What's new
Van's Air Force

Don't miss anything! Register now for full access to the definitive RV support community.

3D (2D) scanning of propeller profile

FinnFlyer

Well Known Member
Trying my luck on this forum.
Looking for help with DIY 3D scanning software.

I have an somewhat unfinished wood prop from 2007 that is rather grossly out of balance. Requires some 60 grams (more than 2 oz) on a prop bolt near blade to get somewhat in static balance. It's lso out of balance at right angle to prop blade axis. If I got it right maple weighs about 0.4oz/cu in. So I need to remove something on the order of 4 cu in on and near the heavy blade hub. I realize that the shape under the spinner is not aerodynamically important and I could just drill a 1/4" hole and fill it with carbide rods. However, seems like a band-aid solution, not handling the root cause of the static imbalance.

So my idea is to mount the prop on an old PSRU (reduction gear drive). Mount a DIY 3D scanner (see attached picture -- less turntable) on a bar above one blade. Take a profile snapshot, rotate the prop 180 degrees and take a new snapshot of the other blade. Comparing the two pictures should clearly show the profile differences on a 2D plot on a laptop at the chosen prop radial. This can be repeated along the prop blade, from hub center all the way out to the tip, by moving the 3D scanner a bit on the bar.

I should then be able to grind and sand down the heavy hub side and prop with confidence.

So, does anyone have experience with DIY 3D scanning software and can point me in the right direction?

I've searched for USB camera data output format but not found any useful hits. Basically I need to subtract or compare two pictures, isolate and measure distance between the two laser lines.

Finn
 

Attachments

  • DIY3Dscanner.jpg
    DIY3Dscanner.jpg
    25.8 KB · Views: 62
Can you rent a white light scanner where you are? This would probably remove a lot of uncertainty.

Probably not. Looking at their prices I expect rental (if even available) will be well over my $100 budget. I ordered a package of 5 line lasers for $13 (just need two). Already have a 640x480 USB camera. I can get a higher resolution camera for $25 or less, if I don't want to use my 4608x3456 Panasonix Lumix camera. Looking for USB camera prices just now led me to the OV2640 Chip, a 1632x1232 B&W CMOS camera chip that looks promising.) Not looking for more than 0.1 to 0.2mm accuracy.

Finn
 
Then I would definitely do some validation scans on some objects with known, precise dimensions. Articles with compound curves would be even better. The set-up you're doing is beyond my experience, That said, a thorough validation plan would be essential to gaining enough confidence (for me at least) to apply any proposed remedy. Keep us informed. Best of luck.
 
ImageJ

Just came across and installed this:
https://imagej.nih.gov/ij/

Can subtract two images:
Open the two images (File |Open)
Process | Image calculator ... , select Subtract
or
Process | Image calculator ... , select Difference

The amount of functions in this program is stunning!

There a batch or macro functionality which I may be able to use to get prop profile differences automatically. We'll see.

Time to start building jig to hold line lasers and camera.

Finn
 
Last edited:
Profile comparison pictures

https://drive.google.com/drive/folders/1V6EdNvlOZEv2rtdx6F9922mPhp4sdhrX?usp=share_link

Finally built a balancing rig so able to rotate prop 180 degrees around hub center pilot hole.

Using a digital camera and a line laser I was able to take pictures of profiles, comparing one blade to its opposite. Took pictures in two step increments from tip to hub. Then compared images using the ImageJ software: Process | Image calculator ... | Average function. Turns out it's really difficult to position blades accurately, turning prop 180 degrees.

Only near the hub did I see somewhat significant differences (pictures 34.jpg thru 40.jpg). But I'm not brave enough to take a sander to the prop based on this.

I measured cord and pitch in two inch increments from tip to near root and they compare nicely within measurement tolerances (1mm and 0.5 deg).

So this method may be valid if one is comfortable sanding a prop.

Edit: However based on what I found, the apparent difference in profiles near the hub does not account for (explain) the needed weight to balance the prop. One side of the prop is significantly heavier than the other side. When hanging it in a string through the center pilot hole it needs a significant weight (38 grams at hub edge) on one side of the hub, not on a blade. So it's got to be wood or glue density or perhaps moisture if it was stored with one prop hub side down for a long time (since 2007). By side I mean parallel to bolt holes.

