Lark Streamliner EV motorcycle Project

[frodus]

I think you're right for discharging. I'm just hesitant for the charging side is all. If there's alarms, and someone is manning the charger until one of those alarms goes off, he should be OK. But I wouldn't do anything unattended if it can be avoided.

[Stevo]

I think you're right for discharging. I'm just hesitant for the charging side is all. If there's alarms, and someone is manning the charger until one of those alarms goes off, he should be OK. But I wouldn't do anything unattended if it can be avoided.

I like that you mention confidence in the discharge phase, because I bet there is no fast way to get out of that cockpit, and no ejection seat!

[frodus]

Let me clarify. It's a race bike. Things get rebuilt, and balanced once back in the shop. Discharge ought to be fine if he sets the LVC on the controller to an acceptable level.

I still prefer a BMS that can alarm on charge/discharge and can cut off the charger if needed, or the controller if a cell reaches a certain threshold.

[Frank]

With an Orion (and Torque) I know lowest cell voltage, min pack voltage, max current, etc. It has an output that can be programmed to signal a alarm or a couple of other options using a variable output.

Sent from my SM-T380 using Tapatalk

[tomdb]

In your latest video you show the VCU, which is an arduino.

Personally I feel, depending on the version of arduino, that these simple microcontrollers should not really be used in a Torque/motion controlling application.

However there are ways of making it 'safer', ways in code and ways with wiring. Would you mind sharing the code and or wiring for your motor setup?

[larkmachineco]

In your latest video you show the VCU, which is an arduino.

Personally I feel, depending on the version of arduino, that these simple microcontrollers should not really be used in a Torque/motion controlling application.

However there are ways of making it 'safer', ways in code and ways with wiring. Would you mind sharing the code and or wiring for your motor setup?

Hey there, unfortunately, I can't share the motor type or code.
But, the arduino is much faster than the motor signal. I am not a software engineer, but I do understand your concern. I got a GEVCU from EVTV and found out that it runs on a modified Arduino. Knowing that, we built a much more simple version that does all the things I need. If the srduino takes a dump or doesn't perform, ill try the evtv version, but I can't imagine it would be any better

[larkmachineco]

As long as you can hear those. I wouln't ever leave the vehicle unattended during charging.

I'd suggest monitoring with something that has an output that can be used to drive a relay to shut things down. All of the battery fires I've heard of are because of overcharging a cell group. Without a means to disconnect the charger if a cell group goes over, there's no safety mechanism and it could be dangerous. As long as you're 100% sure you will be the "means of disconnect" of the charger, that's fine.

Side note: I'm an Sr. EE, and I've worked for Manzanita Micro, Synkromotive and Elithion on BMS systems and done support for them. I've also worked with Orion and built my own BMS. I don't care about balancing (I assume you're going to do that yourself), but monitoring is of utmost importance.

Just be super careful.

I dont ever plan on leaving the vehicle during charging, I have a Manzanita 50A charger. I was planning on having a 1 ohm power resistor on board so I can leak down out of balance cells on the fly. the voltage monitoring is just so I can see which cells are high.
As for discharging, the bike will have effected power and will never be run under 50% (even during testing) I would love to see the voltage profile over the run, but I have to setup a data capture and use the Linduino for that. It's in the cards if need be later on.
LTC6811IG-1 eval board. I have enough of them, I just need to find time to integrate it. I'll find time if the voltage monitoring fails

[larkmachineco]

I like that you mention confidence in the discharge phase, because I bet there is no fast way to get out of that cockpit, and no ejection seat!

Actually, i have to be able to get out of the bike within seconds. You have to perform it at tech. The rules also require you to have 6 lbs of coldfire in the driver compartment. The safety aspect of the vehicle is by far the most expensive and labor intensive. a ejection seat would be cool though!!!

[larkmachineco]

My side note: reverse engineered plenty OEM bms systems, plus had a go at designing my own cell monitoring slaves and now designed my own bms master.

You wanna sponsor me and send one of yours?!?!?!?

[larkmachineco]

I'm gonna play devil's advocate here for a minute. My experience with Formula Ford builds (Skip Barber had his shop in the next town over and a friend worked there) is that the cars were built to last one race, and basically one race only. Tearing the car down and rebuilding it for the next race is what kept my friend in a job. If the man's on a budget, and is careful, I don't see it being impossible to get the most out of the pack without a BMS, with the assumption that between races he's going to have to do a lot of TLC.

I think we tend to think about builds here like we're all building daily commuters, or something that has to behave like an appliance. Race bikes, especially LSR bikes, are a completely different animal.

Thanks for the support. I'm not against a BMS, Im just focused on the essentials, and a BMS is low on the essential list right now. If I can get funding or if it proves to be essential, I will add it.
I'm gonna treat this bike like a pregnant panda! I will be vigilantely monitoring Vsense during charging between runs and making sure that every nut and bolt is secure.

For all of this safety talk and nobody even mentioned isolation detection circuit! (in my opinion, much more essential than a BMS)! THAT'S HOW A FIRE HAPPENS!!!!!!! I got a Bender unit
https://www.benderinc.com/products/g...55-0304-series