Post-015
Feb. 3rd, 2022

Bike upgrades part 2: fixing the seat... with a water-jet.

Yet another indication that something has gone really very wrong.

IMG_2261.HEIC

So today a bunch of parts from McMaster came and it's time — to try to modify the bike seat into something more usable. Before we get into the details of modification itself, we'll talk about what we bought and why and what our concept of this all going together in our head is.

McMaster Parts
1/4"-20 self-tapping screws for aluminum and soft metals
5052 x 1/8" aluminum sheet metal for water-jetting and sand-blasting

AndyMark Robotics Supply
Solf aluminum supporting churro gold color

Amazon
1.25" diameter tube clamping mount
7/8" seat mounting pipe for the seat design below
Large comfortable, retro bike seat

MIT D-Lab
We stole some M5 mounting screws, thanks Jack Whipple! 

So the main idea is that instead of the plastic back clamp that holds the bike seat on our bike, we will replace this with two water-jet plates that attach to the bike frame via clamping mounts that we purchased. The plates will be connected to each other (and support each other) using the churros to bridge them, and the seat mounting pole will also connect to both plates being bolted through the whole assembly. We noticed the tube that mounts the seat is slightly smaller in diameter than the bike frame, so the whole assembly will be a little tapered. This will all start to make sense as we assemble the system (we want to again do very little CAD for this) but first we need to take the seat off.

IMG_2234.HEIC
IMG_2235.HEIC

So here's the obligatory before picture. We removed the three screws that connect the plastic clamp onto the bike frame as well as the screw that holds on the little caddy under the seat. You can see why it was a little flimsy as there's nothing to constraint the rotation except for the pin you see in the second picture and the clamping force of the plastic was not enough. 

The second pictures shows that constraining pin as well as the bare aluminum of the bike's frame. He looks a little naked without the seat so we're going to try to get this done in one day. Bear with me on this, we're going to assemble most of it together without explaining first and then provide a "tour" of the design. 

IMG_2237.HEIC

And of course, here's our starting parts picture with the metal, the churros, the basket (which we will try to keep), the clamps and the new seat on the table in shop. We need to design the water-jet part but first we want to test the clamps that hold the new seat on because if those don't work then we don't really have a design. 

IMG_2238.HEIC
IMG_2239.HEIC

This was a process of trial and error, it turns out the screws included in the kit was just too short to get the clamp around our bike frame which is roughly 1.25" in diameter, but because the rubber needs to compress we need a slightly longer screw. So we went to grab screws from D-Lab and improvise.

Some were too short, some were too long, and then we found a nice goldilocks screw that was just a few mm longer than the one included in the kit. The results of the clamping are in the above pictures. We will have four of these clamps which will be more than strong enough to support our weight.

Screen Shot 2022-02-02 at 9.06.11 PM.png

Because a lot of the design constraints are random in that as long as we have holes to attach the churros and the clamp brackets and seat pipe to we'll be able to hand machine the rest so that everything works, the dimensions of this seat plate were fairly random and so was the whole pattern.

We're going to do the classic water-jet, sand-blast, counter-sink and etc. to get the shape and finish we want. Most of the assembly this time is just kinda coming together in our head but hopefully it'll make sense as we begin assembling it. 

IMG_2241.HEIC
IMG_2240.HEIC

Here are some obligatory water-jetting photos featuring Sydney. The water-jet got a little mad at us during the process but we got the two sides cut out eventually. Things we learned with the water-jet today.

  • If the water-jet stops spitting out Garnet and you have to re-set it, there's actually a feature that allows you to back-track and a tool-path and let you re-cut a messed up section! That was cool to figure out.

IMG_2242.HEIC
IMG_2245.HEIC

The holes that connect to the churros use counter-sunk screws. So before we finish the plate with a sand-blasting job we want to counter-sink those. We marked those out there's like three per side and counter-sunk them by hand with a drill making sure the screws sat flush. 

