[Stoves] FW: ETHOS 2013: Where is the New Data on Stove Performance in the Field?

Crispin Pemberton-Pigott crispinpigott at gmail.com
Thu Dec 6 23:37:42 CST 2012


Dear Frank

 

I am going to discuss the WBT 4.1.2 below, not the UCB-WBT 3.1 which is old now.

 

What matters is what people who select stoves for their promotion programmes think the fuel consumption number is. If people think that conducting 15 back to back water boiling tests will require a wood pile with a mass of 15 times the WBT fuel consumption number, then we should provide a consumption number that is the actual fuel consumed per replication of the test. We speak of ‘dry fuel’ because of course the moisture content affects the mass consumed. Correcting for it is reasonable.

 

Is there any doubt that people think the ‘fuel consumption’ recorded by the end of the test is the amount of fuel consumed?

 

Here is the text from the CREEC laboratory at the bottom of the Quad 2 TLUD test:

 

“The Quad 2 stove boils 5L of water in 27 minutes, uses 636 g of wood to boil and simmer (cook) 5L and has an energy use of 11713KJ.”

 

This gives the impression to any careful reader that the fuel consumed per WBT is 636 g. The actual fuel needed each time the stove is operated is about 1600 g. 

 

Most interestingly the energy consumption is 11,713 Joules. It is energy efficient. The dry fuel equivalent of that amount of energy is indeed 636 dry g of fuel (gum tree). So why the large difference? It is because of the charcoal produced, not burned, and unburnable in that stove. This was what Jim Jetter and others have been discussing with me off-list. 

 

The suggested fuel consumption definition attempts to describe, for many different stove types, how to determine the new fuel you would need per replication to keep on repeating an experiment, in this case the WBT 4.1.2.

 

Because some stoves can use fuel remaining from the previous cooking cycle. In that is the case, the each test should start with that same ‘left over fuel’ from a previous cycle. If you claim that the remaining char can be used in the same stove next time, prove it by starting the test with the typical amount left from an earlier identical cycle. The advantage of this approach is it corrects for fuel heat determination errors that are almost impossible to correct in a simple lab. 

 

Suppose starting a stove wasted, each time, 250 g of fuel just to get the fire going properly.  That is not unused fuel, it is needed each time because of some characteristics of the stove. We count it even though the energy needed to cook on the established fire starts counting when the pot goes on. The ‘energy needed’ is the figure used to calculate the thermal efficiency from the fire to the pot. It is a valuable metric for the design who needs to know how the stove structure is performing. 

 

As seen when calculating the CO2, CO and other gas outputs, the difference between fuel consumption and energy production is real.

 

Simply stated, a stove can be energy efficient but fuel inefficient.

 

Regards

Crispin

 

+++++++++++++

 

Dear Crispin,

 

"Fuel consumption : The mass quantity of new, raw fuel required to replicate any prescribed performance cycle, per replication, ignoring the results of the first cycle in any series of such replications. Combustible material in any condition remaining from one replication which may be used by the same device as fuel in the next should be so used unless it is common practice in the target community not to do so in which case the test may be conducted with new raw fuel only."

 

Not sure if this applies here but it seems to me the error is at the start (heating up stove body, getting the fire lite etc) and at the end (measuring flaming fuel left over). To test stoves we can continue the burn for a long time (simmering and boiling) using a –lot- more fuel reduces these two errors at the ends and may give a better indication of stove efficiency.  If we can’t get the bias and precision needed when trying to simulate a ‘meal’ we may need to go to these longer times – and I think we should.  You say that when you say    : The mass quantity of new, raw fuel required to replicate any prescribed performance cycle. It’s the amount of fuel (time burning) that is adjusted. To determine efficiencies at start and end is entirely another procedure.  As I see it.

 

Frank

 

 

 

 

 

 

From: Stoves [mailto:stoves-bounces at lists.bioenergylists.org] On Behalf Of Crispin Pemberton-Pigott
Sent: Thursday, December 06, 2012 4:12 PM
To: Stoves
Subject: Re: [Stoves] FW: ETHOS 2013: Where is the New Data on Stove Performance in the Field?

