Continue to Site

Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

Logic Design & Switching Theory

Status
Not open for further replies.

PG1995

Full Member level 5
Joined
Apr 18, 2011
Messages
248
Helped
1
Reputation
2
Reaction score
1
Trophy points
1,298
Activity points
3,758
Hi :grin:

My friend is takes a course called Logic Design & Switching Theory. What is it all about? Could you tell me a bit in easy word? Thanks.
 

Ask him for a copy of the syllabus?

If it's an undergraduate class, then the subject matter would be parts like standard logic gates (AND, OR, NOT, NOR, NAND, XOR, etc.), logic devices like multiplexers, counters, and adders. Once you start putting those basic parts together, you can make logic circuits to create decision-based outputs (read from sensors, if some thresholds are met, set some output logic lines, etc).
 

Ask him for a copy of the syllabus?

If it's an undergraduate class, then the subject matter would be parts like standard logic gates (AND, OR, NOT, NOR, NAND, XOR, etc.), logic devices like multiplexers, counters, and adders. Once you start putting those basic parts together, you can make logic circuits to create decision-based outputs (read from sensors, if some thresholds are met, set some output logic lines, etc).

Thank you, enjunear.

The problem is he is out of town and I shouldn't have said he is a 'friend' rather he was. BTW, it was undergrad course.

Does this course also go into technical details how logic gates are formed from transistors etc?
 

How logic gates are made should be part of microelectronics course.
The course you mentioned, should be based on device( and above) abstraction- means it should be considering the MOS as switch.
I can't be sure, but the course should include following:

1) What is MOS and how it works?
2) Basic inverter from MOS.
3) Basic gates from MOS.

If you can refer digital integrated circuit design by Prof Rabey, you can have slight idea what the course should be about.

Regards
Ani
 

Ani, thanks for letting me know that it won't be concerned technical aspects such as how the gates are formed.

I can understand the use of the phrase "Logic Design" which I think means that how logic circuits are to be designed, but what does the phrase "Switching Theory" mean?

Thanks.
 

I told that we can consider my MOS as a switch. now the switch can be switched on and off. if you are electronics student, you will understand switching means charging and discharging of internal and external capacitance.
Now why is this important for logic design? Well.. when we switch on or off any mos, we would like it to be as fast as possible (read clock speed). This speed is limited by external and up to some extent by internal capacitance.
Hence, by knowing switching basics we can modify our MOS CHARACTERISTICS (or simply use off the shelf MOS) to get required speed and functionality.

It's not possible to explain all concepts in a paragraph :oops: , hence i have tried to give you a brief of all concepts while trying to be as clear as possible.. Please prompt me if you didn't understand any point.

Regards
Ani
 
  • Like
Reactions: PG1995

    PG1995

    Points: 2
    Helpful Answer Positive Rating
Ani: Thanks a lot. It was very helpful.

Best regards
PG
 

Thank you, FvM.

Could you someone please recommend a book on LD & ST written for beginners like me who have very little knowledge about such stuff? Thank you

EDIT: Would this book do?
 
Last edited:

Hi again :)

In some of the posts above I was told that the part "Switching Theory" refers to the fact MOS used as a switching device which turns on/off by sensing the input voltage or whatever. MOS stands for 'Metal Oxide Semiconductor". Is this "MOS" a transistor, diode, or what? I think it's a transistor because a transistor can be used as a switch. What is so special about the "Metal Oxide" prefix? Do we use only those semiconductors to switch on/off the circuits which have been made from metal oxide? Please help me with it. Thank you.

Regards
PG
 

Hi again :)

In some of the posts above I was told that the part "Switching Theory" refers to the fact MOS used as a switching device which turns on/off by sensing the input voltage or whatever. MOS stands for 'Metal Oxide Semiconductor". Is this "MOS" a transistor, diode, or what? I think it's a transistor because a transistor can be used as a switch. What is so special about the "Metal Oxide" prefix? Do we use only those semiconductors to switch on/off the circuits which have been made from metal oxide? Please help me with it. Thank you.

Regards
PG

MOS tells you how the layers of the transistor are put together... metal - oxide - semiconductor. Standard BJT transistors (2N2222, 2N2907, etc) are built entirely using doped silicon semiconductors (N-type silicon, P-type silicon, N-type silicon... thus the name NPN, or PNP for the inverse). A MOSFET device (which is a transistor, because it's a FET = Field Effect Transistor) uses a semiconductor for the drain and source nodes, but a metal gate node. The gate metal is separated from the semiconductor by a highly-resistive oxide layer. This is why MOSFETs don't have any gate current requirements (on the order of microamps due to finite leakage through the package materials).

Other metal-semiconductor devices are available, particularly the Schottky barrier diode. This, however, is just a metal-semiconductor sandwich, no oxide between them (so it's not MOS, but a similar concept in terms of the semiconductor physics that make it work).
 
MOS tells you how the layers of the transistor are put together... metal - oxide - semiconductor. Standard BJT transistors (2N2222, 2N2907, etc) are built entirely using doped silicon semiconductors (N-type silicon, P-type silicon, N-type silicon... thus the name NPN, or PNP for the inverse). A MOSFET device (which is a transistor, because it's a FET = Field Effect Transistor) uses a semiconductor for the drain and source nodes, but a metal gate node. The gate metal is separated from the semiconductor by a highly-resistive oxide layer. This is why MOSFETs don't have any gate current requirements (on the order of microamps due to finite leakage through the package materials).

