[Stoves] Using a differential windlass as motive power for stove

[AJH]

On Sun, 10 Dec 2006 22:02:51 +0000, Steve Taylor wrote:

>AJH wrote:
>> I'm not sure on that, I know you'll need to move 400 times the volume
>> of air past the cold junction than water but will the water be
>> intrinsically a better interface? Does it have a different boundary
>> effect? 
>It has a massively higher thermal conductivity for a start, it has a 
>vast heat capacity...are we betting ?

In this instance I doubt the thermal conductivity counts for much nor
the heat capacity it's the ability to carry away the heat, so it's the
heat transfer at the interface that counts. We know from car engine
cooling systems that the water is better able to maintain a uniform
engine temperature but ultimately it's the way that the car "radiator"
transfers the heat to air being moved through it that dumps the heat.

Now you may well be able to reason that water is more effective at
moving heat from a surface, I've already said that you need to move
about 400 times the volume of air compared with water to deliver the
same amount of cooling.

As to betting, que sera sera,  (sp?) my bet will not change the
outcome. If we define the aims as a means of producing sufficient
electricity to power a fan enough to create combustion conditions that
reduce particulates (by a factor of what? say 3 though 9 has been
mentioned) over a similar naturally aspirated stove and with the least
parasitic losses then I still stick with the air cooled system, simply
because unless the water coolant is used in the cooking its heat is of
no use to the majority of our stove customers.

>Pick your metals.

For the experiment or for a thermopile?

I think alumel and niconel are used in thermocouples

I'm at a disadvantage in never having worked in engineering. I don't
know the properties of a lot of these metals. You say to stamp out
large diameter belleville washers is wasteful, also that a lot of the
metals aren't readily available. I worry about corrosion in iron,
though I chose iron and nickel because I thought they were commonly
used materials.

On the other hand I have some nickel welding rods for joining
dissimilar metals, they were very expensive, and some nichrome wire. I
have a sneaking suspicion any slight changes to alloy these metals
will have a significant effect on the voltage produced.

As I said all the heat flux has to pass through the junctions, so if
we optimistically assume the pile will be 1/2% conversion then for 3W
we need a heat flux of 600W(t) through the junctions, this is 1/5 of
the power of a typical 3kW(t) stove. If we don't recuperate the heat
with the incoming combustion air steam then the thermal cost is too
high.

I also don't have much of a grasp as to the size and thickness of the
junctions to provide enough current. I do know that a commercially
available bimetal thermopile is made to hold open gas valves on a
cooker operated away from the grid. Honeywell Q313A generates 750mV
but the current is not specified. 

I have previously tried to contact the thermo electric group at
Cardiff University to see about some expert help but they did not
respond.

Now if the chosen metals are malleable and we need to use strips and
spot welding then a truncated cone could be formed by rolling one edge
of a strip and then joining the ends?

Perhaps we need to form rings of wire, like jewelers do and then forge
them into belleville washers?

AJH