Binky Update!

The feeling of relief was palpable at the shed recently. We’ve been plagued by an issue we didn’t know we had and that has been going on for years unidentified. It hasn’t been a problem up till the advent of our metal solder 3D printing malarkey. Since the very first foray into this idea when initial tests on a cheap 3D printer showed promise, we’ve been fighting a battle with one armed tied behind our back – and we didn’t even know it.

As you lot know, we converted our venerable old Taig mill into a 3D printer because we needed the extra X travel that that allows us. We added a 4th axis to become the extruder part of the set-up and we built an Arduino powered control box to take care of the nozzle heater. So far, so straightforward…

Early on, the Mach3 control board expired – it was after all probably 20 years old and had been abused on a regular basis - and we took the opportunity to rebuild the insides with a new board and a new power supply that allowed us to junk the 240v-110v transformer that it had been supplied with.

Okay, great, let’s now get on with it. Only it wasn’t that simple. We struggled like mad to get consistent results; sometimes it would be absolutely perfect and then the next day you’d spend hours trying to dial the settings back in. It was infuriating. We managed to get the switch panel boards done and filmed and went back to do the main instrument cluster PCB. Again, no consistency. Cue tearing out what little hair we had left at this point, along with a desire to burn the whole place to the ground, pack it all in and go and live on an island somewhere, as a hermit.

We were losing steps on the mill’s stepper motors completely randomly. When the tolerance on the Z height is as fine as it is, any lost step is downright disastrous, but it was happening on all the axes. Then we lost the Y axis completely. F**!

 The cheap new power supply blew a transistor and that left us with 75% of the axes we needed and no way to power the steppers. Brilliant. The control board had been discontinued so we had to think of a different way of rebuilding the mill, again.

We sourced some new parts (4 individual stepper controllers and 4 individual power supplies) and waited for them to arrive. We had the steppers hooked up to 2 x 12v car batteries to give us a stable 24v, the Arduino powered nozzle heater powered by our “scientific” adjustable power supply and what was left of the stepper controller powered by the PC’s PSU just to try and figure out what the hell was going on. Talk about Heath Robinson…

The 12v batteries we hooked up to an old school “start and charge” type battery charger to make sure they didn’t go flat. You know the type – the one with the needle indicator to show how many amps it is drawing. Well, that turned out to be the key to unlocking the mystery.

 While the mill was running through a program designed to check the settings, Nik finished off some wiring by adding some heat shrink to the terminals using the heat gun. Out of the corner of his eye, he noticed the battery charger amps dropped from 8 to 2 while the heat gun was on. “That’s odd” he thought. So, we did some investigation. We plugged a multi-meter into one of the sockets to log the mains voltage and switched various things on around the workshop. The kettle dropped the voltage from 236v to 211v. The big pressure washer did the same. Even the lathe had an effect, especially at start up.

It turns out that there wasn’t an electrical ring main on the socket circuit in the shed. Every single socket was essentially on one long spur, and the shed contains a lot of sockets with a lot of different equipment hooked up to them. The loss of voltage meant that the current draw ramped up as a consequence and the delicate electronics in the mill controls and its power supply simply couldn’t handle it, so that’s why it was missing steps and blowing up controllers.

We installed a ring main to finish off the socket circuit properly and the voltage is now stable despite throwing everything at it to test our work.

Subsequently, the mill purrs like a kitten and once we dialled in the new controllers and power supplies, it’s been absolutely spot on since. So much so that we’ve now completely finished the solder printed PCB boards – to a better standard than we’ve ever managed before – affixed all of the electronic components to the traces, completed all the daughter boards and we’re assembling the instrument cluster for the final time.

God its been a slog and we’re forever grateful to you all for keeping the faith because believe me, we nearly threw in the towel on numerous occasions. We’ve got probably 2/3rds of Binky 39 complete and edited and we’ve got a few final shots of plugging the instruments in to test them along with the pieces to camera to do to tie it all together. Episode 40 will be a continuation focussing on the switch panel and we’ve already shot parts of episode 41 so it’s all systems go, and we believe we’re back on track.

You lot already know that we’re using the colour changing WS2812B LED’s for the backlighting, but keep shtum, because we won’t be letting that slip until episode 40…

Thanks again everyone, we hope to reward your patience with some great content coming regularly from now on.

Cheers,

Nik, Richard and Ben.