Sciencemadness Discussion Board

Voltage Data to text file

uber luminal - 18-5-2005 at 16:50

Has anyone worked with any equipment,( or know of a way to make something work) that takes voltage data at timed intervals and records it to a text file?

Like say I want to output the data from my thermocouple to a PC so I can plot the data. (or other tests like this)

12AX7 - 18-5-2005 at 18:07

Hm, don't know offhand but I'm sure it would be easy with an ADC, parallel port and a quickie QBasic program or something.


Twospoons - 18-5-2005 at 22:06

I'd say this would be exactly what you want.

For $A25 it really doesn't get any easier than that!

[Edited on 19-5-2005 by Twospoons]

Archimede - 31-5-2005 at 18:18

And if you need to measure voltages above 20 Volts you can divide the voltage at the probe by 10 using 2 resistors in series. extending the range to 0-200V.
(you have to add a zero to the reading tho)
Probably values such 10k Ohm and 100K Ohms would work just fine (1/4W).

the schematic is very simple:
test point----100k Ohm---probe--- 10k Ohm--- ground

The cost is around 50 cents for the parts.

12AX7 - 31-5-2005 at 18:28

Well, you need to factor the meter's internal resistance as well. This is paralleled with the bottom resistor of the divider. Then you can calculate the proper ratio (most likely something like 90k to 10k for a 100k impedance, which at 200V would dissipate 0.4W, 1/2 or 1W resistors are needed here).


ELEKTOR (english) - ELEKTUUR (dutch)

Lambda - 1-6-2005 at 04:50

Originally posted by uber luminal
Has anyone worked with any equipment,( or know of a way to make something work) that takes voltage data at timed intervals and records it to a text file?

For many years I have had a suscription to one of the leading "practical" electronic magazines called "ELEKTUUR" being the dutch version. The english version is called "ELEKTOR", german, spannish etc. versions are allso available. These magazines and books are literaly loaded with "analogue to digital converters" (ADC) with the software needed for thoughs who would perfere to stay braindead in respect to programing, designing the printed circuit board (PCB) and schematic design (no insult intended).

I have wants build one with multiple input poorts, and multiple dollars, wich fitted directly into my PC bus connector. I forgot what I payed for it, but the ADC-chipset and the PCB was the main purs buster at about $150-200. VELLEMAN, a belgium firm sells them now for a lot less. If you are interested in such a "self build project", I will make inquiaries for you. However, if you would like to sail on your own steam, check this out on eMule.


You will find complete elektor ISO and NRG CD-ROM images of the complete year publications (12 magazines) of this magazine from 1989 - 2004. Allso parts of 2005 are available.

The 207 gaget books often contain (newest version) ADC's, at a lot cheaper price than I paid for it 8-9 years ago. These chips have really plumitted since. DON'T forget to download the software compilations. Nowadays you should be able to build such a converter for about $25-50 using your RS 232 port.

BEST TIP !!!!, just use your microphone input of your soundcard as a ADC, this is how I measure my loadspeaker designs. Nice waterfall charts, 3-D images you know, really spunky !. I am not shoure (havn't read it yet) but "THE AUDIO COLLECTION" would most probably have a design or two on the soundcard system usage. You will find this book to if you search for "ELEKTOR", it's something like 250-350 mb. I don't know if the software is included, if a RAR or ZIP file maybe, otherwise seperately. I have nearly all the downloads, exept the spannish ones (about 8-9 GB). Just got all the stuff in this week, and havn't had the time to sort it out yet.

If you hit a brick wall during your journey, then I will be happy to dive into this mater for you.

Last but not least, a $5 dollar digital thermometer is a ADC, the digital display can be ripped out and the binary signal may then be translated by software to a RS 232 poort. There are many ways to Rome we would say in Europe.
Wars are won by the tip of a fountain pen, and not with guns !

[Edited on 1-6-2005 by Lambda]

uber luminal - 1-6-2005 at 06:28

yes, I am aware of the mic being used. In fact I know this works, just by plugging a type k thermocouple into a minijack converter and right into the mic port. problem with this is that I fear damaging my equipment.

The area that some of the tests will be done in, will have a lot of noise, which may drown out the signal or damage my stuff. Specifically high frequency from initiating plasma will be used. I have been told that this will damage any computer its hooked up to. I also wonder if it would be difficult to filter the signal without filtering the signal if that makes sense.

