Sciencemadness Discussion Board - Arduino

Arduino

gooby - 12-8-2012 at 03:02

I received my Adafruit Arduino experimenter's kit a few days ago and am very excited about this; getting up to speed on circuit theory so I can know just what I'm doing when I get in there...

Does anyone else have an Arduino board? What have you done with them? Are there any tools I should have above and beyond the stuff that comes with the package (here are the contents: http://www.adafruit.com/products/170). I'm thinking about getting an inexpensive multimeter but I'm not sure what one I should get or whether I really need it for my purposes at this juncture...

Incidentally one of the nice things about Arduino is that you can get it to interface with RS232. A lot of chemical laboratory equipment speaks with RS232 as well. That was the original motivation behind getting an Arduino ... "Santa's little helpers".

zoombafu - 12-8-2012 at 08:00

I have an arduino as well and I love it.. I have mine set up with a probing system (temp and pH). Go watch the videos by Jeremy Blum, they are pretty good at explaining how to use it.

gooby - 15-8-2012 at 08:09

Quote: Originally posted by zoombafu

I have an arduino as well and I love it.. I have mine set up with a probing system (temp and pH). Go watch the videos by Jeremy Blum, they are pretty good at explaining how to use it.

It will be a long time before I can really invest in a chemistry lab (for a number of reasons) but thank you, that's great.

Paddywhacker - 15-8-2012 at 23:44

It's something that I've been wanting to get into. You have seen the Youtube tutorial videos?

A possible project would be an auto-titrator.

A more ambitious one would be an Arduino-controlled fractional distillation setup. Monitor the drip rate to control the temperature. Monitor the temperature slope (and/or the drop count) to control changing the reciever, monitor for abnormal conditions to abort.

Yo-Yo - 20-8-2012 at 06:21

I have several arduinos and use them liberally. The reason I first bought it was to use as part of a diy centilever scale. It is possible with small means to build a balance with quite high presition and sensitivity by bouncing a lightbeam on a reflecting centilever and then reading the reflected light using aphotodiod. Obviously, the amount of light hitting the detector is dependent on the reflection angel and hence on the load of the centilever.

Yo-Yo - 20-8-2012 at 06:23

I have several arduinos and use them liberally. The reason I first bought it was to use as part of a diy centilever scale. It is possible with small means to build a balance with quite high presition and sensitivity by bouncing a lightbeam on a reflecting centilever and then reading the reflected light using aphotodiod. Obviously, the amount of light hitting the detector is dependent on the reflection angel and hence on the load of the centilever.

smaerd - 20-8-2012 at 12:22

Paddywhacker I was thinking of something pretty similar. I was thinking of a way to rig an arduino to automate a reaction to be with-in a certain temperature range using a heating mantle.

Paddywhacker - 21-8-2012 at 23:19

Quote: Originally posted by smaerd

Paddywhacker I was thinking of something pretty similar. I was thinking of a way to rig an arduino to automate a reaction to be with-in a certain temperature range using a heating mantle.

What I like is that you have so many input channels. You can put a thermocouple or thermister at the top of your reflux condenser to switch the power off if it rises, protecting against loss of coolant, condenser flooding and thermal runaway in your reaction.

You can put other thermocouples around to monitor the heating bath and to monitor for fire.

And you can get a nice record of your temperature curves to display in Excel.

Double points if you can write it up here with pictures.

gypsyNo2 - 23-8-2012 at 02:11

This might interest some people

http://myspectral.com/

Enjoy

GypsyNo2

IrC - 13-9-2012 at 18:01

Quote: Originally posted by gypsyNo2

This might interest some people

http://myspectral.com/

Enjoy

GypsyNo2

I have been checking for a while and I still see no software available. Do you have any idea when they will make it available? From the page: "We are working on final touches to the opensource software made with Processing language. We will provide a generic tool, which you can fork and modify for your own applications". Been a while and still no links. All I see is too vague to aid in building one, looks more like an ad for a product they will have for sale rather than a cool project we could build.

As you know this is just a pile of parts with no code for the Arduino. Without any application software for the computer to display data there is even less of an incentive to build this design. Sad because I really want to be playing with the completed project, what a useful tool this would be.

On a side note I have 3 Arduino Duemilanove's and a couple UNO's. As well I have several prototyping shields, sensor shields, 4 servo, 4 relay, as well as you name it about every type of sensor shield out there. My problem is to afford all these toys I must buy them on fleabay from China which means nothing in terms of data or documentation. Things like schematics, and templates for the boards showing interconnections such as on the prototyping shields.

