In the previous video my prototype for a 3d printed machine and I said I was going to dismantle this one and put the electronics over to that. I’m still going to do that but between what I did and what I said I’d do, I decided to do something else and got distracted. And I bought some mgn12 linear slides and I’m actually quite impressed with these. And I think I’m going to change the design yet again, to include these as an option for creating the linear motion. The wheels are great as a starting method to create a CNC machine but I think I just want one decent big CNC machine that’s going to be able  to do everything that I want. But I want to demonstrate that that machine can work, so I’m going to bite the bullet and transfer the electronics over. And at the end of the video I’ll try summaries the pro’s and cons of the methods and materials I’ve use up to now.

There’s not much going over other than the cabling for the stepper motors and the proximity sensors, however the latter will be rewired. When I wired up the previous machine, I added JST connectors which acted as quick release point between the stepper motor wires and the screened cable which ran toward the controller. But I can’t remember how I wired these so I trawled through my paperwork to find some diagrams of what colour cables wired to what.

My stuff is all over the place, I really can’t remember where things are.

Ok in all my wisdom I also bought some pre-wired JST connectors so I should be able to plug those straight in like that and join these together. I assembled my own JST connectors when I wired this previous machine, and it was really fiddly so I’d recommend getting these.

Anyway this is a rough assembly, so there will be no drag-chain and I’m going to use terminal blocks instead of soldering in some places.

Ok I’ve roughly wired everything up using these terminal blocks, as an extension. And I’ve also got these pre-assembled JST connectors which make everything a lot easier.

I should also mention that just in case either my drawings were incorrect, or the colours of the stepper motor wires were not the same as the previous ones, I also checked the motors by holding two of the wired together and trying to turn the motor shaft by hand. When I found two wired from the same motor coil it became harder to turn. I kept the pairs separate and wired as illustrated in my notes.

As my steppers motors are different from the original machine, I’m going to change the switches on the drivers for the new amperage settings. Making sure the steppers are not too over or under powered. Because I’m using the same leadscrews and step settings in the firmware, I’ll keep the steps as they were.

Ok the stepper motors are 3 amps, so I’m going to have to change  switch 1, 2 and 3. That’ll be to 2.8 amp. And that’ll be off, off, on.

I also checked the motors were turning in the right direction before tightening the motor couplers to their leadscrews. If they weren’t I’d have swapped a pair of the wires around at the terminal block.

So I’m happy that all the stepper motors are turning in the right direction. Because the machine is so large and it’s positioned near a wall it was easier for me to take the stepper motors off from the front of the machine, and to feed the leadscrew in.

The x and y axis both have locking collars at either end of the leadscrew –between the motor plates. However the z axis only has one at the bottom.

Ok I’ve put the leadscrew on this side and the opposite side, but I need to push the gantry back just to get the next one in. There’s just a bit too much stuff around here.

There’s a tiny bit of movement and that’s because the locking collars need to be pushed in. It’s a shame there isn’t something that has the same thread as the TR8 leadscrew which you could screw on and then tighten with a grub screw. That would make it a lot easier but it’s not bad – it’s moving.

The other thing I’m going to do is, wire the proximity sensors for the X and Y axis individually. Previously I wired those axis in parallel, but I’m not going to do that again.

So if this was more long term I would have covered this with a bit of heat shrink as well but now I’m just going to leave it as it is.

Ok all the lights are turning on. I’m going to do a danger home. I’m going to press the home button and use this to trigger it. Ok it all works. So now I can let it do a real home.

So I’m still getting error messages when I’m homing. There’s some sort of interference happening somewhere. This is something that started to develop later with the old CNC machine. It’s clearly something in the controller and I’m going to have to investigate this at some point.

So on the motor plates I have an opening for a nut, this holes a low profile machine screw. This screws in place and on the reverse, you place another nut and that tightens up. And essentially what this does is acts like a hard limit which this surface here butts against. In this case the y plate can’t go any further than that face there, and I’ve added this type of fixing on all the plates. When you start using use machine and you go through the process of squaring the x and y axis, what you can do, is bring the gantry forward until it rests somewhere near these stops, you can disable the power to the motors and adjust by hand until you’ve squared everything up. When you’ve installed your proximity sensor you can move the opposite plate towards the proximity sensor in a homing movement. When that triggers you can then re-position your  hard stops so they are a little further forward than the end of the proximity sensor and what that would do is if you accidentally over shoot your mark you can avoid crushing the proximity sensor. And I’ve already crushed one, and I said crush and not crash.

