Fixturing with 5 Minute Epoxy and "Cradles"

Sometimes you have a thing that you have to machine on all of its sides. Here’s an example of how I used minimal cradles to such a thing!

This is a close cousin of the soft-jaw method, but it relies on a small quantity of 5-minute epoxy to hold the workpiece down. For a more production oriented process, a vacuum fixture or an actual vise with a soft jaw set could be used. Here I was making only one and wanted to waste as little time on tooling as possible. It worked great!

This is a mold plant made of 40 pound per cubic foot urethane tooling board (SikaBlock M600) that needed to be machined all over to mold a recess for the keel of a sailboat. I machined this in three setups using scraps of the same tooling board to fixture the part for each operation.

The first step was to model the whole thing. I used Rhino3D for the modeling and MadCAM for the toolpaths. Each setup (cradle and cut on the part) was done with the same work offset.

Key to getting this first flip to work was the use of a tight fitting cradle machined from the tooling board - but only in the corners of the part. This way I saved on cradle material and also had good visibility to make sure the part was precisely located in the cradle before machining the second and third operation.

Here’s the first operation roughed and the finished first operation with machined cradles for the second:

cradle op1and2.jpg

With the cradles machined for the part to get flipped into, it was easy to know exactly where the part is at. I kept the same work offset for each combination of cradle and finish cut. In the 3D model, I turned on the layer with the part in the cradle and was ready to do the CAM work for operation number two. I have found that with high density board like this a tolerance of 0.002” to 0.005” is plenty. To make this offset, I used a negative number in the “stock to leave” setting in the CAM software. You could very easily offset the model by the desired gap and use it to cut the cradle with a solid modeling operation.

To hold the part into the cradle, I used a small amount of 5 minute epoxy on the top edge of each block where I could remove it easily with a razor knife. On the bottom of the part, I used some MDF blocks and more 5 minute epoxy to hold the part in the cradle. Here I knew I would machine that area so could afford to really goop it up with 5-minute - see middle image below.

cradle op2 fin.jpg

Now that five of the six sides are machined, I have to find a way to hold this up vertical to machine the bottom socket. Arguably this could have been the first operation with the sides done second, but I didn’t do it that way. I did it like this:

cradle op3.jpg

Now this got a little dicey. There’s not much of that part in the cradle for the third operation and the fit wasn’t as tight as I wanted - see the blue tape placed on the cradle to tighten things up. To check the location of the part, I used the 1/2” bit placed with it’s center 1/4” from where the edge of the part should be. By tapping the part until it just touched, I was able to locate things side to side. It’s no probe but it works! Once it looked good, I put some 5 minute epoxy on the areas where the tape could keep it from sticking to the part, but where it would chock the fit. I took gentle cuts and it worked out great! My caliper said the slot was within 0.005” of the center of the bottom of the part. For a machinist this would be rough as, but for my purposes here it was great!

And the whole thing was done by quitting time!

I have used the same cradle method many times for machining all kinds of materials on the routers. Because low density materials produce less cutting force than metals, you can get away with a lot more! Here are some examples of me getting away with stuff:


This little boat model needed to be machined on the deck and the bottom. The deck had a more positive locking geometry, so I started there and then machined enough cradle to lock it into place. This was very light foam and I held it down with double-stick tape.


Long ago I machined the crash bow foam for the monster racing yacht Comanche. This cradle was only about 1/4” deep! With the top of the foam piece machined, I routed a cutout in a sheet of chip-board and fit the foam edge into it. I don’t remember how I held it down, but 5-minute epoxy probably played a part.


These daggerboard bearings are machined from a hard Delrin-like plastic called Vesconite. It is expensive and slippery. Here I fixtured these parts for second operations using urethane tooling board cradles. The left-most picture shows a “discontinuous cradle” approach, and the middle image shows a full cradle. The right hand pictures show the completed bearings and the reason for the solid cradle - lots of thin-walled sections to support!


This article contains information that reflects my opinions - I make no promises about its usefulness!  It may contain mistakes (please let me know if you find some!) and will include prejudices based on my limited experience.  If you disagree with anything here, please get in touch. This is not just for me to share what I know, but to learn from others.  I will gladly insert additional information and differing opinions so readers are more aware of the diversity of “right” answers!