[Stoves] Equipment required for testing stoves

[Josh Kearns]

Thanks for forwarding my responses Ron - I had meant to reply to the list.

To briefly address some of your additional comments:

I should clarify that my interest in the char-making TLUD stove is from the
standpoint of wanting to use the char as a locally-produced adsorbent in
water treatment (and then subsequently as a biochar amendment to soil...).
So the sorption performance of chars is my #1 interest, followed by proxy
measures such as surface area, porosity, condensed aromatic C content,
pyrolysis temperature in the char bed, etc., that may relate to sorption
performance.

So I confess I am not thinking first and foremost about cooking convenience
in our little TLUD setup and operation. I mainly just need to be confident
that the chars that come out of our unit under different operational
conditions are representative of chars that would come from other cooking
devices (e.g. those optimized for the user experience).

When I said "there would be no measurable difference between a 1-,2-, or
3-can chimney," I was being a bit flippant. There would likely be
observable differences in temperature and perhaps also in some char
properties. They would not be a lot different though, and you would have to
generate and characterize many samples in order to be confident you are
resolving the differences.

Generating "a lot" of samples gets to be a problem for me since I have a
limited time and resource budget for testing sorption by chars made in
several different manners, from different feedstocks, etc. So I have to cut
down the number of samples I generate with intent to put through testing
because it can quickly get very expensive and time consuming!

But as you suggest Ron, if you had a fan with variable speeds, and you ran
it a 0, 1, 2, 3...n speeds then you would see a continuum of peak
temperatures and a continuum of subsequent char properties. My cheapo
little fan has two modes: "on" and "off." Hugh McLaughlin ran similar char
generation tests with an electric fan at 3 or 4 voltage steps and showed a
progression of sorption capacity according to his GACS procedure. For my
work, at least for now, I can only test the "end member conditions" and
just infer that intermediate draft conditions would lead to intermediate
temperatures and intermediate sorption capacity.

As an illustration: our pine pellet char from this 1-gal TLUD in FD mode
(fan on) produces char that has close to 80% the batch-mode sorption
capacity for uptake of herbicide from simulated natural water compared with
a top-shelf commercial powdered activated carbon. In ND mode (fan off), the
char performs more like a moderate grade traditional cooking charcoal - 5
or 10% of the herbicide uptake capacity of PAC. So from my point of view,
if I add or subtract chimney cans and get a +/- 1% to 3% herbicide uptake
capacity, so what? With a fan we are able to create such impressively
sorptive carbons - so I want to focus more on that.

Hope this clarifies -

Josh




On Fri, Dec 7, 2012 at 6:06 PM, <rongretlarson at comcast.net> wrote:

