Korean floor (hot air radiant)

classicform

hello...I recently purchased a small brick house in Kalamazoo, Mich. that has a very unique heating system and am hoping for some help (I was referred here from hvactalk)

1000sqft home built in 1945. Dual pane fixed glass windows along the south side of the house with the north, east and west mostly brick. Early passive solar design.

The heating was designed as a radiant heat system utilizing hot air circulating in the hollow terracotta floor tiles through the whole house. I've heard this called a "Korean" floor. Not used anymore as Far as I can tell. Maybe a bad idea from the get go. Regardless, it's what I have and what I am wanting to make operational again.

There are 3 return air ducts under this tile that dump into a cement "pit" that has a blower (1800cfm) to circulate air across a heating source.


At some point the original system was converted to a forced air type system utilizing a conventional furnace and creating floor registers by breaking through the floor to allow the heat to escape.

I want to convert back to the original system and this is where my questions come in.

I have had a heating company design and quote a system but I'm feeling like maybe there is a better solution than what they came up with, another company will be out Monday to offer their take on it.

The quote I have in hand calls for a Navien ncb240e tankless to supply dhw and hot water for a heat exchanger to be installed in the ductwork to heat the floor air before it is circulated.

I'm a litttle concerned with the tankless system short cycling due to the amount of temp drop of the returning water not being large enough. The company says that it will modulate down and will not be a problem. Also not so sure about the Navien unit itself...not so good reviews here and there around the web.

Any thought on what to use here? Maybe too vague in my description? I'll add a print if I can...

j a_2

Korea used ondel heat under the floor of the hooches it was a charcoal cylinder shaped bricket..and it heated the torazo floor nicely....but many GIs got carbon monoxide poisoning back in the late 60s and early 70s from sleeping in the huts ..they used a big set of thongs to change it out....man was it cold over there....the heat traveled with the smoke. You do not have a true ondel heat system. But maybe an offspring of it
..I love radiant heat...more than steam. Google ondel heat....very interesting

BobC

I've slept in ondel heated huts on a mountain pass at the 3,200 ft level. It was -30 outside but that nice warm floor kept you cozy. Like JA said it had to be done right or people died.

Was this designed to be heated with some kind of a furnace or was it meant to heat with combustion byproducts - burning wood or charcoal? I would be very leery if it's designed to use combustion byproducts because the chance of CO poisoning is very possible.

Bob.

classicform

Bob, it was designed for a furnace of some type. I don't know what was there originally. Assuming oil due to the year built.

Zman

Your project sounds super interesting.

Your concerns about a short cycling boiler are quite valid. I would be very surprised if the home has a heat loss of more than 25,000 Btu/hr on the design day. With a max turndown of 18k that boiler will short cycle most of the time.

You also want to be sure your water to air heat exchanger is oversized in order perform well with the low return water temps needed for maximum efficiency in a condensing boiler.

What are your domestic water needs?
You could do a small condensing boiler with an indirect.

j a_2

BobC said:

I've slept in ondel heated huts on a mountain pass at the 3,200 ft level. It was -30 outside but that nice warm floor kept you cozy. Like JA said it had to be done right or people died.

Was this designed to be heated with some kind of a furnace or was it meant to heat with combustion byproducts - burning wood or charcoal? I would be very leery if it's designed to use combustion byproducts because the chance of CO poisoning is very possible.

Bob.

Hi Bob, Where, in Korea were you. I was Osan and Kunsan 1971. Got to say I would not trade those times for anything. At 18 it opened my eyes, some good some not good...I know the Nikie Sights up in the mountains was tuff duty...I seriously hope and pray for peace over there, for many reasons....

BobC

Jim,

I was at Salem top site in '69 which was a 6 man microwave site north of Camp Carrol and about 40 miles south of Taejon. I could look down on the city of Gumi from the back door of the comm shack. the site was on the peak and there was a small village a few hundred feet below us.


This image was found online and dates to about 2010, by that time the site was automated and unmanned

We had a great time up there, no officers closer than 26 miles. We kept the gear humming and they were probably happy to not know what we were doing. There was a Nike site due west of us which I discovered and was promptly told to forget about.

