The addendum = 1/dp = .0417, so the pitch diameter is OD - 2*addendum = .5625. Here is where the tooth thickness and space thickness are the same. The pitch is pi/dp = .1309. This would be the sum of the tooth thickness and the space width at the pitch diameter, so the space thickness here is half that, or .0654. I'm sending you a picture.

If you lay your part out in the flat, the helix angle is the diagonal of a box 12 in. by 2.074. Your whole part, unrolled, would be 24 x 37.699 (the circumference is 12*pi = 37.699). If you send me an email, I will send you the drawing. The 'teeth' are something else. You didn't supply any data on them.

The slots are unusual. Most end walls would point to the center (OK the centerline of the end radius). You can just make a ruled surface of each section individually, and cut it with Swarf5ax. You have to join the outer arcs which are in sections. I'll send you my part.

I can't find the file, but if it's a slot around the outside of a cylinder, use ROLLDIE.DLL. Create centerline geometry, preferably at the bottom of the slot. Rolldie offsets each way from the centerline when you use the SLOT option with Auto Reverse.

When you have contours (splines) that are on the surface but are not part of it, you can map them to the flattened surface, such as the edge of a windshield.

You can use FlattenSurf.dll and Map.dll.

d1 6. # Diameter 1 d2 5. # Diameter 2 c1 18.849556 # Circumference 1 c2 15.707963 # Circumference 2 rcone 60.075 # Radius of cone (element of cone) ccone 377.462357 # Circumference of cone a 17.977528 # Angle of flattened cone a/360 = c1/ccone This is the basic pro d1 6. # Diameter 1 d2 5. # Diameter 2 c1 18.849556 # Circumference 1 c2 15.707963 # Circumference 2 rcone 60.075 # Radius of cone (element of cone) ccone 377.462357 # Circumference of cone a 17.977528 # Angle of flattened cone a/360 = c1/ccone This is the basic proportion setup to calculate the flattened cone angle. I'll email the graphic file.

Rotary axis control, axis substitution, SLOT, AUTO REVERSE.

Another comment: the larger the cutter, the better the slot will fit the follower.

Use ROLLDIE.DLL. It can work from the flat. Use only the centerline geometry. ROLLDIE has a SLOT option that offsets from the centerline, such that opposite walls are parallel, for the follower.

Could the groove be an oil groove? Some of them are just a tilted circle, and some are figure eights. In the flat, the tilted circle is a sine curve, and the figure eight is 2 sine curves.

Let me have the measurements over pins, also, so I can check it graphically.

I am working on the solution, but it is messier than I thought. There is a 'solution' in Dudley, "Handbook of Practical Gear Design", CH 8, Drop Tooth Design, Fixed Center distance. It's good as far as it goes, but you need more parameters to plug into GEAR.DLL. I'm trying to provide them all. If you send me the backlash, I can factor that in to the tooth thickness. There's no data yet, I just want you to know that help is on the way.

Axis substitution should work for lettering.

Surfaces that have been extended don't behave well for MAP.DLL. Surfaces with sections that are equally spaced work the best. I'll look at the surfaces if you want. I need a good one and a bad one.

You only have 2 end points? You need 3 points to define a plane, or you need your construction view to be in the plane you want your arc to be in.

The C axis doesn't get its position from the B rotation. Both the C and B axis positions are determined from the tool axis vector VTOOLX. You can't really program either B or C directly. It must be from the toolpath's tool axis vectors.

The specifications are garbled. If this were a gear with PD of 133, OD would be 142.5. If it were an involute spline, OD would be 139.413. You give an OD of 135.75, which doesn't leave much tooth height. Something is wrong. Find out if this is a gear or a spline. Also, a spline would probably have a pressure angle of 30 degrees. I have never heard of a "Space profile angle". It that supposed to be the pressure angle?

If the scallops got smaller at slower feed, it sure looks like you've hit some machine resonance. Servo tuning might fix it. Until then, maybe setting the part up at a different angle might get you out of the resonance zone. Spindle speed probably doesn't affect anything.

It's possible that the router is hitting some resonant frequency. Try dropping the feedrate on the scalloped line.

I'd like to take a look at the part. Maybe I can help. There are several different approaches, depending on the part. Unroll only works for a very special kind of impeller.

A 4-axis post won't see the 38 degree angle. The post must make a rotation about X, then Y, if I understand your setup correctly. That's a 5-axis post. Do you have one?

If you untrim the surface, you can see what the flowline sees. Flowlines will be parallel to the untrimmed surface edges.

Often, lead/lag can be used so you can switch from a ball mill to a flat mill or bullnose cutter. A good example is pipe bending dies. If you set the lead/lag to 45 degrees or so, you don't need a ball mill at all. Some airfoils work the same way, but with 5 degrees lead/lag.

Retired, but busier than ever, Ron. Rick, see if you can guess at the depth and width of each cut. I'll send you a sample, so you can see what I'm getting at.