EPA 2020 Wood Furnace Intertek Data Comparison

[NorMi]

I made up a quick reference comparing the main EPA hot air furnaces in the category by combining the tables from three furnaces. I did this so I could see where I wanted to put the most effort in when building my furnace controller and furnace shed project. I thought it might be useful for others to have all that data in one spot as well, so I made a screenshot of the spreadsheet.

It all looks as you'd expect for the most part. There are two anomalies on the Cat 3 burns that stand out. The first, on the Drolet, it has a higher delivered efficiency than stack loss (overall) efficiency, which I would not think possible, it seems implausibly high. I can't really explain how that happened, but it stands out, anyway. The second, on the Shelter, the Cat 3 delivered efficiency looks implausibly low relative to the other 3 tests. Normally, you'd expect a rising rate of delivered efficiency as burn rate increases, since the blower runs more often. On the Shelter, I'm guessing it is due to the poor controls. To hit burn rate, they leave the thermostat off until it falls below the target load rate, and at that point, the fire is burned too low to recover, which is probably why they only hit 53% of actual instead of the 75% target load- once the fire burns on low for that long, it's too cold to recover in time in that furnace. This is borne out by the equally implausibly low CO emissions for that test - the furnace was mainly coaling instead of burning for the last half if I had to guess.

The other thing that stands out is the Drolet achieves !!incredible!! delivered efficiency at the lower burn rates in particular. Stack loss is close (but not as good) as the Kuuma, but they are getting more heat out the plenum and less is being radiated. Doubly surprising given the firebox window... I think they must have done some fancy insulation work to make this happen. Kuuma is king of the overall burn emissions in general, although they suffer at the lowest burn rates - I think this is because they are able to burn at lower rates than anyone else in general, and delivered efficiency out the plenum just drops like a rock at the low BTU outputs (stack loss is still excellent of course). If they had burned to match the higher demand % of the others, they'd probably posted better overall numbers on that test. Keep in mind the differences in the delivered there, it's not majorly important in practice.

Attached a PDF so the data can be cut/paste more easily if needed. Linked the Intertek test PDFs, too big to attach. Drolet has several invalidated test failures, so you need to be careful when parsing that file, as some of the results are not passing test results, I picked the passing results for the table. Shelter data is from the most recent Oct. 2020 retests.

(broken link removed)
Drolet Heat Commander Report
Shelter SF1000E Report

These are nice guides, but not the end-all be-all. For one, they use HHV, which is basically sandbagging the achieved efficiency, because it pretends that wood moisture doesn't exist or that it's the furnace' fault... The LHV which is more realistic IMO comparing equipment to equipment, like for like brings all the posted numbers closer together - the real world differences are even smaller than they appear here.

 

[JRHAWK9]

You may want to read this about the Kuuma and Cat 1 delivered heat....heck delivered heat period. Believe me, it's not only 34%. It's hard to account for delivered heat when the blower is off.

(broken link removed)

Seeing I am capturing some of the radiant heat off the face of my furnace by pulling these BTU's into my return air, which raises the temperature of my return air, I am literally raising the delivered efficiency of the furnace. My return air is consistently 80-84° when I have a fire going.

 

[NorMi]

Yup, I've read that - one of the reasons I like the Kuuma guys, they have nice writeups on the furnaces and the tech. The delivered efficiency is not the efficiency of the furnace, just what can get "pushed around" via the plenum ductworks. And plenums and ducts radiate heat away as well, so the lower the BTU/hr, the more radiates away before the blower kicks on which triggers the datalogger, and low BTUs/hr even reduce what comes out of the end of the ductwork before it hits the outlet sensor on those artificial tests, due to a larger fraction getting radiated over that distance. It's not a knock on the overall efficiency or how much heat actually is kept inside the home. It's quite useful to measure and know, though! But, the lower the minimum BTU's your furnace can do (which is a GOOD thing), the worse it will "perform" on that particular test.

Another thing the CSA/EPA test methodology doesn't account for or consider at all is OAKs and the added efficiency you get from using those. It's more just a test of the various parameters, the most important imo are the stack loss and the emissions, both of which are hard to test accurately at home. Delivered efficiency is nice to throw in though, since you can measure that one at home easily and keep an eye on your furnace performance. A lot of the install specific stuff that would impact how well the furnace really works in a given application is not accounted for at all in this data.

