# The future in an aircraft charging system

Discussion in 'Hangar Flying' started by Pops, Jun 18, 2019.

### Help Support HomeBuiltAirplanes Forum by donating:

1. Jun 18, 2019

### Pops

#### Well-Known Member

Joined:
Jan 1, 2013
Messages:
6,663
5,547
Location:
USA.
Vigilant1 likes this.
2. Jun 18, 2019

### Swampyankee

#### Well-Known Member

Joined:
Dec 25, 2015
Messages:
1,389
355
Location:
Earth USA East Coast
Thermoelectric generators are hideously inefficient, but waste heat is waste heat. Maybe putting them around an exhaust pipe would be better than a valve cover

3. Jun 18, 2019

### Pops

#### Well-Known Member

Joined:
Jan 1, 2013
Messages:
6,663
5,547
Location:
USA.
Looks like at this time they will not take that high of a temp to use the exhaust.

4. Jun 18, 2019

### Vigilant1

#### Well-Known Member

Joined:
Jan 24, 2011
Messages:
3,551
1,528
Location:
US
Wow, I haven't looked at these in a long time, they have really come down in price.
It looks like it would take a lot of them to provide power at "moderate" temps. If we assumed a hot side of about 300 deg F (valve cover?) and a cold side of 85 deg F (temp inside the cowl? Their charts don't go any lower), that equals 150C on the hot side and 30C on the cold side. From the charts, under those conditions each chip will give us about 3 V and 1.5 A. So, if we ran 4 in series we'd get 12V and 1.5 A = 18 Watts.
But, if we could find a nice spot on the exhaust pipes somewhere that is reliably 575 deg F (300 deg C), two chips in series would give us 14V, and about 6 amps = 84 watts. That's pretty good for the cost (less than $120), weight, and simplicity of this approach. Enough to run a radio, a low pressure fuel pump, lights, and get the battery charged up for another start in an hour or so. Last edited: Jun 18, 2019 Pops likes this. 5. Jun 18, 2019 ### Pops ### Pops #### Well-Known Member Joined: Jan 1, 2013 Messages: 6,663 Likes Received: 5,547 Location: USA. That is what I was thinking. VW valve pans welded up from aluminum, Flat surface instead of rounded. Also could be used under the cylinders or heads to collect the hot air downstream . Grind the fins off the bottom of the case and weld a flat plate on. Square box welded around the exhaust stacks with a certain gap between the stack and the thermoelectric generator for the temp (575 F) needed. Lots of different places and ways. I like the light weight and simplicity. Vigilant1 likes this. 6. Jun 18, 2019 ### Vigilant1 ### Vigilant1 #### Well-Known Member Joined: Jan 24, 2011 Messages: 3,551 Likes Received: 1,528 Location: US Getting the hot side temp right might be fun, a little science project. They can tolerate up to 750F intermittently, but only 626 F continuously. Maybe mount them on a piece of thin AL spaced a bit away from the exhaust pipe (1/2"?) with airflow through the gap. Use a bimetalic strip that would pull the AL farther away from the exhaust pipe if the temps get above about 600 deg on the chip. Edited to add: Oops, I cross-posted with Pops. 7. Jun 18, 2019 ### pictsidhe ### pictsidhe #### Well-Known Member Joined: Jul 15, 2014 Messages: 5,869 Likes Received: 1,488 Location: North Carolina It depends on the material used. Metallic ones are basically a lot of thermocouples in series and can take very high temps, but are lower efficiency. Mini ones are used in furnaces to hold the fuel solenoid open. If you have reliable cooling on the cold side, you can lightly insulate the hot side so it does not reach full EGT. Much safer is to suck up the efficiency loss and heat them with oil. The real downside to TEGs is the cooling required on the cold side. You need a lot of cooling fins, that is not light. They also work badly until the oil gets hot Quick numbers: 150C hot side (oil) 80C cold side 1.64lb heatsink ~10W and maybe 2lb. You'll need a few. 8. Jun 18, 2019 ### Hephaestus ### Hephaestus #### Well-Known Member Joined: Jun 25, 2014 Messages: 763 Likes Received: 155 Location: YMM C'mon - if you're going to scavenge heat - why not do a small Sterling engine exhaust on the hot side, outside air on the cool... You could probably get a bit of real hp out of it if you wanted to waste a few lbs 9. Jun 18, 2019 ### pictsidhe ### pictsidhe #### Well-Known Member Joined: Jul 15, 2014 Messages: 5,869 Likes Received: 1,488 Location: North Carolina Then you need a hefty heat sink and a Sterling engine... 10. Jun 18, 2019 ### BBerson ### BBerson #### Well-Known MemberHBA Supporter Joined: Dec 16, 2007 Messages: 11,405 Likes Received: 2,092 Location: Port Townsend WA And a generator. 11. Jun 18, 2019 ### Pops ### Pops #### Well-Known Member Joined: Jan 1, 2013 Messages: 6,663 Likes Received: 5,547 Location: USA. Run air ducts to the cooling fins like an oil cooler. Short pieces of scat tubing weighs very little. Thin cooling fins like used in an oil cooler maybe .010 . On the VW engine, I always make an aluminum baffle under the fins on the bottom of the case and bring in air from the nose bowl. Just do the same thing 12. Jun 18, 2019 ### Vigilant1 ### Vigilant1 #### Well-Known Member Joined: Jan 24, 2011 Messages: 3,551 Likes Received: 1,528 Location: US It will take some mass in the heat sink to make it work. The chips are small (2.2 " x 2.2") and they'll have a few thousand watts of heat energy going through that square. It will take a thick chunk of copper etc right over the chip to move heat out to the fins. But the available rapidly moving cool air makes things much easier. A heat pipe to a finned plate in a convenient spot would work. A heat pipe can move hundreds of times more heat than a similar size copper rod, and the cool.end could be nice and fat to cover a lot of area. No mechanical bits, but more complex than SCAT tube and a finned chunk of metal. Last edited: Jun 18, 2019 13. Jun 18, 2019 ### pictsidhe ### pictsidhe #### Well-Known Member Joined: Jul 15, 2014 Messages: 5,869 Likes Received: 1,488 Location: North Carolina A few hundred watts of heat through the TEGs, not thousands! It's the fins that get heavy. You need a lot to disspitate all the heat and keep a good delta across the TEG. Cpu coolers would be worth a look. Heat pipes have a power limit above which they run into a wall. You won't know until you try. The heat pipe ones tend to be rather heavy. I found the folded fin heatsinks to be the lightest per dissipated watt when I was playing with computers. As a bonus, they are pretty cheap, too As an idea, the 1.64lb heatsink that I linked to is adequate for 10W of TEG if air is blown over it, it has a fin area of 2.5sqft, so think 0.25sqft of fin per generated watt. Fitting TEGs directly to the sump may work for the hot side. Then clamp them there with suitable heatsinks. I think I'd favour finding some lightweight cpu coolers and then selecting TEGs that will work with them. Fans would be swapped for cool air ducting. It's the heatsinks that are going to be most of the weight. 14. Jun 18, 2019 ### Vigilant1 ### Vigilant1 #### Well-Known Member Joined: Jan 24, 2011 Messages: 3,551 Likes Received: 1,528 Location: US Depends. Per the factory chart and my example case of 84 watts (14V, 6 amps): The factory says they are 5-6% efficient, that's 1,216 to 1,596 watts of waste heat to make the 84 watts. That's 600-800 watts per TEG chip. So, less than my "few thousand watts" and more than "a few hundred watts." But, if we just need/want 7-10 watts of power per chip, and we can operate the hot side in its efficient zone (approx 350F and above), then we'll only need to dump a few hundred watts of heat per chip. 15. Jun 18, 2019 ### pictsidhe ### pictsidhe #### Well-Known Member Joined: Jul 15, 2014 Messages: 5,869 Likes Received: 1,488 Location: North Carolina You don't want to bunch them up together, that makes cooling them even harder. You will need several anyway, so spread them out and give them each an individual heatsink. 5-6% effciency is only if you keep the cold side cold, rather than warm, and the hot side close to max temp. You will need a very heavy and expensive heatsink to keep the cold side cold and running near the hot limit is an inviation to frying them. It's waste heat, take the efficiency drop and use more units in a less efficient (3%), but cheaper (smaller heatsinks), lighter (smaller heatsinks) and more reliable system. Maybe you can run the hot side higher than oil temp from the exhaust, but I would personally experiment on something cheaper to get that right. A dummy TEG with the same thermal conductivity that won't blow a$50 hole in your wallet if it overheats seems the way to do it. A real TEG and heatsink hooked up to an ammeter is the simple way to test possible safe locations.
Meanwhile, you still need power to fly and test...

