Velocity of thermal siphoning?

[Derek]

I've been working on my passive solar heater for this upcoming winter, and have figured out the BTU's and such.  Of course I'd like to add about a 100 CFM blower to it to help with the output.  With a .087 sq ft duct area, this works out to be about 1,150 ft per minute on the velocity.  For figuring cubic feet a minute, you take the velocity times the square foot area of a duct.  

What I'm trying to figure out though, is what is the velocity of the air coming out of a passive solar heater that just uses thermal siphoning.  Now I'm sure it hight depends on the difference in temperature of the air change.  So say a 50 degree difference would cause the air to move at xx feet per minute then.

How can I figure out how many cubic feet per minute my passive solar heater is putting out as is? (without a fan that is)  Its generally heating the air by over 50 degrees from input to output.

Please help, its driving me mad!  haha

[ghurd]

Why not use, say 2 75CFM muffin fans, a thermostat, mosfet... and not worry about it?

Or a 2W solar panel, and 70% that many direct drive peak ma of muffin fans?

Seems like any formula will have exact constants, that the system will not have.

It's the kind of thing that gooks up my plans.

G-

[Derek]

I know that fans would be the best thing.  But in terms of effeciency, I'm trying to figure out exactly how fast heat will rise.  Anybody know how fast heat actually rises?  My best guess is something like 2 feet per minute per degree of difference, but I could be way off.  I'd just like to figure it out, but cant seem to find any info like that online anywhere.

[thunderhead]

You can work out the difference in density, which gives you the pressure; and then from that and Bernoulli's Theorem you can work out the velocity.

The density is worked out for an ideal gas by working out the difference in absolute temperature (temperature above absolute zero).  So for fifty degrees of difference, you might have 300K on the cool side and 350K on the warm side.  If air has a density of 1.33 kg per cubic metre at 300K, it has a density of 1.14 kg per cubic metre at 350K.  That's a density difference of 0.19 kg per cubic metre.

Pressure is given by density times height times the acceleration of gravity.  If your thermal syphon is two metres high and gravity is 10 metres per second squared, the difference of density of 0.19kg/m3 gives a pressure difference of 3.8 newtons per square metre.  For comparison, atmospheric pressure is about 101,300 newtons per square metre.

By Bernoulli's Theorem, pressure is equivalent to density times the square of speed, multiplied by some constant that represents the drag coefficient of your ventilation ducts.  Assuming no drag (the constant is one) the fastest speed of air you'll get for your two metre duct is 1.8 metres per second.  But a more likely figure for the drag is 0.1 or less (the best aerodynamic shapes do well to get 0.3) so a more believable figure is 0.58 metres per second.

That's nearly two feet per second.  Maybe it'll be a little more, maybe it'll be a lot less.

Chimneys obtain better figures than this with increased temperatures (a fire can easily reach 500K) and hot water thermosyphon systems use a much higher density figure (1000 kg per cubic metre), and so a much higher pressure difference.

Better alternatives might be a water system -- a solar panel on the outside wall that drives a radiator on the inside wall -- or, as others have suggested, a PV panel and a fan.

[Derek]

Excellent!  Thats more of the info I was looking for, thanks a lot!  The only thing (and it sucks) is that I'm not too good with metric, and everything else is in our stupid standard measurement (sq ft and inches) so I'm trying to translate it out.

Can you help me with what those numbers would be in square or cubic feet?

Again, I realize that using fans and such will be the way to go in the end.  But obviously the temperature will drop when you use one.  So which way is more effecient in terms of cost and output is what I'm trying to figure out.  Thats why I want to figure out what the speed of the thermal siphoning is.

I do think that running a PV panel and a fan is how I'll end up with the project.  I've messed with that a little bit before.  Although I'd say the thing needs a capacitor for a little extra jump start to get the fan moving initially.  I'm looking at doing about 100 cfm with a fan.  But if I can keep my temperature up while running it, that will be a great accomplishment.  Basically, my current design might be able to provide about 10,000 BTU's per hour.  Sound good? =)

[craig110]

One thing to keep in mind is that the heat you get out of the pipe is limited to the amount of sunshine that is hitting it.  Moving the air faster means that the output temperature has to be lower than without a fan simply because the same solar energy per unit of time is being spread over a larger volume of air.

Regardless, I agree with adding a small fan to it.  I think that heating a room with a lot of warm air will give a better liveability result than heating it with a tiny amount of really hot air.