I found the people to be hard workers and honorable, we trusted them with our lives and were never disappointed. I was there at the tail end of the so called 2nd Korean war when grandpa Kim was causing no end of problems (lot of men killed on the border) while we were distracted by 'nam.

Bob

classicform

Thanks for the comments guys.
My next dilemma with this proposal...
The fan to move the air out of the return pit will be entering another "pit" area with a door that opens to the 12x12" tile duct leading to the water/air heat exchanger.
While I dont have any experience in this type of duct work I do have some with dust extraction for large woodworking equipment and I know that even using a "T" connetion in a run will reduce cfm by about half as opposed to a "Y" connection.
In the image you are looking into the pit that the blower "sees". The door is to the right with the clay tile slightly showing.
How in the heck can I expect to keep 2200cfm here?
Won't it just blow into that pit/room and just kind of stop due to turbulence?

j a_2

> @BobC said:
> Jim,
>
> I was at Salem top site in '69 which was a 6 man microwave site north of Camp Carrol and about 40 miles south of Taejon. I could look down on the city of Gumi from the back door of the comm shack. the site was on the peak and there was a small village a few hundred feet below us.
>
>
> This image was found online and dates to about 2010, by that time the site was automated and unmanned
>
> We had a great time up there, no officers closer than 26 miles. We kept the gear humming and they were probably happy to not know what we were doing. There was a Nike site due west of us which I discovered and was promptly told to forget about.
>
> I found the people to be hard workers and honorable, we trusted them with our lives and were never disappointed. I was there at the tail end of the so called 2nd Korean war when grandpa Kim was causing no end of problems (lot of men killed on the border) while we were distracted by 'nam.
>
> Bob
Bob. Thanks for responding to my post,. And thank you for your service...I as well as you I am sure are praying for a peaceful end to the going on over there. They treated us with the up most respect....

classicform

I'm baaaack..
System has been installed and it is woefully inadequate.
Cant get temp of house above 60 degrees on a 25 degree day.
Guessing that I dont have enough cfm as my initial concerns have been verified, installer states he is only getting 500 or so cfm across the coil/heat exchanger that is circulating water from the Navien.
Some of the floors in the shorter supply runs of the house (lower sqft area to heat) will heat up nicely but the longer run with more sqft and longer return run does not.
Do you think that bypassing the the "pit" area and running duct directly to the coil will help? I see no other way...

Thoughts?

DZoro

How big is the water to air heat exchanger. Like Zman said you will need to over size it. I would be guessing/ estimating a 24" x 24" 4pass heat exchanger at minimum. Understand it's very high in BTU capability, but your really after surface and flow, similar to a heat pump design.

classicform

Its 18 x 20 x 9". There are size limitations due to the area its in however. 18" is about as wide as it can be. Plenty of room to have one longer though, 36" at least.

Assuming stacking two in series (of the same current size) isnt going to help matters?

DZoro

Going longer would be a option. Yes you will need to assist/draw the air through the coil, at or near the coil. Sorry I didn't quite understand how the coil set up was. Will the design allow you pull through the coil with another blower. Some more pictures would help.

mikeg2015

Do you have info on the original blower that was there?

First off, we have one good data point. At 25F OAT you are bout 60% to design conditions or you location. are achieving 60F, maintain 88% of heat loss (60F indoor temp when 25F outside) . SO the system is delivering about 52% of the capacity needed. So increasing to approx. 1000CFM with the same discharge temp will get you there. That's a LOT of CFM for you heat loss.

A Unico high velocity might work best in this application. It has a a oversized 4 or 6 row coil and can deliver up to 1.5" static pressure. The 3 ton model delivers up to 900CFM which with a low approach should get what you need with probably 700-800CFM with better efficiency.

I'll bet unico might work with you on this unique application.

mikeg2015

Thinking more, The best application for this system would be to have a direct fired burner at the plenum, then place Inducer fan at the return point, and exhaust from there using Stainless steel pipe. Then have a PVC pipe for combustion air and draw a negative pressure through the clay pipes to draw in the combustion products. You set the secondary air ratio and flue temps to prevent condensation, so target maybe 160F flue temp for around 86-88% inefficiency. Although clay tiles can handle some condensation. The challenge is evenly distributing the heat. You really need a 2 pass design flow pipe design for it to work well.... meaning pipes go out and come back so the hottest and coolest pipes are adjacent to each other.

