Introduction

Ever since I bought the Evolution 3200 mower, I've been wishing that I could raise the cutting height higher than the robot allows.  I've always preferred my grass to be on the higher side.  So, after two seasons of robotic mowing, I decided to take the plunge and try the modifications.  I had read up on the subject on the Lawnbott Forum on www.bamabots.com.  Another owner had tried a modification similar to what I wanted to do but ran into trouble with the gear ratio's and burning the motors out.  So, I knew what not to try after digesting all that information.

My goals for the project were as follows:

1. Keep the rear wheels in the same spot on the mower with same width but mount them lower so that the whole robot is raised up
2. Keep the front wheels in the same exact spot but just raise them up the same height as the rear wheels
3. Keep the front and rear wheels the same diameter
4. Use a bearing mechanism with very little friction to not put any additional stress on the motors
5. Use a bearing mechanism that would require little to no maintenance or lubrication
6. Make as few permanent modifications to the robot and parts as possible so I can easily put everything back should something not work
7. Learn some things about machining metal and have some fun doing it

There were multiple times during this project that I thought about having someone else machine a part for me or produce a shaft, however, I had so much fun figuring out what to do and how to do it myself that I ended up doing everything.  At the end of the project I also wanted the satisfaction of being able to say that I had done all of the work associated with the job.

Also, I am not an engineer nor did I have any experience machining metal to any extent prior to taking on this modification.  I am good at coming up with ideas and I am creative at getting things done even if I don't have the best tools at my disposal.  For example, I could have used a small lathe for several steps during this project, but I decided that it was too big of an investment especially since I would not have much use for it after I completed this project.  I worked around this and came up with other ways to accomplish the things I needed to do.  Since none of the parts I had to produce were being used in a high rpm, low tolerance type application I didn't have to be 100% perfect on every part.  If I did need that precision then a lathe would have been absolutely required.

Tools and Hardware

I'll have to admit I had a lot of fun buying some tools I did not have in order to complete this project.  I ended up getting the smallest bench top drill press from Harbor Freight.  I used it for several weeks and then realized that I needed a bigger model in order to hold the drill/mill vise that I needed to use.  I ended up going with the heavy duty 3/4 HP 13" drill press from Harbor Freight and that model has worked extremely well for what I needed to do.  Other tools I picked up were an automatic center punch, a centering device to mark the center on round shafts, a circular saw with a worm drive, 6" digital caliper which measures in mm and inches, a rectangle block collet holder, a 5/16" collet and a 12mm collet to hold shafts while I drilled or cut them, HSS Cobalt 135 degree split point drill bit set, Silver and Deming HSS drill bit set for larger sizes, HSS metric drill bit set, a dremel drill stand, dremel thin metal cutoff disks, a carbide tipped metal cutting blade for the new circular saw to cut titanium, a diamond bit set for the dremel, a mill/drill vise with X/Y axis, clamps for holding down large pieces on the drill table, a full face guard. safety glasses,  a hardened steel countersink bit for countersinking titanium, , a tap and die set (inch and metric), a Drill Doctor drill bit sharpener (750X), and metric and inch hex wrenches.

I used some tools I already had including an XPR 400 dremel (I also borrowed my brother in law's dremel so I could more easily switch between tasks, levels, among others.

The hardware and materials I bought were the following:

- 12" x 10" x .22" titanium plate (6AL-4V alloy) which I cut into two equal pieces.  I ordered this from www.titaniumjoe.com.

- Metal spur gears (multiple types but all same size, # of teeth and pressure angle) from www.sdp-si.com

- Teflon bronze flange mount bearings from www.spyraflo.com

- Many stainless steel metric and inch bolts, screws, and nuts from Ace Hardware

- 12mm shaft (13" in length) for new front wheel shaft. I measured the existing shaft and added 49mm for the new shaft.  Purchased from www.vxb.com.

