How do equations play a role in circuit design?

[ChromeBit]

I'm new to electronics, but I'm a physics student. I'm assuming that when someone designs a circuit, they use equations to calculate an end result (e.g. the charge on the electron) and they select components that make these changes in the circuit.

For example, is the equation: ΔE=hf involved in a transmitter circuit?
Is the charge on the electron equal to the frequency/planck's constant?

If I'm wrong please correct me.
Can someone explain how I would work out what a circuit does by using calculations?

 

[phinds]

It's not clear from your question what kind of circuit you are talking about, but for normal circuits, one does not even THINK about "the charge on a electron" ... everything is done in volts and amps relative to the components (resistors, inductors, capacitors, transistors, diodes, etc).

 

[analogdesign]

For most run-of-the-mill circuit design (like a physicist would likely do) people typically use a cookbook approach. They get a circuit topology that does kind of what they need and then they used the supplied design equations to tune the circuit to their needs.

This usually works fairly well for basic requirements but can get you into trouble.

 

[Ryan McCarty]

When designing a circuit you do exactly that - you use formulas and calculations to aid you in controlling the flow of electricity in order to achieve a specific task. However, it is rare that we need to use physics this in depth because for most tasks this level of engineering is not needed. Many, many, circuits can be designed to do numerous tasks using basic electronics knowledge such as how volts, amps etc. work, and how components play into these.

 

[berkeman]

I'm new to electronics, but I'm a physics student. I'm assuming that when someone designs a circuit, they use equations to calculate an end result (e.g. the charge on the electron) and they select components that make these changes in the circuit.

For example, is the equation: ΔE=hf involved in a transmitter circuit?
Is the charge on the electron equal to the frequency/planck's constant?

If I'm wrong please correct me.
Can someone explain how I would work out what a circuit does by using calculations?

See if your university library has a copy of this book:

https://www.amazon.com/dp/0521370957/?tag=pfamazon01-20

"The Art of Electronics" by Horowitz and Hill is a nice intro book to electronics. The first chapters are very basic, and you can keep reading if you like, to learn about more advanced circuit design concepts.

 

[analogdesign]

The Art of Electronics is a nice book, but it can be a bit dangerous. In my experience people who have read it (often physicists) think they "know electronics". They will point you toward the "cookbook" approach I mentioned earlier. It often works but it can be hard to recognize when it doesn't...

Reading the Art of Electronics is an excellent introduction to the field of circuits, then you can graduate to more engineering oriented texts if you feel this is something you find interesting.

 

[ChromeBit]

Woah thanks everyone, you were all really helpful! I'll check out that book also.

 

[meBigGuy]

If you read the wikipedia page on transistor http://en.wikipedia.org/wiki/Transistor you can sort of see the simplicity of describing how a transistor acts as a switch vs. the internal semiconductor physics.

Looking at ohms law on hyperphysics would also help give perspective. http://hyperphysics.phy-astr.gsu.edu/hbase/electric/ohmlaw.html

As said above, generally you pick a topology that can solve your problem and then deal with the design equations to zero in on the performance you need. The design equations are generally approximations and are very high level compared to the base physics.

 

[Baluncore]

For example, is the equation: ΔE=hf involved in a transmitter circuit?

E = h.f is from quantum physics. Yes, sort of, but only for the transmission of light from a light emitting diode. The colour of the light is determined by the difference in chemical energy levels within the diode material.

Is the charge on the electron equal to the frequency/planck's constant?

No. Energy and charge are quite different things.

 

[sophiecentaur]

I'm new to electronics, but I'm a physics student. I'm assuming that when someone designs a circuit, they use equations to calculate an end result (e.g. the charge on the electron) and they select components that make these changes in the circuit.

For example, is the equation: ΔE=hf involved in a transmitter circuit?
Is the charge on the electron equal to the frequency/planck's constant?

If I'm wrong please correct me.
Can someone explain how I would work out what a circuit does by using calculations?

Pretty well never. The photon energy associated with Radio Frequencies is very low. The interactions between the electrons in the antennae at each end and the EM waves in between are, essentially classical. QM has no place in the design of such circuits.

I could also say that design usually starts with some sort of basic layout. A block / functional diagram first, then something more detailed and, later, the component values - based on formulae. But, before all that, the basic specification about what you want to achieve would be involve formulae, related to Communications Theory, establishing frequencies, bandwidth, power etc..

There has been some suspicion expressed about the 'cook book' approach and I agree that it can be very limiting in the end. But there are many 'constructors' who get a great deal of pleasure from only a nodding acquaintance with the theory of what they're working on. But lack of theory can be very limiting and even dangerous.

 

[psparky]

I find that the "physics of electricity" and "electrical circuitry" have absolutely everything in common...yet have absolutely nothing in common.

If you learn electricity in your physics class and then try to apply it to a circuits class, you are going to struggle.
If you learn electricity in your circuits class and then try to apply it to your physics class, you are also going to struggle.