Jamie Hall

First -- yes, the design was and is used in Korea. It was also used in the Roman Empire and before. It's technically a hypocaust, and it's a very very old concept. It is intended for hot air (the Roman designs used a fire). For a modern application, I would use a forced air furnace -- any fuel -- using a low speed high volume blower. Feed through some of the "ducts" formed by the tile, and return through others.

classicform

Thanks for the input guys. Here are a few pics of the install, the original blower, and the new blower. Yes, the ductwork they did is about the worst I've seen.
The blower/return air pit is about 10-15 feet away from the coil.
It blows into that cement chamber with the bypass doors and return air door that I showed above in earlier post.

DZoro

You are correct on the bad duct work, is that 10" round? No wonder you only have 500cfm. You are going to need proper sized air duct to get to your 2000cfm flow. How far away is the blower cabinet from the coil? Which side of the coil is it on, blowing into or away from the coil.

classicform

Blowing into the coil. Approx. 10 feet away. Limiting factor is that the supply to the round ductwork you see is 13 x 13 clay tile.
So, from the blower, air goes into the large cement chamber shown...through the open door to the 13 x13 clay tile which feeds the round duct to the coil.



I was thinking ductwork directly from the blower (bypassing the cement chambet) to the clay tile. That should increase cfm considerably dont you think?
Blower is capable of 2000cfm.

Jamie Hall

A few further thoughts on hypocausts -- like your floor. Perhaps the most important is that they are slow responding. Like, very very slow. They are not intended to vary in temperature much with time. Don't even think of a setback! They are, by nature, even slower than a hot water radiant floor in responding.

It is very important that the air flow be even across the floor. The uneven heating you are experiencing is due to uneven air flow through the entire floor. In your original plans, duct work is shown directing heat to all the channels. As shown, if each of the ducts is fed equally (there are five of them) you will get very even heat across the floor, with the utility room/kitchen area getting slightly less.

You need to recreate that ductwork as closely as possible. I have no idea what has been knuckleheaded in this thing (shades of steam heat1) but until you get an arrangement which directs equal airflow to each of those five ducts, it's not going to work properly. It appears that the original design had a very large plenum which fed the air into those five ducts.

Somehow I very much doubt that your pit arrangement is going to do the job, regardless of how many cfm you put into it.

Keep your velocities down.

I have, by the way, lived in -- and built -- solar boosted houses with a very similar concept. They work splendidly -- when they are done right.

Jamie Hall

I might add, speaking of knuckleheads, that if some bright child somewhere along the line combined those five ducts into, say, two (one going each way) you may be out of luck unless you can separate them again properly.

And again I will note: that plenum feeding them either has to big, with very low velocity in it, or very carefully designed to get equal flow into each duct.

I would also note that, done right, you will find that the solar gain on the floor will probably give you upwards of 75 percent of the heat you will ever need. Just don't put rugs onto that ceramic tile floor -- that's your heat collector.

Zman

Jamie nailed it. Just because the fan says 2000 cfm, that means nothing unless the static of the duct is sized properly.
If you post a sketch of the original duct sizing it may be possible to reverse engineer the original design.
Also, more pictures of the original fan would help it is hard to tell from the picture but it may be designed for higher static than your new one.

classicform

Oh, when the sun is out and shining into the house (wall of double pane windows due south) the furnace doesnt even turn on. It can reach 70 with ease, even on a below freezing day.
Problem is it's pretty gray in Mich. in the winter:)

So you are recommending keeping velocity down? I guess I dont understand the difference between velocity and cfm (even after googling).

There are four runs btw, not five. Regardless, there is not even flow.

Zman

It looks like your new fan is a ECM designed to run a system with total static between 0.1" and 0.7 ". It does have some different speed settings. Have you adjusted them?

DZoro

A simple look at low and high Velocity is look at a garden hose with a spray nozzle on and 5 gpm. and then take the nozzle off with the same 5gpm flow. Picture the difference? In your system you do not have a nozzle, but that ductwork created a nozzle and it wont work. Need to remove that restriction and recreate the flow without the nozzle.

classicform

Zman, the installers have been out numerous times making varying adjustments. They have tried low and high speeds (I think now it is on high). They are not communicating with me very well, it's like pulling teeth, so I'm not positive.

