Author Topic: 1 foot CNC wood blade for a 3 blade 2 foot prop (Read 4624 times)

richhagen · « **on:** February 22, 2005, 11:11:19 AM »

I got a small CNC mill a while back, and finally got around to playing with it to make a small blade. I was aiming to make a blade for a 2 foot diameter prop as the mill has only 12 inches of travel in the X axis. This posting shows the results of the first attempt.

I don't have any good CAD/CAM software to work with, so I basically guessed a foil based on a modified NACA 4412 design, and designed it for a five degree angle at the tip and as much twist as I could get at the root. I created a plot of the airfoil at the root and the tip. Then I calculated a few lines from the root to the tip at various key points around the perimeter and used a spread sheet to plot lines that intersected with those lines at each X position. The outputs of those lines were then transformed via a report in Microsoft Access to a text file in G-Code format. Those files for the front and back of the blade are here. I modified those files to leave a couple of 'pillars' of material at two points along the X axis to act as supports because I was worried that the material might not be held securely in the vise if I removed all of the wood all of the way down. Here are the files for the front and back of the blade if anyone wants to see them.

FRONT: http://www.otherpower.com/images/scimages/742/20050221BLADEFRONT.txt

BACK: http://www.otherpower.com/images/scimages/742/20050221BLADEBACK.TXT

These files were from my main computer, and I think I made some minor changes to them when I actually put them on the computer attached to the CNC mill.

I loaded the program for the front, chocked in the material and homed the machine. Then I started the program.

It started cutting and here it as it just finished cutting the root of the blade. I kept a vacuum handy as it created quite a pile of sawdust.

Here the machine is finishing the front of the blade. You can see the two 'pillars', or blocks of material that I left to act as supports.

After the front side was finished, I flipped the block of wood over in the vise, homed the machine and ran the program for the back side.

All done with the back side, It took quite a while for this to finish a blade. The code could use some optimization, to eliminate some cuts that don't need to be made, and to speed up the cutting on shallower cuts.

Here is the blade out of the machine. I will have some changes to make. The trailing edge was nearly paper thin. I will have to make the blade thicker. Also, I will need to smooth the root into the blade, I'll probably plot another set of lines from about 5/8" to the first profile on the root which was at 2.125 inches on this prototype, and then construct cuts for the positions along the X axis to smooth the transition from the block to the airfoil profile.

Here is an image showing the tip profile and it kind of gives you an idea of the twist. The twist is distributed linearly from the root to the tip. I'm not sure if that is ideal, but it is probably good enough, and it was the most simple way I could think of to do it.

For the next trial, I think I will change the profile to make the blade thicker. It seems very fragile as it is. I may also use a wider block of wood than the 1" by 2" (actually 1.5" by .75") used for this trial. I originally used the 1 by 2's as I could potentially mount three of them on a rack attached to a rotary axis I have on the mill and have enough clearance to flip all three. That way, I could load three blanks, let the machine run, and have three blades waiting when I returned. With wider wood that wouldn't be possible. The blades appear rather narrow to me and off the top of my head, I think they may have trouble at startup. I may, however, make two more blades identical to the first, just to play with them and see if they work. I have a bunch of small NEMA 17 stepper motors, and although I am not certain of the power rating for them, these blades might be a reasonable match.

I'll be out of town till Thursday, and I'll have a pile of work when I get back, so it will be next week at the earliest before I can play with these again. Rich Hagen

ghurd · « **Reply #1 on:** February 22, 2005, 07:05:19 AM »

It looks great!

What is the width and thickness at the tip?

(Every plan I come up with has 1mm thick for the last 2 stations!)

G-

picmacmillan · « **Reply #2 on:** February 22, 2005, 07:28:51 AM »

hi rich...we were thinking of the same thing as we have access to some machines, we were thinking of using maybe a 6 inch fly cutter and just zoom through it....what do you need for a program? maybe we can find someone with lets say(a sample copy).....autocad?..someone i know is using fastcam?...i have auto cad here if you're interested, anyhow, good luck with that, i envy a guy with the tools for the job, you're work is inspiring...pickster

monte350c · « **Reply #3 on:** February 22, 2005, 07:48:24 AM »

Very neat Rich!

I wonder if laminating a blank together out of a few stips of wood with some fiberglass cloth and eopxy resin between them would stiffen things up. You know sort of like a wood and epoxy sandwich.

Very ingenious with excel and access too. Rich-cam.

Ted.

ghurd · « **Reply #4 on:** February 22, 2005, 07:58:25 AM »

Maybe hickory would be better wood for something like this?

Strong, with enough flex to keep it from breaking, not too heavy/dense?

G-

wdyasq · « **Reply #5 on:** February 22, 2005, 08:20:56 AM »

Rich,

Nice work. A different airfoil with thicker profile might help - there are several with Cl/Cd ratios similar to or better than the 4412 IMO. The 4415 is one choice that might be considered a standard.

I have found the tricky thing to do is compensate for flat bits on a constantly changing airfoil, especially if it has a cusp as many of the really modern wind turbines do. I now have a stratigy and am working on larger blades.

If you would like to discuss this further, my user name -at- yahoo.com will get an e-mail to me.

Ron

richhagen · « **Reply #6 on:** February 22, 2005, 11:44:00 AM »

It is 1/6 as thick as it is wide. It is an inch and a half at the root and 3/4ths of an inch at the tip. That makes it about 1/8th inch thick at the tip. -Rich

richhagen · « **Reply #7 on:** February 22, 2005, 11:48:58 AM »

There is a big design engineering show coming to town the beginning of next month. I got a pass to it and a friend and I are going to go have a look around. On the agenda is checking out the CAD type booths for ideas and freebies. The trouble is that whatever I get I'm going to have to learn how to use. I have a program that claims it can convert DXF files to G-code, so Autocad is a possibility for me, but I think I'd have to set a side a good chunk of time to figure it out. Rich Hagen

Ungrounded Lightning Rod · « **Reply #8 on:** February 22, 2005, 11:55:45 AM »

My wife it picking up a copy of Linuxcad as an alternative to autocad for her design class.

I have no idea whether it will do why you need done, and it won't arrive in time for us to check it out and five you a review.

But at $175 the price is right and it is reviewed elsewhere.

gibsonfvse · « **Reply #9 on:** February 22, 2005, 01:04:16 PM »

Well-done on this! If I may ask, what CNC mill are you using? I would expect the trailing edge to be paper-thin, especially if your airfoil profile is like this, and it's not a bad thing, necessarily (sharp trailing edges are good). Maybe a good thing to do would be to strengthen the trailing edge with some layers of fiberglass. Just a thought... keep up the good work!

- Chris

richhagen · « **Reply #10 on:** February 22, 2005, 01:29:23 PM »

It's a Taig mill. I think I need to strengthen the trailing edge a bit, but want to keep the design as simple as possible. I figure that the required strength of the blade material increases with the rotor diameter, and that if Dan and Hugh can use wood on their beasts, then I ought to be able to use wood for a smaller one. If I get a blade I like, I may make a bunch of them, as I have a bunch of small motors collecting dust. Rich Hagen

richhagen · « **Reply #11 on:** February 22, 2005, 01:39:36 PM »

I saw the laminated boards that Jacques and Scott were using, I was wondering how they laminated them so that they were weather proof and wouldn't come apart. The seem to work well and I'm sure that they both did their homework. For these small blades, I'm not planning on keeping it as simple as possible, load a piece or pieces of dimensional lumber and carve out a blade. I think that at this radius most woods would work well, however, Hickory, Elm or Ash would probably make nice strong blades. I have access to lots of scrap pine, so it would be nice if I could use it. Rich Hagen

richhagen · « **Reply #12 on:** February 22, 2005, 02:00:56 PM »

similar to this one:

http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem&category=58254&item=3875320084&rd=1

-Rich

newguy · « **Reply #13 on:** February 22, 2005, 03:07:49 PM »

does anyone use a shopbot cnc

wdyasq · « **Reply #14 on:** February 22, 2005, 05:20:45 PM »

I use a variant. It uses ShopBot drivers and software.

Ron

monte350c · « **Reply #15 on:** February 22, 2005, 08:18:51 PM »

That's a good price.

Just for fun I downloaded the trial version of Rhino 3D a couple of days ago. It's a fully functional demo so you can actually try it out. It's limited to a certain number of saves.

There's also a demo for Rhino Cam which as the name implies will produce G code and toolpaths directly from whatever you drew in Rhino 3D.

I don't have my machine done so I can't speak for how well this will work but it seems quite easy to use and there's a neat cut simulator in the CAM section.

Of course it will be a lot more than $170 ;(

Ted.

Ungrounded Lightning Rod · « **Reply #16 on:** February 23, 2005, 01:01:48 AM »

I tried to respond to this once before but it got lost, so I'll try again.

The twist is distributed linearly from the root to the tip. I'm not sure if that is ideal, but it is probably good enough, and it was the most simple way I could think of to do it.

The tangent of the angle of attack (slope, i.e. opposite-over-adjacent lengths of the right triangle with the chord as the hypotenuse) is inversely proportional to the radius (= constant * 1/r). Not to hard to compute.

A linear twist is far from ideal - especially near the base - but is still a lot better than no twist at all. Much less of the blade is stealing power from other parts and using it to pump air instead of pulling its own power from the air.

If I were doing one I'd let the trailing edge go "off the back" of the workpiece near the hub. Then I'd cut one or more shorter pieces as necessary to fill in the missing part and laminate them.

With a CNC machine there should be no problem getting the layers to line up. Also: You can cut alignment groves and ridges in the mating surfaces to make them self-aligning when you stack them - or cut slots in both for an strip of wood to make an alignment key to do the same thing. Then just spread glue, stack them up, and clamp them (or put weights on them) until they dry. (You can also cut matching bed pieces to stack but NOT glue on, to give you flat surfaces for your clamps or weights.)

An earlier article discussed good choices of glue for laminating props.

Adding the extra depth near the hub by filling in the "missing" trailing edge where the twist becomes extreme also helps greatly with the strength of the blade. The bending forces from wind load are also maximum near the hub. The twist puts extra length along the axis direction (and thus stiffness against bending) right where you need it.

RatOmeter · « **Reply #17 on:** February 23, 2005, 11:36:13 AM »

"I have a bunch of small NEMA 17 stepper motors..."

If I've identified your ebay ID properly, you've got whole shipload of stepper motors. I like stepper motors, but I kinda thought 256 of them is a few more than I could use.

ghurd · « **Reply #18 on:** February 23, 2005, 01:53:44 PM »

I have Full 1" thick x around 2" wide clear hickory, in 12~16" lengths.

I don't have any 'good' stepper motors.

Hmmmm...

G-

richhagen · « **Reply #19 on:** February 24, 2005, 02:07:06 AM »

You are right on the money on the ID. They were cheap enough, and I had an idea for them at the time. Rich Hagen

richhagen · « **Reply #20 on:** February 24, 2005, 02:46:13 AM »

The NEMA 17's are small, but as Ratometer found out I do have a bunch of them. RichHagen (no-spam) at Juno.com

News:

Author Topic: 1 foot CNC wood blade for a 3 blade 2 foot prop (Read 4624 times)