Prop Manufacture

[Rick Willoughby]

I am looking for a CNC milling shop able to produce small one-off props from CAD files.

The props are unusual because they are low power, typically 150W, and very efficient so high aspect blades.

I currently fabricate them from stainless steel using a hand grinder for shaping the foil but am looking for a more controlled method.

I have attached a rendering of the prop required and also a photo of one I fabricated in stainless. (They are not identical designs).

I expect the props required could be milled from 25mm by 35mm aluminium stock and are typically around 400mm in diameter.

Is there anyone out there who can suggest somewhere to start? The props do not weigh much so could be mailed at an acceptable cost from anywhere in the world.

Rick W.

[jehardiman]

What's the problem with the local guys? Any machine shop with a CNC turret mill should be able to produce that simple shape as long as the table is big enough. I've had several HPV props produce that way.

BTW, tell them to run the surface twice if using Al billet, as the blade will relax after the initial hogging out. So cut it about 2mm fat, then come back and make the final.

[Rick Willoughby]

jehardiman

Thanks for the advice on the double pass.

What size prop did you have made?

Can you give me a ballpark cost of having the simple prop milled?

Who made your prop?

What alloy did you use?

Rick W.

[marshmat]

I could recommend a half-dozen shops in Ontario that could whip that thing up to within a couple thou given a few days..... don't know of any in Australia though. You'll want a 5-axis shop with a good machinist; a trained monkey can turn out crap on a cheap three-axis but I suspect you want it a touch better than that.
Ballpark cost? What I'm used to doing is sending the exact same email and CAD drawing to a few different millers, and just see what they say. It depends mainly on the precision you want; cutting tolerances from 0.5 mm to 0.1 mm adds a fair chunk on the price.

[Rick Willoughby]

Matt
Many thanks for your reply.

I have had a shop in Australia respond and am waiting on their reply after sending a CAD file. I have sent them a few file formats. Believe it or not I used FreeShip to generate that image. I have been using FreeShip quite a lot lately and find it far better than Autocad for this sort of work but I do not know what format is best for CNC applications. I simply do not use Audtcad enough to be comfortable with its controls.

I am interested to know how automated the process is to get from a CAD file to a machine guide file. Most of my designs are for one-off use so I am trying to get an efficient process. I have automated optimisation from engine to hull design thru prop design and I want to get efficient prop production.

I expect tolerences to 0.5mm would be acceptable - 0.1mm is better than required. A tolerance of 0.5mm is about what I can get with a grinder and templates starting with 3mm flat bar.

I have someone in Canada (Calgary) interested in a prop or two so a shop in your part of the world, or a bit firther west, would be of interest. Who would you recommend based on price, ease of working with a CNC novice and delivery is no rush.

Do you have any idea of likely cost?

Rick W.

[marshmat]

Any CNC shop worthy of the acronym can mill directly from IGES format. Afraid the toolpath generation is already about as automated as it can be; you can script MasterCam if you really want to but you still need the machinist's eye to ensure it'll work. Usually the CNC shop will do the toolpaths to suit their own equipment, if you give them the geometry file.
There are of course many good small CNC shops; finding big ones is a pain though. In the last year I've dealt with CNC shops mainly for composites tooling; in smaller (metal) sizes we've had excellent parts come back from Alzar Industries in Ottawa, and Woodman Machine Products in Kingston, among others. I'm not familiar with any further west but with the construction/oil boom out there it may be better to just FedEx to Ontario, everything in Alberta's up to the ears in back orders....

[jehardiman]

Originally Posted by Rick Willoughby

jehardiman

Thanks for the advice on the double pass.

What size prop did you have made?

Can you give me a ballpark cost of having the simple prop milled?

Who made your prop?

What alloy did you use?

Rick W.

The largest prop I've personally had milled out for a HPV was a 700mm counter-rotating set. Individual blades were machined and let into a larger hub that you are using. Each blade was ~280 x 180. If was done by a sponsor so it was no cost to us. About 20 hr all told for 8 blades (2 sets) as I recall. The material was just standard marine plate from the scrap bin, most likely 5052.

