Thread Rating:
  • 0 Vote(s) - 0 Average
  • 1
  • 2
  • 3
  • 4
  • 5
[-]
Slideshow

[-]
Stats
» Members: 670
» Latest member: rschoenhard
» Forum threads: 582
» Forum posts: 3,077

Full Statistics

Printing in Flexible Materials.
#1
Smile 
Hell Builders and Droid techs,

Been a while since I posted anything, so I thought I have better do an update.

Progress on my R2D2 has been slow of recent due to other family commitments, but recently I did manage to have R2 move around under his own power, albeit not very well.

The 2 Razor 100W 12V motors were very much under powered using the single belt drive. 
I have a DC-Clamp meter so was able to get current readings of the power going to the motors while R2D2 was driving.


So first a little bit of basic motor 1O1.
A 100 Watt motor running at 12 Volts will draw around 8.33 Amps at full rated load.
Watts Law is given with the formula P = E x I
When a motor first starts, it is overloaded, until it reaches its designed speed, normally, this is only a few seconds and isn't too detrimental to the motors.
If the power is available the short overload can be as high as 5 times the rated load of the motor.
Once up to speed the current feeding the motor will drop off to a current that matches the current load on the motor.
If your motor is under loaded, then the current drawn will be less than the full rated load of the motor. (Marvellous creation the electric motor.)

There are a number of factors that can reduce the current drawn by an electric motor, these include:
  • Motor Controllers
  • Wiring resistance
  • Battery internal resistance
  • Faulty connections

On my first attempt to drive R2D2, I had a current draw of 18 Amps, and R2D2 did not move.
The motor did not turn.
If I gave the droid a push, if would move and continue to move, but the current did not drop.
Based on my experience I was experience I was expecting a locked rotor current of somewhere in the order of 41 - 42 Amps.
So now I was looking for a resistance that could drop the voltage to the motor and therefore the current to the motor.
My speed controller is a IBT-2 PWM controller (not very smart but can handle big currents at a low cost) 
The controller is rated to 43 Amps with internal load limiting, the 18 Amps was no where near the load limit of the controller.
Being the electrician that I am, I did slightly oversize the wires. The larger the diameter the wire, the lower the resistance and so the less voltage drop in the wiring.
All the connections tested as OK and the thermal imager did not show any heat buildup where it wasn't expecting it.
That leaves only the Battery.
In my case a 12V 5 Ah SLA.  The internal resistance was too high to supply that kind of current...  Sad
I should have known better, and now I have done the maths, do know better.
The quick and dirty solution was to add a second battery in parallel to double the current capacity, in truth, I think I could use 4 batteries instead of 2.
First test run saw R2D2 make it half way across the kitchen before pulling a hard left.
Each main motor in my build has its own fuse. The 20 Amp fuse for the left motor had blown.

I replaced the fuse, added the DC-Clamp meter and tried again.
Now R2D2 was drawing 28 Amps for each motor, or at least is did until the right motor fuse blew.
Well, the wiring would handle it for a while (Remember I did oversize the wires Big Grin  )
I replace the two 20 Amp fuses with 30 Amp fuses and gave it another try.
It's still way under powered but I could get it to drive around, and I had to dome turning as well with no problems at all. (10 Amp fuse there)

I installed an MP3 player module and installed the MicroSD card with all the downloaded sound effects.
My R2D2 is currently being controlled by an Arduino Mega 2560 withe FlySky FS-IA6B RX unit connected to it via Serial 2.
This allows me to send 10 channels of control from the transmitter to the Arduino.
Using this I can select a random sound file to be played from one of three groups that I can select with the transmitter.

While R2D2 is far from complete or functional enough for any sort of get together, it was enough to impress my Nice and Nephew  Smile

Back to the problem at hand and the overloading drive motors. 
I really need to change the gear ratio and I think I have too much resistance to rolling in the front Omni wheel assembly.
Its was about then, Michael Baddeley uploaded an Alpha version of his new MkIII geared drive system.
So now I'm back to printing.
Most of this has now been printed out and a dry fit (Still have the new omni-wheels print, but that's just a mater of time.)
I had bearings on hand for the main wheel and as it happens the middle gear.
The Middle gear is difficult to fit (This is an Alpha Version so bug are to be expected)
And the order of assembly is critical. 
Mount the wheel on its shaft first then the middle gear, not easy but can be done.

I needed another tyre printed for this wheel, the last tyres I printed weren't the best so decided to print a new one. (Next Headache)
The roll of flexiable filament I had is labeled as TPE with a print temp of 190-240 C
I have a modified Cocoon Create Touch.
  • I replaced the extruder a while back with a Flexion extruder.
  • Moved the part cooling fan to the back using a duct.
  • mounted a BLtouch Bed leveling sensor to the front
  • I replaced the bed with an Anycubic Ultrabase glass bed. (this has a pourus like surface in the top of the glass)
  • Replaced the firmware with ADVi3++ (https://www.patreon.com/andrivet/overview)

My first attempt to print the tyre resulted in it coming loose from the bed.
I tried a number of other configuration before finally finding one that worked.
All the information I found online talked about the extruder and how flexible filaments tend to find the path of least resistance out the side of the extruder. I didn't have any of those problems, The Flexion extruder solved that for me.  My problem was the print would let go of the bed around 15 layers in of the 136 print.

The solution was: 
set the part cooling fan off
Set the hot end to 240 C
Set the bed temp to 90 C
Retraction I set to off  but have since set it to 3 mm and no Z hop
and the speed is set to 15 mm/s
layer height of 0.25 mm

The results were quite good.
[Image: attachment.php?aid=1425]

So while I had the TPE loaded, I have also printed out a few other parts so i can get the doors working as well  Smile

Parts for the new feet printed so far
[Image: attachment.php?aid=1426]


A closer view of the Middle Gear
[Image: attachment.php?aid=1427]


As can be seen, this is just a dry fit, so no screws yet installed, but already I can see a number of improvements.
The other major change, is all the axle pins are printed.  Will be interesting to see how that holds up over time.
This was printed in PLA.

More updates to come....


Attached Files Thumbnail(s)
           
1
Reply
#2
Nice post.

I am running my droid at 24v with the 100W razor motors. I have four 8500mAh 3S Lipos in series and parallel. It worked fine with the single belt drive design with two 3S batteries in series, but I wanted it to have more capacity and oomph. With just the two batteries I was able to last 10 hours on hard surfaces at DragonCon. I have two more of these batteries in parallel to form the 12V system for all of the electronics.

Eebel
Reply


Forum Jump:


Users browsing this thread: 1 Guest(s)