Also! My big Black Friday splurge was to go ahead and get a foil. I went with this one:
I asked them to send it with the 90cm mast rather than the 70cm.
I tried a couple of paristaltic pumps. Neither of them worked worth a crap.
and this.
The Gikfun does self-prime up to about 6 inches of height, but the other one won’t prime at all.
So after an hour or so of research there is a diaphragm pump called an R385 that might work. I found 2 on Amazon that I’m going to try out. They are probably exactly the same pump, but one costs half again as much so must be better, right?
and this
It does actually look like the motor body is different…
I bought the same pump that is in the bottom photo. It works good, and self primes. I bought a spare because I did not think it would last. So far it still works. I did not use those flexible mounts. I cable tied it to my ESC.
1 Like
What voltage are you running it at for self priming?
12VDC. I have a 1A DCDC converter so I don’t have a lot of overhead. Might have to get a bigger one.
I had problems with the seal on my pelican. It was old or misformed. Emailed them and got a new one for free. No more leakage after that. I always store the pelican with open lid to not compress the seal.
1 Like
I also use the same 12 volt UBEC for my 12 volts that powers the coil on my safety lanyard relay. That is the large relay that disconnects the battery voltage from the ESC when I fall.
Thanks Mac, I’ll pick up of those up (for that price why not?) I had intended to put the DCDC converters on the main controller PCB but offboard works too.
I’ll totally try this. Mine is quite old, but the problem is I got it off Amazon and I think it’s used. Not sure if Pelican will honor that if I’m not the first owner.
The Foil and mast have arrived. It’s heavier than I thought it would be and the wing is HUGE. It’ll be a fine trainer wing but I doubt I’ll get much speed out of it. Also It’s used. But I guess I shouldn’t be surprised considering it was a Black Friday thing and a 2018 closeout. There is no visible damage.
The mast, at 70cm, is clearly not tall enough, so I’ll have to go for the 90cm mast. Trouble is they couldn’t just swap masts so I’ll have to buy it. They offered 10% off, so that’s $150 something - not too bad.
I spent some time tonight CADding up the battery box. I will have plenty of room for the controller board.
The acquisition list is getting scary short. The only part left to get is the waterproof box for the driver. I’m looking at this one.
It’s 9 x 6 x 3. Plenty big for the driver and the water pump.
I’ve also decided that for feedback, for now, I’m just going to put some red and green LEDs in a box with big labels on them and just do threshold warnings. Something like
Remote online 
Flood
Battery Box Temp
Driver Box Temp
Water Flow
Battery Level 
I am looking into using a 7" Paperwhite display like the Kindle has because they are about the only truly sunlight readable displays and I can change the font size so I can read it without my glasses. But I don’t want to drop $80 on the panel until I’m sure I can drive it and so far I haven’t had any luck driving the little one I have in stock.
Soooooo, once I have the Driver Box in stock I don’t have any excuse not to start cutting up the kiteboard.
2 Likes
Do you have a plan on how to implement the threshold warnings? I would like to do a similar thing but I don’t have a plan on how to do it yet. I also suggest adding a loud beeper to indicate the threshold condition. I know I have done damage to things because I was not watching my instrumentation.
As usual I haven’t given it much thought yet.
I am outfitting both the battery box and the motor driver box with this temp/humidity sensor, but the problem is that humidity is not necessarily the same as water ingress. You can be sitting above an inch of water at 50% humidity… So I’ll need to find or make a flood sensor that I can attach to the bottom of each container. That’s really not that hard, a couple of bare wires and a properly biased transistor will do it, but I’ll look for a more modular solution.
After that it’ll just be testing thresholds. Set temp thresholds at 100f to start since it never really gets that hot here in Seattle. The water flow sensors are pretty easy - is water flowing? I bought 2 of them (here and here) for an earlier project.
I totally agree on the buzzer - I got this one. Haven’t actually fired it up but based purely on the size of the thing I’d guess it’s pretty loud. I had originally intended to mount it in the battery box but now that I think about it it makes more sense to mount it in the display box. Unfortunately this one is a 12v alarm and the display box will only have 5v available, so I might have to find a different buzzer. Or put a step-up in.
The display box will probably be this:
and it will be totally self-contained with its own batteries. It will listen to the XBee chatter between the remote and the control board and will light up based on the status messages going across. It will have its own power supply and a lot of velcro on one side so I can stick it to the battery box.