For example one has two switches, two LED's, SMT foil patterns but to know how to actually wire a circuit I have to trace each board out. With crappy vision and tight foil patterns this is nearly impossible. Where can I find useful information since the sellers are virtually worthless on this topic. Once in a while I can look at the auction pages of a dozen different sellers listing similar items and glean some information. I am really tired of searching online because what I find is usually too general to help with a specific item I have.

Still the best way to go if you are getting into this area since for example the $85 package listed by gooby is a lot to pay. If you look on the site for say a single UNO R3 it is twice the cost of my fleabay sellers and they do not give the data I am looking for if I bought their item any more than the sellers in China do. You have to buy the whole package to get the paperwork with the board. I don't know maybe I should bite the bullet and pay twice the cost for another Arduino Duemilanove or UNO in one of the experimenter kits just to get the data they send with it.

Anyone out here with paperwork or software they can post for some of these boards?

I should make a plug for the fleabay seller bunnyhey, so far they do the best at providing what they can if you remember when paying to add a note requesting it. Not to mention some of the lowest prices. The order I received from them yesterday took 6 days coming from China to land in my PO box in Arkansas, not too bad at all.

[Edited on 9-14-2012 by IrC]

watson.fawkes - 13-9-2012 at 18:16

Quote: Originally posted by IrC

Things like schematics, and templates for the boards showing interconnections such as on the prototyping shields.

Duemilanove documentation is on the Arduino site page for it. See the links under "Schematic & Reference Design". I've never needed the EAGLE files (the CAD system); the schematic itself has always sufficed. The Uno docs are on there as well somewhere (Duemilanove shows up easier in a search).

I'd guess most of the shields you've got are simple clones of open source designs. If you just started collecting all the docs for such boards, you should be able to match them up pretty easily.

IrC - 13-9-2012 at 19:14

Thanks watson.fawkes, pretty much I have downloaded local copies of nearly that entire site. My greatest need I have yet to find is templates for two of the prototyping and sensor shield boards. Since they plug on top of the Arduino the pins are easy just by going from the pins on the Arduino. But a template which shows where all the foil paths from the SMD patterns on the prototyping boards is impossible to find forcing me to actually trace the paths one by one. Like I said very hard when my vision is not so good any more.

"I'd guess most of the shields you've got are simple clones of open source designs."

I cannot find a diagram of the sensor shield boards in anything generic on that site which corresponds to the ones I have. This will be much more difficult to trace than the prototyping boards. I just wish the sellers out there would do better at providing documentation for their products.

"I'd guess most of the shields you've got are simple clones of open source designs. If you just started collecting all the docs for such boards, you should be able to match them up pretty easily."

Yes that's something I found out a while back by searching, which has been of great help.

Forgot to add I just received 3 of the Arduino Mega 2560's yesterday. Been thinking what a great board with all the I/O to do something like turn a house into one of those Christmas displays like the Wizards in Winter display I say on Utube a few years ago. I do wish they would get up to 12 bit A/D on board for some of the discriminating metal detector ideas I am playing around with.

The sensor shield board (I bought 2 of them) was only $5.88 free shipping. However nothing in the form of documentation is provided by the seller, forget their name but a different seller than the one I mentioned previously.

[Edited on 9-14-2012 by IrC]

triplepoint - 13-9-2012 at 20:59

IrC said: I have to trace each board out. With crappy vision and tight foil patterns this is nearly impossible.

IrC, have you considered photocopying or scanning the boards? You may be able to blow them up to a more manageable size to see, as well as having a convenient place to write notes. I don't know for sure that it would work, just a thought.

IrC - 13-9-2012 at 21:16

Would be handy if I had a blank board. Not so easy stuffed with pin headers and other parts.

Still, not a bad idea. Think I'll dig out my closeup lens assortment and one of my Sony DSC-F717's. They do closeups well and blowing it up would at least make some of the tracing routes simpler. Why didn't I think of that.

[Edited on 9-14-2012 by IrC]

watson.fawkes - 14-9-2012 at 06:18

Quote: Originally posted by IrC

I cannot find a diagram of the sensor shield boards in anything generic on that site which corresponds to the ones I have. This will be much more difficult to trace than the prototyping boards. I just wish the sellers out there would do better at providing documentation for their products.

That looks like the sensor shield for the Electronic Brick series that Seeed Studio sells. Schematic available on their wiki page for the product.

