3d Water Jet Information

[Navid Shafa]

Originally Posted by stinglikeabee

I think a helical gearset would be a useful and very awesome project. Another possibility could be a countersunk holes (cut through from the non-countersunk side).

Countersunk holes are easy if you need them. I've made Helical gears too, but its tough to hold great tolerances when it comes to 3D cutting gears. Even the slightest offsets can cause meshing problems. Bevel gears and the like are certainly possible. However I'd still recommend purchasing COTS gears, It's likely to be more cost, time, and labor efficient to do so.

[MichaelBick]

Originally Posted by Navid Shafa

Countersunk holes are easy if you need them. I've made Helical gears too, but its tough to hold great tolerances when it comes to 3D cutting gears. Even the slightest offsets can cause meshing problems. Bevel gears and the like are certainly possible. However I'd still recommend purchasing COTS gears, It's likely to be more cost, time, and labor efficient to do so.

Quite a few teams have has sucess with waterjetted/lasered gears. The only difference is that with laser you want to stack up multiple plates instead of having a one piece gear. For a few examples, 148 has run gears like this before and 973 has run many gears like this before, such as gears used on their swerve and 2013 arm.

[Nuttyman54]

Originally Posted by MichaelBick

Quite a few teams have has sucess with waterjetted/lasered gears. The only difference is that with laser you want to stack up multiple plates instead of having a one piece gear. For a few examples, 148 has run gears like this before and 973 has run many gears like this before, such as gears used on their swerve and 2013 arm.

Flat gears and sprockets are pretty easy for a waterjet or a laser cutter because the involute profile is relatively tolerant. Navid is talking specifically about cutting helical gears on the waterjet, where not only linear tolerances but also angular tolerances can cause meshing problems.

[Cory]

I can't see much of a use for 5 axis waterjet in FRC parts. Just have them do simple 2D work for you-still extremely valuable.

[DampRobot]

Originally Posted by Cory

I can't see much of a use for 5 axis waterjet in FRC parts. Just have them do simple 2D work for you-still extremely valuable.

Seconded, normal waterjets are still crazy useful. We like to use the TS waterjet to fab our gussets really quickly, and also for large plates (bellpans, shooters, etc). The main FRC use of the multi-axis head in my mind is to eliminate kerf on stuff like gears and bearings.

[AlexH]

"how to build your everything really really fast" by charles guan might be relevant. even if it isn't, its still a god read...

[AdamHeard]

Originally Posted by DampRobot

Seconded, normal waterjets are still crazy useful. We like to use the TS waterjet to fab our gussets really quickly, and also for large plates (bellpans, shooters, etc). The main FRC use of the multi-axis head in my mind is to eliminate kerf on stuff like gears and bearings.

This is true, but we get along fine without that (for the sake of others reading this thread).

We have reamers for the size bearings we need, and for the few odd size we request they cut slow (this blows the hole out a bit more so account for that dimension ally, but it gets straighter edges). The reamer will make your waterjetter FAR happier. Ours cut 20 37mm holes for our swerve at .8 ipm in 2012 and was NOT happy with us after that (~2 hours runtime). For that run alone, the reamer was cheaper than half the holes cut.

For gears, there are two cases. First is you're running a waterjet gear on a waterjet gear. We run them with opposite tapers so it matches nicely. It feels junky at first, but quickly wears in pretty smooth.

For steel on aluminum, we just let it wear in an the steel wins pretty quick against the high spots in the aluminum and a happy medium is reached pretty quick.

I wouldn't recommend waterjetted gears going a few thousand rpm though.

EDIT: Keep in mind the above numbers, and the affect of taper will vary machine to machine and operator to operator.