Craig

[GaryGary]

Hi,

One thing to be careful about is that the 100cfm fan only puts out 100 cfm at its rated pressure drop.  If you collector is producing more or less than this (which almost certain), then the flow rate will be different than the rated flow.

--

You can measure the velocity on a thermosyphon collector with one of the Kestrel (or equivalent) wind meters.  If you have the thermosyphon collector designed to the usual guidelines (see below), and its around 8ft tall, the velocity at full sun will probably be around 120 ft/min.  Another cheaper velocity meter by Dwyer that works well is listed at the link below -- I think its about $25 -- its a pretty neat little gadget -- very simple.

My thermosyphon collector with full sun produces about a 60F temperature rise with about 120 fpm exit velocity.  Each collector bay is 32 sqft, and has 1 sqft of exit vent area -- so, its about (1sqft)(120ft/min) = 120 cfm per 32 sqft of collector.  

The heat output for this combination is about:

  (120 ft^3/min)(0.065 lb/ft^3)(60F)(0.24 BTU/lb-F) = 113 BTU/min or 6740 BTU/hr

Where 0.065 is the density of the warmed air, and 0.24 is the specific heat of air.

I built mine to the guidelines that the depth of the collector should be about 1/16 of the height (so 6 inchs for an 8 ft high collector), and the entry and exit vent area should be at least half the cross sectional area of the collector (so for a 4ft wide collector that is 6 inches deep, the cross section is 2 sqft, and the exit area and entry area should each be at least 1 sqft) -- more is better.    The flow through absorber also needs to have minimal flow resistance (I used 3 layers of black window screen).  These guidelines come from Steve Baer's book.  I think that if you follow these guidelines, the efficiency improvement you get by adding a fan will be minimal, and probably not worth the bother and expense.  I've done a rough check on the efficiency of mine using the procedure below, and its good.

This is the collector:

http://www.builditsolar.com/Projects/SpaceHeating/solar_barn_project.htm

This is a page that has the air flow meter from Dwyer and other instrumentation information:

http://www.builditsolar.com/References/Measurements/measurments.htm

This is the procedure I used to estimate collector output and efficiency:

http://www.builditsolar.com/References/Measurements/CollectorPerformance.htm

Its not exact, but it works pretty well.

Gary

[thunderhead]

The only thing (and it sucks) is that I'm not too good with metric, and everything else is in our stupid standard measurement (sq ft and inches) so I'm trying to translate it out.

At the risk of kicking off a flamewar between former colonies and Europe -- less than a week before I emigrate to the one sizeable nation that is both -- I would suggest that the metric system is well worth the time to learn.

It's not hard to learn.  In Imperial calculations, you start with the theoretical formula (for example, power equals force times speed) then multiply it by some magic number (to convert pound feet squared per second cubed into BTU per hour).

In metric, you do the same thing, but without the magic number.  So multiplying a force in newtons by a speed in metres per second gives a power in watts, no magic numbers needed.  That makes the calculations simpler, which (for someone as prone to mistakes as me) makes an accurate result more likely.

Metric is Imperial without the conversion factors.  So if you can do Imperial you can do metric.  It's only poor disadvantaged people like me who only do metric.

If I had to work through the calculations in Imperial units, I'd have to look up all those magic numbers.  There are probably people here that are familiar enough to know them off the tops of their heads.

If I just needed the answer in Imperial units, I'd do the whole calculation in metric and convert at the end.  You'll notice, though, that my 2 feet per second agrees with GaryGary's 120fpm.

[Derek]

I wish that we'd just use metric here, or that they would have taught that to us in school, its so much easier!  Good point about working it out in metric, then converting it at the end.  Only problem is all of my formulas are all in standard imperial though too, for all of the other parts of the equations I have.  Plus thats the only measurements anyone understands here too.

I've also debated on trying window screening actually, a few layers of it for more surface area than just a flat surface.  How has that worked out for you, as in whats your temperature input and output on something like that?

[wdyasq]

I see I am not the only one who read "Sunspots".

All systems for fluid flow do better with smooth surfaces and large cross-section piping to transfer fluid. Gasses are considered fluids. Prediction of flow and velocities is "good guess-work" even by air-handling "experts", IMO.

Thermosyphon systems depend on solar radiation, a variable, positioning of panels, another variable, sun position and cloud cover, more varaibles, and the air handling "duct work", an unknown in this case. Fans are a great "CYA" method of insuring a working system.

As all the temperatures and energy for calculation are in "Kelvin", one should realize using outside air may be a viable option. The reports I have read and on my own collectors, I have chosen this approach. Gary, however, has actually calculated, built and measured results.

