[Stoves] Conceptioal Errors and possible pitfals.

Crispin Pemberton-Pigott crispinpigott at gmail.com
Mon May 6 20:23:25 CDT 2013


Dear Erin

 

It is the conceptual traps that have hobbled our sector for so long, even though it is filled with smart, dedicated and persistent people. 

 

Because the number of possible error is so large and the number of sound ideas is small (with respect to testing stove performance) I feel we would get father to cover what is a good idea and why, than to try to list even the main reasons not to do A or B.  However, your proposal is a good start.

 

I have taken a lot of heat on this list and in fact more off the list behind the scenes for concentrating on what I felt for years was a lack of due attention to correcting what in any normal field of engineering study would have been dispensed with years ago. It is the garage mechanic who thinks they have out-invented GM and Toyota. Sometimes they are right, but not very often.

 

Engineering measurements (I use the term but I really mean scientific measurements of the relevant kind) are so well documented and the method of dealing with the issues are so well known that is it surprising to many people who enter the arena of stove development how different stove testing methods are from ‘standard engineering’. I think they think there is some special group of reasons that have been agreed upon by some ‘group’ that knew what they were doing when they wrote the test protocols and developed the spreadsheets that go with them.

 

Of course reality is much simpler than that. People do what they were taught by those who went before. Perhaps this is the time, before there is anything approaching an ISO standard, to discuss the foundational concepts upon which any claims to performance improvement might lie.

 

Without getting into all the types of performance testing one might consider other than a WBT, let’s at least look at the core concepts the WBT accepts, or should accept.

 

I will kick off by identifying what I think are the big ones.

 

There are correct as far as I know (you sometimes have to imagine their opposites)

1.       Most stove developers and program managers assume the fuel consumption to perform a task is of primal, or almost primal importance.

2.       The way an operator uses a stove creates a larger difference in performance than the difference between two stove designs operated ‘identically’.

3.       There are numerous ways to apply the word ‘efficiency’. Fuel efficiency is the most relevant of these for our purposes.

4.       A cold start and hot start test are expected to have different outcomes, therefore one cannot assume from one test anything about the other.

5.       The amount of water boiled during a test is the mass of water placed into the pot to begin with. If any of it is missing later, it must have been boiled.

6.       Cooking involves the energy in a pot, not the energy applied to the pot. Describing a cooking cycle involves the pot and its contents, not the behaviour and power of the fire.

7.       You cannot average ratios or percentage figures. It is mathematically illegal.

8.       You cannot average a pair (or more) ‘averages’.  It is mathematically illegal. In some cases it can be argued that if the things being averaged are for all intents and purposes identical (such as three replications of an identical task). If accepted in some cases, it cannot be accepted in those cases where the purpose of doing the replication is because of the foreknowledge that the result will be different (i.e. a hot and cold start).

9.       There is no such things as the efficiency of simmering.

10.   Placing a lid on a pot does not increase the pressure inside, thus raising the boiling temperature and producing non-representative results. A poor-fitting lid and a tight-fitting lid do not produce different results save in a very windy environment and even then the difference is probably unmeasurable.

11.   The technical performance evaluation of a stove must be done with no consideration of externalities. A comparison can be made with the gas pressure in a tire. An experimenter can check the indicated gauge pressure in the tire and compare it with the rating printed on the side. The experimenter cannot determine what vehicle the tire will be placed on, where it will go, or how fast it will get there. 

12.   There are 4 types of scales: Nominal, Ordinal, Interval and Ratio. Things measured on one type of scale cannot logically be mixed with things from another. Boiling water in multiple pots is a good example of the problem. Heat three pots, bringing one to a boil. The other two are heated somewhat. How much water has been boiled? Answer, the mass of water put into the first pot, only. Boiling is a Nominal scale determination. Hot water is not boiling water. Calculating what mass of water would have reached 100 C if the mass in the other pots had been less is not legitimate. Water heated to 100 C is not boiling. Further, the stove did not boil any of the water in the other two pots.

 

These are incorrect concepts (ditto re opposites)

1.       As many variables as possible need to be controlled (limited) in order to get an accurate test result

2.       Boiling and simmering water represents cooking in most cases and where it does not, it will do as a proxy for cooking ability.

3.       Heat transfer efficiency directly related to fuel consumption.

4.       Emissions per mass of fuel burned for one fuel can be compared with those of another fuel.

5.       A pot is simmered for 45 minutes, there remains 4 litres of water in the pot and the fire consumes 300 g of fuel. The fuel needed to ‘simmer 5 litres’ is 5/4 * 300 g. (This is conceptually incorrect on several levels and needs some paragraphs to cover them all.)

6.       Capturing and measuring 100% of the CO and CO2 from a fire in a hood is a way of determining how much fuel has been consumed, and therefore how much energy has been released by the fire.

7.       In order to determine the energy released during a test with several replications, the energy in the initial and final fuel portions must be determined for each run.

8.       Task-based metrics and engineering performance metrics can be divided, one into another.

9.       The fuel efficiency of some tasks such as drying fish and roasting meat cannot be determined with accuracy because the product is so variable.

10.   The char fraction of biomass fuel must be separated from the non-charred fuel in order to determine how much energy remains in a mass of unburned fuel (this is said with particular reference to pellet stoves that may or may not be pyrolysers. The assumption is incorrect.)

 

That should do for a start.

 

Regards
Crispin

 

 

From: Stoves [mailto:stoves-bounces at lists.bioenergylists.org] On Behalf Of Erin Rasmussen
Sent: Monday, May 06, 2013 7:41 PM
To: 'Discussion of biomass cooking stoves'
Subject: Re: [Stoves] Conceptioal Errors and possible pitfals.

 

Hi all, I'm distracted by one point out of Crispin's last email, and I renamed the subject because I'm hoping you'll help me explore it a little. 

 

I think the point of confusing heat transfer efficiency with fuel efficiency is a good conceptual error to point out because it can, and often does, confuse people who are new to thinking about these things.  And I think it would be a good idea to pull together maybe a  'top 10' of cognitive errors that are out there along with a reasonable explanation.   

 

What are some common cognitive errors that are waiting, like traps, for the unwary stove builder? Do they have relatively simple explanations that we could post on the web site and then use in our own presentations etc.  or direct new people to?

 

Thanks,

Erin Rasmussen

Cooking Stove Website Admin and List Admin

erin at trmiles.com <mailto:erin at trmiles.com> 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.bioenergylists.org/pipermail/stoves_lists.bioenergylists.org/attachments/20130506/c82aa5a8/attachment.html>


More information about the Stoves mailing list