Quik links The BRAK Project What's new Project background Design info Student projects Photo gallery Vendor info Related projects Relevant links Site map Project timeline The Next Generation Brak E-mail Engineering Pacific Page maintained by khughes@uop.edu

Mobile Base Design Details

List of drawings

mbase1a: The bottom plate: this is designed to surround the wheel housings and also support some of the DC batteries.
mbase2a: The middle plate: this is the main load-bearing plate; it rests (via thrust bearings) on the three wheel housings and supports the bottom of the steering shafts.
mbase3a: The top plate, detail 1: this plate supports the drive and steering motors as well as the top of the steering shafts.
mbase3b: The top plate, detail 2: the remaining measurements for the plate.
mbase4a: The shaft assemblies: there are three steering shafts and three drive shafts. The shafts are concentric -- one inside the other -- and separated by roller bearings.
mbase5a: The wheel assemblies.
mbase6d: A side view of the assembled mobile base.
mbase7a: The platform plate (I got off in my naming scheme; sorry); this is the plate which supports the computing platform.
mbase8: The top plate of the computing platform.

Construction specifics

• mbase2: The center opening is important primarily in that the two DC motors extend through it. It was designed with the large hexagonal opening when we determined that the plate would be made from 3/16'' aluminum, which would be stiff enough to not need any central mass. The most difficult thing for us in machining this plate, as well as plate mbase3, was boring the opening for the three steering bearings (F-600-3). The outer diameter of these bearings was 1/8" larger than our cutting equipment and so these had to be bored to their final size by hand. If you have to do the same it is critical that the center of these bearings not be offset. Too much deviation will result in the bearings of the two plates misaligning. We missed by a little and while it has yet to prove to be a problem in operation, it did force us to use a larger cable chain to connect the steering shafts.

  The stand-off holes in the plate were drilled to take a 6-32 screw. We chose not to tap the holes, but that's up to you.

• mbase3: The openings in this plate for the two DC motors are designed, if everything goes correctly, such that the tension of the cable chain (14CCF-90E) between the motors and the primary shafts can be adjusted by pivoting the motor about one of the mounting screws. In practice this has turned out to be a good and bad idea. The good news is the motors can be held in place using only a finger tightening of the screws. The bad news is that the motors are placed too close to one another and when in their proper position overlap. We were able to remedy this problem by filing down the motor mounting flanges. There is sufficient room on the plate to reposition the motors so they clear one another, and when time permits we will redraw the plans to correct this.

  A second item to note is the placement of mounting holes for idler sprocket (14LC28A-13) mounts. We chose a standard length for the two main cable chains (14CCF-270E) and our calculations showed it would be slightly loose when placed around the three shafts. Two sets of holes were placed on the platform to hold a stand-off upon which we would mount the 13-tooth idler sprocket. After assembly we found that the main transmission chain was sufficiently tight while the main steering chain was actually too tight; as such the the idlers were unnecessary. Depending on how your machining of the shaft bearing holes turns out (see the mbase2 comments regarding this) you may or may not need them.

  As in mbase2, the stand-off holes in the plate were drilled to take a 6-32 screw. We used threaded aluminum spacers (one 1 1/2" (J218-ND) and one 1/2" (J178-ND) for each of the 12 stand-offs) to separate plates mbase2 and mbase3.

  Since mbase2 was made from 3/16" aluminum, we decided to use only 1/8" aluminum for this plate. Again, it seems from experiment that we didn't under-design the platform.

• mbase4a: This was the trickiest part of the whole design (so far). For me (personally) it was also the most expensive as we had to have the local machine company (yes, our small city may have only one -- I didn't spend that many days searching) make these parts. We chose to use aluminum for the steering shafts while using steel for the transmission shafts. The real key thing for these shafts is that in our design the transmission shafts support no weight; therefore they are held in place relative to all the other pieces by the retaining rings (Q1-31) on each end. These rings rest against the roller bearings (NK 8/12 TN) inside the steering shafts, so it is critical that there be as little play as possible of these shafts. We obtained very good results with little difficulty.

