Author Topic: Interest survey: engineering graph papers  (Read 11826 times)

0 Members and 1 Guest are viewing this topic.

Online jmwTopic starter

  • Frequent Contributor
  • **
  • Posts: 284
  • Country: us
Interest survey: engineering graph papers
« on: October 22, 2019, 01:29:44 am »
This has been an want of mine for a while: good graph paper pads, in designs useful for electronics and EE pen-and-paper design work: for example, full-page Smith charts and reactance graph paper. I can and have printed these out on a laser printer, but it's not the same in feel as a good engineering pad. Seems it was once a commercial item and there are still a few vintage sheets that appear for sale now and then, but I'd pay to get full pads to keep on my desk. Who else feels the same? Anyone have recommendations for print shops to approach that can do special runs like this?
 

Offline ebastler

  • Super Contributor
  • ***
  • Posts: 6202
  • Country: de
Re: Interest survey: engineering graph papers
« Reply #1 on: October 23, 2019, 06:26:04 pm »
I'd pay to get full pads to keep on my desk.

Best to store them in the top drawer, right next to the slide rules.  ;)
 

Offline edy

  • Super Contributor
  • ***
  • Posts: 2385
  • Country: ca
    • DevHackMod Channel
Re: Interest survey: engineering graph papers
« Reply #2 on: November 04, 2019, 09:54:54 pm »
I'd pay to get full pads to keep on my desk.

Best to store them in the top drawer, right next to the slide rules.  ;)

Yes and if you can bring back some cheap Slide Rules that would be awesome as well. If they can make Soroban's for < $5 shipped from China on eBay, surely a slide-rule could be done for same or less....







Let's bring back VINTAGE TOOLS!   :-+

YouTube: www.devhackmod.com LBRY: https://lbry.tv/@winegaming:b Bandcamp Music Link
"Ye cannae change the laws of physics, captain" - Scotty
 

Online thm_w

  • Super Contributor
  • ***
  • Posts: 6278
  • Country: ca
  • Non-expert
Re: Interest survey: engineering graph papers
« Reply #3 on: November 04, 2019, 11:33:04 pm »
Surely one of these are usable: https://www.amazon.com/s?k=graph+paper&ref=nb_sb_noss_2 ?
I can't comment on the other forms as I've never used them outside of schoolwork. Maybe thread can be converted into a poll.
Profile -> Modify profile -> Look and Layout ->  Don't show users' signatures
 

Offline SilverSolder

  • Super Contributor
  • ***
  • Posts: 6126
  • Country: 00
Re: Interest survey: engineering graph papers
« Reply #4 on: November 04, 2019, 11:43:24 pm »

When I get the very infrequent urge to use graph paper, I print it out using an inkjet.  Laser doesn't make thin enough lines, doesn't print all the way to the edges of the paper, and the laser plastic toner lies on top of the paper, rather than being infused into the fibers - not the same thing at all!

 

Offline CatalinaWOW

  • Super Contributor
  • ***
  • Posts: 5172
  • Country: us
Re: Interest survey: engineering graph papers
« Reply #5 on: February 26, 2021, 10:42:58 pm »
You can buy graph paper of most types from this source. 

https://www.draftingsteals.com/keuffel-esser-paper.html

The minimum buy is a little off putting, but not terrible and it isn't in pads, but that can be corrected with a little ingenuity or a trip to a local print shop.

I don't know whether these guys bought the rights from the people who bought K&E or if they are just pirates, but probably no one much cares.
 

Online jmwTopic starter

  • Frequent Contributor
  • **
  • Posts: 284
  • Country: us
Re: Interest survey: engineering graph papers
« Reply #6 on: February 28, 2021, 09:04:36 pm »
I meant graph paper for visualizing impedance properties that are pretty specific to electronics design. I find it faster to eyeball corner frequencies using a graphical guide instead of using a calculator or sketching up a simulation. Here is that design. I haven't really moved forward as I don't know if there's enough demand, and to do them right requires lithographic offset printing, which has a big setup cost.1183978-0
 

Offline SilverSolder

  • Super Contributor
  • ***
  • Posts: 6126
  • Country: 00
Re: Interest survey: engineering graph papers
« Reply #7 on: February 28, 2021, 09:12:19 pm »
It looks nice - I like it!   Reminds me of the kinds of graphs seen in older instrument manuals.

