Hi please i wish to know if am drafting this graph in the right manner ?

189 22

Best answers

How do I draw this graph correctly?

Your graph should be plotted as individual points first, then joined with a smooth curve of best fit, not straight line segments [#21679631][#21679637][#21679638] The points after 4 kHz were placed incorrectly in earlier versions, especially on the logarithmic frequency axis, so check that each value sits at the correct position before drawing the curve [#21679634][#21679637][#21679641] Do not force sharp corners where the response changes slope; the line should curve smoothly through the transition [#21679638][#21679641] For the cutoff frequency, mark the point where the response has dropped to about 0.7 times the gain at the corner where the two straight-line sections meet, and draw the curve through that point [#21679644]

Generated by the language model.

Report a violation of the law

Reply | New topic

#1 21679627 01 May 2018 20:18

NYAME EPHRAIM MECHANE NYAME EPHRAIM MECHANE

Anonymous
Post #1
21679627 01 May 2018 20:18

Hi please i wish to know if am drafting this graph in the right manner ?

Attachments:

forums-graph-sample-pp-100.jpg (277.07 KB) You must be logged in to download this attachment.

forums-capture-voltage-gain-and-frq-180501-141842.png (101.1 KB) You must be logged in to download this attachment.
ADVERTISEMENT
#2 21679628 01 May 2018 21:02

Elizabeth Simon Elizabeth Simon

Anonymous

Post #2
21679628 01 May 2018 21:02

I don't see where you put all the points but the ones I see appear to be in the right places. The other points should be about on the lines so it looks right to me.
#3 21679629 01 May 2018 21:39

NYAME EPHRAIM MECHANE NYAME EPHRAIM MECHANE

Anonymous

Post #3
21679629 01 May 2018 21:39

Hi Elizabeththank you !
please how do i plot the cut of frequency ?
#4 21679630 02 May 2018 01:13

Elizabeth Simon Elizabeth Simon

Anonymous

Post #4
21679630 02 May 2018 01:13

I assume that by "cut of frequency" that you are talking of "cut off frequency"This gets a bit trickier. See the following Wikipedia articles which do a better job explaining than I can.

https://en.wikipedia.org/wiki/Cutoff_frequencyhttps://en.wikipedia.org/wiki/DecibelFor your graph, you are looking for the point where the voltage is half what it is at the lowest measurement. But this is assuming a smooth curve and not lines like you have.
#5 21679631 02 May 2018 14:40

David Ashton David Ashton

Anonymous

Post #5
21679631 02 May 2018 14:40

Ephraim you still do not have this right.Your gain values for 1, 2, 4 KHz at 100K ohms (3rd column) are 100, 99.6, 99However your red lne shows a value of 100 at 1KHz (correct), about 95 at 2 KHz (wrong) and about 85 at 4 KHz (wrong). You need to plot these points first - I would like to see dots at these points. You then join up the dots with a curve, not with straight lines as you have done.I will email you a copy of your graph how I would like to see it (I will try posting it here as well, but it may be too small).You need to get this right, if I was marking you on this I'd probably fail you....Cheers // David(I have replotted the top line in bright red)
#6 21679632 03 May 2018 00:20

NYAME EPHRAIM MECHANE NYAME EPHRAIM MECHANE

Anonymous

Post #6
21679632 03 May 2018 00:20

Hi David,please take a look at the two graphs i have posted.
ADVERTISEMENT
#7 21679633 03 May 2018 00:22

NYAME EPHRAIM MECHANE NYAME EPHRAIM MECHANE

Anonymous

Post #7
21679633 03 May 2018 00:22
#8 21679634 03 May 2018 02:48

David Ashton David Ashton

Anonymous

Post #8
21679634 03 May 2018 02:48

Ephraim...of the two graphs above, the first one is better. Why? because the value of the 10K gain at 32 KHz is 10, you have it down about 9. But apart from 2 and 4 KHz, I see no sign of you putting points on the graph to show your values as I did with my 3 points, and Elizabeth did with her Excel graph. You do not seem to have grasped this. Let's take 16 KHz. This will be about 2/3 of the way between 10 and 20 KHz on the bottom frequency scale. 32 KHZ will be about 1/3 of the way between 30 and 40 KHz. You don't have to be exact but you should put the points on the graph. I'll send you a full size graph in email and try to explain this.
ADVERTISEMENT
#9 21679635 03 May 2018 14:50