For now, I am trying to get ahold of Labview software, and a 4 channel serial input device. I will find an old PC that I do not care about, and see what happens around high freq start.

Lambda - 1-6-2005 at 06:50

Uber Luminal, you can use glassfiber optics, or even an IR tunnel to transmit to your remote computer. The sensor electronics, can be built into a well shielded metal casing, and spike signals can be quenched by diodes (or even zener diodes). Schotchky diodes are very fast acting, used in switching power supplies. VDR's (voltage dependent resistors, tin oxide based) or even faster acting gas arresters can be used. They do however tend to ignite premeturely in highfrequency electromagnetic fields. A thermocouple has a very low internal resistance, wich would make it less subceptible/sensitive to stray currents. Trying to measure plasma with any solid receptor will undoubtably lead to meltdown of the sensor. These sensors are allso rather slow acting, if high frequency and heat play a major roll in respect to the isolation needed for protection.

Personly I would aim for an IR sensing device. The amature stuff on the market (cheap) goes to about 500 deg. C. In the lab I used IR thermometers to measure rocket propellants and powder, wich accuratly measured temperatures of 3000-4000 deg.C. The price tage on this specialized equipment was mind boggeling. Second place, to a more commercial IR thermometer would be that used in the steal tampering technology. You are talking about plasma. The stuff mentioned above is in comparison icy, icy cold stuff. What I assume you are talking about has a starter at maybe 5000-6000 deg.C.

[Edited on 1-6-2005 by Lambda]

unionised - 1-6-2005 at 12:29

I don't think anyone would be trying to measure the temperature of a hot plasma with a thermocouple- it would boil.

Filtering out high frequency trash from the sigal would be relatively easy. A couple of small resistors and a capacitor would do the job.
The real problem is that thermocouples produce tiny voltages. You would do better to use a dedicated interface (with such nice features as cold junction compensation), or at least to build an amplifier.

Lambda - 1-6-2005 at 13:01

Indeed Unionised, the voltages per K is minute. Nichrome/nickel being in this whole scala of metal combinations one of the higher rated, it's max temperature and voltage output still sucks. Distinguiching between signal and distortion due to extremely high magnetic induction fields (creap currents) will be a very hard (imposible) nut to krack.

Laser or IR detector/measurment strategies may still work best, seeing the envirioment and source being measured. Maybe even magnetic detectors, seeing the strong relation between fieldstrength and current.

I am not yet into hydrogen bomb building, but during the second world war, they allready had quit a good idee, about fusion temperatures, probably due to extrapolation of there calculations. Nuclear physics, and astrophysics reseach manuals would give a pretty good idee on what they use and how the results a interpreted. With a strong emphesisis on interpretation.

A thermocouple can be used, but the heat radiation is focused on this element via mirrors (parabolic mirrors). Archimedis new that, did he not ?.......

Is it the blind man leading the lame man, or the lame man leading the blind man ?

[Edited on 1-6-2005 by Lambda]

Twospoons - 1-6-2005 at 14:20

Having worked with thermocouples in electrically noisey environments I can recommend a few tricks that will help.
1/ twist the thermocouple as tightly as you can. This will help to minimise loop pickup.
2/ thread the thermocouple several times through a ferrite ring core. This helps reduce the common-mode pickup that would be a problem for your computer.
3/ Electrically shield the system as much as possible, by using metal boxes for electronics, and copper or stainless braid over the thermocouple (as much as possible).

uber luminal - 2-6-2005 at 00:59

thanks twospoons for the tips. I will see how this turns out.

Er.. and I never said anything about measuring plasma with the thermocouple, but to correct a statment about boiling the thermocouple... that might be difficult to do.

I use type K, C and D. While it would really depend on where I was measuring in the plasma and how much current was involved, the C and D types might not even melt, let alone boil.

However, the thermocouples will be used to measure environmental temps within the chambers. I do not see it practical to convert voltages to data and then into light so I can transmit them into a computer since the box doing the conversions will likely cost more than the computer collecting the data. (yea, $20)

I think I will just expect to see a void of data when the high frequency is engaged since some of the listed solutions will take me into a price range, where I do not wish to go. (above $30 US)

Tacho - 2-6-2005 at 05:51

Uber luminal,

1) The LM35 is a 3 pin chip (looks like a transistor) thermometer that converts degrees centigrades to milivolts: 23,5ºC becomes 235mV. If you connect it’s +/- to a 9V battery, you can read the temperature in your cheap digital multimeter. It’s range is up to 110ºC.

