More Tweaking

X2 and i took a tour of the village.  Still problems to solve, but so much better. :D

After the jaunt, i decided that i needed to simplify a few things, just to make life a little easier.  I got the right failsafe applied to the 2x32 Sabertooth, so i could remove the relays for the feet.  He cant be pushed now without the battery, but saves any current worries there.

I also reduced the max current to 32A per motor from 64A, but i have already had another trip, that was in special circumstances but i think i might be hitting the 30A constant limit, rather than the 90A peak, so i will try and half the current again, to 16A per motor and see if the ESC protects the battery limits that way. :)

After removing the feet relays, i decided to increase airflow on the box, so got the Dremel out and cut lots of holes. :)  The ESCs should be able to breathe a lot easier now.

The speaker connectors were getting lose, so i added some RCA connectors to the box, and run the speakers through those.  At only about 25W RMS it should be fine. :D  Still sounds ok. :)  The DC converter for the amp had failed, and was the cause of some of the tripping, so that has gone, the amp will have to do running directly at the battery voltage.

Then i decided to look at the PWM controller setup, the way i have been bypassing the power rails has caused a few issues, so i wanted to use code to turn the servos off.  Eventually the code for the body was working well, a little rough, but no relays and better than it was before the change.  So i rewired dome, applied the change in the same way.  Powered it up, but all the servo spots were different!  And 2 of the cheap HP servos were already dead (they had been buzzing hard and i hadnt sorted it). So about an hour later i had all in and the spots and it was still horrible!  It did work but there was still a lot of Buzz, so i added timers for all the servos, now they power off just after they finish travel. :)  Even the HPs are unpowered if not moving.  Works great now, and the wiring is child's play.

While in there playing i edited the way the HPs move in random mode.  Previously, they picked one of 5 points, one of the 3 HPs, then flashed and moved it.  But this always resulted in the max speed of the servo and only a single movement.  I wanted more life in it.  After a little thought, i found a clever little trick.  The Slow Servo library uses the start and end and time for the travel.  But it doesnt know the start, so you have to give it that value.  So by picking one of the 5 spots for the start and one of the 5 for the finish, the servo has to jump onto that path and then will move at the speed depending on travel distance because the time for the motion is set to the same for all random HP moves.  So the original code was only changed by one random and 2 lines, but the effect is so different. :)

Next up i need to get the Vegas kit done for X2.  But i might go back to BHD and apply the Diamond drive, de-buffered PS2 pad code and the servo relay removals. 

Drive challenges and tweaks

X2 is now all wired up and working.  But here is where the fun began. :D

I had a funny noise coming from the drive.  It haddn't happened during testing.  But i had bought a 2x32 Sabetertooth to run it, and eventually i found it was the ESC settings, limiting the current.  So that seemed to solve it.  But no, after a few mins running the problem would return.  It didn't take long to work out the problems, but i had already started down a side mission.

The mix from the stick to the motor speeds on BHD has always been my way,  Basically, adding and removing the X value from centre, from the Y value from centre.  This had a few key points.
1.  Turning was able to be factored so that drive and turn felt very different.
2.  At max foot speed on one motor you cant add to it, so max range driving differs to in range.

So after a little researching and a lot of simple maths it all came down to a few lines. This was the inspiration, and the base for hte method:
http://www.rchordata.com/files/Using%20Diamond%20Coordinates%20to%20Power%20a%20Differential%20Drive.pdf

This one has the following:
1.  Turn and drive are equal, so full spin is full forward and full back on the motors.
2.  This map is true in that it adjusts both motors for better control, especially at top end.

