X-Plane 11 6DOF motion platform

It is good that you cleaned the inside of the carbon tubes, as they often have mould release still on them from when they were made .

I thought it might be a good idea to see how much the motor draws unloaded and test the clamp meter at the same time. At no load and 24V the motor draws 1.8A. As you can see from the images the clamp meter is not very accurate and becomes less so as the current reduces. I didn't pay much for it but it is pretty much useless. I should have purchased shunts instead so I could measure the voltage across them when needed.

Edit: warning - this meter cannot read DC current - the readings were probably AC ripple. If you are going to buy a clamp meter make double sure it can read DC current. Generally, you won't get one for under $40AUS

I have a clamp meter, which I think works fine for initial setup, as I have little interest in small current measurements. Beats the complexity and basically just more things to go wrong in wiring in permanent shunts etc (done this before ). I am fitting thermal couplers to my motors and cockpit remote temp sensors in my upgrade, as this is something that I am interested in monitoring long term. It is also simple/cheap to do and impacts nothing else, if it fails.

Your current readings look to be an order of magnitude off.

...Looks like you're measuring DC current while the clamp meter is in AC mode (does that meter even have a DC current mode?). If that's the case I'm surprised you're getting any reading at all.

For a low-cost low-current DC-capable clamp meter you can do worse than one of these.
I have one in my automotive toolkit and so long as its orientation isn't changed after zeroing it's pretty accurate for measuring small current draws (although it'll also go up to 100A). There's some slight drift over time (as is pretty much inevitable when measuring DC current with a clamp-meter), although only a few mA.

There's a fairly comprehensive review of it on the EEV blog forums.

Good pick-up @DEADBEEF , I'm sure @SixDegreesOfFlight will let us know. I could not find DC current in the ST201 specs .

@DEADBEEF
Hmm...I hadn't thought that it might not be able to measure DC current. The first thing I noticed was the decimal point in the wrong place and that the Mode button didn't do anything. Dammm....just re-read the specs on ebay and it only says AC current, so that now makes much more sense. Pays to read the fine print Good pick up

I don't know if I have missed what you are trying to explain, but my DC capable clamp meter works fine for DC currents.

Nope, you didnt miss anything. I'm from the era where clamp on meters were AC only and just was unaware that there had been innovations. I bought one a number of years ago and haven't needed to revisit. It used to be true that clamp ons were AC only but not any more.

Just checked and clamp on meters can indeed measure DC currents if they are equipped with Hall sensors to measure the magnetic field. Learned something new and deleted the post.

Edit - but it still stands that since Six's meter wasn't one that can measure DC current, what he would have been seeing would be the AC ripple on the wire and the reason the numbers diverged as he turned down the current would probably be because the power supply was able to deliver cleaner DC at low current levels.

That's fine, just did not want anyone to think they only did AC . Technology moves on, but it is surprising how few 'new' innovative products/technologies have been produced in the last 10/15 years - more just improvements on what we already have. I have my theories, but they are not for this forum.

A few more build images...

The first one shows the u joints with their anti-rotate 'washer'. The u joints and the 75mm 'washer' will be removed from the bottom aluminium plate and welded together. What is not shown is the other washer for each of the joints. I am hoping to do some welding this week - maybe even the top frame.

The second image shows the rods connected to the lead screw ball nut and the motors screwed to their mounting plates

The third image shows the rods connected to the bearing plate. At this stage everything was going smoothly

The last image shows the initial assembly of the actuators. Things didn't quite go according to plan at this point. The naughty one in the corner has a problem - the rod sleeve wasn't pushed in the housing enough when the epoxy was applied and as a result is not quite vertical. He will need to do some serious thinking about his attitude - with a file

Well here it is. The belt isn't tight and it's not wired up but I am liking the contrasting colours

The bottom plate will be closer to the motor once the u joint is fitted but far enough from the bottom of the motor to allow the air flow from the motor case.

Great build! I hope to start my own in winter ( well, AUS summer ) and I'm sure your detailed descriptions are going to help.