> Josh and stoves list
>
>    Thanks for the responses.  I add a little below, but I have moved to a
> lower position everything from you that was now clear.  AT first I had not
> realized that your response was sent only to me and not sent to the stoves
> list.  Otherwise I would have just left everything unchanged.
>
>    A few clarifying remarks/questions below from me.
>
>
> ------------------------------
> *From: *"Josh Kearns" <yeah.yeah.right.on at gmail.com>
> *To: *rongretlarson at comcast.net
> *Sent: *Thursday, December 6, 2012 11:55:07 PM
>
> *Subject: *Re: [Stoves] Equipment required for testing stoves
>
> Ron et al., some responses below...
>
> On Thu, Dec 6, 2012 at 1:03 PM, <rongretlarson at comcast.net> wrote:
>
>> Josh  (cc list)
>>
>>
>            <moved RL and JK parts a and b  to a lower position>
>
> *RWL*   c.  You had quite a tall chimney - so the device could not
>> operate as a stove.  Do you have any data on the maximum temperature with a
>> shorter chimney in the ND cases (which might then better simulate a
>> stove)?  This further queried next.   Did you have a means of controlling
>> primary air other than through the chimney height?
>>
>>
> The chimney is not too tall, it just depends on what height you like for
> cooking. You could put a pot on a ringstand at waist height if you like, up
> to you.
>
>     *[RWLc1:  Yes, but the power level will then go way down -
> maybe/probably could operate only in FD mode, because you have such a tall
> fuel load.
>
> *
>    <moved JK para c1.5 on stove heights related to fans to spot below>
>
> Cutting the chimney height by, say, half, would make no measurable
> difference in the char. The real difference is induced by the fan. The fan
> adds something like the equivalent of 30 feet of well-insulated chimney. I
> made the taller chimney mainly to help contain the flames - people at the
> University are already sketched out by what I'm doing.
>
>    *[RWLc2:  I can't yet agree on the "no measurable difference in the
> char"  (and think you say the opposite below).   I concur on the
> importance of adding extra chimney height for safety (and power/speed)
> reasons*.
>        *Have you ever tried a speed controller on the fan - so as to
> simulate the 4-can (and other) chimney heights*?*  That would be most
> instructive on both the production temperatures and the char quantity and
> quality.  I have seen no data of this type.
> *
> * *
>
> Under ND conditions you are at the mercy of the size and shape, and thus
> bed porosity, of the feedstock.
>
>    *[RWLc3:   Yes, agree partially.  If one controls the primary air
> supply you should/might be able to control the (power) turn-down ratio over
> a 2:1 ratio.  Controlling fuel-bed size and shape (diameter and height) is
> certainly available to the third-world cook (or at least the stove
> supplier).  It would be interesting to see if that cook would also learn to
> choose different fuel shapes (porosities - for instance a spherical pellet
> with same total weight might cook differently.)  Probably some big
> cook-time differences (and char temperatures) with (vertically oriented)
> grasses of different diameters.  I have seen almost nothing on this topic
> also*.
>
>
>     *[RWLc4:     More on this below - but I still think the 4 upper cans
> do provide for more power (shorter fuel life) and I (now, still) think this
> parameter will change the char production temperature (and probably the
> properties - with which you agree below).   I hope you or others can show
> more data along these lines - as this is an important new (to me) aspect of
> char-making stoves.  Previously, I had believed (like you) that the power
> level did not impact the char temperature.  Now I am (much) less sure.  I
> have seen many stove videos showing increased power with taller chimneys -
> and am now guessing (based on your two graphs) that the char temperature is
> also (somehow) related.  To repeat - this is a very important point that I
> hope others will comment on.*
>
>
>
>>    d.   Can you supply the weight of the starting and ending weights of
>> the load of pellets in each case.  It appears that the FD power level was
>> close to twice the power level (half the duration) of the ND case.  It
>> would be interesting to see the thermocouple measurements with 0 (?), 1, 2,
>> and 3 extra cans above the secondary air inlets as well as for your case of
>> four upper cans.
>>
>
>
> 1, 2, or 3 cans would make very little difference in temperatures,
> probably substantial differences in char properties, and certainly nothing
> compared to the fan.
>    *[RWLd1   re 1-4 cans and T's:  I still need to be convinced - with
> experimental data.  I agree on use/value of a fan*
>
> A gallon of pellets weighs about 2500 g. I get about 500 g of char under
> ND conditions and 250 g of char under FD conditions.
>
>     *[RWLd2 :   your two power levels were probably about
>
>    a.  ND:   2.5-.5 = 2 kg of wood consumed at 18 MJ/kg over 1.2 hours
> (4320 secs).  This gives 36 MJ/(1.2*3.6  ksec) = 8.3 kJ/sec = 8.3 kW.
> This is very high for a cook stove - of course,  this is great for making
> char alone.  The remaining issue is how low can the char temperature be.
> This is the key char parameter of interest to the soil scientists.
>
>    b.  FD:   2.5-.25 = 2.25 kg .  Over 0.7 hours, this gives about
> 36*(1.125)/(0.7*3.6 ksec) = 8.3 *1.125/.7 = 8.3*1.6=13.3 kW
>
> These kW values are not helpful in comparing between stoves, unless the
> fuel beds (and probably more) happened to be identical or pretty similar.
> Still,  I'll bet that we can give general guidelines for the likely char
> production temperature knowing power level (fuel exhaustion times) and a
> bit more - for any stove.  I have not seen such an attempt.  Anyone?
>
>    Butsince  the temperatures could be so helpful,  I hope anyone else who
