Need guidance with electronics assignment!

chelleteng · 2012-09-13 03:45

Hello,

I am currently studying a course on Attachment not found.Attachment not found.electronics and I need some help & guidance with my assignment just to point me in the right direction into solving these questions. I find these questions very difficult for a degree leveled course, does anyone think so too? I would really appreciate some pointers to help me understand this course better.

Thank you!

Bean · 2012-09-13 04:49

I looked at the document, and it looks like typical problems for an analog course.
What area do you need help with ?

Bean

chelleteng · 2012-09-13 05:06

At the moment, I am having problems with designing the active filter from the given bode plots. I know that I have to determine the transfer functions from the bode plots first, but I am not sure how to proceed from there.

And there is one question regarding the worst case output offset voltage, I am not too sure how to go about that.. Thanks!

william chan · 2012-09-13 06:33

Something is not right.

Those questions should have been for Masters Degree level, not first degree level.

Most electronics graduates I know cannot answer those questions off hand without first researching the internet.

Leon · 2012-09-13 07:05

They are first year first degree level. At a UK university, at any rate.

chelleteng · 2012-09-13 08:56

I would appreciate if I could get some help understanding these problem sets still though.. Thank you..

Bits · 2012-09-13 09:32

I agree chelleteng this is about you needing help not what is being tough in school. There is noting wrong wiht using the internet to get help, in fact it makes you the smart one.

At the moment, I am having problems with designing the active filter from the given bode plots. I know that I have to determine the transfer functions from the bode plots first, but I am not sure how to proceed from there.

I will offer a little help as I see it, but want you to learn so no easy answer from me

. Try google transfer functions, then google active filter. Then let me know what part you still don't understand.

And there is one question regarding the worst case output offset voltage, I am not too sure how to go about that.. Thanks!

What part is difficult, is it your understanding of offset voltage? Or is it the way to obtain the offset voltage given the worst case?

SRLM · 2012-09-13 12:49

I can't help, beyond suggesting that you may also want to try posting in one (not all!) of these reddits:

http://www.reddit.com/r/electronic_circuits/
http://www.reddit.com/r/AskElectronics/
http://www.reddit.com/r/electronics

jmg · 2012-09-13 14:48

chelleteng wrote: »

And there is one question regarding the worst case output offset voltage, I am not too sure how to go about that.. Thanks!

I'd call c) a poorly worded set of questions.

For i) I would instead say look up the typical and worst case ib, Iio Vio
for ii) I would say Calculate the typical Vos contribution due to Vio effects only.

I would add
For iii) and iv) use the cumulative corner cases, ie any error effects sum, not cancel.

for iii) I would say calculate the typical Vos contribution due to Ib,Ios effects only.

for iv) I would say calculate the overall typical and worst case Vos values.

There is also no mention of the precision needed.
Given Typical/worst case values are often > 2:1, and the circuit values given, I would add a precision hint, to allow any brighter students to gain a little time here...
A 1% precision would be a good hint.

chelleteng · 2012-09-13 17:07

jmg wrote: »

I'd call c) a poorly worded set of questions.

For i) I would instead say look up the typical and worst case ib, Iio Vio
for ii) I would say Calculate the typical Vos contribution due to Vio effects only.

I would add
For iii) and iv) use the cumulative corner cases, ie any error effects sum, not cancel.

for iii) I would say calculate the typical Vos contribution due to Ib,Ios effects only.

for iv) I would say calculate the overall typical and worst case Vos values.

There is also no mention of the precision needed.
Given Typical/worst case values are often > 2:1, and the circuit values given, I would add a precision hint, to allow any brighter students to gain a little time here...
A 1% precision would be a good hint.

Thank you for your help! I understand the theory of DC imperfections of offset voltages and currents but I'm still a little unsure about the concept of worst case situations, could you explain to me about it? Thank you.

chelleteng · 2012-09-13 17:08

Bits wrote: »

I agree chelleteng this is about you needing help not what is being tough in school. There is noting wrong wiht using the internet to get help, in fact it makes you the smart one.

I will offer a little help as I see it, but want you to learn so no easy answer from me . Try google transfer functions, then google active filter. Then let me know what part you still don't understand.

What part is difficult, is it your understanding of offset voltage? Or is it the way to obtain the offset voltage given the worst case?

Thank you! I'll look up regarding active filters and I'll get back to you. I'm a little unsure regarding obtaining the offset voltage given the worst case as I'm not too familiar with the concept of worst case.

jmg · 2012-09-13 17:51

chelleteng wrote: »

Thank you for your help! I understand the theory of DC imperfections of offset voltages and currents but I'm still a little unsure about the concept of worst case situations, could you explain to me about it? Thank you.

That's why I reworded the questions

There is production variation spread worst-case (usually tagged MAX) in the Data sheet and Typical (tagged TYP) - but when you have multiple parameters involved, then you need to also consider how they combine/add.

