High side power switching advice.

Thread Starter


Joined Sep 30, 2017
Below is a circuit I want to implement for an automatic engine control board, I have been around the sites and I am fairly happy with my circuit however I would like some advice to some points, some which I am uncertain about, some which I find conflicting information on:


I am looking to make a control board about 2 x 2". The board will be used in conjunction with other electronics and will provide a means to remotely start an engine. I have a 'breadboard' layout and am fairly happy with it, it works fine and does what it should.
So I want to get this compacted onto a pcb so that It can be tucked away to join the other electronics. I've done a design on easyEDA but before I go ahead I just want to make sure Im not doing anything stupid, or if i could improve the concepts used.

1. P-MOS (AOD4185). Used for Reverse Polarity Protection. This chip has a fairly low RDS (15mOhm). I wanted to include this as its easy to get wires crossed but for me its a must. My question here is is this a standard solution for this problem. Its not widely claimed as a one stop solution to RPP - and I suppose that depends on the application but for my application is this the way to do it? Are there specific chips for this problem (or even Ideal Diodes)? Are there any drawbacks - such as back EMF from inductive loads etc when using a P-MOS for RPP?
...And further to that is the substrate conducting the current or is the body diode responsible?

2. My application switches an inductive load (solenoid) which switches another inductive load (starter motor). These types of loads present back EMF and are contained by the use of diodes across the the load terminals. BUT if for instance a diode was to fail or was not included in the first place, as a precaution I would like to include a shunt for any unwanted voltage presented on ground. Now I originally wanted to shunt this straight back to the +Vin but thought better due to the fact that there is a 250 watt starter motor - and wasn't sure if the Protection MOS could handle that kind of kick and secondly there is nowhere for the energy to go except for the buck converter (which i think has OV protection but...), so knowing that there is always a big load connected to D2 I thought better to shunt it there. So my question here is:
a. Could the buck converter absorb the energy kick back from an inductive load - therefore adding some redundancy if nothing were connected to D2?
b. If no loads were connected and somehow a voltage was presented on ground, how would you deal with this situation?
c. Thinking about leakage current and keeping power losses down, what diode (general/schottkey) would you use?

3. This is a dual NPN switcher for the Dual P-MOS high side switches. Bearing in mind that each channel of the Dual P-MOS have different voltages, is it ordinary to switch both transistors with the same signal? I have seen dual transistors with common emitters/collectors, not sure about common bases though!

4. Daft question maybe, But is there any time where you can NOT present different voltages on the separate channels of Dual MOSFET - ignoring common source/drains of course. And would this apply to NPN equivalents?

5. If a load has more than one switch from the same source, and one switch has a greater resistance (in this case the MOSFET) does this present a problem if both are switched at the same time?

Maybe there are some bonehead questions in there but hopefully we can clear up some misconceptions.

Thanks for looking.
Last edited: