Transistor as a switch - low side or high side

Why are you not using complementary transistors for high side switching? I would use an PNP transistor on the high side, to avoid the complication of a Vbe drop, by allowing the collector always to be connected to the load. A similar situation with enhancement mode mosfets, I would use a P channel mosfet. There are considerations of finding fets with the necessary speed or Ids current that may limit the choice, but in general the complimentary transistor is the first choice. Personally, I like to use open collector ( open drain ) transistors for interfacing between circuit boards. This is because the one voltage which is not subject to voltage regulation tolerance is zero volts. The ground can have noise on it, but in general it is a well defined logic signal. For similar reasons, high side switching is a preference. It is possible to design circuits fail safe and to be tolerate to at least one fault. I try and design my circuits so that a simple fault will not let out the magic smoke and force a component replacement. One simple example is when driving an LED, it is always best to connect the current limit resistor to the supply rail and the LED in series to the load. The reason is that an accidental connection to ground by a scope probe will not destroy the current limit resistor, but would most likely instantly destroy an LED connected directly to the supply rail. By using high side switching and open collector drives, it's possible to engineer the interface to survive a short to either rail. This is achieved, simply by dividing the drive circuit limit resistors into two separate resistors. One in the collector feeding the high side switch, and one in the base or gate bias to the high side switch. Both resistors limit the current and can usually survive a momentary short to rail, either because the current flow is limited or because the short is to their respective passive rail. (Nothing happens when an open collector is shorted to ground.) Fault mode analysis is always worth doing on any new design, as minor changes in circuit topology can result in a circuit that is fault tolerant and thus far more reliable. During my time as an industrial circuit designer, I attended many circuit peer review meetings. In these, the designer was required to justify the inclusion of every component and explain why a particular value had been chosen. In some cases, this came down to the component being required, but it's value chosen simply to reduce the number of different values included on the bill of materials.

I wanted a brief into to what low/high side switching was all about and you , nailed it! Many thanks.

Thanks for this. For a future episode, perhaps investigate PNP Transistors and P-Channel MOSFETs for high side switching (and how a logic zero turns them on). For NPN and N-Channel - high side switching requires a voltage HIGHER than VCC to FULLY turn them on. For NPN transistors, it's about 0.6 volts higher and possibly several volts higher for N-Channel (depending on type and current). For reference, you described Emitter-Follower (a.k.a. Common Collector) and Common Drain (a.k.a. Source-Follower) circuits. Again, the downside is you cannot switch fully to the positive rail with these circuits, therefore causing higher heat and a lower voltage than expected. The downside to PNP and P-Channel is they are typically less common and more expensive - although I don't think that was always the case. I remember pulling apart radios in the '60s that had positive ground and PNP transistors (yea, I'm old).

i was just watching a video about 20min's ago switching the load on the high side(collector side) or the low side(emitter side) but it was utilizing a MOSFET INSTEAD.. (an N channel one too.) very well explained Paul.. You hit all the pit fall's one would come across using this circuit. WELL DONE! : )

Brief and to the point. Thank you for sharing !

Thanks. I’ve had issues with this topic. Very helpful.

I always use open collector for variable vcc swithcing, but I have never considered the ground v drop! Very good note I need to start considering

Very informative video. Thanks!

Really useful info. You’d be surprised how often this comes up. Most new users always reach for a npn right away then try to switch an common cathode and wonder why it doesn’t work as expected. Likewise, for PNP. One of your best videos. Thank you

Amazing, I had the question, you had the answer clear and concise! thanks!

great stuff thanks as always

You should also consider using PNP BJT.

Great point about the current going to ground.

Useful to know, Cheers Paul.

You didn't mention a huge part of using a high side switch which is your gate potential has to be a lot higher than your drain since the source voltage is no longer 0 volts. Which is why it's easier to just use a pnp or p channel transistor.

Helpful video

Very informative

Thank you this is going to help me a lot

Nice info, thanks :)

Thanks 👍