After a long day’s work, we all like to relax. Some people read, or play computer games, here’s what I did the other night: I wondered what’s inside a dead Daly 12V 250A BMS, so I pulled it apart.
What I found was interesting (to me at least and given you’re still reading this, maybe you’re in the small minority of people interested in this as well)!
First up, I have to say, it’s a tough nut to crack! The plastic that’s covering the PCB really holds on tight! I discovered AFTER I was done that it’s a thermoplast (meaning it gets softer and melts when the temperature increases), so if I had baked it at 80C, it would have likely been much easier to take apart.
Initially I saw that on the PCB that it says that it’s rated for 300A. So that’s probably the design current and it has been derated. Then I looked up the MOSFETs that are used. There are 40 of them in total, 20 on each side and obviously 20 will be for charging and 20 for discharging.
The spec sheet can be found here: https://datasheet.lcsc.com/…/2208311730_Samwin…
The internal resistance is not as low as some MOSFETs out there (lower is better, less resistance, less heat), but given the number of MOSFETs, it’s definitely adequate! Drain to source maximum voltage is 30V, so you might get away with connecting two of these BMSs in series for a 24V (I’m definitely not recommending this), but any more than that will kill at least one of the BMSs.
The next noteworthy thing is that this isn’t a single PCB, but there are two different PCBs connected with only a few signals to turn on and off the charging and discharging MOSFETs.
Initially I thought this was done so they could easily control other voltages or other currents by having different variations of the boards.
This may or may not be the case, but I found one real reason is that the blue PCB (the power PCB) looks like a normal fibreglass PCB, but in fact, most of it is a solid copper substrate, which is just separated by pieces of fibreglass/plastic into a few sections.
This means that the voltages of all MOSFETs’ drains and sources will be virtually identical without any localised voltage drops.
It also has one added interesting feature: Say your Daly BMS is not getting sufficient cooling and you can’t increase airflow or anything like that, what you can do is oversize your cables so that it conducts the heat away from the BMS.
The downside is obviously that you have to pay for this copper which may or may not be worth it.
The green PCB (the control PCB) has virtually nothing on it. Just one noteworthy IC SH39F003N. I didn’t find a spec sheet following a very brief search, but I did find some information on it and it’s a 3s to 5s controller for a BMS.
The reason that the balance current is so low on the Daly can also be seen, four tiny resistors, one per string and they’re covered by 5mm of plastic, so cooling would be an issue if it was any more than the 40mA(?) that it does do.
One other thing is that while the plastic was very hard to remove, I don’t know if I really believe the claims that the BMS is waterproof, as I could imagine water ingress coming in near the connectors, but that’s merely a theory, I might well be wrong.
So what are my thoughts? I actually think that hardware wise it looks like an okay BMS (though we’ve had issues with a few of them and have stopped using them), but the firmware and app really aren’t great (this can unfortunately be said for most BMSs out there).
All in all, it’s not terrible, it’s not fantastic, but somewhere in between. At least in my subjective opinion.
Thanks to the three people that that made it all the way through this post! 
