If needed, yes ! (Was it ironical though ? 
)
I worked on writing a CANbus (kind off) protocol like a year ago, submitted it to Vedder who will take some ideas for his one. This is what I could do so far.
No. It shouldn’t.
An ESC hasn’t the role of managing a battery.
Like, this isn’t the role of your vaccum cleaner to trigger the fuse of your electrical cabinet at home in case of a short.
1 Like
No. The ESC should not manage the battery. It should manage the power.
Your vacuum cleaner is not responsible for sending you at 60kph down the road. We are talking about user safety here.
When it comes to my safety, i want the motor controller to be the one that cuts the power. In a safe and controlled manner. Therefore, VESC controls power.
2 Likes
This repo looks like a great start. Has anyone been able to create a makefile like in the bldc repo?
Similarity to the VESC build process seems like a reasonable goal.
The current BMS firmware, actually already has a mechanism to do this ( already works for some of my customers ), but I can adopt it to be more vesc specific.
Current implementation takes battery, BMS and connector temperature into account but also cell voltage.
2 Likes
Any way to add cell temperature monitoring?
Is already an option, the current BMS supports 2 external temperature sensors.
2 Likes
Great, sorry i’m not that familliar with DieBie. I should go read some more.
If more temp sensors are needed a small can bus interfaced temperature sensor card could work with some firmware tweaks.
I know aquiring temperature for every P group or even for each cell might be a lot but Lithium batteries are quite sensitive to high temps and the more you monitor them the more you can act preventively in case of temperature spike on one cell/p group.
@JTAG are you still using Keil? And do you have any interest for me and @Pimousse to start a branch using eclipse and starting to add features?
1 Like
If you want / desire more sensor locations, likely you want to know the worst case temperature, if so it might be smarter to put a bunch of diodes parallel en put a diode on each P group, the LTC has a high enough resolution to measure low voltages which can be translated back into temperature.
Yes still using keil, I have seen other environments lately but they were not really quicker to work with… especially the debug side of things is frustrating with the eclipse environments.
You are more then welcome to assist with feature implementations! I think that we should/could first discuss some features and discus what it the most fun / desirable to implement. Beware that I am already doing a large change in the current firmware and for the new version I am switching to the STM32G474 + L9963. But when the time comes ill open a topic!
8 Likes
People are free to do whatever they want, but I’m with @glyphiks on this one. Those of us that have been building battery packs for large off-grid power, or EV’s, know, there’s a practical limit on discharge control by BMS due to the large necessary currents. It’s not to say that BMS’s that directly handle discharge don’t exist, many due, but the failure rates are huge, the ones that can handle high currents are expensive, and necessarily massive due to simple current laws. There are ways to compensate, either huge components and traces etc, or inverting power to very high voltages, but everything comes at a cost, and handling constant high amperage wears on an increasingly complex device.
What’s the advantage of centralizing charging, discharging, balancing, temp disconnect, current limiting, etc all on the same board (single point of failure and added wear on components when directly passing large current) when the current control aspect can be offloaded to a more specialized component?
You can easily still handle LVDS (low voltage disconnect), overcurrent protection, low temp charge/discharge protection (the main reason for temp sensors in BMS btw, NOT for high temp protection like many think, although some have this feature also, most batteries actually perform better as temperatures rise, but charging or discharging below certain thresholds can permanently damage lithium chemistries), by utilizing some type of remote switching mechanism, and I believe this, is appropriate use for a BMS.
The big question is, how best to do it in this circumstance, solenoid, relay, contactor, etc, regardless, instead of passing high currents through the BMS directly you use it to control some switching mechanism that can stop the flow of current if some extreme situation arises, preferably in a programmable manner. This allows you to continue with safety/battery management features, without the heavy, component destroying, bulk creating necessity of directly handling current control. In this type of design you can economically build a small, compact BMS, without compromising safety, or reliability.
If you already have a component that’s handling current control, why you need or even want another source of this, you’re simply introducing wear and a second failure point in series (not redundant independently operating, since either failure results in a non-working device), that doesn’t accomplish anything.
However yes, it does make sense to have a separate backup function to monitor current remotely and shutdown a switching device if it exceeds some unexpected threshold. Easy enough to do with a hall sensor and a relay/contactor/whatever is appropriate for the application. Of course, that’s also accomplish-able with fuses.
TLDR: Definitely give the BMS an ability to switch current remotely as a safety feature, but why on earth would you want to introduce another inline current regulator when that’s the main instigator of wear, size, cost, failure?
Personally I’d much rather see active (energy transfer) cell balancing if we’re adding complexity and cost. Just my 2c.
4 Likes
Anyone know if we met the order quota? i know i signed up for 3. feel like i wont have one for summer :D.
Gotta admit these have been the hardest items to aquire in all of Esk8 for me personally. And man how amazing they seem to be in regards to how quickly they sell/rarely resold.
feels like the Enertion Motor Mount - except much more rare.
6 Likes
Sorry only at 35 pcs at the moment.
1 Like
Do you think 100 or 50 units is a more reasonable goal?
Is a lot of overheads to just do that small amount. Also we will have a lot of supply chain management issues. (ie, start to get suppliers that don’t want to bother, then we have to source all over the place again).
1 Like
What am I buying?
1 Like
I was curious this will work for a all VESC right? I am currently using the VESC 6 right, will this be compatible?
Also is there any available right now, I would like 2 please.
1 Like
Did you read what Sam said?
It will be compatible.
2 Likes