I'd definitely inkjet something like that out on a nice high quality sheet of paper, then I would use my nice medium firm underlay to make it feel even nicer to use than a pad.  Alternatively, one could put it on top of a plain pad, if the "feel" of a pad floats your boat?

Any chance of a PDF of the finished sheet?
 
The following users thanked this post: CJay

Offline gnuarm

  • Super Contributor
  • ***
  • Posts: 2218
  • Country: pr
Re: Interest survey: engineering graph papers
« Reply #8 on: February 28, 2021, 10:02:51 pm »
Wouldn't it be better to have the graph paper on the computer screen in a way for you too draw on it without a pen and ruler and all the added errors? 

It's not often I want to draw a graph that isn't already drawn for me, but when it happens, I don't want paper.  I want it on the computer. 

What I typically end up doing is creating a table of the data in a spread sheet and letting it draw the chart for me.  Tools give me curve fits and many things that would be inaccurate done by hand.  I have even gone so far as to draw a line on top of a chart when it's too much trouble to extract the subset of data to use for the linear regression.  A spread sheet is only a B+ tool for graphing.  A more dedicated tool or a spread sheet addin would be nice.

A perfect example of this is fitting a line to the daily infection rates to estimate when we could be rid of COVID if we just kept doing what we were doing.  The curve showed we would be rid of this disease in the US in just two more weeks.  I don't need graph paper when I can draw this in a spread sheet.  I call it the idiotic asymptote. 
Rick C.  --  Puerto Rico is not a country... It's part of the USA
  - Get 1,000 miles of free Supercharging
  - Tesla referral code - https://ts.la/richard11209
 

Offline gnuarm

  • Super Contributor
  • ***
  • Posts: 2218
  • Country: pr
Re: Interest survey: engineering graph papers
« Reply #9 on: February 28, 2021, 10:09:58 pm »
I meant graph paper for visualizing impedance properties that are pretty specific to electronics design. I find it faster to eyeball corner frequencies using a graphical guide instead of using a calculator or sketching up a simulation. Here is that design. I haven't really moved forward as I don't know if there's enough demand, and to do them right requires lithographic offset printing, which has a big setup cost. (Attachment Link)

I do that sort of stuff in my head.  I can get the order of magnitude and 1 significant digit in about 3 seconds.  You have to remember the inverse of 2 pi is about 1.6 (x10^-1) which is not too hard.  The inverse of 6 is about 1.67, the inverse of 7 is 1.4, so the inverse of pi is right about the middle of those two, 1.6.  The rest is like using a slide rule.  The only thing the slide rule does for you is to give you three significant digits rather than one.
Rick C.  --  Puerto Rico is not a country... It's part of the USA
  - Get 1,000 miles of free Supercharging
  - Tesla referral code - https://ts.la/richard11209
 

Online jmwTopic starter

  • Frequent Contributor
  • **
  • Posts: 284
  • Country: us
Re: Interest survey: engineering graph papers
« Reply #10 on: March 01, 2021, 04:30:02 am »
I've tried both inkjet and "digital offset" printing and both were unsatisfactory. Process color (CYMK) and fine lines are a poor combination because the halftoning process creates the appearance of a broken line where you wanted a fine solid line. It really needs to be run on an offset press with spot colors to look right. PDFs made for print often don't look right on screen, since free PDF viewers don't handle spot color overprinting correctly. These seem like small issues, but having some experience in graphic design, they stick out.  I'm still considering what to do with this project, including the PDFs.

This type of graph paper isn't for plotting data points, it's for understanding systems. I'm pretty fast with LTspice, but I reach for this chart for first-pass understanding of passive networks. I'm not sharp enough to figure out resonance frequencies, characteristic impedance, and Q factor in my head and this gets me the answer quick. What happens if a R/L/C value changes? I can compare the result immediately, compared to several clicks and keystrokes to re-run a simulation. I also like having a physical sheet of paper to markup and come back to later.
 