Mainly I wanted all this data in one spot though, as flipping through the test reports to look for specific data when comparing is incredibly slow and tedious.

 

[NorMi]

Here's what the numbers would look like if we used stack loss efficiency instead of delivered efficiency via the duct blower operation only like is done in the official testing. You can see that more BTU's are really recovered and that in turn lowers the emissions per BTU or MJ of useful heat that is kept in the house. This is sort of the maximum you could expect if you had the equivalent of 8 ft of single wall pipe in your house (that's how high the flue temp probe is mounted on these tests) and an OAK installed, so you're not pumping any of your heat you made with the furnace outside through the combustion air. The pollution numbers per unit of heat are quite a bit better too when you use this methodology which I think is a bit more representative.

Ignore the Shelter Cat3 test being best in that category, that test is not a good test sample of that furnace from what I can tell. The Drolet pollution numbers on Cat3 also went UP slightly vs. the official "delivered output" numbers because they somehow calculated their delivered as higher than stack loss on that test, so throw both those Cat3 tests out as unrepresentative. Kuuma is consistent across the board, and just LOOK at how low the PM is on Cat 2-4 by comparison! Impressive results!

 

[JRHAWK9]

Personally, I'd love to see the results on the Drolet if they didn't instruct Intertek to build a crib.....especially for Cat I and II when there is no call for heat with a loaded firebox. The building of the crib was done for a reason, although that's stating the obvious.

 

[NorMi]

I admit I can't understand the secondary tube design decisions on the Drolet fireboxes. ~50% of the jetting is essentially parallel to the flue gas flow, where the stuff I've read going all the way back to the 1980's indicated perpendicular jet streams created the best turbulence factors and best secondary gas combustion overall. Plus, they concentrate a lot of the secondaries in the back half of the box. You think the cribbing is helping to let the evolved gas from primary combustion flow upwards before flowing forward, getting more secondary exposure with the design they have? Still performs quite decently overall and I wouldn't throw one out of bed, she's a looker, that one! 😁

Shelter uses the perpendicular secondary flow, albeit with far fewer individual jets, and the overall design is just so incredibly simple, it's a good example of how well a simplified, low cost design can do on emissions. The complete lack of a real heat exchanger has meant when I took everything apart to relocate mine, there was just no compelling reason to clean it after 1 1/2 years, no point. It's just a rectangular upper chamber and almost nothing to speak of for accumulated buildup on the walls. Same with the chimney pipes, I was going to clean them since I never touched them since install, but also, no point - almost nothing there.

 

[brenndatomu]

Same with the chimney pipes, I was going to clean them since I never touched them since install, but also, no point - almost nothing there.

The earlier FC/Shelters were self cleaning! Too bad it was all over the basement...and totally unscheduled!

 

[laynes69]

The drolet fireboxes are not parallel. The smoke travels north/south and the burn tubes are east/west. This distributes a broader range with preheated air. Energy king did a north south burn tube and they don't meet epa standards.

Other than the firecheif which I know nothing about their design, comparing emissions you're splitting hairs. Whether the Kuuma or Drolet (SBI) both are extremely efficient compared to the furnaces of yesterday. It's not like one will burn 2 cord and the other 4 cord.

I'm appreciating the Caddy Advanced more now that I'm getting some use out of it. If there's a call for heat, the blower will kick in at around 120 degrees and off at 100 in the plenum. Otherwise, if you load the furnace and the demand is met, the blower shuts off and cycles the blower when the plenum hits around 180/190 degrees and shuts off around 120. This maintains comfort in the home, the ductwork warm and during this more heat in the basement. I've had a firebox almost out (just a few hot coals) and the blower still cycling.

I don't load crib style. I just maintain a small gap in the wood and offset a couple pieces on top. I only do this when the furnace is almost out. The grate air opens, forges the coals the the fire takes off. Once the furnace is hot and time to reload, pull the coals forward and load north south. This allows for a front to back burn.

 

[NorMi]

Oh I might have missed the details on how the secondary tube orifices are oriented on all four tubes on the SBI. I was thinking that the orifices are facing either frontwards towards the door, or rearwards toward the back. On the back 3 tubes, if that were the case, the "front facing" orifices would be parallel-ish to the flue gas stream, flowing towards the door. The rear facing orifices would sort of oppose that stream in a counterflow arrangement, if I'm imagining it right. The front most tube, the jets pointing forward would kinda be perpendicular flow, as the flue gas should turn upwards at that point... Is the front tube where most of the "action" is at?

 

[laynes69]

Oh I might have missed the details on how the secondary tube orifices are oriented on all four tubes on the SBI. I was thinking that the orifices are facing either frontwards towards the door, or rearwards toward the back. On the back 3 tubes, if that were the case, the "front facing" orifices would be parallel-ish to the flue gas stream, flowing towards the door. The rear facing orifices would sort of oppose that stream in a counterflow arrangement, if I'm imagining it right. The front most tube, the jets pointing forward would kinda be perpendicular flow, as the flue gas should turn upwards at that point... Is the front tube where most of the "action" is at?

I see what you're saying about parallel flow, like mentioned it's important to have heated air distributed. No not all the action is up front, many times there's a large fireball under a majority of the baffle. That's where draft speeds are important. Not only to not flush the gasses from the secondary tubes too quickly but also allow time for the gasses to ignite before leaving the firebox.

The other thing to mention about the furnaces is ease and features. I see the fire without having to open and disturb the cycle and ease of operation. It take 1 minute to open the heat exchanger peek, remove and soot and close.

 

[NorMi]

I really do like the window, if not for practical purposes just to satisfy my curiosity, a cool and useful feature for sure! As for heat exchangers, it's always better to have one and have to do a simple clean out now and again! I was just commenting how little there is to do on a furnace with no features or extra engineering to speak of. Comes with a lot of downsides of course- less efficiency, worse heat output control and range in factory configuration, etc. I was originally using it in a shop, so the factory arrangement was ok for that on the Shelter, often burning at the higher rates. Plus, the SBI wasn't released, and the Kuuma was too much $ for my blood for a simple shop install.

When I finish making the changes I have planned I'll post about them in the other thread I made, I should have enough instrumentation along with the CSA loss calculation spreadsheet to be able to provide a reasonable comparison with the original data here. That's probably the only other "advantage" of starting at the rear of the pack, for those of us who grew up with tinker toys and legos, you have almost nowhere to go but up, and it's fun if you like to play in the garage a bit!

 

[brenndatomu]

The other thing to mention about the furnaces is ease and features

SBI should really brag up their easy clean HX setup more...that's a big deal, it really is.

 

[brenndatomu]

If you think about it a bit, even if one furnace shows being a couple % more efficient in testing, in the real world that goes out the window if the runner up has an easy cleaning procedure that can be done in a couple minutes without getting dirty, vs a method that most are only going to actually do once, maybe twice per year...and when cleaning the HX can drop the flue temp 50* (and more) that adds up.

 

[laynes69]

If you think about it a bit, even if one furnace shows being a couple % more efficient in testing, in the real world that goes out the window if the runner up has an easy cleaning procedure that can be done in a couple minutes without getting dirty, vs a method that most are only going to actually do once, maybe twice per year...and when cleaning the HX can drop the flue temp 50* (and more) that adds up.

Yep! Factor in wet wood and it's worse. Fly ash, creosote, soot etc robs a bit of heat from a crucial part of a furnace the heat exchanger. Lab numbers give a starting point, however operation and moisture variables in wood changes everything. The beauty of SBI and Kuuma is the automation to take out some of those variables.

 

[JRHAWK9]

SBI should really brag up their easy clean HX setup more...that's a big deal, it really is.

This is one thing I do like about the SBI stuff. Very easy to clean the HX.

Fly ash, creosote, soot etc robs a bit of heat from a crucial part of a furnace the heat exchanger.

There is no such thing as anything other than flyash in a Kuuma HX though, so just a simple wipe with your hand removes everything from it. I just use a Scotch-Brite pad and wipe over all the metal. No-where near as easy as what you have, but seeing there is nothing but white-ish flyash, it just falls right off.



Yesterday we left town for the day and I had a chance to once again load the firebox full of black locust and leave, hoping to keep the LP from running. Outside temps were low 20's during the day and high teens at night while we were gone. We were gone for a total of 28 hours.

I loaded 93.8lbs of 8 year old black locust at 9:30am on Saturday on minimal coals. Set my draft to around -0.03" or so and walked out the door. I am able to watch/monitor everything remotely.

The blower ran consistently till it finally shut off at 9:08am Sunday morning. I -almost- got 24 hours of straight blower runtime........22 minutes short...dammit! LOL When the blower shuts off, I will typically have some coals, but not a whole lot of them.

