Suggestions for starter motor use with JK BMS

[doggy]

I have installed 8 EVE 280AH LFPs in 2p4s with a 200A (max 300A) JK BMS.

All is good. However, the starter motor is around 1400W or 120A but presumably draws a lot more on start up (cranking a 3cyl diesel on my boat). It is presumably going over 300A because the BMS is triggering a SCP (Short Circuit Protection) disconnect. The interval in the settings is 1,500uSecs or 1.5msecs. I'm reluctant to increase it much and it would probably need to be about one second for the starter motor to start moving and drop the current.

So I'm looking for suggestions on how to handle this. I'm not interested in swapping out the BMS for a relay one because I've done a lot of work to interface 2 MPPT solar controllers, engine alternator regulator and a boost charger to the BMS for another battery bank. Short term I have an AGM in parallel but I'm not going to keep things that way. I'm sure others will have had this issue. My couple of thoughts are:

1. A solenoid bridging B- and P- on the BMS triggered in parallel with the starter solenoid or

2. A bunch of Super capacitors in parallel with the BMS output to sop up the stalled initial starter current.

Thanks,
Dave

 

[Supervstech]

I have installed 8 EVE 280AH LFPs in 2p4s with a 200A (max 300A) JK BMS.

All is good. However, the starter motor is around 1400W or 120A but presumably draws a lot more on start up (cranking a 3cyl diesel on my boat). It is presumably going over 300A because the BMS is triggering a SCP (Short Circuit Protection) disconnect. The interval in the settings is 1,500uSecs or 1.5msecs. I'm reluctant to increase it much and it would probably need to be about one second for the starter motor to start moving and drop the current.

So I'm looking for suggestions on how to handle this. I'm not interested in swapping out the BMS for a relay one because I've done a lot of work to interface 2 MPPT solar controllers, engine alternator regulator and a boost charger to the BMS for another battery bank. Short term I have an AGM in parallel but I'm not going to keep things that way. I'm sure others will have had this issue. My couple of thoughts are:

1. A solenoid bridging B- and P- on the BMS triggered in parallel with the starter solenoid or

2. A bunch of Super capacitors in parallel with the BMS output to sop up the stalled initial starter current.

Thanks,
Dave

I don’t see why a relay on the terminals wouldn’t work.

 

[doggy]

I don’t see why a relay on the terminals wouldn’t work.

Yes, I'm leaning towards that solution.

 

[Ahjwinth]

You can also connect the main current circuit directly on the battery Bypassing the BMS and then leave the circuit of the key switch/solenoid on the starter motor connected on the BMS. When battery voltage then gets to low the BMS will still protect the battery and switch off the solenoid/starter motor but the high current doesn’t go through the BMS And you don’t need an additional relay.
The downside of this solution is that you don’t measure the energy consumption anymore with the BMS and SOC might give a wrong reading. But the same would happen if you install a relay over the BMS.

But this will only work if the BMS is connected on the positive of the battery As I assume the circuit looks like below.

If you would make a small relay connected to the BMS and connect the positive to the ignition switch. You can use the contact to switch the solenoid, this way the negative side of the starter motor can be connected directly on the battery and your still protected and don’t need a large solenoid to switch the full current. Like in the schematic in the attachment, hope it is clear.

 

[doggy]

Thanks for the interesting suggestion.
However it will effectively create some grounding issues for me.

I can increase the BMS short circuit time to one second, which could be okay. However I am a bit jittery about this and so will avoid it. Pushing the starter motor stall current through the FETs is probably not a wonderful idea.

Super-capacitors are doable but a bit messy.

So I have decided to add another solenoid, triggered off the starter solenoid and which bridges battery negative to P- which is also the boat earth. It will miss counting the approx 1AH used to start the engine but this is not a big deal. I'll try to get the path through the additional solenoid to be lower resistance than through the BMS path so that less than 1/3rd of the stall current goes through the BMS. I'd like to keep this below the 300A BMS limit.

I'm also going to insert a time delay so that the new solenoid switches before the starter solenoid. Maybe 200msecs.

I'll report back on how well it works.

 

[doggy]

Okay, all done, tested and working. I left the BMS short circuit protect at 1,500 microseconds.

I added a 200A DC contactor from Battery -ve to P- from the BMS. The contactor can take 700amps "make current" and around 300A "break current". So basically it and its short cabling are bypassing the BMS FETs, the BMS/battery cabling and a 200A shunt which is connected to the BMS P- terminal. Thus the majority (>90%) of the starter currents are diverted away from the BMS and through the contactor.

I also made a simple time delay of 1/2second using a P-channel TO220 FET on a small heatsink which goes to the starter solenoid. That has a 40A startup surge but run current is low.

So, when you push the starter button, the bypass solenoid clacks first then half a second later the starter solenoid and thus starter is engaged. Power for the solenoids comes via the BMS so if any cell is too low or the BMS has otherwise cut off power, nothing will happen. You can override this with the emergency BMS switch but that would not be advisable except in a dire case. There is a small risk to the cells if the starter was engaged with almost totally flat cells and then the BMS bypass happened but I can live with that.

 

[d-d]

I've been looking at a similar project but for a camper. I need to measure the starter current but expecting around 600A and the batteries are Catl 302AH so around 2C for a burst I expect to be ok. I was concerned that shorting the BMS (-ve, p-) might damage the FETs. Sounds like this is ok???