2) The LM 331 is a precision voltage-to-frequency converter. I have also used the old, cheap and widely available 555 chip as a voltage-to-frequency converter, there are plenty of circuits on the net. So, the higher the voltage input, the shorter is the period of the output pulse, because frequency is higher.

3) Unfortunatelly, you will need a operational amplifier between both to bring the voltages to a workable range. I could not get good results connecting the LM35 directly to the V/F converter. Be my guest to try. Maybe you have luck. I usually wire my op-amp to multiply the input by about 10.

Get datasheets using google.

4) connect to an input pin of the parallel port of your computer (use 5V supply to your circuit or optical insulator like the TIL111- otherwise your computer goes kaput). There are plenty of web information on connecting stuff to the parallel port. It’s easy.

5) write a short DOS program using pascal, c or basic (Turbo Pascal is an ancient programming language which supports the parallel port pretty easily, but there are C routines on the net) to read how long the parallel port pin stays “high” in each cycle (measure the period), convert to a temperature number and write it to a text file.

There you go! 3 chips (U$ 5,00), some connectors, wire, and 250 hours of wiring, learning pascal, programming, debugging and cursing Tacho for this shitty idea.

Another shitty idea: Very old PC computers had a game port. This port woul read the electrical resistance between two pins and return a value between 0 and 255. Ancient language "basic" for DOS had functions that would return these values. So, a thermistor changes its resitance with temperature; your program reads the number between 0 and 255, convert it to temperature using this incredibly complicated function, and stores it in a text file.

Good luck.

Archimede - 6-6-2005 at 17:35

Feed us some specs about Voltage/Amps/Frequency/Temp ranges so we can look into the best component for the job.

Defining a problem within it's boundaries

Lambda - 6-6-2005 at 18:14

Originally posted by Archimede
Feed us some specs about Voltage/Amps/Frequency/Temp ranges so we can look into the best component for the job.

I very much agree with you Archimede.

In science, when you ask the right question, you allmost have the solution. I to, find it rather vague Uber Luminal.

If you would be so kind as to be more specific about the conditions and interpretations of your experiment, then we can all attack that bull by the horns.

You will then be able to gain fast and efficient access to your solution.
The biggest inventions, often have the simplest solutions !

[Edited on 7-6-2005 by Lambda]

uber luminal - 6-6-2005 at 18:54

Originally posted by Lambda
Originally posted by Archimede
Feed us some specs about Voltage/Amps/Frequency/Temp ranges so we can look into the best component for the job.

I very much agree with you Archimede.

In science, when you ask the right question, you allmost have your answer. I to find it rather vague Uber Luminal.

If you would be so kind as to be more specific about the conditions and interpretations of your experiment, then we can all attack that bull by the horns.

I appreciate your eagerness to help me, but I don't think I am being vague...

Besides I already told you what voltage, frequency (as noise) and temp range anyway.
If you scroll up, you might see me explain, what I need it for, and what I am using. The temp should not matter for the data input since the heat will not even be near the inputs to a computer. I am using Type C , D and K thermorcouples. Voltages for these can be looked up.

Frequency noise is not really an issue anymore. I will just not use the HF starter to initiate plasma. (i can just strike the tungsten off the copper plate)

Twospoons - 6-6-2005 at 20:44

I suppose the other question that should be asked is - Why such a small budget? Is it a personal project or is there a company footing the bill?

Often overlooked in a commercial R&D environment is the cost of your time, which generally far outstrips the cost of the equipment. Sometimes cheap gear isn't cheap simply because of the amount of faffing around needed to get it to do what you want.

If you are likely to want to do this sort of thing again you might want to get some proper 4-20mA current loop transmitters for your thermocouples. That will make most of your electrical noise problems simply 'go away'.

uber luminal - 7-6-2005 at 06:54

small budget because its not included in the project funds. I agree that my time on the job is worth more than the time it would take to piece something together.

However I am playing with this stuff, on off hours.
Time spent messing around, while learning about it is far better than time spent moping around my apartment reading blogs (or science forums :P) watching TV or whatever. Its hard to go home from work everyday and do something education... It gets old real fast to want to do it. However when you get a little project going on, its quite easy to do.

So overall, yes this comes from my pocket. and I just dropped $2000 USD into a piece of equipment intended to heat things (which I can hardly justify), so I am lacking any real funds of my own right now.