Solenoid Motor

A Lesson 'inEffeciency'

After being inspired by a couple of pieces of impressive craftsmanship like this, I decided for no particular reason to design my own Solenoid Motor. Initially, the concept seemed simple enough, but I soon found some low-hanging engineering flaws with most efforts that had been undertaken by the hobby motorist community (which exists in a thriving sense apparently, and has some very cool works to peruse). Unwilling to accept these shortcomings in my own build, I set about finding ways to reduce some of the inefficiencies inherent in a solenoid motor design. In doing so, I ended up posing myself one of my most challenging engineering problems to date.

The premise of these motors is simple enough. A crankshaft, like those found in an Internal combustion engine is driven by linear solenoids, which act like the pistons in your everyday ICE. But I immediately identified several sources of efficiency loss in most applications of this design that could be improved upon. In non-exhaustive list form they are...

A typical iron core solenoid has uni-directional action. That is, it can only pull the iron core (or plunger) towards the center of the coil. Conventional solenoids may use a spring to return the core to its original position, but this comes with inherent losses in force when moving the core via electromagnet. I immediately recognized the potential to have a dual-action solenoid if the iron core were replaced by a strong rare-earth magnet core. This of course came with its own challenges, as most common magnet materials are difficult to shape, expensive, and can lose their magnetism if subjected to high temperatures or impact. Steps were taken to mitigate each of these.

Most hobby-grade solenoid motors use mechanical switching of the coil windings to actuate the solenoids - that is, something like the brushes in a DC motor (but often more improvised). But this adds some friction, and current losses, which are unacceptable in an already inefficient design. So I sourced some H-bridges and an Arduino to take advantage of more efficient switching and the opportunity to reverse the coil polarity and use my coils to pull and push my now-magnetic cores. This results in what is essentially a '1-stroke' engine. That is - every action of the pistons is powered.

Of course, the coils had requirements of their own. Tolerant of higher voltage, current, and duty-cycles, and capable of rejecting quite a bit of heat (to protect the new magnetic cores). So I designed my own, using square 18 AWG copper wire and bobbins 3D printed from PEEK. Originally I hand wound each one to keep the cost from obliterating my undergraduate-level budget. The casings (which concentrate the magnetic field and increase the strength of the solenoids even further) I cut from 2" -thick low alloy steel using a water jet or machined on a 3-axis EMCO CNC mill.

The first prototype motor was completed in early September 2019. It still had issues though, most notably, a tendency for the pistons to 'jam.' A second version was designed in early 2020, and constructed in April of 2021. It utilized a much shorter stroke to add efficiency at the expense of max torque. Testing on version 2 is ongoing.