books and ways recommendation on gaining electronic understanding

MrAl

Joined Jun 17, 2014
11,389
I guess the nodal analysis is not really that hard either, it gives the same solution:

View attachment 272997
Hi,

Yeah if you know a simpler way you could always use that, but if the circuit is more complicated then you need some form of analysis that in itself is organized so it is easier to apply.
One of the things i always like about it is that we are always forming simple parts like:
(v2-v1)/R1
and summing terms like that. So we dont have to concentrate on anything but one circuit element at a time, usually.

Now here is a point about Nodal for that circuit. It's a minor point for this circuit, but for a more complicated circuit Nodal may be the best way to go about solving it.
The solution i got with my program was:
[v1=(E3+4*E2+4*E1)/11,v2=(3*E3+E2+E1)/11]

where E1,E2,E3 are Vin1, Vin2, Vin3, and v2 is the first node on the left and v1 is the output node.
What is interesting is that we got both node voltages at the same time. IF these were both outputs, we would need that information. What else is interesting is we find that there are no resistors required in the solution and that becomes apparent from the solution, when all resistors are the same value. Did you happen to notice that with the other solution? Maybe you did, but again with a more complicated solution it may be harder to figure out.

BTW we set up 'little parts' like (v2-v1)/R1 one at a time but here are even simpler ways to write out the equations where we only look at a single current contribution at a time, so instead of writing v2 and v1 in the same term, we can break them down further in to one term +v2/R1 and the other term -v1/R1. Seems like a moot point but when we go to write the equations we only have to look at ONE end of the resistor at a time rather than both at once.

Also a little interesting is with E1=11 and E2=22 and E3=33, we get outputs of 12v and 15v, both integers. There would of course be other integer solutions.

Your circuit was interesting because it was a little simple but not too simple.
 
Last edited:

dcbingaman

Joined Jun 30, 2021
1,065
Hi,

Yeah if you know a simpler way you could always use that, but if the circuit is more complicated then you need some form of analysis that in itself is organized so it is easier to apply.
One of the things i always like about it is that we are always forming simple parts like:
(v2-v1)/R1
and summing terms like that. So we dont have to concentrate on anything but one circuit element at a time, usually.

Now here is a point about Nodal for that circuit. It's a minor point for this circuit, but for a more complicated circuit Nodal may be the best way to go about solving it.
The solution i got with my program was:
[v1=(E3+4*E2+4*E1)/11,v2=(3*E3+E2+E1)/11]

where E1,E2,E3 are Vin1, Vin2, Vin3, and v2 is the first node on the left and v1 is the output node.
What is interesting is that we got both node voltages at the same time. IF these were both outputs, we would need that information. What else is interesting is we find that there are no resistors required in the solution and that becomes apparent from the solution, when all resistors are the same value. Did you happen to notice that with the other solution? Maybe you did, but again with a more complicated solution it may be harder to figure out.

BTW we set up 'little parts' like (v2-v1)/R1 one at a time but here are even simpler ways to write out the equations where we only look at a single current contribution at a time, so instead of writing v2 and v1 in the same term, we can break them down further in to one term +v2/R1 and the other term -v1/R1. Seems like a moot point but when we go to write the equations we only have to look at ONE end of the resistor at a time rather than both at once.

Also a little interesting is with E1=11 and E2=22 and E3=33, we get outputs of 12v and 15v, both integers. There would of course be other integer solutions.

Your circuit was interesting because it was a little simple but not too simple.
Thank you for your response. I will agree after hearing your arguments for nodal analysis that it may be the better path for solving various circuits than other methods. I stand corrected.
 

MrAl

Joined Jun 17, 2014
11,389
Thank you for your response. I will agree after hearing your arguments for nodal analysis that it may be the better path for solving various circuits than other methods. I stand corrected.
Hello again,

You're welcome, but also i would not want to see anyone give up any analysis they already know either, probably the more the better so we can check our own results. What i try to do sometimes is after the analysis i calculate the currents and see if each node sums to zero as it should. If not, either myself or my program did something wrong :)
Oh BTW it also comes in handy when you wish to do an all symbolic analysis so you can develop formulae for particular circuits.

Thanks for your well chosen illustrated circuit too.
 
Abdulbadii, if you're just a beginner or experimentor and want to get a good hands-on feel for the basics and what components do, without much theory or math, I recommend "Make: Electronics" by Charles Platt. If you're not a beginner, it might be too basic for you. Good luck - I hope you succeed!
 

DickCappels

Joined Aug 21, 2008
10,152
I think you will gain a lot by following @Papabravo 's advice in post #3. Download a circuit simulator and run some circuit simulations.

Some main advantages of using a simulator over building an physical circuit ar
Making a circuit in a simulator is much faster and easier than building a physical circuit

You can get simulated parts free of very cheaply, plus you don't have to wait for shipping.

