I am starting play with this Ferris wheel design idea with this machine again. I have not looked at this design in a couple of years & want to see if I can get that Ferris wheel idea to work. The outside diameter of the big wheel is 96mm with 46T on the outside & 35T on the inside. The drive gear is 16T & the 2 idler bearing gears are 14T. I am using 4 - 6252RS in each of the idlers. I might could get away with 2 bearings in each, but this will have better support for the big gear with 4 bearings.

I am drawing this design in openscad since fusion 360 no longer allows me to use the gear addon I was using with the personal version. I looked at using FreeCAD to design these, but the inner ring gear might be a bit of hiccup. Here is what the assembly looks like.

 

I ran into a couple of problems with this today. First off, the inner ring gear needed to be opposite direction of outer ring. I had that correct early on, but somewhere along the way I changed it. Anyway, that was an easy fix. I chamfered the outside of the outer ring to cut down on overhang when I get around to printing the wheels. I printed only the part of the plate that had the inner gears for test fitting which showed a couple of things to fix. The drive gear needs to be smaller to get it to fit when putting on the Ferris wheel. I changed that from 16T to 14T, but 12T might be a better choice. There also needs to be an access hole in the Ferris wheel to be able to adjust the bottom 2 M3 motor mount screws. I just drilled one in for testing. Another problem I see is when I attach it to final plate with the wheels, the idler gears & drive gear will need to move the same amount & I don't think a slotted hole for the idler bearings would work. In order to do that, I will need something similar to that test plate added to the outside of the main plate. I will try slots on the test plate 1st but am thinking the bearings might slip in them over time. Not sure how well that will all fit together, but I have not given up yet.

The gearing feels smooth rolling it around. Might make a nice fidget if it doesn't work out for the plotter. Here is what the test print assembly looks like.

 

Probably not. It would be a lot simpler to just drive one wheel with a drive gear & let the other wheel follow. It is more of an exercise to see where this design goes & see if I can make it work.

 

This was starting to give me too much of a headache & was no longer fun, so I have shelved that idea again & will try a direct drive of one wheel with a closed loop belt. I calculated 230mm GT2 closed loop belt using same method I did with the belt before, so it should be correct. If this works, it will cut out the need for the 4 idler pulley assemblies. I will revisit this design when that belt comes in.

 

I ordered 2 more belt sizes, 220mm & 240mm. Looks like I really need 240mm length to mount the motor centered but got a shorter length in case I want to try a shorter belt path. I was using the wrong wheel as a guide, so my 230mm would be too short for where I was putting the motor. Would shortening the belt travel length 10-20mm make much difference in performance or accuracy? Wheels have 88T & I used an earlier version 78T wheel. It is good to have some different sizes to play with anyway. I will have to angle the motor if I use the 220mm length belt unless I change that plate design more. This might be a good opportunity to redraw this plate in FreeCAD. One thing that is a little aggravating in FreeCAD is that all the geometry needs to be tied down. Fusion 360 is more forgiving with this & sometimes quite difficult to get all the geometry tied down in Fusion 360.

This design would allow anyone that built the previous version to use all the previous parts except the 2-wheel plates and cut out all the hardware for the 4 idler bearings. You would also need a different size belt, but that is yet be determined. Here is my 1st draft of using the single wheel drive. Think this is what one or more folks said to try a couple of years ago. Anyway, good to try other ideas. Will be interesting to see if this has better repeatable accuracy than previous version. Looking at a couple of commercial diode laser designs, I see they tie the motors on each side together with a smooth rod. Guess they are only using 1 motor to drive that axis though. I wonder if that would benefit this design? The only way I see to do that is to have a 3d print drive gear with teeth for 2 belts. The other belt would be quite a bit shorter (probably around 100mm length) connected below or above the motor to a geared coupling attached to the rod on each end. I will have to draw something up to see how it looks.

 

I woke up with the idea to use GT2 closed loop belt for the tires instead of O-rings. It seems like those belts would give pretty good grip & I can get 15mm wide belts for the lengths I need. I did a 1st draft of what it might look like to see if it looked reasonable. Probably only need one idler bearing on center above the motor if I close up the motor mount opening at the top. The wheels need some more work before I would print them.
See what you think. I might use the same bearings for the wheels for the idler to give it more support. I currently have the Drive gearing at 88T & the wheel track gearing at 125T which gives me a 79mm outside diameter wheel versus the 59mm diameter before.

 

I redrew the wheels & idler bearing using HTD-5 belt profiles. As you can see from screen shot below, the belt is considerable thicker. I am now using the same 608-2RS bearings in the idler as I am in the wheels. I will test print the geared idler tomorrow to make sure it spins. The teeth are also a lot bigger with the HTD-5. The HTD-5 on the wheels have 38T each & the idler has 22T. I am at around 435mm belt length now & just found a source for that size, so maybe I will not have to change the number of teeth. I was originally looking at 475-500mm long belts.

Here is my latest drawing. It is starting to look reasonable now.

 

Yes, It has total of 6mm of movement up & down. My calculations for 440mm length belt being 439.7mm at center, 436.4mm shortest position & 443.1mm longest position. That should give me enough wiggle room. Thanks for the question as I could have missed that.

Here is a section of it.

 

Not sure I calculated this belt length correctly but can fix it if I am in error by shortening the wheel spacing from 120mm to 110mm. I was using the inside dimension of the belt for the length & looks like it should be the pitch diameter. The length comes out to about 453 using the pitch diameter & my current spacing. I discovered this problem after modeling the 440mm belt length as it fit on this machine.

I initially drew each tooth in a sketch copying them along the straight lines & pulling the arced ones from the other 2 models I had already drawn. The ends did not come out quite right. On my 2nd approach, which was a lot easier, I realized my problem was not using the Pitch diameter. I found the fusion 360 command to copy a body along a path.
The length of the belt along the Pitch length is 439.9897mm. Close as I could get it to 440 by just a little trial and error of adjusting the top idler location.

Here are screen shots showing my method for drawing the belt.
Draw one tooth at the start of the bottom straight line.

Extrude it up 15mm which is the width of my belt.

Create a Pattern along a path (pitch sketch line). I had to play with this command some. Initially setting Distance type to extent & typing in 440. Then change it back to Spacing. The Quantity option did not show up until I did it this way. I used 5mm for distance and 88 for quantity since this has 88 teeth (88 * 5mm =440mm).

I extruded the outer ring of the belt to fill in the gaps of the 88 teeth. I added wider than needed initially because it had some artifacts along the curves otherwise

I then cut the outside of the belt to correct width.

Here is the finished belt

Here is my belt drawn along the pitch line with the dimensions I used. With only a little more trial and error, I changed the idler offset to 56.989 mm which gave me 440.0004mm length.

 

I redid the wheel plate & assembly to use 110mm wheel spacing & looks like it fits together well now.
Here is a section cut showing the belt track with matching gearing.

 

I added a little more character to my wheels by making the cutouts spiral up 72-degrees (360/5). That seems like it should also give the wheel better structure. My timing drive belts came in & the 220mm GT2 belt fits the 20T drive & 80T wheel combination.

I am also looking at adding another idler wheel at the bottom centered between the 2 wheels. That idler like the top one has a slotted hole for the M8 bolt. That might give it some traction by keeping the center bottom of the track in better contact with ground. The outer HTD5 belt should be in later this week. I am not going to print this plate until I test the fit of the HTD5 belt. You might also notice I took the bearings out of the base plate. I am going to see if those bearings are really needed. I designed a HTD5 belt track to print in TPU, but not sure if I will try printing it or not. I added a little bit of tread to the flat outer surface of it.

 

Thanks. That sounds like a good analogy. Think I am also going to keep the bearings in the plate. Seems like it will keep those M8 bolts more perpendicular to the plate with the extra wide metal holding them in place.

I am going to see if I can print a TPU belt today for the tread.

 

I am at 3mm thick out of the 15mm height with the printing of this belt, so looks like it will print. The problem I have found with printing the belt is that it stretches, but this is a different situation than I used it for last time. You can get these HTD5 belts in 20 or 25mm thick. I chose 15mm since it seemed like a lot more surface contact than the 3 o-rings on each wheel I used before. Looks like my belts from China should be here Friday or Saturday.

 

Print came out pretty good for TPU. Since it is dialed in, I started printing another one in case I want to test it before real belt comes in. This took about 2 hours 40 minutes to print at 30 mm/Sec. I have a few blobs on the inside of the belt to clean up a little still. I had to stretch it some to fit since it is essentially 10mm shorter than what I designed for. I did not want to make it the same length since I remembered it stresses a little.

 

Good to know. Perhaps this will be better than the HTD5 belt. I have a yellow TPE from eSun that is a little mushier than the other one I have. The black TPU has a shore hardness of 95A & this yellow one has a shore hardness of 85A. Feels like it might also have better grip, so am printing 2 belt in that also. It is also more colorful with the yellow.

 

I am slowly putting this all together. The Step-down PS for this will be portable using ryobi 18V batteries. I am thinking of putting a switch on this since when I plug the battery in, the display lights up. I will start off using the MKS 8bit board with raspberry pi for wireless for initial testing, since that is how I did it a couple of years ago. I hope to get the DLC32 board working on this which would take the pi out of the mix. the HTD5 belt I ordered is a little loose, so if I use that the idler bearing gear will have to be a little bigger or the wheels. With the DLC32 board, I may be able to define the servo as standard z-axis motor & use more standard gcode.

 

I finally got everything put together for a dry run test today. The wheels are still off after the end of the print. I was hoping they would be more accurate with this belt drive design. I wonder if one double shafted Nema17 motor on one side with an 8mm smooth rod between both sides or use a double shafted motor on each set of wheels with a rod connected between them? Another option would be to put the motors on the outside so the shafts could be linked together, but that would involve redesigning the wheel plates and size of belts. For drawing chalk patterns on the sidewalk, it is probably close enough but would like for it to be somewhat accurate. Any thoughts on this?
Here it is all wired up except the servo is not shown. The wires are not well managed yet. It is really colorful looking.

 

I had not considered the motor mounted above on the end contributing to the problem. That sounds possible. Someone on the v1engineering forum mentioned the treads might be slightly different on one side than the other causing an accumulated error over the duration of the plot. To me that sounds like the most possible problem because this was more off than when I used the O-rings. The O-rings are very consistent, but the TPU printed tread does have quite a few flaws in it. Before I commit to a different change, I have the HTD5 belt which should be on par with the O-rings as far as being symmetrical. I need to run a test plot using TPU tread & then the belt to see what the difference is. There are 2 similar designs on this forum that do not seem to have the problems I have, but there is not much info on how they built them. The wheels for both designs look very similar though, so to me that seems to be the key to making this more accurate. It probably will not be until Monday before I can test this again. If these 15mm wide HTD5 belts work better, but not perfect, the answer might be to use the wider 25mm belts.

 

It runs much better with the HTD5 belts for the tire treads. After running a comparison test between the TPU & HTD5 tread, I decided to bump up the speed to 1st 3000mm/min & then 6000mm/min. I made sure Octoprint had my maximum speed at 6000. It ran just about as accurate at 6000 as it did at 1200. Only a 1-2mm off at the end. The design redraws a couple of circles near the end which shows how close it is. On one of the test runs, it just stopped & I couldn’t figure out why until I looked at the PS display & it was out. I just have to make sure there is fresh battery installed before plotting from now on. One or more folks have mentioned adding some weight to it to give it more grip. I am starting to think that might be a good idea as even the pen is probably giving a little drag on the machine which seems like it could cause it to slip ever so slightly. When using chalk on concrete, that will probably create more drag. I think drag is the right word to use for this situation?

I wanted to see how close to the 6000mm/min or 100mm/sec speed this plot came out to, so I decided to calculate the total movement of the drawing + all the pen up moves and came up with 4315mm. Since the video showed 54 seconds for the plot, that puts it at 79.9mm/sec average speed.

Here is a video of it in action. The outside diameter of the plot is 200mm. I think if it had 25mm wide belt instead of the 15mm it would be even more accurate. I did a video of it running at 1200mm/min

 

Here it is drawing a sandify pattern.

The HTD5 belts could stand to be tightened more, but do not have any more room to adjust them in current design. Maybe that could help the accuracy? Might be better to have a shorter HTD5 belt that just wrapped around the wheels with one idler adjustment pulley between the motor and the wheel that does not have the motor drive belt attached.

I switched from DRV8825 to TMC2208 drivers since the ones I had were drop-in replacements. I had to swap the cable direction & adjust the firmware to use 16 microsteps & readjust the rotation_distance into the 20 range. There is still some rattling noise in the carriage assembly.

I am ready to try chalk next. I have 1st version of chalk mount. Hopefully the chalk will not wear down too much before the end of the drawing. This probably needs a proper Z-axis to be more affective with the chalk.

 

First Chalk test was not very good. Maybe using servo to move chalk down along an arc rather than straight up & down like I saw one other pen plotter do.