Offline SilverSolder

  • Super Contributor
  • ***
  • Posts: 6126
  • Country: 00
Re: Interest survey: engineering graph papers
« Reply #11 on: March 01, 2021, 05:28:18 am »
I've attached a PDF resolution test chart that will outperform any printer I've seen...   the trick is to use vector graphics, I guess - there is nothing inherent in PDF that limits resolution (other than possibly the size of your hard drive!).  PDF is an awesome format.  -  anyway, the attached file is good for setting up your printer to give its best.  It took me a few tries, and I thought it was working OK when I started!

The accuracy of offset printed spot color overlays is not as good as the accuracy of an inkjet print - you can't avoid the paper "breathing" between runs, slight registration differences, etc. etc.  - compared to everything being laid down at exactly the same time.  The type of paper matters a lot, as does the type of ink (dye is better than pigment) and the general state of the printer (rollers, heads, etc. etc.)

The only downside of inkjetting some nice paper with your design is that the dye isn't waterproof, so if you touch it with wet fingers (coffee?  bordeaux? scotch?) things get real messy real fast.

 

Offline Berni

  • Super Contributor
  • ***
  • Posts: 4922
  • Country: si
Re: Interest survey: engineering graph papers
« Reply #12 on: March 01, 2021, 07:04:13 am »
I dunno i never been a big fan of graphs hand drawn on paper. Its tedious to mark down a bunch of points and then draw a line trough them that vary in wobbliness depending on how well you got it right. I prefer computers drawing graphs for me (and even that is not always as simple as it sounds when very large data sets are involved)

I find regular 5mm grid "engineering graph paper" the best for my needs. Works well for drawing schematics with a bit neater straighter lines, its perfect for drawing digital timing diagrams, also works well for drawing data packet structures as grid blocks, visualizing some geometry math etc. But still the stuff i draw on these tends to be pretty sketchy since i never intend these drawings to become actual documentation for something, mostly just visual aids to help me understand things.
 

Offline SilverSolder

  • Super Contributor
  • ***
  • Posts: 6126
  • Country: 00
Re: Interest survey: engineering graph papers
« Reply #13 on: March 01, 2021, 04:01:02 pm »
I dunno i never been a big fan of graphs hand drawn on paper. Its tedious to mark down a bunch of points and then draw a line trough them that vary in wobbliness depending on how well you got it right. I prefer computers drawing graphs for me (and even that is not always as simple as it sounds when very large data sets are involved)

I find regular 5mm grid "engineering graph paper" the best for my needs. Works well for drawing schematics with a bit neater straighter lines, its perfect for drawing digital timing diagrams, also works well for drawing data packet structures as grid blocks, visualizing some geometry math etc. But still the stuff i draw on these tends to be pretty sketchy since i never intend these drawings to become actual documentation for something, mostly just visual aids to help me understand things.

I agree, that's how I do things "in the real world" too.

My HP inkjet has a feature where it generates exactly that kind of graph paper from its menu, you don't even need to use the computer to do it.  A funny remnant from when HP was an engineering company!

Please forgive the crappy photo of the output, it looks much better in real life.  Indistinguishable from using a pre-printed pad.
« Last Edit: March 01, 2021, 04:24:53 pm by SilverSolder »
 

Online jmwTopic starter

  • Frequent Contributor
  • **
  • Posts: 284
  • Country: us
Re: Interest survey: engineering graph papers
« Reply #14 on: March 01, 2021, 05:29:17 pm »
As I said, this isn't for plotting (x,y) data. :) Below is an illustration of what I might use it for, something I find easier than calculating or simulating when I quickly need to know the shape of things or get approximate numbers. SilverSolder - this also shows the printing problem: this was done on a digital press, and the halftoning process has broken the fine brown lines. Same thing happens on the full page Smith charts I made. Everything is vector in the PDF output, it's the printer that is rasterizing to do process color and can't keep up.