IMG_2248.HEIC
IMG_2249.HEIC

Sandblasting! We won't talk too much about this because we do it so often that many of the other posts also contain the sandblasting step. This time Sydney did the sandblasting. We just blasted until there was a uniform coating.

IMG_2250.HEIC
IMG_2251.HEIC

Alrighty back to more substance. While Sydney is sandblasting we went ahead and prepared the screws for the clamping mounts. We used lock washers on the screws because that'll just give us a little extra security and a slightly higher pre-load which should help protect against any vibrations which this most certainly will see.

We "soft" mounted the lower three clamp mounts, we aligned the top mount and clamped it down and will align the rest when the sand-blasted "wing" piece shows up. Two clamp mounts bolt to each wing piece and support one half of the seat frame. They both will bolt together in the back and middle through the churros and the main seat mount (the pipe).

As a side note, we just talked to Levy on FSAE as he was leaving and he suggested anodizing the side plates because Boston weather is awful and it will provide corrosion protection and such as aluminum does corrode still though not as much as steel. This may be expensive so we'll think about it but it's a cool suggestion. 

IMG_2253.HEIC
IMG_2252.HEIC

The next step was to bolt the main side plates onto their respective clamping mounts. We used 1/4"-20 hex bolts with lock nuts and washers to secure them to the mounting structure after moving the clamps into the correct position. We purposefully left the clamps loose so we could adjust them later.

Note the direction of the washers! Everyone on the internet seems to be getting this wrong but washers in fact are directional and the flat side goes against the bolt head and the side with a slight curve (you'll notice the difference if you look at the edges of the two sides) goes against the face of the part. This is because sharp corners produce stress concentrations so when the bolt is applying a force to the part when being torqued down, we want the curved side against the part so the stress drops off gradually.

IMG_2254.HEIC

Now we want to put in the vertical bar that holds in the seat. It's an old style seat that clamps to the bar at the top. The rest of the bar will sit between the two plates and this sandwich of plate-bar-plate will be bolted together as a unit.

We took some super rusty bar clamps off the top of the water-jet and used them to hold this together while we hand-drilled the three holes required in the bar to allow us to bolt this all together. 

IMG_2256.HEIC

Then we added the bolts and washers and flex-lock nuts that hold the whole assembly together and while the clamps were still loose, we adjusted the seat so everything lined up with the frame of the bike. Notice the slight triangular/trapezoidal shape of the structure in this top-view like we referred to earlier in the post.

IMG_2258.HEIC
IMG_2257.HEIC
IMG_2260.HEIC

Now time for the seat! The seat goes on simply by tightening two lock-nuts around a clamp that sits on the pipe/bar we mounted on the bike frame. It uses friction to hold the saddle on and an interesting mechanism using two locking washers to constrain the pitch rotation of the seat.

We actually noticed in this new configuration the bike seat is further over the back wheel than with the old seat, so we decided to add more air to the back tyre to compensate. 

IMG_2261.HEIC

Now here's what the bike looks like with the finished seat. After bolting everything together and sitting on it we realized that it's structural enough to not need the churros or extra bolts going through the frame. We may add them later for aesthetics but for now this is ok to ride around on.

Some Advantages

  • The seat is finally the correct height in terms of setting the seat height to the leg length of the person riding it. 

  • The seat is a lot sturdier than it was before, the metal clamps are doing a really good job keeping everything in place, not rotating or sliding. 

  • The new seat is super comfortable.

Some Disadvantages

  • The new bike is heavier because of the seat and all the metal, it's harder to lift and carry around and we need to come up with a better way to mount the basket in the back or leave it off entirely. 

  • It needs back reflectors or a tail light (don't currently exist).

  • With this design we automatically thought the bike would be able to fold up like normal, but we didn't consider the clamps would interfere with the front bar during the folding process (which they do) so we need to re-think that and possibly come up with a design where they don't interfere. 

Overall though, I think for the next few days we will be very happy riding this bike around. I think at some point we'll need to come up with a better method but for now it's pretty solid and far better than the previous seat. Maybe we'll machine something instead of using water-jet plates?

#bike #upgrades #sort-of #machining-coming-soon