Recent Posts

Hi All,

An updated version of my manual for creating nomograms with the PyNomo software has been released---see

    http://myreckonings.com/wordpress/2009/07/31/creating-nomograms-with-the-pynomo-software/

This Version 1.1 includes the following new and significant features of the latest PyNomo Release 0.2.2:

1.  Automatic spacing of tick marks along scales—more tick marks where space is available and less where it’s crowded.

2.  Drawing of sample isopleths between specified values on scales.

3.  Printing of only significant digits of scale values by default, producing a cleaner-looking nomogram overall.

Ron

Great! Sometimes the polygon transformation tilts a couple of the scales depending on their max/min values, so keep the proportion parameter in your back pocket.

Ron

Ron, I tried the main script polygon statement and it worked. As a result, the relative positions of the axes changed. One may be able to accommodate the desired spacings by using the polygon script in conjunction with adjustments to one or more side axis min/max values. - Dave

Actually, in the example nomogram of three parallel lines on page 12 of the PyNomo essay the middle line is offset slightly from the center, but this was because I included the polygon transformation in the script. Otherwise, there is a block parameter called "proportion" that provides the factor for spacing between lines in a Type 1 nomogram. The default is 1.0 for centering the middle scale ('proportion':1.0,) but specifying a different value in the block parameter set will adjust the spacing (I haven't tried it myself).

Ron

Do you have examples where, in a three variable nomograph, the axes are NOT uniformly spaced? This feature comes in handy, and I have used unequal spacings in manually prepared nomographs. - Dave

Hi Dave,

Wow, you're certainly good at this. There aren't many people who create these things today, and the fact that you once made these by hand puts you into a rare category. I'd be really interested in hearing about any others that you design, so please post when you do. If you have any general tips for other users, or feature suggestions or questions specifically for Leif, you might post them on the PyNomo board of Leif's nomography forum at

    http://www.nomography.org/

Also, if you have any suggestions or questions on the PyNomo essay, feel free to contact me using the Contact form on my blog and I'll email you right back (I'm reluctant to provide an email address here because of my constant battles with spammers here).

I'll be releasing an updated version of the PyNomo essay after Leif finishes up a couple of new features he's working on, and I'll add your recommendation about downloading the full MiKTeX version. Thanks for the heads-up!

Ron

Because you asked: a few examples of nomographs in the field of radiation protection I've done in last month:

1) Co-60 radiography source: dose rate from 1 foot to 200 feet; meter scale option
2) Ir-192 radiography source: dose rate from 1 foot to 100 feet; meter scale option
3) Cs-137 calibration source: 1 inch to 15 feet; cm and meter scale option
4) Field concentration compound nomograph: concentration as function of normalized dispersion factor, windspeed, noble gas release rate
5) Dose rate nomograph: dose rate at point downwind of source as function of field concentration (see 4) and dose rate conversion factor (itself a function of time)
6) Variations on 5) due to accident type.

Had I done these manually I'd still be on the first one.

I anticipate several others, if other company groups are interested, in the areas of Chemistry and Operations (dilution function, reactor coolant boration)

Ron and Leif: I did the MikTek originally as a hands-off install and it went quickly. Later, I did the full install (80M file) which took awhile but gave me no subsequent issues at all. I conservatively recommend that users choose the option that puts everything available into the install.

Insofar as Python and C++, I probably misspoke. My summer intern from RPI had some familiarity with C++ and picked up (as I did) on most of the Python code. My role was to get the entire Pynomo install to work and to generate a few rudimentary 3-axis nomographs for him to sweeten. Later, I asked him to go compound so as to maximize a two-step process calculation process to a one-page method with five axes.

I can tweak the code pretty well, he's gone off to school, but the damage is done - I can use the programs and examples you've provided quite well and quickly. The student found it to be very old-fashioned but challenging and cool. He's never used a slide rule! We installed it correctly in Vista on his laptop before he went back to school. I hope to generate wider interest in my company for Pynomo and nomographs - perhaps as a niche situation.

The only sticky part for me was discovering how to choose the script to get exponential notation for values below 1E-2. Thanks for the leadership on this stuff and the extensive on-line aid. - Dave

Hi Dave,

glad to hear you find PyNomo useful. I guess nomographs are really good tools when lights get out..

Did you mean Python instead of C++?

\leif

Hi Dave,

Thanks for posting about your experiences! Those sound like very interesting nomograms. Can you elaborate on the problems with MiKTeX? I've never installed anything except the Basic version and then let it install packages on the fly as needed, but it sounds like you had to install the full version at the start, which is very large.

I'm using PyNomo extensively at the moment for something else I'm writing, and I'm still amazed by what it can do. PyNomo scripts are really just lists of parameters, and as you say, borrowing from existing scripts rather than trying to remember all the parameters is the fastest way to create nomograms. In fact, it's very rare for me to type a new parameter line from scratch. I'm copying and pasting now from previous scripts and then doing minimal editing for specific differences, and I'm creating a wide variety of nomograms very quickly. Leif has really put a lot of work and care into the software.

Ron