 

Dear Jim

 

I am just getting to your message of Nov 30. I have not been ignoring the conversation. I have just read it and respond as follows:

 

++++++++

 

Dear Crispin,

As we’ve discussed (off the list), I think you and I agree that if remaining char is actually discarded in practice, then the test metric should reflect that. If char is sometimes discarded and sometimes used in practice, then two different test metrics could reflect the two cases.

I agree that two metrics are appropriate. I have suggested a method of determining the fuel consumption in my comment posted on the GACC discussion site where the Draft IWA reporting form is place for public comment. If char or fuel are said to be reusable in a future fire, therefore to be counted as ‘unused’ or ‘unconsumed’ the next test should be started with the fuel remaining from a previous test. The purpose is, with a very simple step, to consider all changes to the fuel and heat content changes in the fuel remaining. The suggest text is:

"Fuel consumption : The mass quantity of new, raw fuel required to replicate any prescribed performance cycle, per replication, ignoring the results of the first cycle in any series of such replications. Combustible material in any condition remaining from one replication which may be used by the same device as fuel in the next should be so used unless it is common practice in the target community not to do so in which case the test may be conducted with new raw fuel only."

Where the remaining fuel or char are not used, or useable, the next test is started without them. In this way the consumption of raw fuel added for each replication is the only thing to quantify. It avoids the need to determine the heat content of remaining char, wood, dried wood or torrefied wood.

There is a third case, in practice.  There are two uses of remaining char that can be considered as candidates for ‘other activity’ meaning it is not part of the cooking cycle. Char in some places in Indonesia is used to dry fuel and typically left to burn. It is part of the cooking task that the fuel should be dried before use, and that drying takes place at the end of the previous burn. The heat value of the fuel is raised. It is not a property of the stove however, so a fuel use calculation would (or should) not include its consumption, but it is no longer available to put into the stove so it is ‘consumed’ by the practice even though it is after the cooking is finished. Another variation on the theme is that food is left on a stove warming for a couple of hours until Father gets home and has a hot meal instead of a cold one, or preempts re-lighting the stove to heat it again. That is clearly an operator behaviour, but again, consumes fuel that remained after cooking was finished. The first example is part of the cooking cycle. The latter is part of the burn cycle.

Following is my response to your message (copied below) regarding our recent article published in the journal, Environmental Science and Technology, available at the web site:
http://pubs.acs.org/doi/abs/10.1021/es301693f

Crispin: “The spreadsheet does not have a place to enter the amount of raw fuel needed to accomplish a cooking task.”

Jim: The spreadsheet does have a place to enter the amount of raw fuel.  The spreadsheet is available at: http://www.pciaonline.org/testing
Equivalent dry fuel is calculated (in the spreadsheet) to provide a more fair comparison between stoves tested with fuels with different moisture content.  It would be relatively easy to use the data already entered in the spreadsheet to calculate fuel use assuming char is discarded.

Could you please indicate the cell which shows the consumption of raw fuel during the completion of the WBT? 

The equivalent dry fuel number is not the fuel consumed. I believe we are already agreed on this.  The derived number is a metric for comparing fire-to-pot thermal efficiencies, not fuel or dry fuel consumed during the test.  I gave the Quad 2 test as a good example of the difference.

I agree that there is a place on the spreadsheet where a number is posted which is titled “Dry Fuel Equivalent” (DFE). I think it is quite in order to examine what that number is, and what the dry fuel mass is the equivalent of.  It is not the dry fuel equivalent of the mass of fuel needed to perform the task and that is of great concern to me – I hope for everyone.  Many, I can even say most people have been led to believe that the DFE is the amount of fuel needed to perform a task, where that mass of fuel has been factored for moisture content. 

Take for example the Cold Start water boiling task. It takes a certain amount of fuel to complete the task: bringing 5 litres of water to a boil. The char remaining is as you point out, often thrown away. So the fuel mass consumed is the raw fuel that went into the stove to get the water to boil, minus the non-char ‘wood’ removed. Fair enough. We are agreed.  The dry fuel equivalent of that number is the moist fuel mass factored for fuel moisture. The calculated dry fuel equivalent (DFE) of the fire heat used to calculate the thermal efficiency which is posted in Cell W26 on each test sheet. There is no cell on the spreadsheet containing the mass of raw fuel consumed, neither calculated nor manually entered from observation.