Map stick input to easy grid -100 to 100 in both axis, including deadzones.
Then constrain the coordinates into a diamond shape.  Using 2 line equations using coordinate geometry to find the intersect, boiled down to the minimum equation.
So if (YDist>(XDist+100)) {  //  if outside top left.  equation of line is y=x+100, so if y > x+100 then it is above line.
OK, the first fun bit. :)  so for the 2 lines this is always true y = m1*x + b1 and y = m2*x - b2
y - y = m1*x + b1  - m2*x - b2  or 0 = (m1 - m2)*x + b1 - b2
We have y = x+100 and y = ((change in y)/Change in x))x
So:   x = -100/(1-(change in y)/Change in x)) and using y = x+100 we can find y with the new x
or:
            XDist = -100/(1-(YDist/XDist));
            YDist = XDist+100;
          } else if (YDist>(100-XDist)) {  //  if outside top right
            // repeat intesection for y = 100 - x
            XDist = -100/(-1-(YDist/XDist));
            YDist = -XDist+100;
          } else if (YDist<(-XDist-100)) {  //  if outside bottom left
            // repeat intesection for y = -x - 100
            XDist = 100/(-1-(YDist/XDist));
            YDist = -XDist-100;
          } else if (YDist<(XDist-100)) {  //  if outside bottom right
            // repeat intesection for y = x - 100
            XDist = 100/(1-(YDist/XDist));
            YDist = XDist-100;
          }
So all coordinates are now in the diamond.
Then we need to map a diagonal 100x100 grid onto the diamond and find the motor speeds.
For the left motor, send ray to y = x + 100 from coordinates along y = -x + b
Find for b, solve for coordinates and resut in y then scale using y = (y - max/2)*2
          int LeftSpeed = ((XDist+YDist-100)/2)+100;
          LeftSpeed = (LeftSpeed-50)*2;
For right send ray to y = -x + Max from coordinates along y = x + b find intersction coordinates and then use the Y vaule and scale.
          int RightSpeed = ((YDist-XDist-100)/2)+100;
          RightSpeed = (RightSpeed-50)*2;
This results in a -100 to 100 range of speeds, so map to usable range.
          LeftSpeed = (map(LeftSpeed, -100, 100, FootMaxReverse, FootMaxForward));
          RightSpeed = (map(RightSpeed, -100, 100, FootMaxReverse, FootMaxForward));

Easy and a completely new mix. :D  Works great, and feels a lot more controllable, especially with the rework on the PS2 buffering issues.  The mix is great.

But that wasnt what i wanted.  I was going to look for a way of driving forwards or backwards on both motors only.  But hey ho. :D  Will try again.

Chis James was in the UK so popped in.  He spotted the supply lead was getting warm, so we upgraded that and it was like a different droid. :D  But then started throwing tracks. :(  And popping wheelies as soon as we added the dome.

The wheelies were a balance and shocks issue.  The suspension is too soft in front when fully loaded, so it is stored energy into the wheelie.  So i decided, for now, to cover the shaft of the damper with rubber washers, so there is some give, but not as much and can run loaded.  Also i looked at the setup and the front height has crept up with the changes, so that was about 15mm too high, so lowered the ride and the adjusted spring and it was a lot better, but still had a tenancy to pop.  So i added about a kilo of Lead to the front of the foot and it seems much more stable.

The shock shoulders rubber bushes have worn already and the holes are now sloppy, they were the uber soft ones, so the 72A are out, and 89A are next. :D  On order. :)  That will help with the rigidity and the wheelies too.

Eventually i spotted it was always the same track when it threw one.  So James and i striped them, got the adjuster moving and set them up properly.  The difference was brilliant.  He now skips over the grass rather than the tracks gripping and twisting, they turn with the feet a lot more than dragging.

I am going to redo the centre, the caster has a lot of slop in the bearing and the noise is pants.

So after a lot more playing, i now have him driving pretty well, and under better control than BHD. :)  The fast and slow modes are setup for in and outdoor use and should work well. :)  I eventually went with the ramping via BHD and a linear map on the drive.  That gave me the best adjustable control via the pad.  But i now have all the failsafes, voltage levels and battery types etc set on the Sabertooth.

The mancave is the ultimate droid tight spaces challenge,  :D   He is now easy to drive even in the tight walkway past the bench, where there is only about 5" clearance total.  I couldn't even get him lined up for it at first. :)

Dome lights and more body work

Steady progress still.

I have wired up and neatened all the 12v sources in the dome and as much of the servo leads as i can.  But to be able to continue on the dome I'm now waiting for servos that should be here by now. :(  I have had him on for the last few nights.  Really pleased with the colour choices.

I have Seglus's Vegas PSI (Mk1) and JoyMonkey's R-Series logics (Mk1).  I didnt fancy having to write code for these, so i decided to use the Rear PSI on the front and the front on the rear, in R2-D2 colours.  With that choice made James and i had a play with the R-Series colours to get something that looked right.