Other metal-semiconductor devices are available, particularly the Schottky barrier diode. This, however, is just a metal-semiconductor sandwich, no oxide between them (so it's not MOS, but a similar concept in terms of the semiconductor physics that make it work).

Thank you, enjunear.

I couldn't exactly understand the underlined phrase. Would you please help me? What are these "package materials"? The ones which are used to glue the main parts together. Thanks.

Regards
PG
 

Thank you, enjunear.

I couldn't exactly understand the underlined phrase. Would you please help me? What are these "package materials"? The ones which are used to glue the main parts together. Thanks.

Regards
PG

A packaged FET is simply a plastic body, or metal can that hold the semiconductor die and external leads/feet in place. These leads are connected to the die by small bond wires, typically gold and 0.7 to 1.3 mil in diameter. These bond wires arch up inside the package, and then connect to the die. Because you have several (at least 3) wire loops inside the package, you can get coupling between them, just like small antennas in air. Beside coupling through electric field propagation (i.e. antennas) inside the package, the material used to to pot, or fill, the device has a finite resistance, so a voltage difference between two bond wires will result in a small amount of current to flow through the highly resistive encapsulating material.

These signal conduction paths will keep you from obtaining an infinite isolation between pins. Since there is a large resistance from the gate to source nodes on a MOSFET, the current that flows between them is very small, but can typically be measured on the order of microamps. The gist of the story is that you will never have 0 conduction between two dissimilar points, but you can minimize the conduction such that it is very, very small, relative to the signals you are measuring/controlling.

Below are some pictures of how die and bond wires are used inside packaged components.
th_131853.gif

glob-top-web_300.gif

**broken link removed**
 
A packaged FET is simply a plastic body, or metal can that hold the semiconductor die and external leads/feet in place. These leads are connected to the die by small bond wires, typically gold and 0.7 to 1.3 mil in diameter. These bond wires arch up inside the package, and then connect to the die. Because you have several (at least 3) wire loops inside the package, you can get coupling between them, just like small antennas in air. Beside coupling through electric field propagation (i.e. antennas) inside the package, the material used to to pot, or fill, the device has a finite resistance, so a voltage difference between two bond wires will result in a small amount of current to flow through the highly resistive encapsulating material.

These signal conduction paths will keep you from obtaining an infinite isolation between pins. Since there is a large resistance from the gate to source nodes on a MOSFET, the current that flows between them is very small, but can typically be measured on the order of microamps. The gist of the story is that you will never have 0 conduction between two dissimilar points, but you can minimize the conduction such that it is very, very small, relative to the signals you are measuring/controlling.

Below are some pictures of how die and bond wires are used inside packaged components.
th_131853.gif

glob-top-web_300.gif

**broken link removed**

Thanks a lot, enjunear. It was a helpful reply.

So, in your view there are two primary reasons. I don't quite agree with the "antennas" reason because constant DC is used to power the device. For antennas to work we would need constantly alternating current. Where am I going wrong? Please help me with it. Yes, I do agree on the second reason of the material used to pot the semiconductor.

Thank you.

Best regards
PG
 

Thanks a lot, enjunear. It was a helpful reply.

So, in your view there are two primary reasons. I don't quite agree with the "antennas" reason because constant DC is used to power the device. For antennas to work we would need constantly alternating current. Where am I going wrong? Please help me with it. Yes, I do agree on the second reason of the material used to pot the semiconductor.

Thank you.

Best regards
PG

FETs are not only used as switches, but can also be used as amplifiers, in which case there can be large (potentially very high frequency) AC signals on the device. Both effects could be present, it just depends on how you are using the device.
 

Thanks a lot, enjunear. You have really helped me without adding to my confusion. I'm much obliged.

Best regards
PG
 

I hope you would have an answer to this one too.

In courses such as circuit analysis one needs to master several techniques such as mesh analysis, nodal analysis, etc. Someone was telling me that logic design course is not difficult in that it does not have such techniques to master. You just have to follow the thread of thought. In plains words, he was suggesting this is not a difficult course. I understand that the term "difficulty/difficult" has more subjectivity to it than objectivity. In your opinion, was that person right? Thank you for your opinion.

Regards
PG
 

Logic design is different in its core concepts because the initial design tends to assume that you are using a low-impedance driver and high-impedance load/receiver, so your signals snap from 0V to 5V/3.3V quickly, and cleanly (not always true in reality). Fundamentally, the concepts of resistance, inductance and capacitance still remain, but are considered to be negligible relative to the speed of the signals.

Logic design courses eventually lead into high-speed digital design, which is where the fundamental circuit elements can really start to affect how the signals look. Logic and electronics design will have different areas of difficulty, but circuits are circuits, so if you try to push a design to it's limits, physics will still win, regardless of how fast it "should" operate.

As far as difficulty goes, that all depends on which topics you can easily wrap your mind around and really understand the topic. But it never hurts to know stuff outside of your core focus area... then, when you are discussing a design issue with another designer in a different area, you can at least follow the basic concepts of the conversation.
 

Status
Not open for further replies.

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top