Logged in users > User Diaries
Bryan's 4Kw motor Conversion Part 2 The testing
SparWeb:
My alternative is to skip the encoders and speed sensors altogether and just sense zero-crossing with a microcontroller.
This works better than typical frequency sensing chips because they are confused by noise, and often prefer the noise over the base signal.
Most microcontroller have frequency sensing functions and they are also fooled by noisy AC, so my trick is to use an Interrupt on each zero-crossing.
If you don't mind programming a micro you can use a simple program (20-30 lines of code) to sense the zero-crossing of the AC tapped from the 3-phase lines.
Once I got the baseline program working, I got fancy and added LCD display and some other stuff. An 8-pin microcontroller worth 5 USD will do if you just want a tachometer.
http://sparweb.ca/tachometer/
SparWeb:
Footnote:
This circuit will spew out tachometer data to a LCD or your computer serial port twice per revolution of the turbine, meaning the data comes in 5-10 times per second in a strong wind. So you need to have a LCD that refreshes quickly otherwise it's a blur. The colorful LCD in the picture above is actually updated every 10th cycle. Even at that its refreshes are too slow in the winter. On the other hand, the serial port is fine with any data rate the tachometer can spit out.
ruddycrazy:
G'Day Guy's,
Well it has been awhile since I updated this post so here we go....
I had a thought on how to regulate the RPM and came up with a concept drawing for a variable drive
So as one can see the centre shaft regulates the speed and I have started to make the pattern of the disks so I can get them cast so the normalizing can start after I rough machine the castings.
Started a new job and got told see those packs of 6" pipe well they are offcuts ( insert Shocked emoji) plenty of larger pipe there so the design for the tower is going to come together. I'm thinking with the pipe length being 3 metres of welding 3 sections and welding in bracing on the joints where the above joint can be a section for mounting the guy wires.
So the gin pole will be the same 6" section and same length so I can just use my tractor to raise and lower the tower.
Now for the smarts for this project I will be making a circuit board with a pic32 as the brain and I'll program MMBasic in it so the programming will be done in Basic where talking it out with the girls can figure out the master plan then I can tech the girls just how to write a basic program.
There is still one task to sort out and thats the wire run and my thoughts of using 400 amp welding cable may be a tad over board so my question is for bringing down around 3Kw of wild AC what cable is the best.
Another task to solve is setting up a prony brake when I do the test and I'm thinking as this is a flange mount motor a solid coupling could hold the weight then a prony brake can be setup BUT as I've only read about this method some guidance will be needed as getting all the data is needed to establish the blade length and profile. I am thinking Sparweb is a guru at this and provide some info on doing this practice as this is something I do want to learn, back in the day when I was doing that engineers degree only to find doing shutdowns was the go I didn't get the learning for that theory.
My goal is get this in the air and finished so I can start the next one using a larger donor motor and get the magnets soon before the China bubble bursts.
Cheers Bryan
bigrockcandymountain:
Hey Bryan, have you thought about using your tractor pto for the variable speed for testing? A 540rpm can usually be slowed down to about 135 rpm by running the engine around 500rpm. Of course you could go lower with a couple pulleys. Maybe I'm understanding wrong. The variable speed is just for testing right?
Tower sounds like it will end up being exactly like mine. It was 3 offcuts of 7" piling pipe that i welded together. They were about 3m each too.
I used 2 awg aluminum wire direct buried for mine. It seems to work fine. I don't worry about a little line loss at high power. The aluminum was about 1/3 the price of copper, so it was an easy decision. I run higher voltage down and only 2kw so your numbers will be different.
Thats about all i can add. Cheers from Canada
Derek
SparWeb:
Better go into a continuously variable transmission like that with your eyes open...
https://www.carfax.com/blog/cvt-pros-and-cons
https://youtu.be/a3PF4fPXe9U?t=18
https://upload.wikimedia.org/wikipedia/commons/transcoded/4/41/Claas_Mercator_12_August_2020_JM_%287%29.webm/Claas_Mercator_12_August_2020_JM_%287%29.webm.1080p.vp9.webm
Things you need to make a CVT work: big opposed forces, thick v-belt
Efficiency looks like 75%
To make it work efficiently... extremely tight tolerances, oil-filled gearbox, steel belt or chain
If you're also preparing to bench test the power, well then you can get things figured out that way, make the blades the diameter you need, and just run a direct-drive to get CP about 40-50% efficiency. The variable transmission will waste about 20% so it would have to boost your prop match to near perfect over a very wide range to be worthwhile.
Controlling something like that from the ground also has its own challenges. You've got the control circuit on the ground, and the machine up the tower. Need some skinny wires in between so it invites fidgety things like slip rings.
It would be cool to see, however.
Navigation
[0] Message Index
[#] Next page
[*] Previous page
Go to full version