Universal testing machine

During the 2020-2021 pandemic I decided to build my own UTM (Universal Testing Machine). The progress and learned lessons from this project are documented on this page.

Solidworks design overview updated 09-05-2021

Initially efforts were mostly focussed on the clamps of the UTM.  Above the as build model of the clamps. The idea of this type of clamp is that it's self tightening. The design is slightly different from the industry norms because of the limited machining capabillities in my backyard. 

My first proto looked quite different from the final result and turned out to be a learning experience that didn't produce a useable casting. However, using proper casting techniques and simulation I got a nice result

Redesign, optimised for casting using the ten rules from John Campbell. This design was a lot better for casting indeed. However if I had to make something like this again I would make the seperation between both halves on the bearing surfaces for the wedges. 

Obligatory credits to my friend Wytse de Vries for 3D printing the plug. Thank you so much!

My first idea  to mill the cast parts down was making a small simple milling machine from a dremel power tool and some linear guides.  This was no succes because there was too much flex in virtually everything. Not unimportant either the dremel had a rediculously low material removal rate.

After the dremel mill I tried using a router with some straight planks as guides. This worked surprisingly well. Ofcourse the tolerances are awfull but I feel like it's good enough for my hobby project.

For the wedges I didn't spent as much time on optimizing the gating system because of the simplicity of the part. The relief angles were removed using an angle grinder and a lot of patience.

C45 hardened and tempered jaws

Karate chop those threaded rods to the right length! Unfortunately these threads were metric coarse thread and load cells use metric fine thread.. sigh

Minilathe in action

Finished clamp

Frame

Drilling holes (left).  Placing rivet nuts for the lineair guides (right).

Hot glueing the frame together, turns out stick welding stainless steel nicely is not something you learn in half a day.

This week it's been a year since the start of the project and the UTM has given it's first (albeit not very acurate) outputs.  Hooking up the electrical components was very easy because I've used them before and a Load cell and potentiometer are relatively simple sensors to begin with. (11-7-2021)

extensometer proto number 1. as a first test I made a simple setup with a homemade load cell I made some other time. Unfortunately I could not get a large enough mechanical advantage on this load cell and had to live with a maxiumum of about 70 microstrains, while the strain gauges can handle a lot more and should get a lot more for an accurate signal. Nevertheless this first test showed promising results for making an extometer using strain gauges. 

A second test with strain gauges directly glued onto leaf springs showed much better results however I also learned that it's very difficult to make an accurate/precise extensometer and I decided that I would rather focus on other projects that are more fun and relevant right now. 

I first placed 4 strain gages on one spring and covered them in epoxy. Unfortunately this attempt showed a large amount of creep (change of measured value over time). I'm not sure if I did a bad job in gluing the gages or if it's really because of the epoxy layer but for my second attempt I only covered the solder contacts with a small amount of hot glue. With this attempt the creep was much less. My testing method was clamping down the spring and putting coins against the end of the spring. 

19-10-2021, 

School is taking a lot of my time and I have moved to a different city which also cost some of my time. As a result this project has not really seen any work and for now this project is onhold/discontinued. This is fine considering the design is flawed in many ways and I have already learned a lot while making everything. Yet I feel like it would be satisfying to atleast finish the extensometer clip and plot some more stress strain curves. 

28/11/2023 Recently I bought a 3D printer and naturally wanted to check the strength and layer adhesion of my prints, for this purpose the tensile testing machine came in handy.