1184488-0
 

Offline gnuarm

  • Super Contributor
  • ***
  • Posts: 2218
  • Country: pr
Re: Interest survey: engineering graph papers
« Reply #15 on: March 01, 2021, 06:33:05 pm »
As I said, this isn't for plotting (x,y) data. :) Below is an illustration of what I might use it for, something I find easier than calculating or simulating when I quickly need to know the shape of things or get approximate numbers. SilverSolder - this also shows the printing problem: this was done on a digital press, and the halftoning process has broken the fine brown lines. Same thing happens on the full page Smith charts I made. Everything is vector in the PDF output, it's the printer that is rasterizing to do process color and can't keep up.

(Attachment Link)

That's a very nice image, clear, no distortion... except for those black dots and lines someone drew on it.  What's up with that? 

Seriously though.  What is that diagram telling you that you didn't already know from knowing the schematic?
Rick C.  --  Puerto Rico is not a country... It's part of the USA
  - Get 1,000 miles of free Supercharging
  - Tesla referral code - https://ts.la/richard11209
 

Offline SilverSolder

  • Super Contributor
  • ***
  • Posts: 6126
  • Country: 00
Re: Interest survey: engineering graph papers
« Reply #16 on: March 01, 2021, 06:51:31 pm »
I'm warming up to that tri-axial graph!  -  so you plot the impedance of L, the impedance of C, they cross at resonance, and R sets the "roof" of the curve - very neat!  :-)

I printed out your screenshot from before, just for comparison.

Of course, I discovered that one of the print nozzles is not working (it is visible in the vertical brown lines at high magnification), so the results are not quite as good as they can be - but close enough that I'll upload it anyway.

The printout looks much better in "real life" than on the scan.

 

Online jmwTopic starter

  • Frequent Contributor
  • **
  • Posts: 284
  • Country: us
Re: Interest survey: engineering graph papers
« Reply #17 on: March 01, 2021, 07:29:33 pm »
Right, in series impedances add, so if one is much greater than the other, you can take the maximum as an approximation (x + y ≈ x when x >> y). With parallel impedances, it's the opposite: you take the minimum (x || y = xy/(x + y) ≈ xy/x = y when x >> y). The approximations are reasonably good until x and y are closer than a decade.

I learned this technique for sketching Bode charts without needing arithmetic from a power electronics course. If you want to go beyond impedance and look at the transfer function across a voltage divider that yields H(s) = Z2 / (Z1 + Z2), you can compute the division by taking the difference between two lines drawn on the paper (since it's a log scale and log (x/y) = log x - log y).
 
The following users thanked this post: SilverSolder

Offline gnuarm

  • Super Contributor
  • ***
  • Posts: 2218
  • Country: pr
Re: Interest survey: engineering graph papers
« Reply #18 on: March 02, 2021, 01:42:41 am »
Right, in series impedances add, so if one is much greater than the other, you can take the maximum as an approximation (x + y ≈ x when x >> y). With parallel impedances, it's the opposite: you take the minimum (x || y = xy/(x + y) ≈ xy/x = y when x >> y). The approximations are reasonably good until x and y are closer than a decade.

I learned this technique for sketching Bode charts without needing arithmetic from a power electronics course. If you want to go beyond impedance and look at the transfer function across a voltage divider that yields H(s) = Z2 / (Z1 + Z2), you can compute the division by taking the difference between two lines drawn on the paper (since it's a log scale and log (x/y) = log x - log y).

The trouble is impedances only add as vectors, not scalars.  That's why the impedance of series capacitance and inductance goes to zero at resonance, they are equal and opposite, so canceling.  That is also why the maximum approximation is good far from resonance, the magnitudes are unequal enough that the difference is approximately the greater one.  If the real component of the "real" devices becomes significant, all bets are off because the phase of the current or voltage is no longer close enough to the ideal. 

This is not limited to reactance.  The same thing is true for resistance.  The larger resistance in series dominates and the smaller dominates in parallel.
Rick C.  --  Puerto Rico is not a country... It's part of the USA
  - Get 1,000 miles of free Supercharging
  - Tesla referral code - https://ts.la/richard11209
 

Offline SilverSolder

  • Super Contributor
  • ***
  • Posts: 6126
  • Country: 00
Re: Interest survey: engineering graph papers
« Reply #19 on: March 02, 2021, 03:52:13 pm »

I guess the point is this is a "sketch", not a precision calculation like you would do on your computer?