The expectation that someone cloning an open source design to say so explicitly in not based in reality. They're putting in as little effort as possible already. Their whole business model is about appropriation; it's not like they have any interest at all in touting that, or even disclosing it, for that matter.

IrC - 14-9-2012 at 15:56

Thank you very much. That information for the sensor shield is exactly what works with what I see now is a clone. I had no data whatsoever until you gave that link and after a little study I see it perfectly matches what I bought. What you say makes sense and explains why they suck at providing data for their products. Now I can take full advantage of the features on the two I have. No doubt it does not make the original designer happy but I am glad they put out that Wiki page.

watson.fawkes - 15-9-2012 at 06:35

Quote: Originally posted by IrC

No doubt it does not make the original designer happy but I am glad they put out that Wiki page.

A specific strategy to consider. When you buy future things in this area, you could select a product from the manufacturer's site, check the documentation to ensure that it's what you want (and that you have a copy in the first place), and then go shopping for a clone. This is a perfectly rational strategy to optimize the gap you've already experienced, given the facts on the ground about how these things are designed and manufactured.

Now on the other hand, there are voices from the OSHW (open source hardware) movement that consider cloning a Bad Thing and are trying to exert some kind of moral suasion not to do it and not to buy it. This is nothing but a modern day temperance sensibility, a kind of abstinence teaching. We all know how those have all worked out in the past. These attitudes just aren't well grounded in harsh reality. When you publish not only schematics, but also board art and a detail bill of materials, it's a fantasy not to expect clones.

The interesting question, to my eye, is whether this model is sustainable. The presence of clones means that you shouldn't expect revenue past an initial period of product introduction, before it's know whether the idea has traction or not. And there's a middle period (where we are now), where brand awareness gives a certain amount of revenue in the presence of clones. At the very least, it's clear that any sustainable pattern here has a large churn of individual products. So the more specific question is whether there's enough revenue in the early stages to pay for the engineering. It seems that there may be.

The further irony is that Seeed Studio is located in Shenzhen, a likely location for where the folks that cloned their board are. And even more, Seeed has their own versions of the Arduino, some plug-compatible with standard-pattern shields (Seeeduino), some in different form factors. Their company itself leveraged itself off other people's work, including not only the hardware design, which they modified, but also the software, which they use in its entirety.

smaerd - 20-9-2012 at 06:07

So I put some time into thinking of the idea I had earlier(maintaining a reaction temperature in +/-5*C). Here's what I came up with but before I consider investing I'd love to hear a little input from the people on here

.

I decided I would put a relay after the rheostat/glas-col controller and before the heating mantle. In another circuit I decided I would have a slide potentiometer in parallel with an arduino and a stainless steel thermistor probe(which is in a flask's side-arm or condenser via thermometer adapter) . Thermocouples in general need an integrated circuit to work with arduino's which would add another 20$ to the design whereas a thermistor just needs some code tweaking.. The arduino controls the relay in the other circuit.

How it should function:
A slide potentiometer will be calibrated to degree's Celsius and have a scale next to it so the temperature can be set. The arduino reads the resistance of the potentiometer and stores this value as the temperature to maintain. It then reads the thermistor and if the temperature exceeds the value by 4-5*C it opens the relay for the heating mantle effectively cutting the power. If the temperature is lower then or equal to the set temperature value it closes the relay or does nothing if the relay is already closed. It could also be programmed to track temperatures maybe, such as if a run-away is occurring and the temperature leaps 10*C in 1 second it could cut the power and maybe sound off a cheesey buzzer until a button is pressed saying that it's okay.

I was thinking it would be nice to set up a little LCD to display the current temperature and the set temperature range but I have never worked with anything like that before. It would probably help with debugging as well.

My biggest concerns are, would using a relay to cut power be enough? Would a glas-col rheostat be able to handle this? I would have ideally used a voltage controlled resistor but looked into it and it seems mythical, then switch gate capacitors flew over my head so I figured off and on was more with-in my range.

[Edited on 20-9-2012 by smaerd]

watson.fawkes - 20-9-2012 at 08:20

Quote: Originally posted by smaerd

a stainless steel thermistor probe [...] Thermocouples in general need an integrated circuit to work with arduino's which would add another 20$ to the design
[...]
My biggest concerns are, would using a relay to cut power be enough? Would a glas-col rheostat be able to handle this? I would have ideally used a voltage controlled resistor but looked into it and it seems mythical

I don't want to be insulting, but there are two pathways here. (1) Get temperature regulation as simply as possible, or (2) Learn how to build temperature regulation so that you can do more than simple COTS (commercial off the shelf) solutions. In this case, trying to roll your own to make something that's cheaper than COTS is not worthwhile. For you, I recommend option (1) because it's clear to me that you don't know enough about the field to make option (2) anything but very frustrating. You might consider option (1) as the very cheapest education course on temperature control that it's possible to purchase.

Option (1). The COTS solution is three parts (a) a commercial, panel-mount (1/16 DIN), Chinese-made PID controller from eBay. These can be had for $40 or less, hardly more than an Arduino, and they already have a display and software that works. Part (b) is a temperature sensor, either RTD or thermocouple, that's compatible with the PID. Inexpensive thermocouples can be had for less than $5, though you might want to pay more for longer lead length and longer probe length. Part (c) is a solid state relay, also Chinese-made, for $6-8. Those are the main parts, though you also need a case, switch, etc. Nevertheless, this is easily the cheapest entry, assuming you value your time at more than $0.25 per hour.

Option (2). Design and build one. Only do this if you need more than simple thermostat control. There are various reasons to do this, but if you don't already know the reason, you probably don't have one. You haven't stated one, certainly. And you have greatly underestimated the utility and importance of calibration in a piece of gear like this. But you have stated one feature that's not in standard controllers, namely, runaway sense. One feature does not rise to a proper reason, in my book. And frankly, this feature seems like a kind of false insurance, since if you're not anticipating a runaway, you're probably also not anticipating many other serious problems. But whatever. If you want to try this, I'd recommend the osPID project, an open-source PID controller. You can augment its existing firmware with your feature. This will be much cheaper than redesigning a controller from scratch.

And now for the newbie concerns:

Never use a mechanical relay for variable-heat applications. They'll wear out very quickly. That's the reason for using an SSR with the PID controller.

As long as your supply voltage works with your mantle at 100% duty cycle (that is, just plugged in), switching it on and off won't be harmful. There can be lifetime issues with switching certain kinds of heating elements, but the nichrome family aren't one of them.

As for variable resistors, OUCH! Using a resistive ballast for variable heating is just crazy. Simply apply Ohm's law and estimate how much power would need to be dissipated in the ballast resistor.

smaerd - 20-9-2012 at 17:39

No insult taken, rather, much appreciated. I don't want to de-rail the thread(as now my interests are not arduino based) but I highly value your advice and will be looking into your suggestions.

The runaway idea was just something else I thought off while posting. Cutting the power during a runaway isn't going to do much anyways. This isn't an idea born out of necessity but it would be a fun project and would be useful for maintaining reaction temperatures assuming it works. Would be nice for reactions in DMSO as it decomposes before boiling and fiddling with the rheostat and checking the thermometer can be a bit tedious at times.

Was checking out those solid state relays the other day, looked like a good solution.

About the voltage controlled resistors this will make you laugh I was originally researching MOSFET's for this

, only took a couple calculations to realize how far into the wrong field I was. Anyways, thanks again, really appreciate you giving good hints!

IrC - 20-9-2012 at 20:37

You might take a look at these links.

http://www.twopossibilities.com/2011/01/26/homebrew-temperat...

http://arduino.cc/playground/Main/BarebonesPIDForEspresso

http://fermentationriot.com/arduinopid.php

http://letsmakerobots.com/node/30810

smaerd - 21-9-2012 at 11:46

Very cool IrC. the http://fermentationriot.com/arduinopid.php is exactly what I was talking about minus the ethernet interface and add an LCD. Go figure people already had a PID library for the arduino. The graph looks pretty good definitely within the range of error I was looking for.

@Watson.Fawkes: Where can I learn some more about actual PID's?

watson.fawkes - 21-9-2012 at 12:47

Quote: Originally posted by smaerd

Where can I learn some more about actual PID's?

The proportional-integral-derivative terminology comes out of control theory, which I'd recommend reading something about (like the linked Wikipedia article) first, since it gives the overall setting and will allow you to make easier sense of PID control specifically. It's still an active area of research, and it has the lovely property that starting with very simple ideas you quickly reach very deep mathematical questions. (At least that's lovely for a mathematician.) Control theory has many applications, starting in the 19th century with the analysis of mechanical governors and still vital today in the design of switching power supplies.

After that, the Wikipedia page for PID control is as good a starting point as any. It's probably more than you need right now, and it has plenty of references.

IrC - 21-9-2012 at 17:22

http://www.ebay.com/itm/Digital-PID-F-C-Temperature-Controll...

Will not last too long but I thought you should look at this controller.

This alcohol sensor for Arduino is interesting. They have other gas sensors as well.

http://www.ebay.com/itm/MQ-3-Gas-Sensor-module-Alcohol-Ardui...

LPG, Propane and Hydrogen

http://www.ebay.com/itm/MQ2-Gas-Sensor-module-LPG-Propane-an...

Methane, Butane and Propane

http://www.ebay.com/itm/MQ-5-Gas-Sensor-module-Methane-Butan...

carbon monoxide

http://www.ebay.com/itm/MQ-7-Gas-Sensor-module-carbon-monoxi...

Pollution Sensor Module

http://www.ebay.com/itm/Pollution-Sensor-Module-/13054531050...

Smaerd if you can get a copy of the book listed below project 22 titled 'LCD Thermostat' is perfect for you. Code to run it listed. I scanned pages 125 to 131 but cannot post it here as each scan is around 2.5 mb in size. Lower resolution scans are not worth using as it is too hard to read.

"30 arduino projects for the evil genius"

There is a fairly fast torrent for it out there.

[Edited on 9-23-2012 by IrC]

IrC - 23-9-2012 at 11:50

No idea why more are not discussing this most interesting topic but on that note I will go back to ragging on how little info and support the bulk of the low cost sellers on fleabay provide.

Does anyone actually know what the pollution sensor is looking for? PDF link to info provided for it.

Attachment: pollution sensor.pdf (194kB)
This file has been downloaded 3842 times

watson.fawkes - 23-9-2012 at 20:18

Quote: Originally posted by IrC

Does anyone actually know what the pollution sensor is looking for?

Most all of these inexpensive gas sensors use catalytic oxidation and temperature differential as their physical basis. What the catalyst preferentially oxidizes in ordinary atmosphere determines what the sensor will pick up. If you used a selective catalyst, you get a specific result with less chance of false positives. I don't know much about the catalysts, except that (1) they are often compounds or mixtures I don't see used in other contexts, and (2) they're deposited in thin layers in very small amounts so that expensive catalyst precursors still make inexpensive chips.

The TP-4 sensor part in that PDF looks like it may use a broad spectrum catalyst. The only specific things I saw were hydrogen and alcohol (presumably ethanol) and hints that it may pick up nitrogen oxides and ammonia.

IrC - 9-10-2012 at 12:35

Maybe it's just me but it is a little depressing there is not much interest in the Arduino at SMO. Hopefully this project will spark some. You will need to get Eagle to study the circuitry and you can download a 22mb zip at github of His project from a link on the page.

Ardustat "The Ardustat is my (in progress) open source successor to Jonny Galvo. It uses a standard USB-Arduino unit with a custom designed daughterboard to provide:

A galvanostat with a current range from .1 microA to 10 mA
A potentiostat with a potential range from 0 to 5V with 5 mV resolution.

The Ardustat is a two electrode system for the time being, primarily for characterizing capacity and power delivery of small batteries, supercapacitors, and DC energy harvesting devices such as photovoltaics and thermoelectrics."

http://steingart.ccny.cuny.edu/ardustat

arsphenamine - 9-10-2012 at 18:35

Quote: Originally posted by IrC

Maybe it's just me but it is a little depressing there is not much interest in the Arduino at SMO.

SMO cut a deal with Arduino.cc.

They don't host chemistry forums and SMO doesn't host Arduino forums.

Adafruit, google, makezine, github, et.al., operate under similar constraints.

IrC - 9-10-2012 at 19:49

While I am guessing this is meant to be ironic I can see a flaw in the logic. The Arduino is a goldmine for chemistry with the incredible number of shields already on ebay for very low prices. Monitoring PH, temperature, ion concentration, current flow in an electrolysis cell and on and on. Combined with SD card shields for data logging, Xbee for talking to computers in other areas, etc., controlling things like current and voltage as in a plating or chlorate cell where voltage is reduced when temperature rises and so forth. All while a log file is being written to a 2 GB SD card. The list is so endless I would not know where to begin in describing the possibilities.

Edit to add a few examples:

http://www.robotshop.com/bnc-ph-sensor-shield-arduino-3.html

http://practicalmaker.com/category/store/arduino-shields

http://www.reefsanctuary.com/forums/diy/73570-arduino-ph-orp...

A post about PH measurements combined with other parameters and His site below.

http://saltydogaquariums.webs.com/apps/blog/entries/show/790...

Does not take much thought to consider his aquarium dosing system could add various amounts of different chemicals to an ongoing reaction while other parameters are being monitored, you could even add programmed stop and/or alert functions.

As I said the list is unending.

http://practicalmaker.com/category/tags/macroduino-code

You can even find schematics and code if you look at His macroduino code pages.

[Edited on 10-10-2012 by IrC]

arsphenamine - 9-10-2012 at 22:19

Quote: Originally posted by IrC

While I am guessing this is meant to be ironic I can see a flaw in the logic. The Arduino is a goldmine for chemistry with the incredible number of shields already on ebay for very low prices.

Intrinsic merit is necessary but insufficient.

It's about numbers, numbers of people, critical mass of involvement.

Perhaps the number of people interested in chemistry is not only numerically smaller but proportionally smaller than those interested in more general programming+maker projects. There is limited overlap of the populations.

If there is a particular Arduino-based project that you could put to work, then do it. Take pictures as you go. Post them.

I'm no Kevin Costner fan, but if you build it, they will come.

For the record, I _am_ an Arduino fan.

[Edited on 10-10-2012 by arsphenamine]

watson.fawkes - 10-10-2012 at 07:42

Quote: Originally posted by IrC

The Arduino is a goldmine for chemistry with the incredible number of shields already on ebay for very low prices. Monitoring PH, temperature, ion concentration, current flow in an electrolysis cell and on and on.

The larger issue is that there's been fairly little interest on this forum for instrumentation and calibration. Occasionally there the over-ambitious fellow who wants to talk about something that never comes to fruition, e.g. notably NMR. There's a small cadre that are trying to get surplus instruments running, but they aren't hacking the insides of them.

Take pH, for example. Actually building a pH meter is rather more multidisciplinary than it might appear at first glance. pH electrodes are a classic low-current source. Ideally, you want to measure them at zero current, and that's hard. This is required for decent accuracy, and if you don't need that, well, there's no particular reason not to use COTS equipment. Zero current measurement requires you to eliminate or mitigate all source of stray capacitance, since that capacitance must be charged (requiring current) in order to equilibrate the voltage at the sense amplifier. Elimination means knowing something about the physics of capacitance. Mitigation means building a guarded electrode, which requires a guard amplifier, guard rings on the PCB (if you're doing a bang-up job), and shields within the probe cable. Proper sensing of thermocouples has similar issues (though not as severe). Related to this is the reason that precision RTD's use three wires (and high-precision use four), because you can't drive too much current through them lest self-heating cause inaccuracy. These details matter less in specific than the fact that the overlap of chemists and people-who-know-instrumentation-electronics is really quite small.

And much less instrumentation, the dominant trend here is not in building apparatus in general. A few people have experimenting with this. At this writing, for example, Magpie's phosphorus experiments are active, and they're as much about equipment as reactions. This kind of experiment, however, is not the norm here, where most people do their work in standard glassware.

I'm not bewailing this situation, particularly. I am personally quite interested in instrumentation, but I haven't had the attention to pursue it lately. Personally, I think there's a good piece of low-hanging fruit in an open-source melting-point apparatus. The key to making this useful is that webcams have gotten dirt cheap and image manipulation software on personal computers is easy to use. Point a camera at the sample, watch for the phase change, and shut off automatically. Almost as easy to do with the basic setup is DSC (differential scanning calorimetry) which gives you heat capacity etc. of the sample in addition.

IrC - 10-10-2012 at 13:05

Good points all. I wonder though if hyper accuracy is so mandatory for the do it yourself chemistry experimenter running a reaction that is not so critical. The inexpensive setups I have looked at do a decent job controlling the PH of aquariums for example. Mainly I was just trying to stir up some interest in the subject. I have only recently acquired several Arduinos and a few dozen different shields to experiment with. I am trying to teach myself enough coding to create my own useful devices which right now relate to metal detector discrimination and Geiger counters. Some of the language seems common sense as I have worked with basic for many years, and some I am having difficulty with. Right now I am playing with code to pulse a MosFet to drive the high voltage thereby eliminating some circuitry.