David Ashton David Ashton

Anonymous

Post #9
21679635 03 May 2018 14:50

Ephraim...I found the following tutorial on graphing which may help you.It does not have anything on how to plot the points, which seems to be the main thing that you have trouble with.But it does have some good points about other errors you have made, and it is worth watching.Cheers // David
#10 21679636 03 May 2018 23:15

NYAME EPHRAIM MECHANE NYAME EPHRAIM MECHANE

Anonymous

Post #10
21679636 03 May 2018 23:15

Hi David thanks for the reply, please kindly go through the new attached graph
#11 21679637 04 May 2018 02:53

David Ashton David Ashton

Anonymous

Post #11
21679637 04 May 2018 02:53

Well that looks nice. But your points are in the wrong places after 4 KHz.... At 8 KHz your red point is at 7.5 KHz and there is no green point. At 16 KHz your red point is at about 13 KHz - should be more to the right - and there is no green point.
At 32 KHz your red point is right, but there is no green point.
Both your 64KHz points are at about 44 KHz. At 128 KHz your red point could be a little bit more to the right, but your green point is at about 80 KHz. At 256 KHz your red point is right but your green point is at about 180 KHz.Please see email just sent and see if this helps.I get the impression you plotted the lines first and then put the points on? It should be the other way round. And don't worry if your points are not exactly on the line, they may be slightly off. the line should be a "Best fit" with points above and below the line if they don't fit exactly.
#12 21679638 04 May 2018 03:33

Elizabeth Simon Elizabeth Simon

Anonymous

Post #12
21679638 04 May 2018 03:33

And don't forget that the result should actually have a curve up where the lines join.
ADVERTISEMENT
#13 21679639 04 May 2018 21:29

NYAME EPHRAIM MECHANE NYAME EPHRAIM MECHANE

Anonymous

Post #13
21679639 04 May 2018 21:29

Hi David thanks for the reply,please take a look at the correction
#14 21679640 04 May 2018 21:31

NYAME EPHRAIM MECHANE NYAME EPHRAIM MECHANE

Anonymous

Post #14
21679640 04 May 2018 21:31

Hi Elizabeth
thanks for the reply.
#15 21679641 05 May 2018 05:49

David Ashton David Ashton

Anonymous

Post #15
21679641 05 May 2018 05:49

Ephraim that is better , you are getting this now, but:- Your 256K points are still not right, one is at about 210K and the other about 230K- but I see you have put dots at the frequency axis to refer to, use them!- all your points from 8K up are a bit too much to the left, but not much, this is always a problem wehn getting used to Log graph paper- you still have sharp corners where your lines got from flat to downwards sloping. Elizabeth told you above to put a curve on these. Listen to her, she knows more than I do!But I'd probably pass you on that, it is fairly near to what you should have.
#16 21679642 05 May 2018 11:00

NYAME EPHRAIM MECHANE NYAME EPHRAIM MECHANE

Anonymous

Post #16
21679642 05 May 2018 11:00

Hi Elizabeth,please i posted the graph correction on David blog, please go via and also please kind me on the to side curve cheers Ephraim
#17 21679643 05 May 2018 13:04

patricia biem patricia biem

Anonymous

Post #17
21679643 05 May 2018 13:04

the graphs are more explained plotted ...great job...thanks
#18 21679644 05 May 2018 18:36

PeterTraneus Anderson PeterTraneus Anderson

Anonymous

Post #18
21679644 05 May 2018 18:36

For each curve, note the frequency where the two straight lines meet. Calculate or measure the gain at this frequency, and plot the point on the graph paper. Note that the point's gain will be about 0.7 times the gain where the two straight lines meet. Draw the curve through this point to connect the two straight lines.
#19 21679645 10 May 2018 01:30

NYAME EPHRAIM MECHANE NYAME EPHRAIM MECHANE

Anonymous

Post #19
21679645 10 May 2018 01:30

Hi peter,thanks for the comment, please i will appreciate if you can help me with demonstration, regarding your explanation
#20 21679646 13 May 2018 20:01

PeterTraneus Anderson PeterTraneus Anderson

Anonymous

Post #20
21679646 13 May 2018 20:01

Let's stick to the blue gain-of-ten curve and points for this discussion. I forget whether you are simulating or measuring your points. I suggest you simulate or measure the gain and phase-shift at 10, 15, 20, 30, 40, 50, 60, 70, 80, 90, 100, 150, 200 kHz, and plot these points on your graph. Do you see why I am suggesting this?What do the terms "closed-loop gain", "open-loop gain", "loop gain" mean to you?Do you know how to do a paper-and-pencil analysis of your circuit?We need to know what you understand, so we can build on that.
#21 21679647 13 May 2018 23:18

NYAME EPHRAIM MECHANE NYAME EPHRAIM MECHANE

Anonymous

Post #21
21679647 13 May 2018 23:18

Hi peterthanks for the reply, i build and simulated the circuits below are the result for both the 100 and 10 kilo ohms, my next task is to graph the results and identify the cut off frequency thanks
#22 21679648 16 May 2018 23:00

NYAME EPHRAIM MECHANE NYAME EPHRAIM MECHANE

Anonymous

Post #22
21679648 16 May 2018 23:00

Hi peter please look at this attachment, do i need to do the calculation and connect both the voltage and frequency respond like this?
please kindly show me a demonstration on how i can connect the two curves.
#23 21679649 17 May 2018 05:30

PeterTraneus Anderson PeterTraneus Anderson

Anonymous

Post #23
21679649 17 May 2018 05:30

Hi NyameFor 100Kohms, simulate Vo/Vi gain at 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16 kHz and plot the points.For 10Kohms, simulate Vo/vi gain at 30, 40, 50, 60, 70, 80, 90, 100, 110, 120, 130 kHz and plot the points.These points will show you the curves between the straight lines.
Create an account, log in here. You will receive points by participating in discussions.
Join this discussion.

Install Elektroda application

Reply | New topic

Report a violation of the law

Topic summary

✨ The discussion centers on correctly drafting frequency response graphs for electronic circuits, specifically plotting gain versus frequency for different load resistances (10K and 100K ohms). Key issues addressed include accurate placement of data points on logarithmic frequency scales, proper representation of gain values at specified frequencies, and the importance of plotting points before drawing smooth curves rather than connecting points with straight lines. The concept of cutoff frequency is clarified as the frequency where the output voltage drops to approximately 70% (or -3 dB) of its low-frequency value, with references to Wikipedia articles on cutoff frequency and decibels. Participants emphasize simulating or measuring gain at multiple frequencies to obtain detailed data points, then plotting these points accurately on the graph. The need to create smooth curves that reflect the actual frequency response, avoiding sharp corners, is highlighted. Suggestions include using graph paper with logarithmic scales, marking points precisely, and understanding terms like closed-loop gain, open-loop gain, and loop gain. Tutorials and example graphs are recommended to improve graphing technique. The overall goal is to produce clear, accurate frequency response plots that correctly identify cutoff frequencies and reflect the circuit’s behavior under different loading conditions.
Generated by the language model.

Home page
/
Forum
/
Circuit Design
/
EEWeb Topics
/
Hi please i wish to know if am drafting this graph in the right manner ?

FAQ

TL;DR: For Bode plots, mark the corner where straight lines meet, then place a point at about 0.7× the corner gain and draw a smooth join; “Draw the curve through this point.” This nails the −3 dB cutoff. [Elektroda, Anonymous, post #21679644] Why it matters: Students and hobbyists often misread log axes or connect straight segments, which hides the true cutoff and gain behavior.

Quick-Facts

Cutoff (corner) gain ≈ 0.7× the intersection gain; that’s the −3 dB point to plot. [Elektroda, Anonymous, post #21679644]
Plot points first, then connect with a smooth curve; avoid sharp corners. [Elektroda, Anonymous, post #21679631]
Log-scale placement drifts right at higher frequencies; expect slight left-bias errors when new to log paper. [Elektroda, Anonymous, post #21679641]
Around the corner, sample densely (e.g., 10–200 kHz list) to reveal the curve. [Elektroda, Anonymous, post #21679646]
For two load cases, sweep 4–16 kHz (100 kΩ) and 30–130 kHz (10 kΩ) in small steps. [Elektroda, Anonymous, post #21679649]

Quick Facts

Cutoff (corner) gain ≈ 0.7× the intersection gain; that’s the −3 dB point to plot. [Elektroda, Anonymous, post #21679644]
Plot points first, then connect with a smooth curve; avoid sharp corners. [Elektroda, Anonymous, post #21679631]
Log-scale placement drifts right at higher frequencies; expect slight left-bias errors when new to log paper. [Elektroda, Anonymous, post #21679641]
Around the corner, sample densely (e.g., 10–200 kHz list) to reveal the curve. [Elektroda, Anonymous, post #21679646]
For two load cases, sweep 4–16 kHz (100 kΩ) and 30–130 kHz (10 kΩ) in small steps. [Elektroda, Anonymous, post #21679649]

How do I find the cutoff frequency on my hand-drawn Bode plot?

Locate where your straight asymptotes meet. At that frequency, plot a point at about 0.7× the intersection gain. Then draw a smooth curve through that point to connect the flat and sloped regions. This gives the −3 dB corner visually and accurately. “Draw the curve through this point.” [Elektroda, Anonymous, post #21679644]

Is cutoff half voltage or −3 dB (0.707×)?

Use −3 dB for voltage gain, which equals about 0.707× of the flat-region gain at the corner. Half voltage (0.5×) is −6 dB and would mark a different point. Plot the 0.7× point at the corner to show the correct cutoff. [Elektroda, Anonymous, post #21679644]

Should I connect points with straight lines or a curve?

Place data points, then connect them with a smooth curve near the corner. Avoid sharp corners where flat and slope meet. “Don’t forget that the result should actually have a curve up where the lines join.” This better reflects the real filter response. [Elektroda, Anonymous, post #21679638]

What frequency list should I measure or simulate to see the shape?

Sample densely across the expected corner to reveal curvature. A useful sweep for one case is 10, 15, 20, 30, 40, 50, 60, 70, 80, 90, 100, 150, 200 kHz. Record both gain and phase if available, then plot points before drawing the curve. [Elektroda, Anonymous, post #21679646]

How should I sweep for 100 kΩ and 10 kΩ loads?

For 100 kΩ, sweep 4–16 kHz in 1 kHz steps. For 10 kΩ, sweep 30–130 kHz in 10 kHz steps. Plot Vo/Vi at each test frequency to reveal the curved transition between asymptotes for each load. [Elektroda, Anonymous, post #21679649]

I keep misplacing points on log paper—any tips?

On a log axis, equal distances mean equal ratios, not equal Hz. New users tend to place points too far left at higher frequencies. Mark decade and mid-decade ticks first, then interpolate visually before adding dots. Expect small errors and use best-fit. [Elektroda, Anonymous, post #21679641]

Do I plot points first or draw the lines first?

Always plot your measured/simulated points first. Then draw the smooth best-fit curve through them. Avoid drawing lines first and forcing points onto them; that hides real behavior and errors. [Elektroda, Anonymous, post #21679637]

How do I correct a graph where my 8–256 kHz points are left-shifted?

Re-mark exact log ticks for each target frequency.
Re-plot each dot directly above its tick using your gain table.
Redraw the best-fit curve; let dots sit slightly off the line if needed. This fixes systematic left-bias. [Elektroda, Anonymous, post #21679637]

My reference values: what gains should I show at 1, 2, and 4 kHz (100 kΩ)?

Use your tabulated gains as anchor dots: about 100 at 1 kHz, 99.6 at 2 kHz, and 99 at 4 kHz. Plot these explicitly as dots before drawing. Misplacing them skews the curve and the inferred cutoff. [Elektroda, Anonymous, post #21679631]

What’s open-loop vs closed-loop gain in op-amp plots?

Open-loop gain is the amplifier’s gain without feedback. Closed-loop gain is with feedback applied via the network that sets your target gain. Use closed-loop plots to read bandwidth and corner frequency of the finished design. [“Operational amplifier”]

What is loop gain, and why measure it?

Loop gain is the gain around the feedback loop when you break it for analysis. It predicts stability margins from its magnitude/phase Bode plots. Designers use it to assess phase margin and ensure robust behavior across loads. [“Loop gain”]

How steep is the roll-off after the corner?

A typical first-order slope is about −20 dB/decade after the corner. Plot a straight asymptote with that slope, then add the curved join near the corner using your data points. Use denser sampling to confirm the actual slope. [“Bode plot”]

Can I use Excel or SPICE to speed this up?

Yes. Create a frequency table and plot gain in Excel to validate hand plots. Or simulate a SPICE model to generate the same point list before transferring to log paper. Always transcribe dots first, then draw the curve. [Elektroda, Anonymous, post #21679646]

What’s a common failure that ruins accuracy?

Connecting straight segments without the transition curve masks the true −3 dB point. Another failure is drawing lines first and forcing dots to match. Use best-fit curves through correctly placed dots. [Elektroda, Anonymous, post #21679631]

Generated by the language model.