The other thing to mention are these locking collars. When you first start to move the machine around these possibly push the bearings in a bit and you get a little bit of movement. So after you set up the machine, you want to push these back. And I’m doing this with my elbow. I’ll slacken this off and then push that in that direction with my finger and now that movement has gone. But I feel like over time you still have to check and maintain that tension. It’s just one of the limitation of leadscrews.

The other thing I’ve added to the design is these bracket mounts and these fit under here. To add a little more rigidity. You can see the holes are slightly elongated slots and what you can do with this, when all the machine screws holding the C beam are slackened, you can actually pivot. And that’s actually quite a substation bit of movement. And the value in that is that you can eliminate the nod in the spindle which often happens.

So I’ve made this simple jig with an off-cut of c-beam and and a bolt threaded through it. I’m just going to raise the bolt up until I get 90 degrees. I’m just going to go back 0.1 degree, because the weight will drop it back down after I tighten everything up. I’m going to do the same on the opposite side. Ok I’m happy with that. And this is when I’ll tighten these additional brackets.

I’m just going to move this across that way and see whether the angle changes at all. That’s gone to 90. It’s teetering on 89.9 – 90 degrees.

So two machine screws here, and the ones on the opposite side allow the z axis to pivot this way around. I position that here and zero it up, and then ideally after that you’ll level your waste-board and recheck everything this time paying more attention to the bit as opposed to measuring other parts of the gantry.

So I don’t have a fixing method as of yet so I’m just going to use these clamps.

I’m going to do a test cut – a hello world sort of message.

After probing the v-bit, and doing a G92 command to set the z – zero height I start my first cut.

Remember I also haven’t levelled my waste-board – you really need a dust shoe to level your wasteboard otherwise you’ll make a massive mess – and this machine doesn’t have a dust shoe.

So I want to just recap on some of the features I’ve introduced in the design, the first thing to mention is the pivot plate and the pivoting mechanism I’ve incorporated into the gantry. This allows me to rotate the spindle along the x and y axis, so that when the spindle moves up into the z axis there’s no travel. With other entry CNC machines and hobbyist machines, there’s not really an easy way of doing this. What ends up happening is you notice that your machine isn’t cutting entirely square, and you have to dismantle it up to a point and shim different parts of it and to then repeat your cuts and see if that’s made an improvement. The problem with this method is it involved a tremendous amount of time and a tremendous amount of trail and error so the pivoting mechanism is easy to access and a lot quicker to adjust.

The other things I’ve incorporated into the design of the plates is hard limits. These have two purposes. The first is to prevent the plates crushing the proximity sensors and the second is to help square the x and y axis. The third thing I did was add mounts for the proximity sensors. This felt like a good idea at the time but in retrospect I could have put these elsewhere. So the proximity sensor which fitted on the motor plates facing towards the y plates could have gone on the y plates facing outwards, and the sensor on the y plates facing towards the gantry could have also gone on the rear of the gantry for example facing towards the y plates. And I think that would have help to reduce the amount of wiring overall.

The forth thing was to use leadscrews for this machine and not belts. When I first got the x carve that used belt and that performed really poorly. The fifth thing was to increase the rigidity of the gantry and I did this by adding a piece of 20x40mm aluminium profile at the bottom of the y plates. And that help square the c beam to the y plates.

The sixth thing I wanted to incorporate into the design was , to minimise the amounts of wheel I used without effected the performance of the machine. The cost of this really mounts up and I wanted it to be quite affordable. And the last thing was, I wanted the plates to be produced in different ways. So I’ve built variation of this machine using birth ply, using acetal and now using a 3d printer. And designing nut traps into the plates has meant you can use these material and avoid using other material like aluminium and steel which are harder to machine.