I started wondering about using the ADC in voltage regulation schemes. This lead me to wonder about monitoring the GM tube for operating characteristics where the circuit would adjust to whatever type tube it saw. Going further I started being curious about closely monitoring where on the curves the tube ran as a way to determine the actual energy. This would make it easy to differentiate units from counts per minute over various sources with differing decay rates and energies. Still far above me in coding but I can see the concept and that it can work. As I went along I started thinking about so many other ways it could be useful in chemistry. For example the member wanting a thermostat controller with LCD readout.

arsphenamine - 11-10-2012 at 10:57

Quote: Originally posted by watson.fawkes

Related to this is the reason that precision RTD's use three wires (and high-precision use four), because you can't drive too much current through them lest self-heating cause inaccuracy.

It is common practice to separate sensing leads from drive leads.
The topic is frequently covered in analog power suppy regulation design texts.

Platinum devices typically receive a tiny excitation current to minimize self-heating.
For Pt100 RTD's, this usually translates to 20 mV, 200 μA, levels which, using 6 foot long leads,
are usually down in the noise floor of lesser devices and which make a separate sense lead mandatory for useful measurement.

Quote:

These details matter less in specific than the fact that the overlap of chemists and people-who-know-instrumentation-electronics is really quite small.

Worse, there are many chemists with a negative interest in crufting up their own apparatus, to judge from my discussion of the book Building Scientific Apparatus with one of the university analytical chem prof types. _Physical_ chemist types are a different story.

On the practical side, if you have a hobbyist level of electronics proficiency
(read schematics, do Ohm's Law calcs, solder without destroying things),
you can read a few dozen application notes and get usefully familiar with
issues in small signal measurement.

Analog Devices, National Semiconductor, and TI have a plethora of app notes with worked calculations.

There is a good deal of hobbyist-level pedagogy published in the late 80's.
Search initially on the words "instrumentation electronics design" and refine as desired.

*sigh* I used to do a lot of that stuff and had forgot until you reminded me.

watson.fawkes - 11-10-2012 at 19:31

Quote: Originally posted by IrC

I wonder though if hyper accuracy is so mandatory for the do it yourself chemistry experimenter running a reaction that is not so critical.

In my own experience, the difference in skill between someone that can design a decent measurement circuit and one that can design an accurate measurement is just not that large, that is, not large compared to the not-insignificant body of knowledge needed to design any circuit at all. And I mean analog circuits; digital doesn't count here.

One simple exercise is to drive a high-current LED from a low-power CMOS signal using a single NPN transistor. Run the transistor in common-emitter configuration at saturation. In addition to the digital output, you need a base resistor, a collector resistor, a transistor, and an LED. It's a good exercise because you really need to understand what's happening in the transistor, then compute the currents, voltage drops, and finally resistor values. Anybody who can do this can learn to understand instrumentation electronics by applying effort.

The hard part comes when you have to start doing design with noise budget. That's the difference between ordinary accuracy and high accuracy.

Arduino Due, 32-bit

arsphenamine - 20-10-2012 at 09:41

Massimo Banzi leapfrogged the 16-bit Atmel cpu's, has finally released the
Arduino Due based on the 32-bit ARM SAM3X.
The Due is compatible with existing 3.3 Volt shields.

http://www.adafruit.com/blog/2012/10/04/due-arm-powered-ardu...

Much IO, more bits, more memory, fast enough to do audio DSP in real time.

NOW how much would you pay?

$49

Not a Bean: When following "Arduino", Due is pronounced DOO-eh,
even if you aren't Italian.

IrC - 10-12-2012 at 01:44

Smaerd, I found this version of a temperature controller which seems very easy to build, complete with schematic and ready to use software. Maybe you can use it.

http://www.myplace.nu/avr/thermo/index.htm

smaerd - 10-12-2012 at 05:59

Thanks for looking out IrC. My next pay-check is going to parts for the DIY roto-vap I've had haulted for about a year now, and then tax-returns and such will probably go to a PID controller+thermocouple. That's a really nice set-up though it's definitely something to keep in mind I'll do some research on cost and see what's most appropriate(sadly $ is one of the biggest factors in my projects). One of my upcoming projects requires stable temperature control so this does need to get done.

IrC - 10-12-2012 at 09:57

I liked the use of LED digits because unlike LCD you can see the reading from further away and at greater angles, even in a room with poor lighting.

Oscilllator - 23-12-2012 at 14:48

The only thing stopping me from buying many many arduino boards is that I cant think what I would actually do with them once I got them

arsphenamine - 24-12-2012 at 11:30

Quote: Originally posted by Oscilllator

... I cant think what I would actually do with them once I got them

In order of increasing importance, you only need one Arduino, one sensor, and an itch to scratch.