Finn
 

Attachments

  • P1020763.JPG
    P1020763.JPG
    157.5 KB · Views: 38
  • P1020764.JPG
    P1020764.JPG
    155.5 KB · Views: 28
  • 40.jpg
    40.jpg
    40.3 KB · Views: 32
Last edited:
I applaud you for the process you have figured out on how to get your prop balanced. The pictures are really cool. It is more than I would take on with my knowledge.

It may be a good idea to give Catto or another wood prop company a call and ask them. I know the people at Catto are very friendly and always help me when I had a question. They may be able to give you info on how they do it.

I will follow this and hope to learn a thing or two!
 
I applaud you for the process you have figured out on how to get your prop balanced. The pictures are really cool. It is more than I would take on with my knowledge.

It may be a good idea to give Catto or another wood prop company a call and ask them. I know the people at Catto are very friendly and always help me when I had a question. They may be able to give you info on how they do it.

I will follow this and hope to learn a thing or two!

Thank you.

I have watched several youtube videos with the Culver Prop woman balancing a prop. She puts it on a balancing stand and is magically able to determine which prop blade is thicker and where (typically near the hub) and is able to get it perfectly balanced by sanding down the high spot(s).

Finn
 
https://drive.google.com/drive/folders/1V6EdNvlOZEv2rtdx6F9922mPhp4sdhrX?usp=share_link

Finally built a balancing rig so able to rotate prop 180 degrees around hub center pilot hole.

Using a digital camera and a line laser I was able to take pictures of profiles, comparing one blade to its opposite. Took pictures in two step increments from tip to hub. Then compared images using the ImageJ software: Process | Image calculator ... | Average function. Turns out it's really difficult to position blades accurately, turning prop 180 degrees.

Only near the hub did I see somewhat significant differences (pictures 34.jpg thru 40.jpg). But I'm not brave enough to take a sander to the prop based on this.

I measured cord and pitch in two inch increments from tip to near root and they compare nicely within measurement tolerances (1mm and 0.5 deg).

So this method may be valid if one is comfortable sanding a prop.

Edit: However based on what I found, the apparent difference in profiles near the hub does not account for (explain) the needed weight to balance the prop. One side of the prop is significantly heavier than the other side. When hanging it in a string through the center pilot hole it needs a significant weight (38 grams at hub edge) on one side of the hub, not on a blade. So it's got to be wood or glue density or perhaps moisture if it was stored with one prop hub side down for a long time (since 2007). By side I mean parallel to bolt holes.

Finn

Is the Face of the blade painted with matte black paint like most props?
At our prop shop, on fixed pitch props that do not provisions for balance weights or lead to be added, we use paint to correct balance. Works very well without removing material. Sherwin Williams Polane T is what is on most smaller GA propellers and is used on all Hartzell, McCauley and Sensenich props leaving the OEM's. The more thinner used the thinner the coat of paint. Scuff the existing paint with Maroon Scotch-Brite or similar so the paint adheres. The more paint applied to the outer stations of the blade have the most effect on balance = bigger arm.
 
Is the Face of the blade painted with matte black paint like most props?
At our prop shop, on fixed pitch props that do not provisions for balance weights or lead to be added, we use paint to correct balance. Works very well without removing material. Sherwin Williams Polane T is what is on most smaller GA propellers and is used on all Hartzell, McCauley and Sensenich props leaving the OEM's. The more thinner used the thinner the coat of paint. Scuff the existing paint with Maroon Scotch-Brite or similar so the paint adheres. The more paint applied to the outer stations of the blade have the most effect on balance = bigger arm.

No, back not painted black yet. Note that a significant part of the unbalance was side-to-side and not blade-to-blade so paint would not help there.
I ended up drilling a 5/32" inch hole close to the edge of the hub which I filled with carbide rods. That got it closer to static balance. When I get back I'll do a series of flights with test weights added to dynamically balance it. Then I'll do some at altitude cruises to determine how much, if any, I need to reduce pitch (and/or) length.

Thanks for the tip on the paint!

Finn
 
Back
Top