The book Practical Electronics by Scherz and Monk states (3rd ed., p 48):
A ground bus, or bus bar found in breadboards... serves as an adequate substitute for a single point ground.
Is this statement correct?
Electrical Engineering Stack Exchange is a question and answer site for electronics and electrical engineering professionals, students, and enthusiasts. It only takes a minute to sign up.Sign up to join this community
No, the statement is not correct. A single point ground is used specifically to avoid daisy chained returns, because it addresses problems caused by them.
X is not an adequate substitute for Y, when Y is specifically designed to fix issues with X. At least, not in those circumstances when those issues are a problem.
If bussed returns were absolutely adequate, there would be no point in star grounding, because star grounding is more difficult to lay out and takes up more PCB space, and creates more messy point-to-point hookup cabling when it's done between devices in a chassis.
It is much more tidy to daisy chain devices. Imagine if every light in a Christmas tree had individual wiring all the way back to the power supply. But chaining, convenient though it may be, introduces parasitic interactions between circuits or devices. You can have a ground loop even on a single circuit board. For instance a the power output stage of an amplifier can generate large currents which can appear as a voltage on the reference ground of a sensitive input stage, because the return path happens to be shared.
The only real problem with that statement is that it doesn't give any context. It certainly isn't true for high frequency circuits, but perhaps they figure that if try high frequency circuits on a breadboard you'll have bigger problems anyways.
Basically there is so many other crappy aspects of breadboards (trace to trace coupling, additional capacitance, poor connections, loops of wires ...) that this is the least of your worries for doing performance circuits.
But people do circuits on tose all the time and get results so there obviously is a usefulness to them. In this case it really isn't an issue with whether or not it is bussed, that bus in there can probably handle amps it is the connection to the bus that is the issue.
Most of the frequencies that you deal with on circuits that can run on breadboards are modeled by lumped circuit elements so the distribution of wiring plays less of a factor.
The author should have really stated the limits and their assumptions to avoid that being too broadly applied.
I'm going to play the devil's advocate and say that it is correct, within the context of a breadboardable circuit. As nice as a star ground topology sounds, a breadboard has no practical way of keeping return loops small. The bus bar is thick and relatively wide, and will have low inductance and dc resistance compared to the wires in a star topology. Mildly high frequency power related returns can be localized with capacitors. So yes, I think it is correct, within the context of a breadboard that lacks power planes and alike, and with reasonable inductance on pretty much any connection you make. So I'm going to go on a limb and say that if the circuit works on a breadboard with a bunch of grounding wires, it will also work with the bus bar, therefore it is an adequate and far more convenient replacement.