Long Term Regen

Support for creating custom user applications on the VESC.
RoderickRead
Posts: 2
Joined: 19 Jun 2017, 22:17
Location: Isle of Lewis

Long Term Regen

Postby RoderickRead » 28 Oct 2017, 20:31

Has anyone used the VESC for a regeneration focussed project yet?

I've built a highly unusual autogyro kite turbine "Daisy" and publish it open source hardware too.
So I'd really like a good open source regen controller.

Ideally, I'd use the extra smarts in the controller to balance how fast the kites are spinning with how much power can be taken out....
I'd also like to be able to take a signal to say quickly stop regen spin fast until an over-twist fault condition disappears.

I'm going to try and charge an ev car with an improved kite turbine this year... I used to charge my ebike with the old one.
roddy
windswept-and-interesting

rew
Posts: 893
Joined: 25 Mar 2016, 12:29
Location: Delft, Netherlands.

Re: Long Term Regen

Postby rew » 29 Oct 2017, 07:21

Indeed very interesting!

Yes, you'd need some Maximum Power Point algorithm in there. The difficult part is how long you need to wait to get a valid measurement of the new situation. Otherwise, increasing the load will always seem advantageous because you are extracting kinetic energy from the moving parts....

For the controller there is no difference between powering the motor one direction and having it move in the other or not.

The kites you have seem to be the cheap ones you can get for about $10 in the "everything" stores. One of the parameters of such a kite that is important is the glide ratio. These are REALLY bad in that respect. Peter Linn is a famous kite maker. But IMHO his kites are better than the dollar-store ones, but not by much. Ozone (coming from the paragliding side of parafoils) makes kites that are much much better.

Get three "Little Devils" or something like that. (The smallest are called "Imp" nowadays.)

But be careful. Three of those little devils will create a pull on your installation that you haven't seen before. With the wind that I saw in one of your videos I would estimate: easily 300kgf.

Have you heard of the kite-power project from Wubbo Ockels? "Laddermolen". A friend worked on that project and borrowed my Little Devil for a while. They broke my lines rated at 400kgf total. (just adding the line strengths doesn't give the breaking strength of the whole. But if we assume most of the force on the front lines at 120kgf and an even distribution between left and right, they achieved > 250kgf at least.... With one kite....)

pf26
Posts: 263
Joined: 28 Mar 2016, 14:37
Location: FR Valence

Re: Long Term Regen

Postby pf26 » 29 Oct 2017, 13:36

I use a VESC for unusual power generation as well (prototype of this thing: http://monopale.forumcrea.com/viewtopic.php?id=14, vids here, formerly using another controller: https://vimeo.com/182373528)
I wrote a VESC custom application to calculate the generator braking torque in real time depending upon speed and angle reached by the blade and some parameters - which I can change remotely via CAN bus and that are stored in non volatile memory.

Instead of a MPPTracker, you could program a braking torque law that varies with the RPM squared or some other calculation to make it stable and near the optimum.
Otherwise, you could implement a smart MPPT that takes into account the kinetic energy variation (based on RPM changes and the inertia of your system).

You can also use the CAN bus to make the VESC application communicate and react to external signals, or acquire the digital/analog/i2c signals directly with the VESC.
There is a relatively steep learning curve to install the toolchain and understand the VESC firmware architecture, but then the possibilities are huge.

RoderickRead
Posts: 2
Joined: 19 Jun 2017, 22:17
Location: Isle of Lewis

Re: Long Term Regen

Postby RoderickRead » 29 Oct 2017, 19:54

Hi pf26 and rew,
Thank you both so much for your replies.
rew
It's nice you've also heard about kite power rew. Even though my gear looks scrappy, it's only a proof of concept using simple 2 line kites for control reasons. The simple no controller action and the tensile torsion to ground methods are both impossible according to the authoritative academic texts from AWESCO (TU Delft and Uni Freiburg...) Yeah sharper driving wings on the rings will make a huge difference. But I'm playing safe as it's already as powerful as me. I've been working on this a few years now. I sewed my lift kite on Peter Lynns home sewing machine.

pf26
Thank you so much for the reassurance this will work as a generator control.
Crikey that's one very clean, mighty strong yet surprisingly smooth tacking wing.
Certainly sounds powerful in the video.
It's really good to hear there are so many control configuration options available. As I improve the sensing capabilities, adaptability will be crucial. The turbine max efficiency dynamic will essentially come down to keeping the Tensile Torsion Lines as close to an optimal angle with respect to the Power Take Off wheel as possible. and then accounting for some override maximum operational condition parameters... Thank You.
I have a lot of reading, ordering and testing to do now.
Would love to have a look at any custom generating code you recommend.

Thanks again both
rod

pf26
Posts: 263
Joined: 28 Mar 2016, 14:37
Location: FR Valence

Re: Long Term Regen

Postby pf26 » 30 Oct 2017, 10:46

You can see 2 application examples here http://omegawatt.fr/paille/app_example.zip
One controls a generator (and acquisition of sensor), the other one the wing pitch motor (and acquisition of IMU sensor).
Rather poorly written and commented sometimes in french - but I hope it helps.


Return to “Custom Applications”

Who is online

Users browsing this forum: No registered users and 1 guest