I suppose that nowadays, using a computer to do this accurately will probably be faster than making a sketch!  :D
 

Offline CatalinaWOW

  • Super Contributor
  • ***
  • Posts: 5172
  • Country: us
Re: Interest survey: engineering graph papers
« Reply #20 on: March 02, 2021, 09:58:16 pm »
There are reasons for using graph paper to this day.  While it may not match your style, it is a tool us old timers grew up with and like.  And even for you young guys, for ad hoc measurements it is often faster than setting up an automated measurement and generating the data fully automatically, or hand measuring data and entering into a spreadsheet.  If you are plotting analysis results the computer is the way to go, but getting data into a computer is a non-trivial exercise unless you are repeating the same set of measurements over and over.  Again, horses for courses.
 

Offline SilverSolder

  • Super Contributor
  • ***
  • Posts: 6126
  • Country: 00
Re: Interest survey: engineering graph papers
« Reply #21 on: March 02, 2021, 10:07:25 pm »

I like using pen and paper for drawing/sketching ideas, it is much faster than using a computer.  Once you have a design in mind, the computer is a good tool to actualize it.
 

Offline gnuarm

  • Super Contributor
  • ***
  • Posts: 2218
  • Country: pr
Re: Interest survey: engineering graph papers
« Reply #22 on: March 04, 2021, 06:37:18 am »

I guess the point is this is a "sketch", not a precision calculation like you would do on your computer?

I suppose that nowadays, using a computer to do this accurately will probably be faster than making a sketch!  :D

To make a pencil drawing on graph paper, the first thing I would need to do is to go out and buy a pencil.  Then I'd have to dig through the basement to find the graph paper that must be squirreled away somewhere, maybe with the 8008 computer board?

I get what sketches are for.  And I lament they aren't quite as easy to do on a computer as they are to doodle on paper.  But I just don't have a use for them on paper anymore.  It's a bit like entering contact information in my Palm pilot.  No way to get it into my computer from there. 
Rick C.  --  Puerto Rico is not a country... It's part of the USA
  - Get 1,000 miles of free Supercharging
  - Tesla referral code - https://ts.la/richard11209
 

Offline Berni

  • Super Contributor
  • ***
  • Posts: 4922
  • Country: si
Re: Interest survey: engineering graph papers
« Reply #23 on: March 04, 2021, 07:24:12 am »
There are reasons for using graph paper to this day.  While it may not match your style, it is a tool us old timers grew up with and like.  And even for you young guys, for ad hoc measurements it is often faster than setting up an automated measurement and generating the data fully automatically, or hand measuring data and entering into a spreadsheet.  If you are plotting analysis results the computer is the way to go, but getting data into a computer is a non-trivial exercise unless you are repeating the same set of measurements over and over.  Again, horses for courses.

I don't need to set up automated SCPI remote control to my instruments to be able to use a computer to graph stuff.

I can still just open up an excel sheet and manually punch in numbers from my multimeter. And i often do that when i have small numbers of data points to work with. Then once they are in a spreadsheet i can easily also do math on all of the data points at once. No need to retype them into a calculator to do math. Then at the push of a button turn those numbers into a graph.

What i find paper useful for is quickly sketching ideas down to get the big picture of my idea. Like drawing a quick rough schematic so i can look at it and imagine current flowing around it to see if it could work in theory at least.

But when it comes to anything bigger i will instead go draw on a computer. Start up CorelDraw and have at it. You can draw pretty fast in on a computer once you learn how to properly use a vector graphics tool. On top of that you also get the benefit of digital design to be able to move things around, copy paste, scale things, quickly make arrays and grids etc...

So yes pencil and paper is still an awesome tool for quickly putting data from your head into the physical world. We will definitely keep using it far into the future, but for some things we found better ways of doing in the last 50 years.
 

Offline SilverSolder

  • Super Contributor
  • ***
  • Posts: 6126
  • Country: 00
Re: Interest survey: engineering graph papers
« Reply #24 on: March 04, 2021, 01:02:18 pm »
Yep, you just knock up a sketch with a charcoal pencil, then enter it into the computer using your mouse.   Oh, wait....

 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf