Moisture Content and Efficiency

[Nofossil]

I'm lifting this from another thread. There's been some great discussion about this topic, and I think it deserves a thread of its own.

I've been guilty of promoting the value of really dry wood for gasifiers. While it's necessary to get them started, it may be overrated. I did some detailed system efficiency calculations and found virtually no difference between wood at 30% and wood at 20%. Based on that and many related posts, I decided to do a real (gasp) analysis of the energy loss from moisture in wood. Here it is - let the fur begin flying.

From the initial post, we have 94 lbs wood at average 30% moisture = 65.8 pounds ‘bone’ dry at 8600 BTU/lb = 565880 potential BTU

Calculating the energy lost to the water in the wood:

It takes 1 btu per pound to raise water 1 degree. It takes 970 BTU to boil one pound of water that’s at 212 degrees. It takes .5 BTU per pound to raise steam 1 degree.

We have about 30 pounds of water in our wood. If the wood starts out at 70 degrees and the steam goes up the flue at 600 degrees, it will require the following energy which will be lost to the system:

30 lbs water, 60 to 212 degrees = 4560 BTU
30 lbs water to 30 lbs steam = 29,100 BTU
30 lbs steam 212 to 600 degrees = 5820 BTU

Total loss = 39,380 BTU at 30% moisture. This represents 7% of the energy available in the wood itself. Thus, I’d expect a 7% loss compared to bone dry wood. Wood at 20% should represent a 4.6% loss.

The theoretical difference between the energy loss due to water at 30% vs. water at 20% is less than 2.5% of the available energy, assuming the same piece of wood in both cases.

Not a big number. I should have done the math earlier.

Of course, if you flip that around, this 7% number could account for a sizable majority of your combustion efficiency losses if you’re somewhere near 90%. From a system level, it’s not such a big deal. According to these numbers, I need to burn another tenth of a cord if it’s 30% instead of 20%.

Based on a post by slowzuki in another thread, I'm also calculating non-water flue gas losses.

Air has a specific heat of .24BTU/lb. It takes about six pounds of air to burn one pound of wood, although the boiler fan may drive much more than that. Making a wild assumption that the specific heat of all the non-water flue gas is about that value, we can calculate heat loss up the flue.

For the original example, 6 pounds of air per pound of wood = 455 pounds of flue gas at 600 degrees = 59,000 BTU of non-water flue gas loss for another 10%. At this point, we're down to around 80% depending on the unburned fuel losses. I actually got 56% measured, so I'm losing energy elsewhere.

Here's the beginning of a breakdown of losses:

Starting BTU value of wood: 565,880
Water related losses at 30% moisture: 39,380
Non-water flue gas loss at 600 degrees: 59,000

At this point, we're down to 467,500 BTUs or 82% of the original potential.

 

[wdc1160]

Dry wood was not the only difference. For this burn, the house, outside temp, hat water tank, and storage tank were all cooler. The burn was longer as well. with about 50% more actual fuel.

This was the driest wood I’ve burned this year. It created another problem: The delta T through the boiler was so high that when the controller tried to keep the inlet above 140, it caused the outlet to exceed 180, which shuts down the EKO fan.

I think this means that I need to either install my three speed Grundfos which at high speed pumps more than my Taco 007, or I need to throttle back the EKO when burning wood that’s this dry. Am I on the right track here?

Folks at greenwood tout that their boiler needs thicker logs/splits otherwise you may loose some gas up the pipe(regardless of moisture). I guess I am throwing this out as another idea/variable onto an already messy wood effciency discussion

 

[Nofossil]

Folks at greenwood tout that their boiler needs thicker logs/splits otherwise you may loose some gas up the pipe(regardless of moisture). I guess I am throwing this out as another idea/variable onto an already messy wood effciency discussion

I'm pretty convinced that the optimum settings for small chunks is very different than large chunks. Of course, I was pretty convinced that dry wood was a lot better, too.

 

[Eric Johnson]

I think we run the risk here of giving the impression that it's OK to burn green or wet wood. I would say that while you can always find examples to the contrary, most people who heat with wood will have much better results burning the driest wood they can get their hands on. As a practical matter, I suspect that's usually higher than 20% mc in most cases, and probably higher than 30%. Drier wood is superior in every respect, in all wooburning situations.

 

[kuribo]

6 lb of air to 1 pound of wood would probably be what is called the stoichiometric ratio, or the "perfect" amount for complete combustion. In actuality, the real volume of air supplied, especially when using a blower, is many times that. Might be a better approximation to find out what cfm your blower is providing...

 

[Nofossil]

6 lb of air to 1 pound of wood would probably be what is called the stoichiometric ratio, or the "perfect" amount for complete combustion. In actuality, the real volume of air supplied, especially when using a blower, is many times that. Might be a better approximation to find out what cfm your blower is providing...

Yeah - ideal for gasifiers is supposed to be something like 1.6 times stoichiometric. Don't know how close to that the EKO is. I should at least use that in the calcs above.

Eric - I would agree that drier is better, and I suspect that there's a point where efficiency takes a big hit and emissions go up. However, 30% really does seem to be nowhere near the penalty that I expected. It's worthwhile to explore that, since I among others have been spending a lot of time and effort trying to figure out how to get to 'really dry'.

I don't expect to do any tests with wood at 50%, for instance, as I'm sure it would be terrible. On the other hand, I'm a lot less concerned about being two years ahead than I was.

I may have mistakenly led some to think that gasification was too much hassle because of the imperative to have really dry wood. I'm repenting that particular sin insofar as the data supports it.

 

[Jags]

I totally agree with Eric, but it is also interesting to know that you "loose" only 2.4% of the net "heating" energy going from 20 to 30% wood.

I have often suspected the argument of "but wet wood (fuel in stove) consumes alot of the energy that could be produced as heat" - to be somewhat over exaggerated.

But with that aside, all of the positives of having seasoned wood still stands.

 

[Eric Johnson]

I understand what you're saying, nofossil, and I'm not trying to diminish the value of what you're observing and documenting. It's just that I'm intimidated by all you numbers guys, and I needed to find something contrary to say.

Jags--I think nofossil's analysis is limited to a downdraft wood gasification boiler. It might be a lot different with a wood stove or other type of boiler. Or not. Like everyone else, I would have thought that there would be a notable difference between 20% and 30% mc.

Parenthetically, I've always found that with the same setup: wet wood/dry wood; cold winter/mild winter--didn't make a big difference. We always burned about the same amount of wood every year. I think it all averages out over the course of a season. I just stayed warmer and spent less time diddling around with the boiler when I fed it dry wood.

 

[Jags]

Eric - it would be a very interesting experiment indeed. I would love to do it, but I don't have all of the gadgets that NoFo has. But I suspect that an EPA cert stove with cat or non-cat re-burn technology would have similar results as a gassifier (although maybe not as efficient), because that is what the reburn tubes or cat is doing, reburning the gases. So you could be very correct, but I still think that the argument of "using alot of energy to boil off the excess water" is a little over rated.

 

[Nofossil]

Parenthetically, I've always found that with the same setup: wet wood/dry wood; cold winter/mild winter--didn't make a big difference. We always burned about the same amount of wood every year. I think it all averages out over the course of a season. I just stayed warmer and spent less time diddling around with the boiler when I fed it dry wood.

That's interesting. My experience has been that my wood use tracks degree-days almost perfectly. The attached graph shows a purple line that predicts what I should burn based on degree-days (uses actuals to date, historical averages for the future). The orange line shows actual usage. There have been only two weeks that deviated significantly, and one of them was due to being away for several days..

 

[Eric Johnson]

In our case, I think it was made up for by staying warmer during milder temps. That's in two different houses with undersized wood-fired boilers (conventional indoor). With all kinds of wood heat, I think that's a common occurrence.

I thought I remember you saying in a post that your wood usage (paltry as it is) doesn't vary much year-to-year. True, we haven't had a really cold winter in awhile--at least not like back in the Old Days.....

 

[Buck1200]

nofo- does you chart show that you have done a heat loss calc on your house and arrived at what is basically a reset curve as used in most boiler controllers now? If so, I'm curious to know what your design day loss is. You're in a good position to add fuel to the debate that those calcs are pretty heavily padded. I'm just curious.

 

[Nofossil]

nofo- does you chart show that you have done a heat loss calc on your house and arrived at what is basically a reset curve as used in most boiler controllers now? If so, I'm curious to know what your design day loss is. You're in a good position to add fuel to the debate that those calcs are pretty heavily padded. I'm just curious.

I could go back and look at the design loss calcs, but that's not what I did here. Instead, I looked at the degree days and the total wood consumption for 2006-2007 and calculated wood per degree day. I then plotted a historical average year week-by-week assuming the same fuel consumption. That became my expected wood usage. As each week goes by, I plug in the actual degree days, the expected consumption based on that number of degree days, and the actual wood consumption.

I calculate actual wood consumption based on 285 hours of 'boiler above 140 degrees' per cord burned. Remarkably consistent and accurate.

 

[mikeyny]

Not to throw a monkey wrench in the works but??
I understand degree days very well. It is an easy concept to grasp to calculate fuel usage if you are using those costly, but predictable fossil fuels. How do you calculate usage when using wood?. I suppose if you are using the same species of wood it might be more accurate.

 

[Nofossil]

Not to throw a monkey wrench in the works but??
I understand degree days very well. It is an easy concept to grasp to calculate fuel usage if you are using those costly, but predictable fossil fuels. How do you calculate usage when using wood?. I suppose if you are using the same species of wood it might be more accurate.

Simple calculation: the 2006-2007 season had 6177 degree days, and I used 4.5 cords of wood. That's 1597 degree days per cord.
Next simple calculation: My boiler was above 140 degrees for 1282 hours during the season. That's 284.96 hours/cord.

Based on that, I can see if the same numbers hold true this year. The graph above shows a pretty good correlation.

If I burned wood that was radically different from one year to the next, it might affect the numbers. I'll be burning a lot more poplar next year, so we'll see. All in all, it's amazingly accurate so far.

 

[jebatty]

When trying to get a handle on how much wood is burned as related to different variables, it might be important to use weight standardized to an assumed m.c., say 20%, rather than to use cord measurement. A cord itself is variable by its nature. Weight also would eliminate the variable related to wood type, as all wood has the same btu/lb.

 

[Nofossil]

When trying to get a handle on how much wood is burned as related to different variables, it might be important to use weight standardized to an assumed m.c., say 20%, rather than to use cord measurement. A cord itself is variable by its nature. Weight also would eliminate the variable related to wood type, as all wood has the same btu/lb.

I'm weighing wood right now in order to do some detailed efficiency and loss analysis. I'm WAY too lazy to do that for a whole season's wood. To make matters worse, my wood is a mix of many species, sizes, and moisture contents.

I'm really trying to do two things here.

At a high level, I want to make sure that I have a wood harvesting, processing, and storage plan that will give me enough wood of suitable quality to keep me warm. Toward that end, I'm trying to plan and analyze consumption at a gross level - will I need 4 cords, or should I plan on 5 or maybe 6 in case we have a cold snap? Am I on track to finish the year without running out? This is where cords and degree days seem to be appropriate.

Second, I need to feed my obsession with understanding everything to three decimal places. Is there a way to improve my system? Where am I losing potential performance? Understanding at this level requires more meticulous data gathering, which is not sustainable, by me at least. Here I use pounds and correct for moisture, averaging moisture readings for each load.

 

[MarcM]

Nofossil, everything I saw in your calculations looked correct to me, although the new school engineer likes looking at everything in SI, I was able to deal Certainly using a volumetric measurement (cord) for a wildly variable density solid like wood is enough as it is to make even the most jaded engineer keel over. But I would never weigh all the wood burned in a season either. It requires enough effort just to do everything else.

At any rate, your assumption for the specific heat of all other gases except water vapor doesn't look to be a bad one- Air is mostly nitrogen which doesn't participate in a complete combustion reaction, just absorbs heat, and the specific heat capacity is 0.241 BTU/lb-R (or F, but I like rankine). Other gases one would expect to see as the result of an efficient combustion reaction are all in that neighborhood, CO2 ~ 0.2, CO ~ 0.25, O2 ~ 0.22. http://www.uigi.com/physical_prop_e.html

However, the catch is, you're probably underestimating the amount of water vapor in your flue. Not only will you have water vapor from the moisture content of the wood, you'll have an appreciable amount from the combustion reaction itself, which if perfectly efficient only produces CO2 and water vapor. A balanced combustion reaction of celllulose (C6H10O5) yeilds five molecules of water vapor and three molecules of CO2.

Also, running at full vent, your combustion reaction is still not 100% efficient. I'm not crystal clear on what the 8600 lb/BTU represents... does this take into account the heat needed to raise the temperature of the solid wood to the point at which it pyrolizes and also the heat needed to raise the wood gas further to whatever the temperature is inside your combustion chamber?

 

[kuribo]

"However, the catch is, you’re probably underestimating the amount of water vapor in your flue. Not only will you have water vapor from the moisture content of the wood, you’ll have an appreciable amount from the combustion reaction itself, which if perfectly efficient only produces CO2 and water vapor. A balanced combustion reaction of celllulose (C6H10O5) yeilds five molecules of water vapor and three molecules of CO2."

Also the water vapor in the primary and secondary air.....

 

[MarcM]

"However, the catch is, you’re probably underestimating the amount of water vapor in your flue. Not only will you have water vapor from the moisture content of the wood, you’ll have an appreciable amount from the combustion reaction itself, which if perfectly efficient only produces CO2 and water vapor. A balanced combustion reaction of celllulose (C6H10O5) yeilds five molecules of water vapor and three molecules of CO2."

Also the water vapor in the primary and secondary air.....

Good point, probably a small contribution with dry winter air, however definitely shouldn't be overlooked.

 

[jebatty]

I'm weighing wood right now in order to do some detailed efficiency and loss analysis. I'm WAY too lazy to do that for a whole season's wood. To make matters worse, my wood is a mix of many species, sizes, and moisture contents.

Agree with you 100%. Do the detailed test on a typical wood mix. Then make a rough conversion of weight to cords. End result is amount of wood needed. Or, after a year or so of burning with typical outdoor temperatures, estimate cords burned, multiply by 3, and each year keep this on hand, burning oldest wood (3rd yr) last. Worst case is burning 2nd year wood, best case is no need to cut so much wood for next year.

 

[jebatty]

I’m not crystal clear on what the 8600 lb/BTU represents…

Good question, other than the info states that 1 lb of bone dry wood is 8600 btu, which BTW is about the same as coal, peat, grass, etc. Rule of thumb for wood burned in a stove, air dry, is about 6500 btu/lb.

 

[kuribo]

"Good question, other than the info states that 1 lb of bone dry wood is 8600 btu, which BTW is about the same as coal, peat, grass, etc. Rule of thumb for wood burned in a stove, air dry, is about 6500 btu/lb."

I think that is a good approximation, but if you look at the literature, the figure varies widely, all else constant, with species. Hickory/locust and the like are said to have 8600 btu/lb, while alder, poplar, etc., are around 5500 btu/lb.......Pine and some of the other softwoods are around 6000 btu/lb.

 

[Nofossil]

Thanks for all the commentary. There are a bunch of assumptions and simplifications that I've made, and I'm feeling a bit more comfortable that they're pretty much valid.

I'm assuming that the water vapor that's a byproduct of combustion can be lumped with other flue gases without invalidating the specific heat of the flue gas mix.

I'm assuming that the 8600 BTU/lb for bone dry wood doesn't include the heat energy to raise it to pyrolysis temperature, but I'm ignoring that energy anyway. It would only be a couple hundred BTU per pound at worst. I do assume that the 8600 BTU includes the energy to vaporize the water that's a byproduct of combustion, since that's what happens during combustion.

Here's a screenshot of the spreadsheet that I use for doing my efficiency analysis. I know, it's Excel. I'll do this part on my Linux box when the Open Office spreadsheet can do ODBC properly.

Green boxes are per-run data. Blue boxes are constants that I might need to adjust. The refresh button extracts all of the logged data for the selected time period (in 30 second increments) from a database server, does the calculation, builds the pie chart, and inserts the graph.

Comments are welcome. If anyone would like to have a copy of the spreadsheet itself, it's here. It's not well documented, and it won't refresh since it won't have access to my database server. However, feel free to use it, tear it apart, or improve it.

 

[MarcM]

Can you break out your calculation of the heat lost for to non water vapor flue gas for me?

Also, you're right not to inlucde any energy loss to vaporize water that's a byproduct of combustion, because it is created by the combustion in gaseous form. No latent heat required.