Fronts have a Green/Orange fade with white spots.  To try and mirror the Green/Orange PSI.  Not a great vid, LEDs dont like being filmed. :(

Then the rears are Blue/Purple and Red.

I think it suits him.  The lights are off, or not installed on the screenshots, so i figure this gives us free reign.

The diffusion on the PSI is a trial and error thing i did a while back.  It is a 1mm layer of plastic from a Hobbytronics delivery box, then OHP inkjet film with the honeycomb print, then 10mm (2 x 5mm) light box perspex. Boards sit on the back, spaced off  with, and held in to the silicone hose (my PSI holders) by large O rings, smaller diameter between pcb and perspex and thicker to retain it. Looks great in person, but tricky to film.

I then mounted the servo for the big door.  Using a modified version of the same servo holders i had used in the dome.

After that i finished wiring up the UAC servos to a communal point.  And glued it in place using HP40.  Seems nice and solid, and thanks to a little shaping and the pull of the glue it all lines up well.  A touch up of the paint and it looks like it was always there. :)

CBI door servo and PCB next.  Then id like to look at the front foot one more time before testing.  I don think the positions are optimal at the moment.  Maybe time to start modelling that in Cubify.

Utility Arm Carrier and Arms

The Master setup has already failed.  The dremel shaft eventually failed at too low force. :(  But only in the anticlockwise direction. :(  So I have ordered a alternative to try.  I also found that i can use a bearing in the arm, that may help too. :)

I had built the UAC to replicate the JAG V5 UAC points that i needed, but from CNC cut styrene.  It took a few goes to get right. I glued the carrier up with the arms installed without spacers, but due to the amount of slop the first 2 attempts were binned, so i replaced the bearings, and used a nice flange bearing instead.  

These are a few pics of my final prototype:

I then redid in black and added an LDP / Vegas lights strip carrier. So all glued up and skins added.  I removed the arms and painted.

The complete reassembly was a pain.  Having to thin the spacers to the right heights, have all elements aligned to be able to do up the all the grub screws, and insert the spacers, and the belts, and the servo, all inserted in the right order, the right way up.  So all in, but scratched all over. :(  And not clearing nicely, and rather strangely, sitting differently, so a little filling later they were opening and closing pretty well, paint was touched in, and all working well.  Good enough for this job. :)

And the Vegas lights should be better concealed on X2 :D

This is probably easier on an Ali frame, but without being able to strip the UAC the install was very tricky and tight.  Now it is installed, the Arms open beautifully and the mechanics of the mech are superb. 

I spotted i had missed the front power coupler.  So masked and painted that too.  So the whole droid is now externally done. :)  

James is working on the dome motor hub for the A&A gear and the pitman i mounted in the top ring.  I hope he will update you on that when complete. :)

Not too much more to go. :D

Master Utility Arms

These have been bugging me for a while.  I ordered the Servo City shaft adaptors and they took nearly a month to come.  I had already bought the recommended HS-65MG servos so finally once i had the dome done, i could crack on.

I had broken one of the Dremel hand shafts not long ago.  So i got the new one running and cut up the old one.  The centre was just about perfect for the flex shaft in the arm.  So I cut 2 35mm sections and cleaned up the ends.  Then drilled out the brass couplers provided and the servo adaptors to take the 3mm shaft.

The brass bits were great, so easy.  The servo couplers were a nightmare.  The Allen key provided didn't fit the grub screws and the screws didn't fit the adaptor.  So i ended up just gluing the shaft into the adaptor, but the hole is tight and almost threaded, so should hold ok.  But only after a lot of playing.

Grease was needed on everything, it really helped, and also the plastic bush when fully pushed on flared due to the ali of the paw doing so.  So that was trimmed by about 1.5mm in length, with that it rotated quite easily under servo power.  But took a bit of playing to find the right answers.

So with that i painted one arm and sorted the other's mech.  Then painted the 2nd arm. :)

Today i reassembled the painted arm.  The long grub screw for the paw to bush didn't want to go in, so i cut it short and it only grips one side.  It runs beautifully. :) Shame about the wiring colours, i will have to connect these to correct extensions in the body, but it was the only wire of that thickness i had.

2nd one can wait another day. This one is a day older and the clear is still a little soft.