This introduces another concept some might call worst case, but I prefer to call a "cumulative corner case", to avoid confusion with a MAX worst case.

Let's say we have an imaginary OpAmp with Typical Vos of 2mV, and a Iib of 1uA and a Ios of 20nA, those can have any sign, but a cumulative corner case considers them to add in the worst possible way.

Then I can type this into my calculator (to ~ 1%)

2m + 1u*(110k/2) - 1u*(0k/2) + 20n*(110k/2) ans = 0.0581 => 58mV cumulative corner case, Typical column.

and that might also give numbers like this, from the MAX column.

4m + 2u*(110k/2) - 2u*(0k/2) + 60n*(110k/2) ans = 0.1173 => 117mV cumulative corner case, MAX column.

The question of how might one change/add any circuit passives, to improve these values, is left as an exercise for the student.

william chan · 2012-09-13 19:19

Seems like analog theory and knowledge is losing its relevance amongst product designers nowadays.

For example, if you need an audio amplifier in your product, would you design one with op-amps and transistors or would you use an amplifier IC?
Same thing goes with telephone hybrid circuits, radio freq circuits, etc.

Heater. · 2012-09-13 19:26

In our same office building is an office full of application engineers from MPS. They help MPS customers implement designs using MPS switch mode power supply and class D audio amp chips. All digital right?

Turns out these guys know thier analog inside out, they have to, it's that aspect of design the customers have most trouble with.

chelleteng · 2012-09-14 06:51

jmg wrote: »

That's why I reworded the questions

There is production variation spread worst-case (usually tagged MAX) in the Data sheet and Typical (tagged TYP) - but when you have multiple parameters involved, then you need to also consider how they combine/add.

This introduces another concept some might call worst case, but I prefer to call a "cumulative corner case", to avoid confusion with a MAX worst case.

Let's say we have an imaginary OpAmp with Typical Vos of 2mV, and a Iib of 1uA and a Ios of 20nA, those can have any sign, but a cumulative corner case considers them to add in the worst possible way.

Then I can type this into my calculator (to ~ 1%)

2m + 1u*(110k/2) - 1u*(0k/2) + 20n*(110k/2) ans = 0.0581 => 58mV cumulative corner case, Typical column.

and that might also give numbers like this, from the MAX column.

4m + 2u*(110k/2) - 2u*(0k/2) + 60n*(110k/2) ans = 0.1173 => 117mV cumulative corner case, MAX column.

The question of how might one change/add any circuit passives, to improve these values, is left as an exercise for the student.

Oooh okay, I think I understand slightly better now. So, just so that I'm on the right track - the worst-case output offset voltage that I could obtain from the circuit is the highest offset voltage calculated from however the combination of the input bias and input offset currents are.....? Although, why is it that the typical values are used to calculate part iii and not the MAX values?

chelleteng · 2012-09-14 07:02

And regarding Question 1b that asked to design an active filter circuit that corresponds to the respective Bode plots; I did some reading and found that these plots shows the characteristics of a low-pass filter... However, I'm still a little unsure on how to obtain the transfer functions from these plots. Do I have to work backwards or is there another way to go about it?

jmg · 2012-09-14 14:23

chelleteng wrote: »

Oooh okay, I think I understand slightly better now. So, just so that I'm on the right track - the worst-case output offset voltage that I could obtain from the circuit is the highest offset voltage calculated from however the combination of the input bias and input offset currents are.....?

Yes, that I call the cumulative corner case - the worst possible summing of all errors.

chelleteng wrote:

Although, why is it that the typical values are used to calculate part iii and not the MAX values?

Good question, that is why I said the questions were poorly worded.
If I was answering this I would give both cumulative corner case answers, for the TYP and MAX columns and let the Prof decide what he really meant.

In a real design you probably want both

- here the typical cumulative corner case shows a flaw in the circuit, and gives what you should expect on a multimeter, and the MAX cumulative corner case is what you might put in a spec sheet, or set your QC limits to, in production, or use for a Trim-Range calculation. You would rarely (if ever) expect to actually 'see' a MAX corner case, unless a part was damaged/faulty.

chelleteng · 2012-09-14 18:18

jmg wrote: »

Yes, that I call the cumulative corner case - the worst possible summing of all errors.

Good question, that is why I said the questions were poorly worded.
If I was answering this I would give both cumulative corner case answers, for the TYP and MAX columns and let the Prof decide what he really meant.

In a real design you probably want both

- here the typical cumulative corner case shows a flaw in the circuit, and gives what you should expect on a multimeter, and the MAX cumulative corner case is what you might put in a spec sheet, or set your QC limits to, in production, or use for a Trim-Range calculation. You would rarely (if ever) expect to actually 'see' a MAX corner case, unless a part was damaged/faulty.

Alright! I get it now! Thank you so much!

Do you think you could guide me with these other questions as well? Especially the one with the Bode plots that I've mentioned my problem above..

Need guidance with electronics assignment!

Comments