- 5/16" bearing shaft (36" in length) from www.speedymetals.com

Initial Design

 I talked through several design ideas with my dad and my brother-in-law.  The options we discussed were a chain drive system, a friction drive system, and a spur gear system.  The biggest things in my mind were to select a drive mechanism that was easy to maintain, wouldn't get clogged with grass clippings, didn't need to be covered up, didn't need to be lubricated to any extent, and was very sturdy.  After drawing up several designs on paper, I decided to go with the spur gear system.  The plan was to attach a spur gear to the existing motor shaft and then a gear to the wheel with a third gear inbetween these gears to allow the wheel to rotate the same direction as the motor shaft.  The motor shaft gear and the wheel gear would not touch with the middle gear transferring drive power from motor gear to the wheel gear. I did not use any CAD software to do this design.  I didn't want to spend the time to learn the software, but looking back this might have been a good idea to do.  My approach was basically to execute my plan step by step and adjust the parts as needed as I went.  I purchased hardware as I went and did quite a bit of research online to find what I needed.  Initially I was thinking I would have to custom design the gears, but I ended up finding the sdp-si site that had gears that would work for me in stock ready to ship. I needed some type of metal plate to be able to bolt on to the existing motor mount and also to support the new axles and bearings on each side.  I ended up deciding to go with titanium.  First of all it sounded cool to use, a challenge to machine, and I had never ordered it before.  I found a site online (www.titaniumjoe.com) who deals in scrap titanium plates and rounds.  I went back and forth with them several times and finally settled on a piece that I thought would work.

Working with Titanium

I spent quite a bit of time reading up on how to drill and cut titanium and how to live through it.  I have to admit I was a bit concerned about this step.  I needed to cut the titanium plate I bought in half and I needed to drill many holes in each plate.  One post I read was quite funny - it said while cutting titanium to wear many layers to protect your skin from red hot titanium shards, wear multiple face and eye protective layers, or better yet, have someone you don't like do it for you and you stand 40 feet away when they do it :).  One post I read on drilling titanium said that if you have the right drill bits you can drill through it like butter.  All info that I read said low RPM, high torque and constant feed rate were essential to getting through the material successfully and without damaging the tools, blades, or drill bits.

So, I followed all this advice (except for the part about having another person do it of course).  I purchased a circular saw with a worm drive and a metal cutting blade with carbide tips.  The worm drive produces lower RPM's but very high torque.  Man, you can really feel the power of that thing when you turn it on.  I wore a ski jacket with a hood, eye protection, and gloves while cutting the titanium.  It was extremely loud and shot sparks which were actually small flaming pieces of titanium for a good distance in my garage.  I went slow and steady and used lots of cutting oil and in about 25 - 30 minutes, I had cut through about 6" of the material - SUCCESS!  I now had my two plates that I would use as supports for the new axle shafts.

I used the motor mount plates on the robot to trace the mount screw holes to the titanium plates.  I then drilled those holes out and tested the fitting to make sure they fit well.  The person who said that you could drill through titanium plate like butter was correct and I was more than pleasantly surprised.  You definitely need drill bits with a 135 degree angle and high speed steel cobalt bits worked the best for me.  The bits up to .5" cut through the metal with ease and produced long strands of thin titanium as they cut through.  I had to also cut holes up to .75" in diameter straight through which presented a challenge.  The good drill bit set I bought only went up to .5" so I had to order larger bits online since I couldn't find what I wanted in the local hardware stores.  I ordered a HSS Co .75" bit and it worked fairly well.  It didn't cut as fast or as easy as the other smaller bits, but it did the job.  I did find that it dulled out pretty quickly so I bought a drill bit sharpener which worked very well (Drill Doctor 750x).  Sharp bits are a must!!  I did eventually wear that .75" bit out - got tired of sharpening it very frequently, so I bought a set of Silver and Deming bits from Harbor Freight.  They were HSS and went up to 1".  I didn't have high hopes for them working well, but to my surprise the .75" bit in this set actually cut through the material easier than the HSS Co bit I had used earlier.

When I started drilling the titanium plates, I was using a small bench drill press from Harbor Freight. The drill worked ok, but the drill table was so small I couldn't attach a decent size drill vise and have it be steady and level.  I even tried taking off the drill table and using the base to secure a vise, however, the column would tend to flex when I applied pressure to get through the material.  Most likely I was using bits that were not sharp enough causing me to use excessive pressure to cut through the material.  I also did not like the ease of accuracy with drilling holes in the spots I wanted them in.  I tried a laser device to center the bit and that worked out pretty well.  In the end I still felt constrained with this bench drill.  I decided to take the plunge and go for a larger bench drill press.  I went with the 13" heavy duty 3/4 HP model from Harbor Freight.  I also wanted to be able to move pieces around with more accuracy on the drill table so I could drill in the exact spots I wanted to.  I purchased a mill/drill vice from Harbor Freight and this tool has been great!  I can make very fine adjustments to ensure I am drilling on center.

In order to drill the holes for the two new axles, I measured 49mm from the center of motor shaft hole straight down.  I then measured 45mm from each of those holes to find the center of the hole for the axle and gear that would allow the new wheel location to spin the same direction as the motor shaft.  After all those holes were drilled, I put the bearings on the titanium plate and marked the holes to secure the bearings. In order to give myself as much room as possible, I also countersunk all the holes that were going to secure the bearings.  That way I'd have nothing sticking past the face of the titanium plate.  I found a countersink bit at www.wlfuller.com - high alloy steel.  They didn't have the exact size I needed which was 10mm, so I ordered a size bigger and just didn't let the bit go all the way in.

Front Axles

I took apart one of the front wheel casters and discovered that the existing axle shaft was supported by 2 12mm ball bearings, one on each end of the Lawnbott chassis.  There was a rubber stop that fit against the bottom ball bearing and against the bottom of the shaft.  The shaft also had an e-clip at the top which held the shaft in place.  After some thought, I decided the easiest way to increase the height of the shaft would be to cut off the original shaft leaving 3mm of the shaft on the caster to support my new shaft.  I bought a 12mm bearing shaft from www.vxb.com and cut it to the length of the original shaft minus the 3mm I left on the caster, plus 49mm.  Since the ball bearings are 12mm, I needed to take some material off of the new shafts so they would fit into the bearings easily.  Since I didn't have a lathe, I put the new shafts in my bench drill press and used fine sandpaper to sand the shafts down.  It took about 5 minutes per shaft and they fit easily through the bearings.  I also had to replicate the scoring of the shaft at the top for the e-clip so I could secure the new shafts in place.  A lathe would have been perfect for this, but since I didn't have one, I rigged up a way to do this with my XPR400 Dremel with a Dremel drill press attachment.  I secured the Dremel drill press attachment to my workbench next to my vise.  I then attached the Dremel right angle attachment.  In order to do this I had to modify the drill press stand slightly to allow the right angle attachment to secure to the XPR400.  I removed a spacer from the drill stand and then used 4 large metal washers to set the dremel the right distance away from the right angle attachment so when I screwed them together it would be a tight fit.  Without the washers, the Dremel was loose in the drill stand.  Once that was done, I needed something to keep the axle shaft in the same horizontal position while I spun it so I could get a perfect concentric circle around the shaft for the e-clip.  I took two short pieces of 1"x2" wood scrap and secured them at a right angle with screws.  I bored out a hole in the middle of one of the pieces almost all the way down so I could slide it back and forth on the workbench to cut different length shafts.  I'll call this L-shaped wood a shaft stop. To cut out the circle for the e-clip, I placed the collet block with a 12mm collet horizontally in my vice and positioned my shaft stop so that the metal cutting blade on the dremel was in the right position on the shaft.  I left the collet loose so that I could rotate it freely.  Keeping the shaft pressed against my shaft stop, I turned the Dremel on, used one hand to control the depth of the Dremel drill press and the other hand to slowly rotate the shaft in the collet holder.  After several rotations and lowering the drill press slightly each time, I have my circle bored out on the shaft.

Now, on the original shaft, the rubber stop sat against the bottom of the shaft and since I was now increasing the length of the shaft, the rubber stop had nowhere to sit.  In order to give the rubber stop something to rest on, I bought a couple of steel spacer tubes (inner diameter larger than 12mm).  I cut these to the exact lenth that I needed and then fit these over the axle shaft.  Now I had a place for the rubber stop to rest on.

The next step was to secure the new shaft to the caster.  I drilled a 5mm hole in the end of each shaft and then tapped a 6mm thread in each shaft.  I then drilled a 6mm hole through each wheel caster so I could screw the shaft and caster together.  I secured the new shafts to each caster with a 6mm screw.  One didn't have a tight fit so instead of messing with the tap (I had already gone as deep as I could), I just used a washer to allow the screw to tighten all the way.