We need two numbers: the dry mass of fuel used by the stove, and the dry mass of fuel that would be the equivalent of the heat energy number used to make the thermal efficiency calculation. The difference between the dry mass of fuel consumed and the dry mass equivalent of the energy generated (nominally) by the fire can be as much as 100%.  The difference is caused by how the charcoal produced is treated.

In summary, the spreadsheet does not include a place to record the total fuel consumed by the stove to perform the task.  The number I seek is the total fuel consumed. The DFE is not that number. The DFE is always a lower number except in special cases. The result is that the DFE number, bearing that name, is misleading.  It is not the mass quantity of fuel required to complete the task. 

In the case of TLUD gasifiers producing significant amounts of char, the difference between the two is substantial as evidenced by the Quad 2 test.  Because the DFE number is habitually misinterpreted to mean the fuel consumed performing the task, the stove technology selection process is skewed to favour stoves that produce ever-larger amounts of char in the belief that the stove uses less raw fuel. 

Crispin: “In the paper it is labeled 'overall thermal efficiency', but is it actually a proxy for the heat transfer efficiency…”

Jim: Overall thermal efficiency is not a proxy for heat transfer efficiency.  Please see a discussion of this in the Supporting Information (available on the journal web site), Section 10, Page S19.

The sections reads “10. Combustion Efficiency, Heat Transfer Efficiency, and Overall Thermal Efficiency. MCE (modified combustion efficiency), CO2/(CO2+CO) on a carbon basis, is a reasonable proxy for combustion efficiency (CO2 carbon/total emitted carbon), and MCE indicates how well fuel is burned, i.e., how much of the potential energy in the fuel is converted to heat and radiant energy. A full accounting would include all the carbon in the products of incomplete combustion, and actual combustion efficiency would weight the products of incomplete combustion by their remaining potential chemical energy15, 24. HTE (heat transfer efficiency) is the ratio of energy delivered to the cooking pot versus the total heat energy released from the fuel combustion. OTE (overall thermal efficiency), which directly relates to fuel consumption, therefore is 

OTE ~ MCE x THE”

This is a clear and accurate description of how to generate a heat transfer efficiency number showing the relationship between the fuel heat potential and the pot as a receiver factored for the thermal efficiency of the stove structure, factoring in the combustion efficiency of the fire, disregarding the unburned hydrogen, and disregarding the thermal mass of the pot. I find the label confusing as the Overall thermal efficiency will be a system efficiency number based on the consumption of raw fuel and any work done.

This method of calculating the heat transfer give a useful design metric.  I always prefer to include the Cp of the pot as per the Indian and SeTAR thermal efficiency tests to increase the accuracy of the result. 

Crispin: “The statistical significance (or rather, the confidence we can have in the result) is undermined by the use of an averaging technique that is not accepted by statisticians as valid.”

Jim: We performed a minimum of three replications for each stove/fuel combination we tested, and we reported results as the average and standard deviation of replicates…[snip]

I understand that. I was referring in my message to the Berkeley paper and I also see the technique repeated in Footnote 5 on the Draft IWA reporting sheet. I was referring to the determination of the precision of the test method, not your results. Do you agree that the comparison of averages of triplicate tests is not a valid method for determining the test procedure’s precision? 

Crispin: “A peer reviewer might have noticed, for example, that the heat value of rice hull char was credited with a heat content of 29.5 MJ/kg. This is far from the actual heat content which is closer to 12-14 MJ (an error of ?100%).”

Jim: The heat value of rice hull char was not credited with a heat content of 29.5 MJ/kg. As we described in the article and supporting information, heat of combustion for fuels and remaining char was measured using ASTM Standard Method D4442-07 (bomb calorimeter).  Measured heat of combustion values are shown in Figure S6 in the supporting information.  The heat of combustion for rice hull char from the Mayon Turbo stove seemed high, but we had a second sample analyzed with nearly the same result.

I stand corrected.  I did see the test of the fuels and I agree with the method used and also recommend it as ideal. I was looking at test results and the value of 29.5 was indicated for the rice hull char heat value. I will try to locate the sheet where I saw it. I must have confused it with something in the Berkeley paper. Possible?