Each of those modules contains many individual chips in series. Look at the performance graphs to get an idea of achievable power per module. When sizing a system, don't forget that to halve the temperature delta on the cold side, you will need to double the size of the heatsink. Except that they don't quite scale like that, so a bit more than double.

14v 6A will be expensive and heavy.

Last edited: Jun 18, 2019
16. Jun 18, 2019

### Pops

#### Well-Known Member

Joined:
Jan 1, 2013
Messages:
6,663
5,547
Location:
USA.
I should have worded it different. Didn't mean to say to use Scat tubing, just the idea of high pressure air piped in to the cooling fins like the use of scat tubing to a oil cooler.

I like the idea of the chips mounted to the VW square valve cover pans with heat sink grease and thin copper folded fins on the chips and an aluminum air shroud to direct the air over the fins of the chip. Used on VW exposed cylinder installation like a J-3 Cub
Also the used of a wrap around box on the exhaust pipes for the higher 350 F temps.

17. Jun 18, 2019

### pictsidhe

#### Well-Known Member

Joined:
Jul 15, 2014
Messages:
5,869
1,488
Location:
North Carolina
Hmmm, would it be possible to machine some fins off a VW head and replace with a TEG and heatsink? Size it well, you could improve the cooling and get free electricity.
I have an idea for moderating the exhaust temperature. Mix it with air before it gets to the TEG. You don't want whatever the TEG is mounted to to ever get above it's max temp. That will include heat soak when the cooling air stops.

Pops likes this.
18. Jun 18, 2019

### 12notes

Joined:
Aug 27, 2014
Messages:
692
445
Location:
Louisville, KY
If the mixture is slightly rich, adding air to hot exhaust might raise the temperature, as it provides oxygen for the unburnt fuel. Not sure on the exact temperature for that to happen, but I think that's how the air pump in automobiles brings the catalytic converter to operating temperature faster.

My non-professional opinion is that mounting it on a plate with short standoffs might be a better method for reducing the heat to tolerable levels. It wouldn't be good for corrosion purposes, but stainless standoffs to an aluminum plate would work well from a thermal conduction standpoint, as the aluminum plate would raise to an almost even temperature while the stainless standoff would limit heat transfer from the exhaust, different size standoffs would allow you to "tune" the amount of heat transfer.

Last edited: Jun 18, 2019
spaschke and Pops like this.
19. Jun 19, 2019

Joined:
Jan 1, 2013
Messages:
6,663