You can see voltages and current that would be impossible to measure on a physical circuit, and the components can be perfect (ideal) if you want. For example the signal at an opamp's input -try that with a 10 meghom resistor in the feedback path of a 100 MHz signal.

There are many good SPICE-based simulators, LTspice being one of them, and it is free.
Download LTspice

LTspice is used by many people on this website and that makes it easy to get help with it.

Horowitz & Hill:
It is good for starting out but does not help much when you need a new concept for your circuit or system.


Another good simulator is Texas Instrument's TINA
Download TINA

And then there are some other very good ones that cost money.

If you don't have one already, get a good scientific calculator. I have some little ones that only cost U.S. $2 or $3 and they work just fine. With the calculator you can check the results that you calculate to see whether the agree with the output of SPICE>
 

MisterBill2

Joined Jan 23, 2018
18,176
One caution about most simulators is that parts do not fail when overloaded. THis means that an impossible circuit may work very well, but in the real world a transistor will fail if you pass one amp thru a 50 milliamp device. Some much more expensive simulators may catch that, but not the simple free ones.
So just because the simulator works does not always mean that if it is built with real parts that the results will be the same.
 

panic mode

Joined Oct 10, 2011
2,715
unfortunately electronics is rather broad field with a lot of topics. depending on your goals and available time, you may have to pick and choose things you care about. the nice thing is that it can be a lot of fun. and one can have a lot of fun without calculus. and it helps to team up. check if there are groups or individuals close to you that share same interest.
 

dl324

Joined Mar 30, 2015
16,845
I recommend "Make: Electronics" by Charles Platt
From what I've seen of his style, it's awful. Instead of teaching people how to read and create readable schematics, he uses that awful pin order wiring diagram method with colored wires. It looks very childish to me.

Before long, they'd be drawing things like this:
1659712241585.png
I doubt that anyone would be able to determine timer functionality at a glance; as they could from this:
1659712319510.png
IIRC, Platt doesn't use component designators either.

He should be showing them how to form leads properly too:
1659712531823.png
And I disagree with the notion that you learn by "burn things out, mess things up".

Here's a schematic from that book someone sent to me:
1659712934299.png
No component designators, and he's using an NPN transistor to drive an LED high side. And poorly at that... He should also show capacitor polarity.

EDIT: After giving it some thought, I guess not so poorly... He still made poor choices regarding flow. For some reason he likes to put the power connections on the top instead of on the left where it would make more sense.
 
Last edited:

MisterBill2

Joined Jan 23, 2018
18,176
Capacitor polarity is optional when only describing how circuits work. It is required when presenting circuits to be built. How many simulators demand polarity for capacitors??? None of the cheap simulators, it seems.
And vert true about wiring diagrams. They may be useful for those folks wiring something, but useless for all others. Hard to follow and confusing to all.
 

DickCappels

Joined Aug 21, 2008
10,152
You can simulate an electrolytic capacitors in a simulator like LTspice but placing a diode across the capacitor. The only thing missing it the alert when the current becomes huge.
 

MrAl

Joined Jun 17, 2014
11,389
One caution about most simulators is that parts do not fail when overloaded. THis means that an impossible circuit may work very well, but in the real world a transistor will fail if you pass one amp thru a 50 milliamp device. Some much more expensive simulators may catch that, but not the simple free ones.
So just because the simulator works does not always mean that if it is built with real parts that the results will be the same.
That's a good point. Also along those lines is the voltage (or current) levels that may be present intra-circuit that we dont look at sometimes such as the output of a gain block that we didnt need to look at because we mainly were concerned with the output levels. The output of the gain block could go up to thousands of volts with a 5v system in order to push the output to the right level so solve the equations, and we would never know without looking. So it's good to look at all the nodes before calling a circuit 'verified' and certainly before calling it 'ready for production'.
 

Exjay

Joined Nov 19, 2015
166
I was once in your shoe. My best bet is getting books by David Buchla and Thomas Floyd. They both write a series of books in electronics that are very more understanding for beginners. Also, get books written by Louis Frenzel. The art of electronics is like a reference book to most electronics books. Studying it alone would not help you. It has Laboratory manual to be used with the book. There is even another one titled Learning the art of electronics.

Study the book in that order and get the practical manual. Get ready to get your hands dirty doing the circuits in the manual
 

MrAl

Joined Jun 17, 2014
11,389
You basically convinced me, but I still don't see that many advantages in Nodal Analysis. I see something in common with matrix analysis - you might want to point that out.
Hi,

Yes the nodal equations can translate right into matrix form, row by row.
That's often the way it is done anyway.

BTW when i quote some of these things i presume that the user also has math software that does matrix algebra and integration and stuff like that. These days i cant see doing anything without math software it just helps so much especially if you want to experiment with circuits in pure math. Doing that leads to profound discoveries in circuit theory that you cant usually get from a simulator.
 
Top