MOSFET switchover for E-Bike controller

[d123]

Hi,

"I think I'll remove the diode from the Charge path and put it before M3 Source pin." - Best not remove that one, both diodes need to be in their respective places.

If yi-yun placed the word "language" in an international language or a picture of a flag (so that every human can guess what it's for) on their website, they may get more English-speaking traffic rather than landers and leavers... I looked at their site for a technical document but the language barrier makes that impossible, and other online marketplaces do not provide anything beyond sales-level overview technical details. Thanks for posting the image of the part, anyway.

R.E. last point: Not sure, maybe adding another boost converter might help but I think a gate driver IC (as suggested by Klaus higher up in this thread) will make life far less arduous for you to deal with level-shifting issues if they arise.

If you can get this to work using relays and omitting semiconductors which appear to be adding a lot of unexpected issues, I strongly recommend it.

 

[deepak4you]

I suspect that even if we remove the PMOS on the charge side before the buck/boost, and add an NMOS instead in between ground and buck/boost, it will still not work? I don't like the idea of keeping the buck/boost floating, but just an idea.

 

[Kajunbee]

@d123 I imagine there are several mosfets in the controller. Is it possible that the regen voltage he is seeing is through the body diodes of the fets. If so would this damage them overtime. Impossible to answer I know without a schematic. But should it be considered before damage is done. Hopefully this isn't a completely ignorant question.

 

[d123]

@d123 I imagine there are several mosfets in the controller. Is it possible that the regen voltage he is seeing is through the body diodes of the fets. If so would this damage them overtime. Impossible to answer I know without a schematic. But should it be considered before damage is done. Hopefully this isn't a completely ignorant question.

Given that you seem rather more informed and versed than I am in general regarding electronics/circuits, I can only say I have no idea but that's an interesting point I was unaware of. I'm struggling here a lot, if you hadn't already noticed .

- - - Updated - - -

I suspect that even if we remove the PMOS on the charge side before the buck/boost, and add an NMOS instead in between ground and buck/boost, it will still not work? I don't like the idea of keeping the buck/boost floating, but just an idea.

I don't know, my head is beginning to hurt in novel ways . I genuinely don't understand what that would achieve, sorry. Does the Buck/Boost have an enable pin or a similar function/feature?

 

[deepak4you]

The buck/boost doesn't have an enable input.

The theory about body diode conducing in reverse during freewheeling is compelling indeed. My head has been hurting for the last couple of months. I think I'll stop the public torture and rest my case by deciding to use relay contacts to keep it simple and simple. Thanks a ton for your inputs people, truly appreciated.