even 500uA is too high of a leakage (well in our book), those smd cap if not properly glued down, will break off in a high vibration environment.
Really like your circuit, very clever.
Little tip while qualifying your switch for inrush, only mount a single fet. If that single fet is able to handle the entire inrush of the circuit, you’re on your way to bullet proof.
Yeah for the passive drain I like to look at worst case, someone rides their 10s2p battery to dead with no discharge protection and leaves it plugged in. Those things are like 5Ah capacity, assuming lower cutoff is like 3.1V then you have about 10% of battery left (0.5Ah) so 1000 hrs or 40 days until you start doing some damage to the pack.
This is what the design looks like right now, any other thoughts before I order up the PCBs and start making prototypes? @Pedrodemio@AlexBE@shaman@Deodand@Blasto
On the surface mount pads that aren’t high-current, if you put a ring of no copper around them, with only a little piece connected, it makes the heat flow much better while soldering because the giant plane doesn’t try to suck heat from the tiny pad. For the high-current pads, you can’t do that.
I’d add independent gate resistors. Like a 10 ohm between them to damp parasitic oscillation. I didn’t think they would matter with the slow turn on of our linear charge circuit but experimentally it seemed to drastically reduce failure rate.
Easyeda does not allow me to selectively choose which pads to attach thermal reliefs to :(, so I’ll have to bite the bullet and solder tiny pads to huge ground planes. Shouldn’t be a problem with the TS80, might be a problem during reflow??
I just realized that when the primary MOSFETs are turning on, Q3 is driven into forward active mode (similar to the linear region of a MOSFET). I think this is why I didn’t see problems with failures during testing of the initial prototype, and also why I couldn’t get it to turn on with larger sized zener diodes. This occurs because as the collector current through Q3 increases, R1 drops more voltage, thus causing the emitter voltage of Q3 to fall, restricting base current.
I can’t really move the components around much more, unless I want to redesign the PCB, should I go for that instead? I don’t think I’ll be able to shrink the PCB much more either.
I will rename the pad and order up the PCBs tomorrow. I have enough parts for three prototypes. Will let you guys know when those are ready. I will be keeping one for myself to test and will see if I can get the other two shipped to you guys for testing.
I can send one to you. I might have screwed up and not have the connectors until next week. And I leave for a two week vacation then -_-.
I might be able to ship it to you next week, but it may not have the connectors. Would that be alright? It needs a two pin and three pin connector, the spacing is 2.54mm. I intended for it to use right angle connectors, not sure if there is enough clearance between components for vertical connectors.
It is a prototype, so I can’t guarantee it won’t explode. I would highly recommend inlining a fuse.