R2-X2's Dome

We have the Crash dome with the carbon fibre panels.  It is a really nice strong piece, beautifully made.  It is very hard to tell it is a fibreglass dome.

But offering up the standard hinges, the results weren't great.  So we decided to use the CNC and try to make some to suit the thicker dome and the lack of inner layer on the doors.

Starting with the point we wanted the pie door to stop, we designed a new style of hinge, with much greater opening and clearance of the opening.  We then did the same for the lower doors.  After doing the 9 doors, but just hinges, i dropped the dome, busted 3 hinges and proved my dome strength theories and my hinge weak point ones too.  So i redesigned the profile to have holes not grooves, Mk2 is SO much stronger.  I then sorted the servo arm and connection and ran a few tests.

Then i got all 4 pies reinstalled and setup the servos, at that point i couldn't resist another video. :)

So 9 servos, hinges and doors sorted.  But i had a carbon top panel, so it would be rude not to add it as a door too, right?  ;)

James cleaned up and assembled the other 2 HP servo mounts and i cut the top mount to allow the top hinge to clear.  I then test fitted all the logics and any other dome bits.  Once happy i started preparing all the elements and getting ready to paint.

Painting took almost a week.  The carbon panels are a lot of work to get a good finish.  Several filler and primer cycles eventually got them to a pretty good standard.  Then i gave the white bits 4 coats of stain, and the green/black 3-4 of colour and at least 2 thick clear coats. 

The dome stripes was the tricky bit.  I mounted the dome to the Rockler and made a bracket to hold the pencil.  Then slowly rotated the dome marking as i went.  Then using flexible masking tape i masked all the bits to be painted green, filled in the rest with blue masking tape and set to it.

Painting went well, stripes are great, but the blue masking leaked.  So I had to clean quite a bit of the white, but luckily you would have a hard time spotting the fix.  The rear PSI area could be better, but im happy with it, overall it is really crisp and clean.

So here he is, all elements now glued in and ready for the wire up and electronics, :)

Body and UAC work - Delayed update

This is all work i did a while ago, but forgot to update the blog!  :D

I moved the pivot points on the front suspension, it did the opposite of what i expected. :(  The change in wheel size and addition of the caster has altered the physics. :D  So the caster needs to be moved forward and i think the shock needs to go back to where it was.  But currently he isnt riding like he should.

So i stripped the ankle, remounted the caster and lowered the mech.  It is riding a little high, but it is so much better.  New legs for the caster and a slight height adjustment should get the ride height about right and give a little more firmness to the shock effect.  But we might test with it as it is.

Also need to rework the drive mounts a touch, half the bolts have now threaded on the small depth plate, may look at adding some inserts.  But that can wait.

Back on with the main build and continued working up.  I cut some styrofoam and made a battery holder, and bonded that in.

I then added the speaker back mounts and velcros for the Boom Mat covers.


Then i did the volume knob and octo port.  Same setup as BHD.  But all bonded into place. :D

I was then thinking about the electronics.  I didnt want to add too much weight, but wanted it neat and easy to work on.  So i decided to mount all the electronics into a box.  This may change, but it will be ok for now.

I might redo this.

The UAC (Utility Arm Carrier) has been bothering me.  I have the RotoPod mechs and Master arms, with the twist paws.  So it had to be set well and be a reliable set up.  The mechs are designed for the COM8 frames, so i borrowed the JAG plans and manually transferred them to Cubify.

I cut out the old one. and enough of the top ring to allow the new one to be fitted.

After a few test runs i ended up with a nice solid frame and mount.  But the bearing were nasty, so i ordered some more, they rob a little space, but still fit fine, it is a little tighter.  The assembly isnt the easiest, but is possible. :)

I glued it all with the arms in place, so they sit pretty straight, there maybe a small amount of adjustment of the skins needed but dry for they seem fine without paint.  Im yet to sort the spacer heights, so fun still to be had, but I haven't looked at the master arms yet so will wait to bond it to the droid.  I hope all 4 servos and the LDP lighting will be tested before it goes in forever.

