- Tue Jun 04, 2019 11:28 am
rudy wrote:And you should have capacitors at the IR4426 from Vss and gnd. And get the IR4426 away from the ESP12. And the traces are to thin. I would turn U1 90 degrees and place it about where you have D1. Right side of the chip drives the mosfets, left side gets the signal from the ESP12.
As far as clamping protection. I see that you have nothing in series with them. What will be the series impedance ? Did you not understand my previous explanation about the inductor?
Ok I'll do these modifications but why is it too thin ? Is it related to current ? Because there should not be a lot of current.
For the inductor, it is only there to protect the system when the spark will occur many times per second or can it also be seen as a spike protection ?
I have to put it in serial before regulator's input ? So I should calculate it for what frequency ? Because as it should only make DC pass through, I don't know.
Pablo2048 wrote:Some more tips:
1. you are dealing with motorcycle power (aka Automotive) so take a look at properly designed Automotive power supplies for CPU - there need to be some huge filtration and protection first - your C1&D1&D2 combination is definitely not enough. You have to handle high frequency noise, voltage spikes, ... so there is probably need for some big/small capacity capacitors in parallel & some inductor in serial.
2. connector P7 (sensor) - what is this? What is the signal shape on this input? What are the voltage levels? It is connected directly to the ESP without any signal conditioning which is no-go in automotive tech.
P7 is connected to a Hall effect sensor which senses crankshaft position. It gets 3.3V and returns a 3.3V peak square wave. Cable is approx 1 meter long. Is it again a noise problem ? What should be added to be automotive compliant ?