger, you could run more like 4 gpm at a delta of 60 and still get 4*60*500=120,000 btu through it.
Also be sure to find an HX with low head loss characteristics so you do not compound your 1" pex issue.

I usually use the Flat plate select sizing program.
Here is an example of what a program like that looks like.
BTW, the exchanger in the example would likely work for you...

EYoder

A large Delta t is your friend here. I wouldn't feel stuck on 20.

GroundUp

Keep in mind that the thermostatic valve will do anything in its power to keep boiler return water above 150 degrees, as they are designed to do. The 6048 has a horrible habit of the actual SWT being nowhere near the temp on the digital display and with the absolute max of 7 GPM you'll flow through that loop, the mixing in the 393 gallon water jacket will be poor to horsesh*t. So let's say the controller is set to 185 (which is about max, anything higher will boil above the loading door), the actual SWT to the building will be roughly 170. In order to keep RWT above 150, the TV will only allow 70,000 BTU through- and these are absolute perfect conditions. There will be a differential in the controller of at least 5 degrees, so that drops to 52,500 usable BTU at the bottom of the cycle. A band-aid for this is to pipe another circ, or "shunt pump" just between the extra set of supply and return ports on the boiler to aid in mixing the water jacket which will tighten the delta between the controller and SWT closer to 10 degrees instead of 15 and may allow slightly higher temp setting without boiling due to the better mixing.

Basically this is destined to fail as-is. The building will fail to heat in extreme cold if you do install the TV as Central recommends. If you skip the TV, the building will heat but the boiler will shock itself to death and rot out in a few years due to condensation in the flue. The only way to make this right is to bury another set of lines that'll carry enough heat. No matter which way you go about it using the existing underground, the failure will be your fault according to the customer. I've screwed with dozens of these 6048s and their crappy installs (usually recommended and/or performed by the dealer) and know enough about it to recommend you just walk away and let someone else screw it up.

I should also add that unless the underground lines are the green Thermopex with the closed cell foam that Central sells, there will also be drastic ground loss between the boiler and the building, 3-5 degrees at 7 GPM (this is assuming the circ he provides will even allow 7, which is doubtful) which further reduces your available BTU

EYoder

Two other ideas I had-
Put the flat plate in the OWB to keep the OWB loop short. Or pipe it P/S in the OWB so the Delta t can be whatever it wants to be on the underground lines.

gpjazz

Now with all the comments regarding the destined failure, I'm getting nervous. But if I proceed and use a heat exchanger and plumb in all of the in-floor radiant side of the job, at least that will be done without any issues. As far as that goes...
Now, if the wood burner side of the system is gonna fail in some way, then the customer will just have to pay the price to correct all the issues.
So my question now is... Would there be any reason why the garage side radiant would need any changes/corrections made?? Aside from not heating properly due to lack of btu / gpm from the source, I don't think any of my work in connecting the radiant side of the system would be an issue...

hot_rod

What is the pump that is included in the OWF? With a bit of info we could crunch some numbers and see what you could reasonable get from that unit with the line size you have.

If you can run ups around 170- 180, and the slab only requires 100F of so, an injection mix from the OWF to the shop may work, A load calc on the shop would tighten up the numbers.

You are stuck playing with the hand you are dealt, you and the owner need to be clear on what is expected.

Seems odd that a dealer would sell a 500K unit with those long runs of 1" pex, for a building with maybe a 80K load. Maybe that person should guarantee performance.

Gordy

Been following along. I think a lot of written CYA documentation is in order, with concerns, and recommendations before proceeding. Not verbal. Things change if a verbal communication is agreed upon, and things do not turn out as the owner was hoping.

gpjazz

hot_rod said:

What is the pump that is included in the OWF? With a bit of info we could crunch some numbers and see what you could reasonable get from that unit with the line size you have.

If you can run ups around 170- 180, and the slab only requires 100F of so, an injection mix from the OWF to the shop may work, A load calc on the shop would tighten up the numbers.

You are stuck playing with the hand you are dealt, you and the owner need to be clear on what is expected.

Seems odd that a dealer would sell a 500K unit with those long runs of 1" pex, for a building with maybe a 80K load. Maybe that person should guarantee performance.

I'm not sure exactly what pump he has for the OWF at this moment... and I'm also not sure how/why these materials were selected. I don't believe the pump was "included" from central boiler, rather I assume whoever sold the equipment to the homeowner selected the pumps. Pretty sure the homeowner had this material for a couple years already and just never got to building the garage until now.

gpjazz

I ran the load on the garage with Wrightsoft and I came up with 56,414 btu load for the space