> has used thermocouples can also report in on their results - especially if
> anything was variable and the (relatively constant) interior fuel bed
> temperatures changed appreciably.   I have certainly seen plots like yours
> - but nothing to show how temperatures changed - and as dramatically
> (600-900 C) as did yours with the same stove operated differently.  I am
> guessing that your system could get as low as 400 or 500 C - with either
> fewer upper cans or a slower fan speed.  Better understanding this ability
> to produce chars of different character should be a hugely valuable
> experimental result.  I think this can be done with one thermocouple, not
> needing 3-4.*
>
>>
>>       In other words - how low a temperature char could be obtained?  I
>> have seen no data on char temp as a function of power (primary air flow
>> rate) - so your geometry offers a nice (non-stove) way to prepare chars
>> which should have very different in-soil characteristics.
>>      In a stove, changing the power levels (fuel duration time) through
>> changes in primary air supply (changing the turn-down ratio) would have a
>> similar effect - and I have seen no previous data on this phenomenon.  Is
>> such data out there?
>>
>
>
> I have not seen or worked with a stove that has a primary air damper. I
> have worked with feedstocks that, if too fined grained, reduce draft too
> much. Then the TLUD doesn't work - it smolders and smokes badly.
>     *[RWLd3:  Agreed.   Not every combination of stove and fuel is going
> to work.   I think the ability to change primary air supply is hugely
> important to a cook (and to optimize chances of excelling in the water
> boiling stove tests).  This is not to say that in your (char production)
> situation that controlling air is important  But you probably can help
> those of us trying to improve and understand char-making stoves.  I don't
> understand enough of your work, but believe you also would like to know
> more about the production temperatures for the char you are producing.* ]
>
>
> You can call our awkward laboratory experimental TLUD setup a "non-stove"
> if you want, but it produces chars that are physico-chemically
> representative of the range of chars that come out of TLUD cookstoves.
>
>    *[RWLd4:  I am complimenting you on your work - which is not
> stove-optimization oriented.   I believe you would only be able to cook
> satisfactorily if you did two things - drop to 1 or 2 upper cans (partly to
> get the power level down) and b) control primary air.  You are presenting
> results that should be very helpful to both char-cook-stove designers and
> char-users - and that I have not previously seen.*
>
>    *I hope others will join in if they have any char-bed (not flame)
> thermocouple data - preferable under different operating conditions - such
> as Josh has shown.*
>
>
>      *<moved part e  to below   -     and snipped 4-5 earlier posts.  To
> understand my thermocouple questions, see Josh's late afternoon (Colorado
> time) message from Wednesday which shows a photo and two graphs.>
> *
> Ron   (end of Friday responses)
>
>
> Moved portions from Josh today (no need to respond to these)
>
> *RWL
>
> *
>
> *On Thu, Dec 6, 2012 at 1:03 PM, <rongretlarson at comcast.net> wrote:
> *
>>
>> Josh  (cc list)
>>
>>    Nice data - thanks for sharing.
>>
>>    Several questions:
>>
>>    a.  In the ND case, I am surprised that the upper thermocouple showed
>> a sharp drop about 25 minutes prior to the almost identical drops of the
>> lower two.  It is almost as if it fell out - as the dark black curve came
>> back near the end..  Your explanation?
>>
>
>
> *Josh*
> a.  This is indicative of the thermocouple shorting out. When the probes
> malfunction the loggers will record nominal values of about 80C. Several
> firings eventually damage probes, even those rated for high temperatures,
> so unfortunately I see this kind of thing frequently.
>
> N.B.: One things I've found over the years, especially with monitoring
> traditional charcoal kilns, is that even high temperature probes with
> kevlar insulation around the wires break down quickly - not because of the
> heat but because chemicals in the smoke break down the fibers. (This is a
> serious bummer when you're doing research on a shoestring budget and probes
> cost $40 a pop.)
>
>  *RWL*
>
>>
>>    b.  There was somewhat similar behavior in the FD case, but no
>> "falling out".  I would have expected all probes to have dropped in
>> temperature about the same time and manner - why the upper probe behaving
>> differently?  Faster cooling due to secondary air nearby?
>>
>
> *Josh* b
> This is because by the end of the run the top probe was no longer embedded
> in the char - the char had subsided below the level of the top probe. The
> other probes were still buried in char and so were insulated during the
> cooldown phase.
>
>     <Above, in first part of this, are RWL responses to most of c and d,
> leaving just these two moved down>
> *
> *
> *Josh*
> c1.5   One reason the unit is tall is because I had to move the fan far
> enough away from the reactor so the heat wouldn't burn it out. For cooking
> you would possibly solve this by putting the fan off to the side. For me it
> doesn't matter if it's awkward or inconvenient to cook on, I just want to
> generate and test the char.
>
>
>  *RWL*
>
>    e.  There were several cans below the fuel can with the 3-4
> thermocouples.  What was their purpose?
>
>
> *Josh*    Answered above.
>
>
> End of RWL response
>



-- 
Josh Kearns
PhD Candidate
Environmental Engineering
University of Colorado-Boulder

Director of Science
Aqueous Solutions
www.aqsolutions.org

Mobile: 720 989 3959
Skype: joshkearns
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.bioenergylists.org/pipermail/stoves_lists.bioenergylists.org/attachments/20121208/a6d00224/attachment.html>