It's finally time for another update! To summarise the last month, I have finished the throttle and machined the front cap, as well as started machining the tailcone. I have also decided to use lithium batteries instead of sealed lead acid (SLA) batteries to save weight and end up with a much more compact unit. Changing to lithium batteries reduces the total weight of the scooter to just under 18 kg, as compared to 25 kg with a smaller capacity SLA pack. The length is also reduced from 695mm to 528mm.
My initial plan was to use four 12V 7.2Ah SLA batteries because they are cheap and readily available. However, upon further research I found that lithium batteries will not be as expensive as I first thought. Another point is that cheap SLAs doesn't like to be deep cycled and they suffer a lot from voltage sag under high load. This means that from a 350Wh battery pack I would probably only get half of this capacity and a limited number of cycles. The battery pack I intend to use is a Li-NiCoMn pack from BMSbattery.com. This chemistry is not as volatile as LiPo and is cheaper than LiFePo4, at least the packs I have seen. Lithium packs are still sensitive to overcharging, overdischarging, cell imbalance and too high discharge rate, so a battery management system (BMS) is needed. Fortunately, this is built into the pack I will use, but I will still need to be cautious not to put too much strain on the batteries. The pack is 48V 10Ah, giving me 480Wh, something that is more than most entry level technical dpvs. In the picture above you can see how it looks with the new battery pack. Notice how the length is drastically reduced compared to earlier.
One problem with changing battery technology is that I need to add weight into the front cap in order to balance the scooter. The motor and magnetic coupling add a lot of weight to the back of the scooter, but the heavy SLAs would counter this weight when placed near the front. The new pack is too light to counter all this weight, so I need to add almost 1.5 kg to obtain balance.
This is how the finished throttle turned out. The rod sticking out will later be cut to length and press fit into the hole. The lever for on/off operation is made from 316 stainless steel plate. I made the lever as two separate parts and welded them together. Finally, the part was polished using a small dremel.
The two parts before welding. The weld did not come out great, but turned out nice after some grinding. I used a mill to drill the two holes in the lever and made the rest of the shape by grinding freehand.
Once the throttle was completed, I started machining the front cap for the scooter. Apart from the tailcone, this is the biggest machined part in the project, and a very nice challenge. Also, I deliberately decided to machine the front cap before the tailcone in order to get some practice machining larger parts. In the 3D model above, you can see how the battery pack is integrated into the front cap. This means that the battery pack comes out when you remove the front cap, but it can be easily removed from the cap so that the scooter can be stored with the cap in place when the battery is charging. The two cylinders between the acrylic plate and front cap are balancing weigths.
The piece started as a 230mm diameter and 70mm long cylinder of 6082-T6 aluminium. First I machined the O-ring grooves, as well as the inside. I had initially planned not to machine the mating surface of the tube that the cap will fit into, however I found out that this is necessary in order to get a good seal as the surface on the inside of the pipe is simply too irregular.
Once the seal grooves and inside had been turned, I flipped the part around and turned the taper on the outside. In the 3D render above, you can see I have incorporated a D-ring into the frontcap for lifting the scooter. This feature would be possible to machine in a lathe using a 4 jaw chuck and moving the part eccentrically in small increments. My original plan was to do it this way, however it would take a lot of time and there would be a great potential to screw up, so when the foreman in the workshop offered to help me doing it in a CNC-mill, I said yes.
Here the front cap is set up in the mill. The machine is quite old, so there are some difficulties related to importing G-code from an external source into the machine. Therefore, the code for the circles was programmed directly into the console, and the rest was done by manually jogging the axes. There were some problems with aluminium gumming to the endmill, so one of the surfaces ended up with a pretty bad finish. I will probably remove this later in a lathe.
This is how the front cap turned out. A D-ring will be installed in the slot in the eccentric cylinder, and a plate will secure the D-ring. It will later be anodized (or perhaps just painted) for protection against corrosion.
As soon as the front cap was finished, I turned my attention to the tailcone. This is the biggest machined part on the scooter, and I have been thinking a lot on how to machine it without doing any errors. In the picture above you can see the stock being cut to length. This particular stock was left over from the hovercraft project, and from the day I first saw it I thought it would be perfect for this project. The piece I cut off weighs about 20 kg and the final part will be about 3 kg, so I will have to make a lot of swarf.
Here the stock for the tailcone has just been faced off. I set up the part in a four jaw chuck since the stock was heavy and was sticking out a lot. Since there is a lot of material to remove, I also wanted to be able to take heavy cuts without fearing the stock would come loose.
After facing, I turned the OD and O-ring grooves. Next, I started working on the inside. Here I am using a 52mm drill, the largest I could find in the workshop.
The lathe I'm using is a Tos Trenchin SN40, made in Czechoslovakia in the 70's.
I have now worked on this project for nearly 6 months, and I am finally starting to see the end of it. I still have to finish the tailcone and machine a couple of brackets, as well as the inevitable modifications that will have to be done once the scooter is ready for testing. I still hope to be finished by this summer.