@FistMeNaruto
Thank you! Make sure you take the time to plan out the whole build and costs - it will be worth it in the end especially the choice of motor. Read through @wannabeaflyer2 thread so you are across all the aspects. I have a detailed cost spreadsheet that might help you on page 1 and images of the parts as they came in. All the best

I worked up the courage to actually turn the thing on



A brief test of a completed actuator using the SMC3 Setup and Monitoring Utility. I followed the instructions in the tutorial here.

To set it up I hand wound the motor so the lead screw nut was roughly in the middle. I did the same with the pot - I put a mark on the pulley wheel and turned it 5 rotations and then attached the belt. After turning the power supply on I turned the battery isolator switch to allow the Sabertooth 2x60 to get power. The relay was already energised but the 60A breaker was still open. I fired up the SMC3 utility and put everything at zero as the tutorial suggested. Then I engaged the breaker and checked that the motor wasn't moving - it shouldn't with a pulse width modulation of 0. I increased the PWM setting and the motor began to turn but it was moving away from the blue line so I switched the power wires on the pot and tried again. Now it was working fine. It was squeaking though which means the 3D print is rubbing on the aluminium frame. I will adjust this later.

The two limit switches are wired in series with the coil of the relay. If the ball nut reaches them, they will cut the power to the relay coil causing the contacts to open and break the power to the motor. They have extra wire on them so I don't have to solder them again when the actuator is mounted. The relay is driven directly from a 12V DC power pack which powers the Arduino which you can see plugged into the Arduino jack. The 12V is picked off before the Arduino regulator.

As you can see in the video - with a PWM of 100 it still moves very fast. To be honest it scares the hell out me

Great to see you have it working . I would have liked to see a bit more speed on a small load, but all that really matters is how fast it is under a full load.

@SeatTime
Thank you! There are a few more tests to go yet. I too am interested in how fast it can go. I have a 10kg bag of rice that I will tie through the rose joint eye and see how it affects the responsiveness. One thing I am not sure about is how to control the amplitude of the waveforms coming from the SMC3 utility e.g. the saw tooth / square wave. It 'seemed' that as I turned up the PWM the stroke also got longer but maybe that was my imagination. What I am hoping to do is reduce the amplitude of the square wave and turn up the PWM and record the responsiveness over say 200mm with my phone in stopwatch mode. I can then calculate the speed from the video

You should be able to fix any tracking issues by tuning the PID. One of the reasons that I still use a 1280 controller is because it's PID works very well.

@SeatTime
Once all the actuators are working I will experiment with the PID control - which I know can take a long time to get right. So at this stage I am not concerned with the fidelity of tracking the waveform but rather the amplitude of the waveform itself coming from the SMC3 utility. Is it something to do with the limit setting?

I have just finished the sixth actuator. They are all now assembled. The one not in the image is on the test bench. I have figured out how to fix the squeaking which is caused by the V Slot sides being a fraction too close to each other causing the 3D printed ball nut housing to rub on it - but only on one side. The top bearing plate had one hole which was about 0.5mm too close to the centre. It appears to be same on all of them so it is probably my template measurement rather than the drilling. I simply enlarged it slightly with the next drill size. One now is much smoother and quieter and have to do this now to the other five.

I painted the frame on the weekend using two cans of satin spray paint. The British Paints Paint & Prime was applied directly to the bare timber and it worked really well. One can per coat and I did two coats separated by 1 hour. The other two cans I got will be used on the top frame.

Hey @SixDegreesOfFlight, very cool and very beautiful!

You said you were using a pulse width value of 100 and we're surprised how fast it was moving for that being a low setting. I'm not familiar with the Arduino software. What is the range on pulse width values? I was tuning up my JRKs this weekend and coincidentally also setting up with a max pulse width of 100 (out of 600) and even though it was slow-ish, my output arms were still peppy enough and had enough force that they would break things if they ran away. These actuators can put the fear in you! I'm sure my gearmotor output arms would break fingers no problem if they got in the way. I haven't run the pulse width max up to give them full range but can only imagine.

Also, does the Arduino code have any safety setting to shut things down if it loses the feedback pot signal?