That is a big heat sink!jpgey wrote:Also, don't forget to add a heatsink on the SSR as it may be hot with high power bike
https://endless-sphere.com/forums/viewt ... 8#p1120590
Bare with me ... electronics is not my strong suit.amberwolf wrote:You could add one easy enough by finding the FET gate pin.
To use a switch to turn on the FETs, you can add a resistor in series with a switch that goes to the same voltage on the BMS that the cell-monitoring-electronics uses to drive the gate.
To use a switch to turn off the FETs, either add switch that grounds the FET gate, or add a switch in series with the gate (you'd have to cut a trace somewhere between gate and the BMS control circuit that drives it) and add a resistor from gate to ground, so that opening the switch grounds the gate and turns off the FETs.
Yes it is overkill, but it was easier for me to clamp this heatsink. Also this is for a 7000W setup.LewTwo wrote:That is a big heat sink!jpgey wrote:Also, don't forget to add a heatsink on the SSR as it may be hot with high power bike
https://endless-sphere.com/forums/viewt ... 8#p1120590
I was thinking that just clamping it to the frame might be adequate but if it is going to produce that much heat then it is wasting so much power that the idea is a non-starter.
Yes, this one you don't have to mess with as long as your charge and discharge connections are separate.LewTwo wrote:The Mosfet in the upper right corner goes to the charge connection. It is a RU8590R.
Sounds normal; there will be a 12v source somewhere on the BMS for the gate drive, probably sourced at the 3rd cell's positive if it's not LiFePO4, and the fourth cell's positive if it is LiFePO4. It could have a 3-pin (or other, smarter) 12v regulator (usually marked 7812 or similar) that produces this voltage instead, but probably not--this is why the bottom few cells can get killed or unbalanced vs the rest of them on packs left alone a long time without being used.That big blue wire is the negative side of the 58 volts power going out. Between it and the gate pin I have a 11.65 volts with the negative side on the output wire.
also normal; there will be a voltage across them when the bms is "off", though any current draw should collapse this.The source and drain pins are at the same voltage level as the output wire.
The unsoldering gate pins is easy, and reversible enough. I'd go with the other method below, though.If I am reading all this correctly then what I need to do is:
... unsolder or cut the gate pins (this would effectively switch OFF all the mosfets).
,,, connect the gate pins together with a wire (it could be something like 18-20 AWG because the gate does not use much power)
... add a switch between the wire and one of gate pin pads (each of those pads has a small surface mount 1/8 watt resistor next to it... the label reads 101).
This is the "easiest" place to put the switch--just cut the trace somewhere, and put the switch across it.Edit: The trace on the back with two hole under the left mosfet is the one that feeds the gates. It has a corresponding trace along the edge of the top of the board that the resistors are connected to.
I leaning toward putting a switch in place of the trace. I think that I can use the ends of the trace as surface solder pads and connect two small wires to run to a key switch. The 100 ohm (thank you for the explanation) resisters will still be in series with the gates.amberwolf wrote:This is the "easiest" place to put the switch--just cut the trace somewhere, and put the switch across it.Edit: The trace on the back with two hole under the left mosfet is the one that feeds the gates. It has a corresponding trace along the edge of the top of the board that the resistors are connected to.
The only thing you might have to do, if it's not already built in there somewhere, is a resistor to ground (battery negative) (like 10kohm or so, maybe higher), as a pull-down to prevent the gate from "drifting" to a partially-on state. The board may already be wired like this, or the FETs may be of a type that this won't be much of a problem, so you might not have to worry about it at all.
Anyway, either method is pretty simple to test, and if it doesn't do what you want, simple to undo.
Ah ... that makes a bit of a difference. I am looking at less than 1000 watts (17 Amps) max draw. I think I prefer modifying the BMS if I can and the more I think about it, I am leaning toward replacing that trace on the back with a switch.jpgey wrote:Yes it is overkill, but it was easier for me to clamp this heatsink. Also this is for a 7000W setup.
I don't have a diagram, but there ought to be quite a few under "precharge" or "anti-spark" or similar, here on ES or on google, possibly even on our wiki if anyone got an article up there.LewTwo wrote:Next question is if one only wants avoid a big spark when the battery is plugged in then is this the correct wiring diagram?
+1. search for "inrush limiter" "anti spark" or "precharge" and you WILL find more infomation than you can takeamberwolf wrote:I don't have a diagram, but there ought to be quite a few under "precharge" or "anti-spark" or similar, here on ES or on google, possibly even on our wiki if anyone got an article up there.
That works for loose packs and cables but does not lend itself to a pack that has a fixed connector.amberwolf wrote:... But before you go looking, basically you split either the positive or the negative wires from the battery and controller into two where they meet, so you now have three connectors between batt and contr. One of the split wires doesn't have to be very big, as it doesn't supply much current; it's going to be used for the precharge.....
Thank you 'inrush' led me to the 2012 thread but alas it is 23 pages long. Going to take me a few days to work my way through it.izeman wrote: search for "inrush limiter" "anti spark" or "precharge" and you WILL find more information than you can take
there are a lot of possible ways to do it right and trouble free, plug & play.
I like that idea. I have collected all the bits to do including a key switch. However I want a plan "B" in case I screw up or something goes wrong (that could never happen ...). That means having a source for another 14S BMS board. That is something I have to research .... Yo Google ... here I come ... again.amberwolf wrote:Personally, if I had a BMS on my pack, and I wanted to avoid sparks, or have a power switch, I would just use the switch on the BMS FET gate(s) to do it. ...