Ego Street Scoota LiFePO4 Lithium Upgrade
Here's my upgraded "Ego Street Scoota". It started life as a 48V SLA bike, but after several upgrades it's now fully converted into a Lithium Ion based system using LiFePO4 Headway cells.
The pack is a 24s2p, nominal 76.8V, built out of 48 Headway LiFePO4 38140S 12Ah screw-tap cells. It weighs just over 19Kg, versus the 48Kg my SLA bank, so I've shed nearly 30Kg! I elected to mount mine vertically. It wouldn't fit flat as it was too wide, so I sacrificed some seat space as before. The advantage of this arrangement, though, is I can more easily remove it if needs be. I also have enough space down the side to slide the BMS down alongside it so it is flush with the battery case, which can be used as an extra heat sink to protect it if it ever gets too hot shunting if cells are particularly out of balance. It usually runs cool as a cucumber though, even at 9A!
Here's the final arrangement. I had to take a chunk out of the seat, but I'm quite happy with the result Smile. My Lyen's controller is en route back to me, so I've only had chance to use it with my cheapo controller, but I'm delighted by the improvement to its all round performance.
Here's my painstakingly constructed Goodrum Fechter BMS doing its thing... I'm using a BMSBattery EMC-900 with it, which gives me the aforementioned 9A, though I've wired it all up with nice, thick wiring so that I can take advantage of the 20A charge rate that the GF Zephyr is capable of handling, if needs be. I used an ATX connector, as a 25-pin serial couldn't accommodate my 1mm2 tap wires.
I'm delighted with the improved charge rate, which is almost four times what I got with my 2.5A SLA charger! A bit of weatherproofing is in order, now, to protect the bank and battery compartment from the wet when poorer weather arrives.
The Range Test
I've now had a chance to give the setup a proper range test. I have my Lyen's 12-FET controller set to 45A, which is pretty hungry as far as power consumption goes. Though the Lithium pack is nominally 76.8V, in actual practise the pack spends most of its discharge between 84V and 77V, with their voltages quickly collapsing once they pass below the nominal 3.2V per cell level.
The CellLog monitor for my 24s LiFePO4 pack
Range
Over a couple of days, I charted the distance I travelled through 9 different journeys, all of them in city conditions (with a good amount of stopping and starting at lights). The bike made it 27 miles in all before the weakest cells passed below 2.5V and I called it a day. I'm confident that if I kept to a lower speed control setting, and did a long uninterrupted out of town run between cities, I could greatly improve on this range. Bear in mind also that this was using my power-hungry LYEN edition controller on its highest setting, which cost the bike about 40% of its range when it was running on SLAs.
For my next range test, with my charger in the top-box, I'm going to see if I can traverse the 35-40 miles to a nearby city, then recharge and return.
Re-charge time
From completely flat to a full charge, the pack took 2 hours 50 minutes to fully recharge from my EMC-900 charger, which delivers 86V at 9A. I've wired the Goodrum-Fechter "Zephyr" BMS so that it can comfortably support the full 20 amps that the board is capable of managing, and so have the option of upgrading to a more powerful charger.
The charge setup, running at 9 amps
Based on this, I could probably get the recharge time down to 1 hour 15 minutes with a suitable power supply unit.
Full details of my upgrade are available on my blog:
Building a LiFePO4 Lithium Pack
Building the Goodrum-Fechter "Zephyr" BMS
Monitoring Pack Cell Voltages
Nice build and nicely documented! Good job.
I like those cell log's; might have to add some to my bike.
Motorcycles: 2011 ZEV Trail 7100, 84V, 60AH, 60+mph, Cycle Analyst, TNC throttle, modified charger. 2013 Kymco GT300i
Bicycles: 2017 Sondors Thin
Cars: 2016 Leaf SV, 30KWH pack. 2007 CR-V
Solar array: 5KW. Cost per lifetime KWH produced $0.073
Bi
Thanks! Yes, the CellLogs are excellent little units and really versatile. Very cheap too!
I'm thinking of getting a batch and stocking them in my online shop.
Zenid's Ego Scoota Blog & Upgrade Guide http://zenid10.wordpress.com/
In reading more about the your battery pack on your website, it was interesting to see a few cells near the positive end of the string were depleting faster than the rest of the pack. Do you think this is only an issue with Headway cells or also Lifepo cells? I wonder if you set the capacities that way or if the end of the string gives up electrons easier??? I guess you could move those cells and see if the issue goes with them. Anyways, I seem to remember Matt saying he setup his pack per capacity of each cell (similar AH cells together) and could switch in some extra AH capacity to the leaker cells to extend range. Would it help your range to add some small additional AHs in parallel to your weaker cells?
Yes, those cell logs are cheap, I can pick them up at hobbyking.com for about $14. Looks like I will be adding three of them (or just one and move between three connectors) for my 24 cell GBS LiFeMPo4 pack as a winter project. That winter project list keeps growing.
Motorcycles: 2011 ZEV Trail 7100, 84V, 60AH, 60+mph, Cycle Analyst, TNC throttle, modified charger. 2013 Kymco GT300i
Bicycles: 2017 Sondors Thin
Cars: 2016 Leaf SV, 30KWH pack. 2007 CR-V
Solar array: 5KW. Cost per lifetime KWH produced $0.073
Bi
I think this is an issue with ANY cells that are connected in series. Due to the electrochemistry of the cells and the way they deliver charge, the positive terminal of a cell - and therefore also that of a string of cells in series - gets depleted first, and a slight gradient of charge exists from one end of the pack to the other, which either diminishes as the cell rests, or is accentuated by high-demand discharge.
That's how I understand it anyway...
No pack will ever be perfectly balanced and some cells will always be slightly weaker than others. It a question of whether the difference is sufficient to drag down the performance of the pack as a whole. Also these discrepanceies tend to appear exagerated when the pack is particularly low, as the cell voltages will all 'cliff-dive' as they pass below their nominal level, with even small differences becoming amplified exponentially. I wouldn't try to fix weak cells by trying to patch on others in parallel, as this would be way more trouble than it's worth and liable to cause more problems than it solves. If you have a bad cell, it's best to just replace it.
Zenid's Ego Scoota Blog & Upgrade Guide http://zenid10.wordpress.com/