Physics explains it all perfectly, but for some reason doesn't help much in circuit design.
Not sure what the disconnect is, but there is definitely a disconnect~! No pun intended.

 

[meBigGuy]

The disconnect is simple. All circuit equations are approximations derived from other approximations and the base physics. Cicruits deal in a world of of high tolerances. +-10% is many times close enough (1% can be considered precision). Transistor parameters, like beta, can easily vary 5:1. Circuit topologies that are sensitive to these tolerances are naturally avoided. Things are "detuned" to give adequate performance. Specifications need to be relaxed.

The physics are precise, circuit design is an art based on approximations and trade-offs.

 

[psparky]

The disconnect is simple. All circuit equations are approximations derived from other approximations and the base physics. Cicruits deal in a world of of high tolerances. +-10% is many times close enough (1% can be considered precision). Transistor parameters, like beta, can easily vary 5:1. Circuit topologies that are sensitive to these tolerances are naturally avoided. Things are "detuned" to give adequate performance. Specifications need to be relaxed.

The physics are precise, circuit design is an art based on approximations and trade-offs.

I was trying to say something like that...thanks for choosing the correct words~~~!

 

[jim hardy]

Physics describes behavior of charges and fields in the whole universe.

Electric circuits on the other hand are restrained to inside of wires and mostly metallic devices, which comprise a very small subset of the universe.
So our everyday circuit formulas have become a sort of shorthand suited to their task.
Ohm's law can be used quite effectively by someone having no familiarity with Maxwell's equations. We electronics folk work in a small world, really. But it's a really fun one.

When i went to engineering school, 1960's, we studied Maxwell's equations in Modern Physics class. I do not remember their being used in any EE course. But that WAS a while ago...

old jim

 

[sophiecentaur]

I was trying to say something like that...thanks for choosing the correct words~~~!

I did like the way you put it though.

 

[meBigGuy]

I liked it too. It was right on.

 

[sophiecentaur]

The disconnect is simple. All circuit equations are approximations derived from other approximations and the base physics. Cicruits deal in a world of of high tolerances. +-10% is many times close enough (1% can be considered precision). Transistor parameters, like beta, can easily vary 5:1. Circuit topologies that are sensitive to these tolerances are naturally avoided. Things are "detuned" to give adequate performance. Specifications need to be relaxed.

The physics are precise, circuit design is an art based on approximations and trade-offs.

What you say is very true and accounts for why people who are not in the business seem to think it's so confusing. Fact is that you couldn't design the simplest amplifier circuit if you felt it necessary to study each component in 'physicist' detail. This middle path is something that Engineers have found to work well in all fields, not just electronics. Engineers invented the Black Box.

 

[Okefenokee]

Honestly, it's getting to a point where can you just look up an inexpensive all-in-one integrated circuit to do what you want. If, for example, you want DC to DC conversion there are thousands of chips for that.

Microprocessors are made now that have their own internal timers. That eliminates many of the external components and PCB layout issues that you used to have to deal with. So many circuit questions on this board could be easily resolved with a microP but there is a learning curve to using them. I got past that learning curve and now I might use a tiny microP where in the past I would have used a 555 timer.

 

[jim hardy]

So many circuit questions on this board could be easily resolved with a microP but there is a learning curve to using them. I got past that learning curve and now I might use a tiny microP where in the past I would have used a 555 timer.

I appreciate that μp's make it easy on the designer.
But not so for any poor fellow who might be tasked to repair the thing. Only the programmer knows what that micro is supposed to be doing.

The few embedded micros I've designed, i included LED status lights to give a clue as to program status and I/O troubles. That concept seemed foreign to some programmers I've worked with.

So please - make your microprocessor leave a clue as to why it quit doing whatever it was doing before it quit doing it ... (is there a grammarian in the house?)

old jim

 

[meBigGuy]

We are doomed to a future where intern programmers will decide how we will interact with the world. This is true at every level, from the gmail user interface (I have a list of 43 terrible things about gmail that google will never fix) to what my microwave does when I press 1. I know what sort of crap goes on inside an embedded uP because I'm guilty of designing the hardware that requires it (which results in removing as much as 1 cent from the BOM). An LED to tell you something when it doesn't work? Fuhgeddaaboudit!

 

[jim hardy]

We are doomed to a future where intern programmers will decide how we will interact with the world.

That's why i have a '68 Ford pickup truck.
No ECU
No Metric Bolts
No Problems !

No payments, either .

 

[psparky]

That's why i have a '68 Ford pickup truck.
No ECU
No Metric Bolts
No Problems !

No payments, either .

Post some pics of that bad boy if you don't mind.

 

[meBigGuy]

I have a 69 landcruiser FJ40. Used to have an electronic ignition (aftermarket) but of course that failed on a trip through Yellowstone. Had to rewire on the road. The old systems still used the points, so I just re-wired the coil. But, I'd rather have a later model with disc brakes. Put that baby in low gear and you can drive through places you wouldn't want to walk. But, not a useful as a pickup so I have a 1 cord capacity trailer for it.