The new front shafts were now complete!

Gears and Bearings

I spent quite a bit of time researching what bearings and gears to use for the modifications.  Like I mentioned above, I wanted a bearing that required little to no lubrication, could withstand the environment (dirt, grass, wet, cold, hot), very sturdy, and had very little friction.  I just came across the company Spyraflo in my web searches and they seemed to have the solution that best fit my needs.  I bought a couple of different bearing types so I could try them out and evaluate which one I wanted to use.  The bearing types I bought were needle roller and teflon bronze.  There was also a choice of self align bearings which could actually move 5 degrees around to match an off-angle shaft.  The needlle roller bearing requirements stated that the shaft had to be extremely hard but the teflon bronze bearing could tolerate a hard or soft shaft material. The teflon bronze version had a very small tolerance for the diameter of the bearing - it had to be between .3106" and .3115".  In the end I chose to stick with the teflon bronze bearing with a flange mount.  While the needle roller bearing provided the least resistance, it would get a ton of dirt and grass material buildup in the bearing mechanism and probably require cleaning from time to time.  The teflon bronze however, would not get any buildup and was impervious to dirt and dust.

The gears also required quite a bit of research.  At first I was considering having some gears custom-made to the exact specifications that I needed, however, I quickly realized that if I went that route it would be fairly expensive.  After quite a bit of searching online, I came across a company called SDP-SI.  They actually had a ton of gears in stock made from many different materials, both metric and inch, and many different sizes. I decided that a good height to shoot for would be to raise the robot close to 2 inches.  I settled on gears that were 45mm in diameter (including the teeth).  That would mean that I would set the center of the new wheel location about 49mm away from the center of the current motor mount so that the gears would not touch.  49mm is about 1.93".  Another decision was what material to use.  I ended up deciding to go with carbon steel gears and stainless steel gears.  The last decision point was what type of hub to use.  For the gears that would fit over the robots wheels, I needed a gear with no hub.  In fact, I was going to need to bore the gear out to fit over the wheel hub and also over the set screws in order to allow the wheel to sit close enough to the robot. The two other gears in each assembly needed to either use set screws to attach to the shafts or it needed to attach using some other method.  On SDP-SI's website, they were advertising a new shaft-lock design called a Fairloc hub.  This hub has no set screw but a screw that serves to tighten the hub around the axle shaft.  The hub has several cuts through it that allows the screw to tighten and clamp 360 degrees around the shaft.  I was a little skeptical at first but when I got them and tried it, it was perfect!  You have to be very careful to not tighten one of these hubs down without a shaft in it though because you can bend the Fairloc hub and then it is very difficult to insert a shaft again. For the third gear that would sit on the shaft that would allow the new wheel shaft to rotate the same direction as the motor shaft, I decided to go with a regular set screw design.  The Fairloc hub gears were fairly expensive at about $42 a piece so I decided to go a litle less expensive for the others.  So, I ordered gears with a regular hub and one small set screw and decided that I would drill and tap a larger set screw in the hub myself.  In order to get the gear in the material I wanted, I had to go one millimeter down in the gear width than the other gears.  It is identical with 45 teeth, a 20 degree pitch angle, 45 mm face width, but the gear width is 5mm instead of 6mm.  The set screw that came with the gear was 1.34mm which was so small you could barely even see it, so I removed that, drill two 4mm holes and tapped the holes for a 5mm set screw. I now had all the gears set for both gear assemblies. Note that since all the gears are the same size, same number of teeth, same pitch, and same pressure angle, I am not changing the gear ratio of the motors at all.

After assembling the gears, axles, and bearing to the titanium plate and doing some testing with the robot, it was apparent to me that the bearing was not going to withstand the weight nor the back and forth movement of the wheel axle.  Basically what was happening was that the bearing was starting to come loose from the flange mount and was allowing the wheel axle to have way too much play. So, to give more support to the rear wheel axles, I decided to use a double bearing setup.  I used some small scrap pieces of titanium and gave myself enough room so that I could point one bearing through the titanium plate pointing towards the wheel and the other bearing pointing towards the blade. This is a much more stable bearing setup and should provide the necessary support without putting too much stress on the flange mounts. Here are a couple of pictures of the updated bearing mount design.

More to come . . . . . . .