My first was a cheap gaussmeter.
Arduino+display+sensor was $51.50

Solomon - 25-6-2013 at 15:43

I have an Arduino mega 2560. I have designed several programs. One was a simple code that ran an RGB led in a fade mode while simultaneously operating a speaker that you could adjust the sound wave it produced via a potentiometer. You can also use Arduino to make your own satellite.

Best arduino/chip

plante1999 - 30-7-2013 at 05:09

I want to automatize some things in my lab, temperature monitoring etc. I would like to know what is the best "kit" to do so, most functionality etc... I'm not really good in theses things, so if anyone can help it would be appreciated.

Thanks.

Funkerman23 - 30-7-2013 at 06:18

http://www.sciencemadness.org/talk/viewthread.php?tid=23062

http://www.sciencemadness.org/talk/viewthread.php?tid=16440

while I can't help you just yet the two threads above might help. There was another excellent thread on this but my crappy sleep seems to have removed its name from my memory. If I find it I'll post here.

[<a href="u2u.php?action=send&username=bfesser">bfesser</a>: fixed internal link(s)]

[Edited on 30.7.13 by bfesser]

IrC - 30-7-2013 at 11:54

Most usually pick the UNO R3 such as:

http://www.ebay.com/itm/Arduino-Starter-Kit-w-Uno-R3-Ultraso...

However in my mind the Mega2560 is a better choice because you do not so quickly run out of I/O:

http://www.ebay.com/itm/arduino-mega-2560-r3-starter-kit-mot...

You can also just buy the Arduino and carefully pick the other devices based upon your goals, say pick the shield most useful to go with your 2560. This would save you money compared to a kit price which likely includes items you do not need. However if starting out and learning to code is your goal I for one would pick the kit. In fact even though more costly the 'non clone' choice for about $130 is best as that 170 page book included is worth it's weight in gold.

http://www.ebay.com/itm/The-Arduino-Starter-Kit-Official-Kit...

Just keep in mind it comes with a UNO R3. Only about $18 to add a Mega2560 clone later on when you find you need more lines for I/O.

http://www.ebay.com/itm/ATmega2560-16AU-ATMEGA16U2-Board-Fre...

Probably the best route, total investment around $150. Later you can dedicate the UNO to some lab controller and still have the kit with the 2560 for learning/playing. Bear in mind sketches designed for the UNO in the book/kit may need alterations to work with the Mega.

Myself being cheap I just bought the clone and spent time studying at http://www.arduino.cc/. However I strongly suspect if I had bought the official kit I would have learned at a much faster rate.

[Edited on 7-30-2013 by IrC]

smaerd - 30-7-2013 at 12:22

I can't say what is best or what is worst or anything like that. I can say I'm very happy with my arduino leonardo.

My advice is this, research before you buy anything. See what's been done before, what the limitations are of arduinos, stuff like that. Then buy the board, several little bread boards, jumper wires, and lots of little pieces for your projects. Good idea to have an assortment of capacitors, and resistors. The rest depends on what you plan on doing really.

Also don't forget this - http://ruggedcircuits.com/html/ancp01.html
This quick article saved me 25$!

[Edited on 30-7-2013 by smaerd]

bfesser - 30-7-2013 at 13:12

<strong>plante1999</strong>, Adafruit has a nice <strong><a href="http://www.adafruit.com/products/68" target="_blank">Starter Pack for Arduino (Includes Arduino Uno R3)</a></strong> <img src="../scipics/_ext.png" />.
You could also check out the small <a href="https://www.tindie.com/products/category/platform/arduino/" target="_blank">Arduino clones and accessories on Tindie</a> <img src="../scipics/_ext.png" /> (although, the new site doesn't load properly for me).
And, of course, SparkFun has a large <a href="https://www.sparkfun.com/categories/103" target="_blank">selection of shields</a> <img src="../scipics/_ext.png" />.

I have an <a href="http://arduino.cc/en/Main/ArduinoBoardUno" target="_blank">Arduino Uno</a> <img src="../scipics/_ext.png" /> R3 that I'm very happy with. But, if you'd like to go a little more advanced, I suggest a <a href="http://www.adafruit.com/products/1014" target="_blank">Raspberry Pi</a> <img src="../scipics/_ext.png" />—which can be interfaced to an Arduino for added I/O (among other benefits). I've been wanting to try the <a href="http://wyolum.com/projects/alamode/" target="_blank">AlaMode</a> <img src="../scipics/_ext.png" />, but don't have the funds yet.