Alarms get annoying pretty quickly, so I imagine I’ll make it beep 3 times when one or more fault conditions happens. Then it won’t beep again. Maybe a different algorithm for low battery.
This too will probably be tuned with experience.
I’m always open to suggestions! You appear to be about 2 years ahead of me in skill and experience 
UPDATE
I tested the buzzer - it’s a warbler - exceedingly annoying, deafeningly loud. Should be perfect
2 Likes
So the waterproof aluminum box came in for the driver.
The only parts still outstanding are the water pumps and they should be here by the week of the 25th. I’ve dedicated that whole week to working on the eFiol and can no longer find any excuses not to start cutting apart the board. It’s funny how hesitant I am to cut up a $40 board.
I still need to order the longer mast before I put the thing together, but the biggest issue to work through is building the infrastructure inside the board to attach the mast to. I need to spend some time going through other build threads to see how other people have done it.
I also found out quite by accident that uxcell, the manufacturer of the above box, makes everything. Just type “uxcell” into amazon and you’ll get a phenomenal variety of things from gearmotors to couch covers… Specifically they make a variety of incredibly inexpensive aviation connectors. I need something like this to run the control and sensor lines between the battery box and the driver box. I was going to use DeviceNet, but this is a lot less expensive.
Quick update, this pump came in and it does indeed draw a 6’ head, so almost all the components are in stock to build out the battery box and the controller box. I realize now that I need some more quick-disconnects for the inbound and outbound water lines.
1 Like
Made some cool circuit boards, bought a lot of wire. (25’ each of 8AWG.)
The one on the right is the Receiver and sensor module that goes in the battery box, the one on the left is the sensor module that goes in the driver box. The two can talk via a serial link.
I know what you’re wanting to ask, “but Doug, why is the radio on the board that goes in the battery box? shouldn’t it be on the board that goes in the receiver box since that’s where the ESC is?” Well, I’ll tell you, the ESC box is aluminum and the battery box is plastic. so from an RF standpoint I’d have to put an external antenna on the ESC box, which I didn’t want to do. But the PWM signal is direct, not interpreted, between the two boards, so hopefully it’ll work.
Both modules have temperature, pressure, humidity sensors and the ability to drive 12v pumps. The receiver board also has a voltage divider so it can monitor the battery level and sound the alarm if it gets too low
There will be a third board that will comprise the User Entertainment system, giving me sunlight readable red and green LED indication of the state of the system and the battery level. Might have to be Red and Blue, though. While red is a great sunlight-readable color green not so much.
I’m thinking about adding a tilt/pitch/roll sensor to the Receiver board to cutoff the motor if the angle of the board exceeds some limit, like 30 degrees. The Swordfish X instructions warned specifically against cutting power when the motor was spinning, so I’m trying to find additional ways to idle the motor before my hard cutoff trips.
The build starts in earnest this weekend and, notwithstanding a few hopefully short diversions, goes all the way through Jan 1. A few straggler parts will hopefully be here tomorrow.
But I’m also going to be upgrading the 2 Esk8’s in the same timeframe since i finally got my dual FOCBOX.
Also I invested in an incredibly sophisticated last-line-of-defense water ingress containment system
4 Likes
Well, I’ve been quiet for some time, not because I’m not working on the machine, I am, but because I’ve had to write a lot of code for it lately, and that’s my least favorite part. I have one more hard problem to solve, then all the code should be in place, and the last of the circuit boards should be here in 2 weeks. Of course in the meantime I thought of several upgrades I really should have thought of earlier.
So I’ll fill in a bit more later, but today’s job is to get all these things into this box:
Also, I did the initial design and a test print of the prop extender:
It fits really tight, so I need to loosen up the tolerances, and while the slots are all in the right place they don’t chamfer correctly to meet the case of the motor so I need to fix that. I also need to make a similar slot-cover-er for the wire end.
I also realized today that I’m going to need some sort of water filter to keep the crap out of the cooling water inlet. So I’m going to design that into the nosecone.
This weekend I really am going to cut up the board - for better or worse. It kinda depends on the weather - I don’t want to cut it up in the shop, but I don’t want to cut it up in the rain.
2 Likes
What are the specs of the propeller in the picture?
With the 80100 APS 80kv version?