Dennis, thanks for the explanation. Unfortunately I can't increase size of the tile duct to the coil so I guess I'm stuck with a nozzle. It was designed that way though however. It was built differently than the plans show but I can't imagine it was changed without the architects input (or possibly changed by the architect during the design phase of the house).

mikeg2015

I wish I had a better sense of how the plenum is configured. Any way to add 5 ducts going to each clay tile then add a damper to balance the flows?

I'm not quite sure why you need a ton of CFM. you only need maybe 30k BTU, 700-800 CFM should get it done. More airfow just wastes energy.

Also I'm not sure the Navien vents are piped correctly. You can't have that many elbows with 2" depending on the total length of pipe. Hard to tell from the photo. They sometimes have a minimum straight pipe requirement off the top too.

Jamie Hall

@Zman asked if you have a sketch or as-built of the ducting along the north wall (and, for that matter, the return collector along the south wall). That would be very very helpful for us to figure out how to set this thing up.

I suspect that where your installers are having problems is that they are accustomed to conventional forced air systems -- which this arrangement is not. The objective here is not to move the room air around and through the furnace. The objective here is to move air -- quite a bit less of it -- through the floor to distribute the floor heat evenly, and to supply a limited amount of makeup heat (as @mikeg2015 pointed out, probably not much over 30K BTUh) to the floor when the sun doesn't do it's thing. In this way it is much more like a hot water radiant floor, with the only real difference being that you are moving air rather than water.

It is not a difficult design -- no a particularly difficult concept -- but it is certainly not a conventional forced air design, either, and your guys may not understand it (and no blame to them -- it's unusual and I doubt they've ever seen the like).

classicform

I'll drawsomething up tomorrow afternoon. Until then here are some missing portions of the print (I just noticed were cut off)

Jamie Hall

Ah good. If it really was built that way, and all the ducting does feed from or return to those big plenum chambers, it should work very nicely. You may have to adjust the manual dampers slightly to get even heat, but you shouldn't have to change them much.

That said.

Note the dimensions of the plenum chambers. They're big. They're also none too big. It is essential that your air be fed into, or drawn from, chambers of at least that size, and the ducts leading into or drawing from the two chambers should be kept away from the under floor ducts, and not directed at the underfloor ducts. Also, the area of the return or supply main ducts must not be less than three quarters of the total area of the return or supply ducts under the floor. Bigger -- and slower -- is better.

classicform

More info. Hope this helps
Appears the plenum on the suppy side of the coil is not being utilized (opened up with round going through it)

Jamie Hall

Well... it would have worked, had it been done the way it was intended... with a cute little metal duct in it, unlikely. What is that duct supposed to do?

I have a further thought as well. When it was designed (and probably built!) there would have been a thermostat to turn on the circulation blower at some moderate space temperature in the south room -- probably around 75 -- to circulate air in the duct system to transfer heat from the floor in that room to the floors in the rest of the house. I'll bet it's not there, but it needs to be.

classicform

Thank you again for your thoughts and insight on this. As it is now, the fan runs continuously, even without heat call.

Jamie Hall

classicform said:

Thank you again for your thoughts and insight on this. As it is now, the fan runs continuously, even without heat call.

That's OK, too

classicform

So going forward with this.
What would be your plan of attack?

Re-construct the plenum area on the supply side (on the outfeed side of the coil)...maybe dropping the coil down to a lower position to allow added room for an even larger plenum (albeit "L" shaped)?

Does the coil size need to be increased as well?

In essence, then, the blower in the return pit would be "pulling" the air through the floor and loading up the two plenum chambers (one large cement with the damper doors before the coil and the other a sheet metal plenum on the supply side of the coil).

Jamie Hall

I can't guarantee it without seeing it -- but that sounds like a good approach. The blower really doesn't care whether it's pulling or pushing.

The coil size is set by how much heat you expect to get out of it -- as has been noted earlier, 30K may well be ample. What you do need to do, though, is make sure that there is enough room, and an expansion, so that you have low velocities in the plenums themselves. This may take some creative sheet metal work...