jonescg wrote: ↑Mon, 09 Dec 2019, 20:04
Well done! Very smooth and quiet.
Did you end up sealing the battery pack with silicone or something?
You must be stoked
Yes very happy with the way it all went except that I need to replace one of the cell groups so once I’ve got the replacement cells I’ll rebuild the pack (again!) with silicone and thread lock. Hopefully for the last time ... until I upgrade
Hoping to get everything ready for brake testing and engineering in January, then I can sell my petrol scooter!
After the last lot of testing there was a fair bit of work to do getting ready for brake testing and inspection. Replace dead battery cells, wire in the emergency cutoff switch, replace oil in forks and add spacers, etc so after all that the bike pretty much looks the same but a lot has happened! Last Thursday took the bike out to the drag strip at Eastern Creek where an engineer put some instrumentation on the bike and did a series of brake tests. Everything went well so I just need to provide some documentation on the conversion and get it officially weighed then it should be good to get certified. Looking forward to getting it on the road so I can play around with the speed controller settings to get it dialled in.
OK so it's been a while, again! The bike has now been registered for about 3 weeks and I'm having a lot of fun on it. Now that I'm getting to do longer rides, I'm finding new issues that I'll need to work through but overall I'm really happy with the way it's turned out. Will now start saving up for a bigger and better battery pack!
I managed to convince the family to go for a drive with me on one of my rides so I could get some video.
Thanks! Having lots of fun. Still things to sort out but at least I can have a bit of fun in between fixing/improving things.
My biggest constraint is range which at the moment is only 25km. The 10Ah cells I bought on eBay and the 15Ah cells I got second hand. I’m not even sure if the 10Ah cells are original Headway cells or some other brand! I’ve got a couple of CellLog 8S from my previous project that I’m using to log rides and work out which are the weakest cells. So one by one I’m replacing the weakest cells but I’m going to run out of spares soon. So my dilemma is do I keep buying replacement cells and slowly fixing the weakest link or is it better to save up and replace the lot. If I do that I’d like to use all 15Ah cells and possibly go to 24s3p instead of 24s2p. The bigger capacity should handle the current draw better and increase my range to around 50km assuming the current cells have lost some capacity. Or do I go to a new chemistry altogether? Lots to think about.
Here’s a log of the middle 8 cells in the pack during one of the test rides clearly showing cell 8 dropping away before the others.
D25D1BF3-FA09-4FBB-A543-9F3D13044C85.jpeg (394.34 KiB) Viewed 1477 times
In a bike I would think going the tesla route with bucket loads of 18650s would be best use of space but they are harder to build a pack out of.
Looking back at you battery, it's pretty square so maybe pouch cells would be a good fit.
In terms of chemistry I like NCM, it has a good life and is not too expensive or explodey and most EV manufacturers use NCM so it must have something going for it
I recently added a brake lever to the bike to control the amount of regenerative braking to use. Previously I was just using the percentage setting in the Kelly controller to control the amount but it was either just right at lower speeds and not enough at higher speeds or vice versa. This way I can control it myself. Just need to adjust the amount of air in the hydraulic reservoir to make it feel like the other brake lever.
Also I've managed to get some A123 20Ah cells to upgrade my battery pack. These looks like pretty amazing cells compared to the Headways so I'm really looking forward to getting these in the bike and seeing how they go. I built some cardboard boxes to try out different configurations and although I could do a 23s3p pack, I've decided to go with a 26s2p pack instead. The 2p pack can be made with all the cells stacked neatly together whereas the 3p pack would have to be an odd shaped box with the pack split into 3 sections. Plus I have enough cells for the 2p version already. Here are some photos of a test box to check everything fits into the box and the box fits into the bike!
available space.jpg (160.75 KiB) Viewed 1196 times
Hey Andrew,
I should be able to ship the buslinks and PCB over to you by the end of this week. Looks like a great fit!
If you want to cut the current 30s2p boards down to 26s2p you can do so with a Dremel and a mini cut-off wheel. Shame the controller can't take more volts!
Rusdy wrote: ↑Mon, 04 May 2020, 15:05
What did you end up doing with the coolant inside the frame? Just leave it empty? Interesting place to store coolant I reckon...
I just left it empty, yeah I was pretty surprised to find that! I thought about cutting off the tubes that come off the frame where the hose attached but since I'm avoiding altering the frame as much as possible I left them on. Plus if I removed them I'd have to weld up the holes which I also wanted to avoid.
OK so I've finally received a battery capacity tester I bought on AliExpress a while ago and am starting to do some testing. My plan is to test each cell so I can pair them up into equal size parallel groups before putting them in series. The cells are A123 20Ah so their nominal voltage is 3.3V and 1C is 20A ... is this correct? When I set the discharge current to 18A, the voltage drops to 2.5V which is the voltage I was planning on using as my LVC so the test says "Complete" straight away which is a bit surprising. So either I'm doing something wrong or the cells might be a bit tired? As a workaround I've been testing at 10A but I'm also not sure if this is the correct discharge current to be using. I saw a video on YouTube that said you get the best results at 0.2C
This is the setup I’m using inspired by someone on Endless Sphere. The tabs are pretty well clamped down and I get the same voltage drop if I put the alligator clips directly onto the tabs.
79ABBF45-1069-4482-9A95-AF831F3B25B0.jpeg (294.67 KiB) Viewed 877 times
OK so I just used a multimeter to test the voltage on the tabs and also where the wires connect to the PCB board and there’s a 0.3-0.4V difference when discharging the cell but they’re the same at rest. Any ideas what’s going on or how to fix it?
It’s only a $50 device so maybe that’s my problem
I may not be able to measure the actual capacity but hopefully I can I still use the results to pair the cells into 2p groups. Better than nothing I guess!
