Making Runtime Plots

[pryan]

Hello,

I'm interested in making runtime plots for my lights. I understand that I need a light meter with an output. I'm currently thinking of a MeterMan light meter that has an analog output.

How do I go from that to a plot? I understand that I probably need something like an oscilloscope that has an RS-232 output so I can hook it up to my computer. I don't want to spend $500 for a scope that can do this, though.

How do you guys do this? What hardware do you use? What software do you use? I see a lot of plots that use ScopeView, is that free? A quick Google for ScopeView didn't turn up anything relevant. A few different CPF searches came up empty.

Thanks,
Patrick

 

[gadget_lover]

There are several light meters with RS232 outputs. They usually come with the software to plot the data.

Daniel

 

[pryan]

Any recommendation for a decent light meter with RS-232 output and the software for somewhere in $0-$200?

 

[Roy]

What I use is a Recording DMM with a serial output. It came with a logging and plotting program. For the light meter I use a EXTECH mdl 401021 Lightmeter Adapter. My adapter is calibrated such that at 1 foot one mVDC equals one Fc. A version that is calibrated in LUX is also available.

Here's a picture:

Look here for mor details on the Extech.

The DMM is a Radio Shack model that is out of production....but they probably have a replacement.

 

[pryan]

Have the RS-232 on the DMM would be far more useful than only on the light meter. I'll poke around for a DMM with an RS-232 output.

 

[Roy]

An advantage to a recording DMM with a serial port, is you can use a simple solar cell (Radio Shack solar power demo kit) as a light detector. That is what I used for the first year of my plotting...just remember that the vertical axis of the plot will be in mA and not LUX or Fc.

 

[pryan]

Well, I just ordered a MeterMan 38XR DMM and a MeterMan light meter. I'll post my first plot here. /ubbthreads/images/graemlins/smile.gif

I was thinking it would be very cool to also collect voltage, amperage and temperature as well as lux, but I don't think I can afford that many DMMs with RS-232 outputs. /ubbthreads/images/graemlins/smile.gif

 

[Heath]

Radioshack sells a cheap ($50) meter with RS-232 output here.

 

[NewBie]

I use one of these light meters, http://www.extech.com/instrument/products/400_450/407026.html

Check Newark, Mouser, Allied Electonics for pricing, it varies.

The bad thing about solar cells is typically their open circuit voltage down 0.64% per degree C, and the current goes up by 0.1% per degree C. Shinning any sort of powerful beam at them causes a very significant significant error to develop, most especially with incandescent sources.

Additionally, many of them are non-linear in their output, see http://downloads.solarbotics.com/PDF/sunceramcat.pdf page 17 graph 2. If you look at page 9, you'll note that the current is highly non-linear as well as you get a very significant error between incandescent and fluorescent sources.

In many light meters you will find devices such as these:
http://www.vishay.com/docs/81519/81519.pdf
That are linear over seven decades. Additionally, they have a color correction filter to approximately correct for the human eye response.

If you look at page 3, you will see a short circuit current line that is very linear. The proper way to utilize a PN junction for the most accuracy is to run it into a transimpedance amplifier (op-amp with no input resistor, and just a feedback resistor to set the gain). On a side note, for speed of measurement you reverse bias the photodiode, which drives the "plates" further apart (or widens the no-mans land), which reduces the device capacitance.

Another device is found here, http://www.osram.convergy.de/scripts/product_family.asp?CLSOID=10069&FAMILYOID=20009

 

[Roy]

Lets not make it any harder to get started in measuring RUNTIME than necessary. If all you want to do is measure the BATTERY RUNTIME, then the photocell and logging DMM is all you need. If you want to add light output vs time to the mix, then yes, you will need something better and much more expensive than a simple photocell.

When we talk about making "runtime plots", we need to understand the difference between battery runtime and light output vs time plots. "Runtime" can be calculated from both types of plots.

 

[NewBie]

Yes, but unfortunately, for comparision purposes, light outputs based on a solar cell are inherently not accurate.

It further complicates things when comparing two different flashlights, especially incandescent.

45 minute runtime doesn't mean much. Some LED flashlights will run for a very long time at reduced output. So, what is the threshold for saying the runtime is over? If you pick 50% light output, and used a solar cell, and someone else utilized a light meter, and you did the measurement with a M6 flashlight, a fella could see differences...

 

[NewBie]

I forgot to mention, an incandescent's output shifts to longer wavelengths as the battery drains.

As such, a solar cell might show a flatter runtime, since the solar cell is more sensitive to longer wavelengths. On the other hand, a Lux meter has a filter in it, that filters out the longer wavelengths, thus the output would drop faster with the light meter as the bulb shifted to longer wavelengths.