As far as complaining about Metric/Imperial, get over it. One can actually buy books old enough to have measurements in Imperial only. Then one only has to wonder, when a gallon of water is mentioned, if it is the 3.63 Kg "American gallon" or the 4.55 Kg "English" gallon. Then you can apply the "Imperial English" conversion to the "American Imperial" figures you have measured (- gee was that multiply by .25 or divide by 25.4 or .3937? or subtract 32 and multiply by 5/9 and add 283... couldn't be the last, 283 is a small block Chevy.) Convert to metric on the front end and forget it.

Ron

[GaryGary]

Hi Derek,

"I've also debated on trying window screening actually, a few layers of it for more surface area than just a flat surface.  How has that worked out for you, as in whats your temperature input and output on something like that?"

I think that the two or three layers of screen works well.  I have several identical collector bays, and tried 1, 2, and 3 layers of screen in side-by-side tests -- the best appears to be between 2 and 3 layers.  I've seen a lot of debate on the merits of different absorbers -- furnace filter media, polyester felt, expanded metal lath, solid metal, etc.  My own prejudice (not based on much) is that there is less difference between these materials than what the proponents of each would like you to believe.  One thing that is important is that if its a thermosyphon collector, and the air flows through the absorber (which is a good way to go), then the flow resistance needs to be low so that you get good flow.

For mine, normal operation in full sun in winter conditions is typically about 60F inlet and 120F outlet.  But, temperature rise is not a good measure of collector output or efficiency.  Its the product of (temperature rise)(flow rate) that determines the output.  You are almost always better off with more flow rate and less temperature rise -- this reduces losses out the glazing, and improves efficiency.  If you are getting a larger temperature rise that 50 or 60F, you collector flow path may be to restrictive, and you need more flow rate to keep the temperature rise (and losses) down.  The link in the post above on measuring efficiency of collectors gives more detail on this.

Hope that makes sense?

Gary

[GaryGary]

Hi,

"I see I am not the only one who read "Sunspots"."

Right -- along with the Shurcliff "100 Inventions" book, its one of my favorites.

"All systems for fluid flow do better with smooth surfaces and large cross-section piping to transfer fluid. Gasses are considered fluids. Prediction of flow and velocities is "good guess-work" even by air-handling "experts", IMO."

Good point -- especially for the thermosyphon, smooth flow passages should have some payoff in better airflow.  Large entry and exit vents definitely make a difference.

"Thermosyphon systems depend on solar radiation, a variable, positioning of panels, another variable, sun position and cloud cover, more varaibles, and the air handling "duct work", an unknown in this case. Fans are a great "CYA" method of insuring a working system."

My (slight) objection to fans and controllers is that its just more stuff to take care of, and people tend to just let things go when the fan stops working.  The thermosyphon collectors just keep working.  

The nice thing about fans is that you don't have to cut so many holes through your walls

"As all the temperatures and energy for calculation are in "Kelvin", one should realize using outside air may be a viable option. The reports I have read and on my own collectors, I have chosen this approach. Gary, however, has actually calculated, built and measured results."

I guess this depends a lot on climate, and what you want to use the air for, but it does not seem like a real good bet in Montana when is -20F outside

Gary

[Derek]

Currently what I'm doing is having the air flow through thin tubes made of either plastic or aluminum.  Of course they are black.  What this does is actually allow the thermosiphoned air to remain seperated from the internal box temp.  Basically, if it clouds up just a bit, the box holds heat for a while, and transfers it into the tubing.

If you have window screening, I assume you have the hole cut in front of the box on the bottom and on the back top area so that the air passes over all of the window screening?  Or maybe flip the front/back holes, but is that the general idea?

My thoughts on using window screening is that I would have the entire box at a 45 degree angle, and then the window screening would be parrallel to the ground every inch or two, like stairs.  So the angle would allow the sun to hit all of them, and the air passes over each of them.  If you have a 10x10" area of window screening (100" sq), is it safe to assume that maybe only 50" sq allows for air flow through it?

[wdyasq]

Derek,

If you haven't read "Sunspots", I can't recommend it high enough. It is very basic with explainations and tests to backup what Steve Baer sugests and implies. I think the ONLY place I will disagree with th book is on rocks as a storage medium. It may work in a very arid climate BUT- there have been some mold problems in more humid climates.

The research I read on using outside air stated as much was lost by putting already heated air in contact with frigid glazing cost a lot of energy. The Higher "Delta" made up those losses quickly, if I remember the reports properly.

Ron

[GaryGary]

Hi Derek,

"If you have window screening, I assume you have the hole cut in front of the box on the bottom and on the back top area so that the air passes over all of the window screening?  Or maybe flip the front/back holes, but is that the general idea?"

The air enters to the south (sun) side of the screen, then flows through the screen and  and out an exit vent that is on the north side of the screen.  The diagram in the article shows how its arranged.  The entry vent is on the back of the collector, entering air flows under the screened area to get to the south side -- see the diagram in the article at the link below.  I think that forcing the air to flow through the absorber makes for good heat transfer from the absorber to the air.

Diagram here: http://www.builditsolar.com/Projects/SpaceHeating/SolarBarn.pdf

"My thoughts on using window screening is that I would have the entire box at a 45 degree angle, and then the window screening would be parrallel to the ground every inch or two, like stairs.  So the angle would allow the sun to hit all of them, and the air passes over each of them.  If you have a 10x10" area of window screening (100" sq), is it safe to assume that maybe only 50" sq allows for air flow through it?"

As I recall, the window screen is more like 80% open.  You can work it out by measuring the screen wire diameter, and then counting the wires per inch, and then calculating the blockage.

I'm not sure how well the stair arrangment would work -- it might be fine.  It does seem like a lot more work though to support all those separate screen stairs?

What advantage do you feel it has over the simple arrangment of a couple layers of screen parallel to the glazing?

Not sure if this is an option for you, but mounting a space heating collector vertially (on a south facing wall) has some advantages.  It collects about the same amount of energy in the winter as a tilted collector, but it collects much less heat in the summer than a tilted collector -- so it does not tend to overheat as much in the summer.  If you have a snow field in front of the collector, the vertical collector actually collects a fair bit more energy than the tilted collector due to the sun reflected off the snow.

Gary

[Derek]

Gary, here's what I was thinking on the stair step approach to the window screening.  If you have all sheets parrallel to the glazing, then I feel like the air will flow just fine through the center of the box in a near direct line between the input and output ducts and not do the extra work to go out and around and across the side edges of the screening.  So I'd think there would be some lost heating to that.  

With the stair step, I'm thinking that the heats natural convection would be more easily harnessed too.  Plus, each level of screening is going to be in 100% sun, and increasing the amount of maxed out heat temps the air will pass over.

On screening parrallel to the glazing, the front screening will be the hottest obviously, and then the 2nd and 3rd ones will not be absorbing as much infrared because of the first screen blocking the light.  With the stair step approach, I'd think the air would pass through much more maxed out heated surface area, as in each screen will put out the max temp it can.  Where as the 2nd screen is less than the 1st, and the 3rd less than the second.

If the whole unit was constructed at about a 45 degree angle where I live, every screen would actually be in 100% full sun all of the time if the proper measurements where applied.  But not only would each step be in 100% sun, you could still harness your 2-3 screens along the back for hours where the sun is lower in the sky.

Basically, its "a semi-solid vertically corrugated aluminum screening system".  I think that will be the official title of it....  But the idea is not to block some of the rear layers and provide More surface area then, what do you think!?!?  The whole thing is just an idea on paper for me, but will be trying it out I'm sure here shortly.  Anyone have any good or bad thoughts about the idea?

[GaryGary]

Hi Derek,

I'm not sure if the stair step arrangement would work out better or not -- it might.

One thing I would recommend is that if you go ahead and build the stair step arrangement, that you also do a 2nd one with just the 2 layers of screen parrallel to the glazing, and do some side by side measurements.  This way you can pass on what you find out, and really understand which works better and by how much.  The 2nd collector could be very simple and cheap -- kind of a throw away.

You would need to measure entry and exit vent temperatures and the airflow velocity in order to see which collector is putting out the most heat.

On the issue of the air flowing primarely through the center of screen layers that are parrallel to the glazing, I think its more likely that (since the exit vent is right at the top) that the flow might concentrate more at the top area.  But, if the screen offers some resistance to flow, this will tend to spread the flow out more evenly -- not sure what is really happening.  I have a new smoke generator that is intended to allow you to see where the flow is going -- I might have a go at trying to use it to see where the flow is going and how even it is.  Its an interesting question.  I guess another way to get an idea would be to measure the absorber temp at a number of points -- if its hotter in some places than others, thats an idication that not as much flow is getting to the hotter areas.

Gary

[Kwazai]

http://www.engineeringtoolbox.com/convective-air-flow-d_1006.html