The house was 72° when the blower shut off, almost 24 hours after loading. When we got home today at 1:30pm (almost 4.5 hours after the blower shut off), the house was down to 66° while the outside temp was 28°. I did complete my objective though, as the LP furnace did not run as I had it turned off. I can turn it on via my phone, but when I looked on the way home I knew I was going to be good leaving it off. I lit another fire soon after I walked in the door.

Here are the screen shots of my flue and supply temps during that whole burn.

Flue (note flue temps never reached 350° internal at any point during the burn, except right after I loaded):

Supply:

As you can see in the beginning, even though I had the computer set at minimum burn, my draft lowered and with the computer on pilot, I still had some pretty good off-gassing happening early on from the fresh load. This led to some elevated firebox/supply temps, but my flue temps remained pretty low. Once the off-gassing was under control and the firebox temps fell back down to normal (for the computer being set on minimum burn), then the supply/flue temps came back down to what they typically will be with the computer on low and on pilot.



Here are my previous two postings about having extended burn times with a firebox full of BL. The wood I burned all three times were of the same haul. I ran a normal -0.05 to -0.06" draft back then on the ones below.

84lbs:

Kuuma Vapor Fire 100: Supply and Stack Temps vs Time over 20 hours

I decided to buy a -GATEWAY- for my Thermoworks Smoke which I use to monitor my supply and stack temps. This sends temps to the cloud every minute or so and allows one to export the data to a CSV file to play with in a spreadsheet. Supply temps are taken in the plenum, but the HVAC probe I...

www.hearth.com

99.4lbs:

Kuuma Vapor Fire 100: Supply and Stack Temps vs Time over 20 hours

I decided to buy a -GATEWAY- for my Thermoworks Smoke which I use to monitor my supply and stack temps. This sends temps to the cloud every minute or so and allows one to export the data to a CSV file to play with in a spreadsheet. Supply temps are taken in the plenum, but the HVAC probe I...

www.hearth.com

 

[NorMi]

Wow! Nice work! Actually this inspires me to wipe out my heat exchanger today since I have it apart and I'll have some fancy datalogging stuff on mine soon - I can perhaps measure the difference on heat transfer over time as the soot deposits form! I love geeking out on the data collection stuff

 

[JRHAWK9]

We've had some cooler temps here lately (got down to 7° last night with below 0 WC) and I am seeing some very good results from the lowered draft....in flue temps, supply temps and house temp. Went to bed when it was 16° out, woke up to 7°. The house was 71° this morning. I loaded 50lbs last night at 10pm and left the computer on minimum burn. When I woke up this morning 9 hours later I glanced at my remote temp monitor and it was showing 120° plenum temps yet. I went down the basement and saw the computer on '2' with firebox temps still in the high 800's ('3' is fully open, which is does more towards the end of the burn, when firebox temps reach ~825°). Looking at my graph, it looks like it went to '2' that time only about 30 minutes before I got up. So I opened up the door and there were -plenty- of coals. I raked them forward and turned the computer to high (which opens the damper to '3') while I got ready for work to burn down some of them so I didn't load on too many coals before work.

Here's the draft I have been running.



I have my BD set for ~ -0.04" and using my key damper set at the position below to achieve the draft you see above. It's a little over half closed, as vertical would be half closed seeing my stove pipe runs at a 45.



When I adjust it a bit less to where it doesn't seem to effect draft much, it will still drop my flue temps vs running it open. Thinking, seeing it's located right after the collar, it helps deflect some heat and keeps it to be delivered to the house vs up the chimney. When I adjust the key damper as shown above in order to pull the draft down that -0.01", it helps keep my flue temps even lower.

I'm wondering if I can get away with this due to how dry/old my wood is.

 

[woodey]

I'm wondering if I can get away with this due to how dry/old my wood is

Exactly what I was thinking.

 

[JRHAWK9]

This past Sunday we were going out of town for 24 hours to the Packer game so I loaded the Kuuma with 94lbs of Black Locust and put the computer on minimum burn before we left at noon on Sunday. Based on the outside temps, I knew it was not going to handle the full heat load of this house for the full 24 hours we were going to be gone. I was just curious as to how long it would be able to fend off the LP furnace and how much it would run with both furnaces "dancing" once the LP did furnace did kick on. I actually timed Sat night's load (loaded 59lbs of oak at around 11pm Sat night....so 13 hours later is when I loaded the 94 lbs of BL) absolutely perfectly and was able to let it burn down to limit the coals I was loading on so I had more room to load more wood Sunday before we left. It was a low of -5° early Sunday morning. The house was 70° when I loaded the 94lbs and when we left at noon on Sunday, outside temps were 9°. Sunday night it dropped -3° again. It was -1° at 4am Monday, 6° at 9am and 16° when we got home at noon on Monday morning.