Crispin: “If the Journal provides us with the spreadsheets from the actual tests as supplementary material for the published article (normally required for publication), we will be able to reproduce the work mathematically and see what the effect of making the necessary corrections would be. I expect there would be a significant change to the ratings given to all char-making stoves…”

Jim: In the supporting information, we provided detailed information on the stoves, fuels (including moisture content and heat of combustion), cooking pots, operation (including fuel burning rates – see Figures S14-S16), test protocol, and equipment used.  I believe the information is sufficient for others to reproduce the experiments.  We can calculate fuel use with the assumption of discarded char, if there is a need for these data.  There would certainly be a significant increase in fuel consumption for char-making stoves if we assume the char is discarded.

Discarded or put to some other use.  Your offer is greatly appreciated. There is some confusion about the raw fuel consumption of a stove. So some effort was put into creating a definition of fuel consumption (above). The idea is to properly characterise the raw fuel consumption each time a stove is used – off-take from the fuel pile. At the moment there is a significant difference between the Specific fuel consumption and the fuel drawn from the store, per replication, where unusable char or wood remains. Because of the heat treatment of fuel remaining (by the fire), the wording in the Fuel consumption definition above (perhaps with the sentence following it included) was crafted to try to consider the change in fuel heat that will be used in the subsequent fire. The advantage is that only raw fuel need be measured as old fuel is consumed in the subsequent replication, and new heat treated fuel remains behind after each replication.

Crispin: “What we have never seen in a Journal is a review of the procedures and mathematical methods conducted by an independent lab to put some confidence intervals on these WBT results.  Dr Penn Taylor's Masters and PhD theses have not been published. They address the issues and give a precision of about 50% for the WBT 3.1 (which was under review at the time).”

Jim: Robert Pendleton Taylor’s thesis is available at the web site:
http://lib.dr.iastate.edu/cgi/viewcontent.cgi?article=1534 <http://lib.dr.iastate.edu/cgi/viewcontent.cgi?article=1534&context=etd> &context=etd
>From Page 65 of the thesis: “Taking these errors together, the existing UCBWBT should be considered to have a minimum method error of between five and ten percent. Depending on the fuel used, the method error may be as high as twenty percent, as in the case of a fuel with a high ash content. If the test is altered to properly account for ash, the minimum method error drops to about five percent. The UCBWBT currently reports the uncertainty of fuel use and other results as the intra-test variation between three trials. It is not uncommon to see uncertainties of one percent or less reported for factors such as fuel use and energy efficiency. The sources of error identified in this thesis indicate that even if the repeatability of the test can be brought to such a fine level, the actual deviation of the test results from true values is much more likely to be on the order of ten percent. If results of the current UCBWBT are being used to compare two stove designs, the relative error in thermal efficiency, specific fuel use, firepower, turn-down ratio, and any emissions factors expressed on a per-energy or per-mass-of-fuel-consumed basis should be assumed to be ten percent, regardless of that cited as the intra-test error.”
Some of these sources of error have been addressed since the thesis was published in 2009, but many of us are working to continue to improve existing methods and to develop new methods (e.g., methods for plancha stoves and charcoal stoves).  USEPA QA personnel (independent from our project) are working on an uncertainty analysis including error propagation for the equipment and methods we are currently using.  The methodology may be useful for other labs using different equipment and methods.

Dr Taylor says a lot more than that in his thesis.  I asked him during a discussion of this work to estimate the total error in a UCB-WBT and he replied that it was about 50% and described them. The errors are discussed in the text.

The citation above says that the 5-10% error is a minimum, meaning the best possible case. That is not the precision of the test method, which it was my intention to indicate, based on his work.  It is, in his considered opinion, the minimum error. Considering all the errors he identifies and quantifies produces a much higher number.  He shows some char heat values on page 49 indicating a variation of 35% in char heat content. Page 50 shows a variation in the ratio of fuel heat to char heat from 1.04 to 1.75. [This variation is a large source of error in the calculation of the heat value in char remaining in the CCT 2.0.] 

>From page 51, highlight added

“Table 6.2 very clearly shows that the matter removed at the end of a test which visually appears to be charred fuel may have a fairly large range of calorific value depending on its temperature exposure history inside the stove. Looking at the extreme cases, assuming LHVchar=LHVraw f uelto be equal to 1.5 when it is actually equal to 1.75 will result in an estimate of remaining energy that is actually 14% too low, and assuming LHVchar=LHVraw f uel to be equal to 1.5 when it is actually equal to 1.04 will result in an estimate of remaining energy that is 44% too high.”

