Forced Panel Ventilation

[Madscientist267]

In reference to http://fieldlines.com/board/index.php/topic,144030.0.html I wanted to expand on the idea but in a slightly different direction, so that's why the new thread for it.

I'm looking into force ventilating the mobile panels to suck out that extra 15 watts, since I'm really noticing the effects of heat and summer is just around the corner.

I have the main fan/duct assembly made, and will be powering it from an auxiliary panel. Gets good airflow in full sunlight, and not too shabby flow in moderately overcast skies too (where you can still make out the solar disk).

The fans are 80mm, and there are two of them in a 'jet' configuration, with buffer chambers both before and between the fans to help optimize their operation:


End View. The fan is recessed to reduce turbulence since the table will be blocking a band in the center of the intake.


Side View, detailing the construction. The 'ducts' are old fans with the blade/motor assembly removed. I hang on to them specifically for this kind of thing.


My problem comes in the form of determining the best method for routing the air behind the panels. I've decided on suction rather than pressure, since I can control the air flow pattern more closely.

I am either looking at slots at either end of the table to let a 'band' of fresh air in, or multiple different size holes to provide more uniform cooling:


The smaller dots are a little difficult to see in this picture, but they are toward the inside (near the fan, orange circle).

I am leering away from the slots because I am thinking that as the air moves toward the center of the table, it will be warmer, and therefore would reduce the cooling of the inner cells of the panels, defeating the purpose.

What do you guys think?

Steve

[Madscientist267]

Wow, nobody? LOL

No yay, no nay, no [censored]? 

Hmm... Guess I'll run with it...

Steve

[zap]

Wow, nobody? LOL

No yay, no nay, no [censored]? 

I read it yesterday...
I figured you already have entirely too much fun as it is... 

Actually, my thought was that a 'net gain' would be pretty much out of the question everything considered... but since you were going through with it then it was going to be fun watching... and waiting for the outcome.

[DanG]

In a perfect example heat moves equally in three ways - conduction, convection and radiation. Glass PV panels have low conduction and a smooth surface unable to transfer much heat to the air surrounding it. That leaves only radiance cooling with heat leaping away in the infra-red like a floodlight.

If you have armored the panel back then about 2/3rds of the panels total absorbed heat is being trapped and bounced back at the panel - the other 1/3rd might make it out the front to beam to the sky. Open up the back (if you have the cells encapsulated) and let it radiate that away, you'll never reclaim the fans parasitic energy.

[Madscientist267]

LOL Now you tell me...

After I already started on it.

I figured it would be a marginal increase, but when I saw the 4+ amps out of the buck the other day, I got excited, and came to the conclusion that this was the only missing piece to the puzzle.

It is finished now, I'll get pics put up in a few...

The fan runs from an auxiliary 8W panel, so there's no loss either way, other than some time and effort on it.

It's hard to say right now if there is any real improvement or not, I won't know until I see it perform in various conditions.

The fans are moving heat out of it, as the air is warm leaving the exhaust, but how much good it is doing is still left to be said.  

Oh well, it was fun and I (possibly?) learned something from it... haha

Steve


EDIT - The pics:


The holes all marked, ready for tooling.




The holes drilled. I went with a slightly different pattern than originally envisioned; and ended up revising them once again on the fly. I somehow missed the outer set of holes while I was drilling, and so rather than enlarging the outer sets, I drilled another set next to them where the bigger holes were supposed to go in the first place. Oh well.




To do this properly (or as close to as possible), I had to carve out a majority of both junction boxes. They weren't really able to do their job as designed anymore anyway, so since this is a portable application and would not be left out in inclement weather, I did away with them as much as I possibly could in favor of cooling...




Close up of one of the modded J-Boxes.




A closeup of the fan assembly mounted on the back side of the table. Doesn't work too bad, and can pump some decent air, but as DanG pointed out, a good portion of the heat is radiant, so without some kind of heatsink backing on the panels, the gains are minimal.




Here it is operational, with the auxiliary panel that powers the fans sitting on the left.



I don't know that I would do it the same if I did it again, time will tell. I wouldn't have bothered except the 8W panel is so small, it doesn't really find much use for very much else, and it would run the two fans (in series) rather well, so I ran with it.

Somehow, I envision a water cooled layout in a future version of this, but I doubt I would bother on another portable system. I could use that surplus heat for something else, but water cooling is clumsy in a portable implement, so the pros wouldn't really outweigh the cons, methinks.