[Rick Willoughby]

I have had some feedback from one shop so far. They are working on methods to support the blades. I have also accepted that it will be more economic to hand finish rather than taking the machine down to close passes. This is not an issue because I normally hand finish my stainless blades and the material is much harder than aluminium.

I also found there is some dimension mismatch in the iges file I produced using FreeShip so would like to understand this.

As a matter of interest I have found some software from Delcam that allows conversion of most 3D formats to other 3D formats. I have not purchased it yet but it operates on a voucher system per conversion on an ad hoc basis or you can just use it to view any particular format without converting at no cost.

Rick W.

[jehardiman]

Originally Posted by Rick Willoughby

... They are working on methods to support the blades. I have also accepted that it will be more economic to hand finish rather than taking the machine down to close passes. This is not an issue because I normally hand finish my stainless blades and the material is much harder than aluminium....

Cutting through at the tip should be the last cut. As a prop needs a ball mill to get the proper shape (sized to the root fillet for stuff this small BTW), there will always be some ripple that needs to be taken out.

[Time]

Is this for the pedal the ocean project? the kayak "within"?

[jehardiman]

Originally Posted by Time

Is this for the pedal the ocean project? the kayak "within"?

Not likely at 150 watts. Normal HPV cruise output is ~400 watts. A prop as small as 400mm dia and shaped like that is way too small for meaningful propulsion in open water.

[Rick Willoughby]

Time asked: is the prop for Within.

Yes. One or more for the current boat and possibly a few more for the boat that will eventually be built to do the ocean crossing.

I also want to determine if it is a worthwhile method for my own applications and possibly others if it can be done efficiently.

I accept about a 1% loss of efficiency by designing for ease of fabrication from flat bar. I want to see if a milled prop can be made without this compromise.

There seems to be blade support issues for such thin blades using direct milling methods that may involve their own compromises. I use an MA409 foil that has a 7% thickness to chord ratio.

Rick W.

[Rick Willoughby]

The engine for Within has a continuous daily rating of 150W. It wound down to below 100W after 16 hours or so - getting low on fuel. The boat is fitted with an SRM metered chainring so the data is accurate. Part of the compromise for the drive leg was to include the metered chainring to confirm performance predictions.

My engine has a continuous rating of 130W when well tuned. Can be cranked up to maybe 630W but only for a few seconds. It is 1951 vintage. Later models can normally do better.

Human engine with a continuous rating (more than an hour or two) of 400W is getting into superman category.

Rick W.

[jehardiman]

Originally Posted by Rick Willoughby

Human engine with a continuous rating (more than an hour or two) of 400W is getting into superman category.

300W for a semi-pro or pro cyclist for 8 hour is well documented.

from the wikipedia site

20-40 W - BioMed: approximate power consumption of the human brain
100 W - BioMed: approximate average power used by the human body
500 W - BioMed: power output of a person working hard physically
745.7 W - Units: 1 horsepower
900 W - BioMed: power output of a healthy human (non-athlete) averaged over the first 6s of a 30s cycle sprint. [1]
up to 2 kW - BioMed: approximate short time power output of sprinting professional cyclists

from a HPV site at Ohio University



Can't find the recovery power curve set I used to have, but the push-rest-push-rest average curves are have much more energy under the curves than the steady state curves.

[Rick Willoughby]

John
The curve is a good one as it fits closely with my experience and data I have personally collected from others.

Basically at the 8hr level (near enough to a daily rate) it is 75W for average and 300W for first class athlete. Like I said 400W for a daily output is superhuman and well behond what a 60s model engine can produce.

The design point for Within that I was asked to work to was 150W. This was about the power level achieved for the first 5 hours during the 24 hour record attempt. You can see the fatigue setting in in the attached chart.

Rick W.