Fixing the Variac……..Part 2


A few months ago we did a blog post about our variac. It was tripping the circuit breakers on our distribution board and we made some mods to the variac to reduce the inrush current. Our story generated a lot of interest and was picked up by Hackaday. We used a 2 terminal NTC inrush current limiter from Ametherm.

Following our story, the good folks at Ametherm and Rhopoint Components (our local distributor) were in touch with us. They gave us some great technical advice and did some calculations for us. They have a whole range of limiters and it turns out that we had actually chosen quite a good model for our application. They kindly sent us some samples.

There are many ways to soft start a load. We originally thought about adding a series power resistor and a shorting relay. We also had some suggestions around Triac and IGBT solutions. Each of these methods have their merits and can allow fine grained control of switch-on. Unfortunately there was very little room inside the variac case and we didn’t want some kind of external box. So we looked at these relatively simple 2 terminal limiters.

Carefully chosen, these components are an elegant solution. They can probably last the life of the application with no maintenance. Placed in series with the load, they have a relatively high resistance at room temperature which limits the initial inrush current. When current flows through them they self-heat and the resistance falls very quickly, so they have little effect on the normal operating current.

We have now done a rev 2 of our mod. This time selecting the model MS32 5R020 limiter. It has a lower cold resistance of 5 ohms which should keep the operating temperature down a bit. Originally, we connected the limiter using PVC insulated wire, solder and heatshrink. Given the relatively high temperatures, we have replaced this with tubular glass fibre insulation and ceramic connectors. We should have no temperature worries now.
One other gotcha we found, is that our distribution board circuit breakers are actually B-curve and not C-curve as we earlier thought. That makes them even more sensitive to surges. When we did our original mod, we didn’t do a full load test. We did a quick run with our heatgun (about 9 amps). This time we did a full load test. At 12A of output current and an ambient temp of 20 deg c. The body of the limiter measured 200 deg C, which is hot, but below the 225 deg C max rating of the device. This was quite an extreme test, but we’ll continue to monitor and do some tweaks if possible.
If you’d like to take a look at our variac, or do something else with electronics, our regular Electronics, Microcontroller and IOT evening takes place every 2nd Monday. Check out the events section of our website for details.