Oh well, it was interesting, and it MIGHT give me some gain, will have to see when summer rolls around full throttle and run some tests to really find out.

Steve

[Volvo farmer]

I think you would get more watts and be dollars ahead if you just added the small solar panel to the larger array, rather than fool with this fan-cooling idea.

[Madscientist267]

Ahh, except that the large array is in series, which then goes thru a buck converter which then charges the 12V battery.

OTV from the panels is only 39.5V thanks to the idea that I had to bypass two cells in the one panel (one of the cells is very obvious in the pics). Since they don't match, I ran them in series instead of parallel.

I theoretically could tie them in parallel without the cell bypass being too much of an issue, albeit at the cost of being off-Ipeak, but it wasn't worth it because the only cable I have long enough to reach outside is a 100ft 16 EDIT: 14/2 extension cord that was modified for the purpose (ends replaced with my standard DC connectors). I get a fair amount of drop in it at 1.75A, I'd lose even more if that were doubled up.

I am slightly ahead by going with series and a buck converter; I've seen over 4A@13.8V (~55W+), and that's running through the long cable. This is more than I would get if the battery were right next to the panels in parallel (they are rated 1.72A Ipeak each). Not bad at all considering the panels are collectively rated at 60W.

So, since the nominals are different, I can't tie it in parallel, and if I were to put it in series, my current output would fall to just over 10% of what it is now (array cops 1.92A Isc (at the panels), aux panel can do 250mA).

Kinda caught between a rock and a hard place, see?  

Steve

[rossw]

In that case, why not rig yourself up a real basic tracker. Use the 8W panel to run the tracker and forget the fans. You'll pick up more power over the day than your 8W worth of cooling will add, methinks.

[Madscientist267]

Have considered it... Even though I manually track them about once an hour or so.

Problem of course with tracking is added weight and complexity, which contradicts portability and simplicity.

Haven't figured out a way to get both worlds to play nice together.

I wasn't super impressed with the performance today, but there were a few things going against it too.

My "pumping" battery (used as a giant stiffening cap after the buck) was significantly discharged because I had forgotten to bypass a diode to let the grid charger keep it up (the buck is parasitic with no input, and the weather here lately has had MPD to put it nicely, sigh). When I connected the panels today, I was getting somewhere near an amp or so (and feared having damaged the panels). Intuition told me to yank the fuse for the pumper, and the current tripled. It didn't hit me til much later that the pumper was simply recovering (rather than damaged somehow as originally thought).

Without that battery in the line right after the buck, I lose quite a bit of charge current to parasitic induction in the wiring headed for the main battery, so I wasn't able to see full output from the system anyway, regardless of panel temp.

Also, the afternoon sun is not as energetic where I live because of my juxtaposition to a paper mill, where the gaseous waste proves regularly to be my arch nemesis. 

Tomorrow, we are supposed to see clear skies and cool air, so tonight the system is stabilizing on the grid and tomorrow the games begin! 
Steve

[mikeburg]

Good Idea, I just wanted to ask if the Fan, Duct,Fan was a Forum idea as I use them for bring warm air down from the ceiling during the winter months. I use a 5" square tube with the single fan at the bottom. I never thought to double them up but I am impressed!
    thanks,
Mike

[Madscientist267]

Naa, but that's not to say that someone hasn't used it before... ?

When the fans are stacked, the pressure increases, but there's a catch. The air has a tendency to spin inside the fans, so when the next one in line catches the air, the turbulence reduces the efficiency.

My solution isn't ideal, because the air doesn't get completely straightened, but there is a noticeable difference with the gap than without.

They also make a lot of noise unless they are gapped out, due to the blades coming in close proximity to the motor mounts from the previous fan.

Ideally, two fans that spin in opposite directions but blow the same direction would be best, or a fan designed something like this inside (and kinda what gave me the idea):


I call this one "the hurricane"... 120mm of pure BLOW!  



Here you can see the stationary vanes that straighten the air back out as it exits the fan, the rotating blades are below.


Unfortunately, this thing draws WAY too much juice (3A+ @ 12V!) to have used it in the panels (not to mention it has a pesky 'I start when I want to' problem). I use it when I need to get a lot of air somewhere for a reasonable amount of power, I just have to give it a spin to get it going. Not sure what the CFM is but at 14.4V, it will refuse to stand up on it's edge without being anchored down...  

Steve

[wooferhound]

You will get a lot more flow if you put the air straightening vanes between the 2 fans.