This is the last prototype, before i redid it in black. :)

I had the CBI panel in black, but added top and bottom parts to make a shallow pan for it.  The panel sits about 1-2mmm from the edges and so it quite prominent in the space.  I like it. :)

For the big door i wanted to leave the options open.  I added a pan surround and bonded it in, then made my own blank SK10 panel surround and a back plate.  I did a horrible job on the glue, so painted it, and wedged it into the back.  I will replace this once i know what im doing with the touch screen pad.  Again my gluing was horrible on the surround.  But i found some carbon  fibre vinyl and covered the mistake quite nicely.

So body almost done.  Some electronics and some testing then we can bond in the UAC and the top ring and get testing. :D  There has been more weight added than id like, but hey ho. ;)

I really don't like painting.

I had avoided painting BHD, but there was no getting round it this time.

James and I had already painted the body white.  And James had done a great job. :(  But we decided against R2-D2 so it had to be redone.

R2-X2 was the chosen droid.  So i went to halfords with our chosen paint code, only to find it was nearly 3 times the price to specify the code!  They have a stock range too, they were much more reasonable, so after a while pondering different options i decided on the Rover Brooklands Green and a Mercedes Pearl Metalic Black.  Not a bad match in the shadows. :)

I started by getting Colin over.  He gave me a painting 101 lesson and left me to it. I started with the feet and legs.  Over about 3 weeks i have been slowly prepping and painting parts.  Some parts took 4 or 5 cycles of sanding filling and priming to get right, but it was worth the effort. :D

I managed to clear Oxfordshire of that colour in that time, James even cleared out the Reading store.

The paint finish is not mirror smooth, i have some minor clear wrinkling and the can fresh orange peel is still there, but he is fine for our needs.  The body had major issues first time, so i sanded the lots and started fresh, 2nd time is much better.  Im still learning. :D

I got all the legs painted up first, so they could harden for a little before assembly.  I spent quite a while before painting making sure everything would easily fit when painted.  It paid off, they slipped together beautifully. :)

Then James came over and worked his magic with the airbrush on the Horseshoes and the shim details, to make the right bits silver. :)

The horseshoes and boosters currently click on with magnets.  For an event droid this would be sound, but i think i will pop some silicone in the holes too eventually, just to give them a little more grip for off road use. :D

  

The leg to body hubs are 3D printed, so i covered those with brushed aluminium vinyl, added a small strip of that for the leg groove and the legs are 90% done.  Just the USD to finish and then glue the last bits in place at the end.

I ordered some nice black hardware for the leg to ankle and battery box hatches.  Loads better. ;)

There were a few masking dramas and pulled paint, so some parts were re-cleared but it's hard to spot and i can always touch it up later.

So with the legs painted the body got done.  Happily most of the bits needed arrived over the last few weeks, so i was ready to install a lot of the body bits.

The pocket vent holes were not the right radius, so i had to remove the surround top and bottom on the body so it could sit in right.  The back is perfect, the front does poke out a little.  But good enough.  So with pocket vents done i dry fitted power couplers, coin returns and octos.  Then looked into the side vents and speakers.

I bought some 100mm speaker rings in 18mm mdf.  I shaped these to fit the side vent backs.  Then marked and cut the holes in the back to expose the speaker through it.  Using 8 CS wood screws and a ton of powerflex superglue i screwed them together.  Then filled any gaps and filed and painted them in black.  The speakers will mount directly to the backs. :D

I then opened the holes to fit and moved on to the doors. :D  

The big door was the challenge here.  I had cut the door out assuming we would make a door.  So the cut-out wasn't great.  But needs must and it was the best i had.  I squared it up and fitted the trim with lots of glue, then flattened the back as best i could with sand paper.  Eventually i glued on a 3mm styrene back and filled and primed it all. :D  I think i will paint the back black along with the hinge. :D

  

The centre vents almost friction fit. :D  So easy job there. Then i ground out the knob of the octo port as i had for BHD; that will be the volume.  I hated attacking Dave Shaw's amazing part, but needs must.  Once all dry fitted i painted the required and started bonding them in with HP-40. Should take the bumps without being totally rigid.  Fingers crossed.

  

 

I am waiting for my Coin Slots (in the post) and i need to look at mounting and setting up those master utility arms and mech from IA parts.  And hinges need mounting.  But good progress.  Almost time for the dome.  That is it for now.  I may slow for a few weeks and look at BHD.  But X2 is really coming to life and almost able to undergo the next round of testing. :D