 

Figure 6.2 shows, after detailed discussion, that the mislabeling of the fuel remaining as ‘wood’ or ‘char’ creates another error. He shows -6% to +12%. With some fuels separation is easier than others, however it is the precision of the test method we are considering so this is another systematic error.

P. 53

“When sorting and massing the unburned fuel and char fractions remaining at the end of a test section, the current UCBWBT protocol does not properly account for ash content, but instead counts ash mass as char mass.”

 

He provides a Figure 6.3 showing error potential of 0-20% for the mislabeling issue.

 

P. 57

“Figure 6.5 represents a test run using oak wood with an ash content of 1.5% as the fuel. In this case, the failure to account for ash in the supposed char fraction results in an overestimation of the actual amount of char by 20.3%.”

 

P. 57

“Figure 6.7 shows a similar situation using an agricultural residue pellet as the fuel, but in this case the error in the energy calculation is much larger at 15.7%. Such an error in the calculation of energy used in the test will cause fuel consumption to be understated by the same 15.7%,

and will cause the thermal efficiency to be overstated by 15.7%.”

 

This single error, examined in an actual test, produces an error of 15.7% without considering other issues. In considering all the issues identified by Dr Taylor I find that the figure of ±50% is supported by his work and may in fact be modest!

 

It is imperative to consider the range of fuels and stoves tested over the past few years which have been rated using this protocol (UCB-WBT 3.1). Considering all the errors he identifies, the total is far in excess of the 5-10% minimum mentioned on page 65. It is quite possible it was more than ±50% in numerous tests. 

 

Dr Taylor also considers the implications of the production of large amounts of char on performance rating. See Figure 6.8 where an error of 43.2% is indicated for the calculated heat liberated by burning grass pellets, again well in excess of the 5-10% minimum.

 

Perhaps you would agree with the wording, “Dr Taylor demonstrates numerous, significant errors which, if one considers their influence on the final calculated metrics, can exceed 50%. In one test he showed an error of 43.2% on a single metric.  It is his opinion that even under perfect conditions the error is never less than 5-10%.”

 

As demonstrated prior to the review of the WBT 3.1, the combination of errors in the formula calculating the thermal efficiency ranges from zero into hundreds of percent depending on the fuel moisture and char produced. As the ‘Specific fuel consumption’ is calculated from the same numbers that give the thermal efficiency, the effect on the final performance rating is equally large, and as suggested previously on this forum, may explain why there is so much variation in apparently identical tests conducted using the 3.1 protocol. 

 

Crispin: “Until the process of correcting the WBT is formalised the sorts of problems described above and in previous communications will continue to dog the stove community. In the meantime it is perfectly reasonable for stove projects to develop and use alternative test methods that are scientifically sound, properly documented and independently validated.”

 

Jim: With support and leadership from the Global Alliance for Clean Cookstoves, I expect we will soon be engaged in a more formalized process for developing and improving test protocols and standards.  We have challenging work to do, and I think this is an exciting time to be working together!  Thank you for constructive comments.

I appreciate that you took the time to consider my comments and to research the matter thoroughly. I do not want to take up too much additional time with the WBT 3.1 unless it is to show that the same original errors relating to the calculate fuel consumption are still present in the CCT 2.0. It is still used in some quarters (UNFCCC/ CDM) and has a low precision. The problem formula is in Cell D17 of each test sheet and in the consideration of what happens to the char remaining after a test.

Something we can do with the errors identified by Dr Taylor is to perform the following calculation:

Total Error = Square root of {[(Error 1)2 + (Error 2)2 + (Error 3)2 + …(Error n)2 ]/n}

This is calculated only after the propagation of each error to the final numbers has been assessed. It would be a valuable exercise.

I look forward to our future discussions of stove test methods, the sharing of insights and the deepening of our understanding of this subject.

Best regards

Crispin

 

PS I again apologize for the delay in responding – I have been preparing for and traveling to a conference.

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.bioenergylists.org/pipermail/stoves_lists.bioenergylists.org/attachments/20121207/fc179969/attachment.html>


More information about the Stoves mailing list