Anyway, I set the LP stat to come on at 62°. Looks like about 4am Monday morning (16 hours after loading) the LP came on for the first time and both furnaces started "dancing together". I remotely bumped the LP stat up to 70° at 9am ~3 hours before we got home. The LP furnace ran on/off for about 45 minutes total from 4am till 9am. The LP ended up running for 225 minutes total while we were gone. I shut the LP furnace off when we got home and raked the coals forward and re-loaded. I had plenty of coals left to re-load on after 24 hours.

A: LP first kicked on (Kuuma blower goes to low speed tap instead of being speed controlled when LP is running)
B: Remotely set LP thermostat to 70° when we left Green Bay. (Looks like the controls system was busy from the time I did that till we got home.) The LP was running continuously, but the Kuuma blower was kicking off/on via the low speed tap seeing it was not being speed controlled due to the LP blower running.
C: Re-loaded Kuuma when we got home and shut the LP off. The Kuuma took over from here.

So, it costed us about 3 gallons of LP to be gone 24 hours in single digit/sub-zero temps.

 

[JRHAWK9]

The above post is also with my draft pulled down to -0.03" when on pilot by using my KD. As I'm finding out yesterday and today, it's best to keep my KD open and my draft a consistent -0.05" with my BD when it's colder out. I get much better heat (BTU/HR) with the higher, more consistent draft. Now, how much a difference it would have made LP use wise if I would have had a -0.05" draft over the entire burn....probably not much. The house would have stayed warmer longer, but I probably would not have had the same burn time. It was a case of lower BTU/hr over more time vs higher BTU/hr over less time.

 

[JRHAWK9]

I know I posted this before, but this is what our Kuuma VF100 is heating.

We don't have a very large house, square footage wise, but do have a fairly decent amount of volume based on the footprint. It's a 32'x42' footprint, but around 32,000CF including the walk-out basement. We have very tall (25'+) 12/12 pitch cathedral ceiling and no attic. The basement is heated solely off of the radiant from the Kuuma. The 1/3 of it where the furnaces are stay right around the same temp as the rest of the house, the other 2/3rds of it is a bit cooler, but is still heated enough to where the other half can work on her crafts down there.

The following three photos were taken back in Feb '20 right after we received 9" of snow. After I blew the driveway, I snapped these photos only a couple hours after it stopped snowing. Was 22° outside. There is supposed to be 9" of snow on the roof. Garage pic as a reference, as it's not heated. Front and back roof of the house. Losing a ton of heat along the whole length of roof peak. Our indoor humidity levels are also low. We also have a few spots along the INSIDE of the exterior walls in all three rooms that will literally form frost on the INSIDE when it gets below zero. These spots are at corners and are all located at the celling of the main floor, below the loft.

Front of house:


Rear of house:


Garage:


Can also see snow missing along the peak from this photo taken the year before shortly after a snowfall:


Inside ceiling:


Having said that, here are the results from the coldest night we have had yet this winter. It was -5° outside and 73° in the house when I loaded 65lbs at 11:15pm last night and set the computer at medium-high. For reference, minimum burn firebox temperature setpoint is 1,080°, medium is 1,180°, high is 1,280° and full out wide f'n open given 'er all she has is 1,380°. I had it set at 1,235°. Got down to -16° this morning. Woke up to 69° in the house. I did some raking of the coals forward a few times but didn't do any of my coal burn down "tricks" I've used in the past, just turned the computer to high to feed the raked forward coals more air. I re-loaded this morning at 9:40am, house at 68°, -6° outside. So that 65lbs heated this place for 10.5 hours in sub-zero temps the whole time during the night, which is the hardest part of the day to heat.

It's colder mornings like this where having the ability to automatically burn down coals like the HC and CA have would come in handy. Although, not automatic, my method works fine though.

Here are my supply temps throughout that burn that last night:
A: loaded
B, C, D: turned blower off to rake coals forward
E: reloaded