basically, you can turn on manually balancing via an app, and every voltage limit is configurable, cool little thing
theres a pdf about it explaining everything
@SimosMCmuffin Do you happen to have any more pieces left? I thought I wouldn’t be able to allocate any time for testing it but things are getting a bit easier now. I’d like to do some DAVEGA developments against the FlexiBMS. I also plan some maintenance/rework on my battery pack and the FlexiBMS would be a great fit for it.
3 Likes
My life is fucking crazy at the moment dude, and likely will be for the next few months, i’d be happy to send you mine once it gets here if Simo doesnt have any unallocated
2 Likes
That sounds like a very efficient way of getting stuff from Finland to the Czech Republic. Ship it to Australia first! 
Seriously though. Thanks for the offer. I’ll happily take you up on it if Simo is out of stock.
3 Likes
i know right. Last resort but the offer is there. I have a build ready to drop it into but i doubt im gonna find the time to put it through its paces
1 Like
You’re in luck,
One of the testers backed off on their original 2 boards to just 1 board, so I have currently 1 un-allocated board available.
I’m still waiting for a reply from another tester who had 2 boards allocated for them, but hasn’t paid and hasn’t responded to my query about the payment for the boards. Forum shows that he has been “seen” in the last 16 hours, so I think he should have seen those PMs to him. I’ll allow him a couple more days to respond, but if the radio silence continues, I’ll free both of those boards back to the available lot.
EDIT:
I have had queries from multiple people about getting boards for testing, so I’m going to be talking about via PMs about possibly getting boards.
3 Likes
Great! I’m up just for one. I’ll PM you.
Report on the Postal employee strike, it’s over.
They are predicting couple weeks to completely clear up all piled up in- & outgoing shipments.
https://www.posti.fi/private-news/english/current/2019/20191127_strike_is_over.html
4 Likes
4 Likes
@SimosMCmuffin this may be a stupid question that I’m missing the obvious answer but is there a reason the charge solder pads are smaller than the battery pads?
For headeres it seems a bit random if there coming out the top like a VESC. This would be food for low profile builds siting it next to the battry
Or 90deg headers coming out the sides witch is better for building in to a pack and sitting on top of the cells.
Struggling to fined dimensions of space that would be required in either of the out comes I think the first one is about the size of 2*18650 cells
3 Likes
Testing and FW development has been going on. Improvements and fixes have been implemented and plans for new features are in the works. People are finally starting get their boards, thanks to the postal office strike. There’s been active discussion going on in the PM group chat for the testers.
Got the USB DFU bootloader working on Sunday, but it seems that there is a silicon based bug/glitch in the L43x-series that causes the USB connection to drop out during image download, but the download can be retried again until it completes succesfully. I’m looking into different solutions to see, if it would be possible to make it more reliable. If the download fails and you un-plug your BMS, it will have a corrupted FW after that, so it’s pretty important to complete the download, but as I said at this point you just try it again and again until it completes succesfully.
https://sourceforge.net/p/dfu-util/tickets/40/
I have also gathered some changes to implement for the 0.5 HW with probably the biggest one being to change the 5V buck IC (MAX15062) to another with an easier to solder/pick&place package for more reliability. The current IC’s package is so small it’s starting to become more of a hindrance than a worth-it space save.
The real problem is that there is an increased risk of a solder-bridge forming between pads 1 & 2 (at least in my hand assembled boards, can’t say if it would be problem for automated solder paste stencil machine + pick&place machine, but I want to minimize the risk), which essentially causes the battery voltage to directly connect to a pin on the MCU and let’s just say at the higher S packs, you might get some light show out of there. I tested the boards before I sent them out to the testers, but at least @mackann has reported one of his 2 boards getting this short. Also HW change coming up to protect the MCU even IF the solder bridge or something else were to short out the battery and enable signal pins.
With that said, I’m reserving the 2 un-allocated boards for spare boards (sorry for the people who volunteered for them in PMs) in case someone gets a HW problem on their board, so I can send them a spare. I’m still a bit miffed about Mackann’s board, as I had done basic tests to them and it hadn’t shown any problems then… I’m gonna design a basic HW tester for future production boards that’ll allow me to test them in a semi-automatic way and run through a batch of tests. Most likely it will have a known good FlexiBMS Lite board testing against the DUT (Device Under Test), along side current, connector and comms test and then calibrate the ADC gains and offsets for the charging currents and charger & pack voltages.
I’m starting to feel confident enough about the HW design with the upcoming changes that I can start thinking of getting a batch of them made, at least once I figure out a solution for more reliable DFU firmware flashing.
I’ve been thinking about where to make the zero production batch and I have decided to go with a local nearby fab first, so I can be available in-person if need be. It’s going to be more expensive both in start-up costs and per board assembly costs, but I want these boards to be quality.
The biggest hurdle to overcome before getting to the zero batch though is going to be setting up my LLC company to have my personal assets separated from the company’s assets. Now the biggest headache is going to be accounting, but I’m looking into different solutions for actually running it legit…
The boombox project has also been progressing nicely.
9 Likes
No any special particular reason, but the traces were a bit better/easier to route with their current position and sizes. I also wanted a little bit of distance from the connector housings next to them.
The design allows the end-user choose whether they want vertical connectors or horizontal connectors, or even no connectors at all, if they’re soldering wires directly to the board for example. They can decide which configuration will/would be best for their intended application location.
2 Likes
Have you considered asking for the insight of our man on the inside? Maybe @jeffwuneo could point you to a good and reliable fab line. Really glad to see this progressing
We can make SMT in house. But too busy now. After spring festival can helping.
8 Likes
Always balancing can mess up the pack if one parallel group is busted.
1 Like
As a thought with the trampa VESC 6 mark 3 having a hibernate mode as well as unity and flipsky switches on ESC will the flex light be able to activate hibernate when it detects the vesc has been turned off or entered hibernate mode? Same for when PTS will that trigger a wake up in the BMS? I’m presuming this would be a software thing. I know the DieBieMS used the 5v from the VESC back EMF to a extra pin on the BMS to turn on for PTS
Moro Simo,
Is it anyhow possible to join the party already?
Might there be someone who got one but does not have the time to test it?
Same here. If there is a spot open, let me know.
Get in line folks I think I’ve got first dibs 
1 Like
I guess I am in line now then ^^
What is their eat cost?