  The overall shape of the steering shafts may or may not be as critical, depending on whether you use our design or come up with a better one (and if you do, please send it to me so I can incorporate it into these plans). What I'm referring to is the central portion of the shaft, the bulge in the center. One problem of the design for which I didn't have a solution was how to attach the wheel housings to the bottom of the steering shaft. I wanted some way to attach these which wouldn't be permanent. One idea was to have the top opening of the wheel housing "keyed" to slip into a notch in the shaft bottom's side wall and then held in place with a nut or pin. When the shaft was made, I was told that due to the thinness of the side wall this would be risky. Instead it was decided to try to use the tension of a fastening nut on the bottom of the shaft, pressing against the wheel housing plate and thrust bearings (B5-10), to hold the wheel housing in place. So far this seems to work, but honestly at this point we haven't been able to test it much.

  Sorry, now how does that little story relate to the shape of the shaft? It is now very important to have a flange on the shaft to rest against the steering shaft bearing in mbase2 since it is really this flange and the fastening nut which are squeezing the thrust bearing (B5-10) and steering bearing (F-600-3) between them to hold on the wheel housing. If we think of a better design (or if you do) then the flange may not be as important. It is also important to make sure the flange is not round but has some surfaces which can be gripped by a wrench, since you need to hold it still while tightening the fastening nut.

  This final point is actually something I discovered was important after having the shafts machined. Be sure to have the face of the shafts flattened where the set screws of the sprockets and collars will be tightened. You don't want to tighten them down and marr the smooth surface of the shaft over which you want the sprockets and collars to slide. Really, you don't want to do this.

• mbase5a: Each is made from one top piece and two side pieces, fastened however you want. I had originally planned to use an a piece of angled steel bolted together, but someone suggested welding them and this worked fine. Be sure that the holes for the wheel shaft bearing (F-165) are aligned properly. We used 1/8" aluminum on these (again, maybe overkill but it does work). The wheels were machined from a single piece of aluminum with a set screw in the collar to hold the wheel in place.

• mbase1a: The bottom plate: this is designed to surround the wheel housings and also support some of the DC batteries. We haven't made this part yet as we don't have the DC-DC power supply working (so we don't have the batteries and don't need support for them yet). We plan to make it from plastic and mount it to mbase2 using 12 3" spacers (two 1-1/2" aluminum spacers (J218-ND) end-to-end).

• mbase7a: The platform plate is still being designed. It will also probably be made from plastic and supported by 2-1/2" spacers. The critical part in its design will be that it is easily detachable from the computing platform.

• Miscellaneous: there are some subtle, tricky, or otherwise worth-noting points to make relating to the machining or assembly of other pieces.
  • The 60-tooth chain sprockets (14LP5A-60) we chose have no hubs and must be mounted on something which can secure them to the shafts. Our transmission shafts are 5/16" so it was easy to find hubs (PH4-61) for them, but since our steering shafts are 3/4" we could find no hubs that large (without them being just huge); instead we used 3/4" bore collar clamps (CS-57). For both of these you have to be careful when attaching the hub/clamp to the sprocket or you will end up with a wobble.
  • This point bears repeating: be sure to have the face of the shafts flattened where the set screws of the hubs and collars will be tightened. You don't want to tighten them down and marr the smooth surface of the shaft over which you want the hubs and collars to slide. Really, you don't want to do this.
  • The thrust bearings have a 1" inner diameter while the steering shafts have a 3/4" outer diameter. I chose to do this since the wheel is not centered below this shaft but 1" off to the side. This does leave the problem of how to center the bearing about the shaft. The easiest solution for us was to use a small piece of 3/4' ID PVC pipe which is sanded down to fit inside the bearing. It's important that the fit not be snug or else the bearing can't do its job.
  • Some of the aluminum spacers are female on both ends and need inserts to attach to one another. We cut the heads off 6-32 screws and epoxyed them into one end of the female/female spacers to fix this problem. There are other solutions to this problem but this way we don't have to worry about losing the inserts and paid less than buying all male/female spacers.
  • If you haven't done so already, you should read the section on the "steer drive" phenomenon in the base theory page. If you're determined never to read that, then all I can tell you is don't be surprised.

Parts list

Aluminum part drawings

• Adobe PDF format
• Postscript