I am currently working on a project for University, and would appreciate some guidance for it.
What we need to build is a Radon Detector. Our design is pretty simple - we're building two Ionisation Chambers, on that detects alpha particles and one that doesn't. How these chambers work is that they output a current signal that is proportional to the amount of radiation present in the air.
The idea is to collect the two current values from the two chambers, take the difference between them, calibrate them with some formula (found through trial and error) and get a value for the concentration of alpha-particles in the air in Bq/m3, and thus, the Radon level in the air.
What I need is some sort of interface or chip that can take in these analogue, current values, allow me to use them in some calculation that I programme into it, and then output the resulting number in a format that could be read and displayed on a simple LED display.
It doesn't sound very complicated, but I am completely lost, coming from a Mech Eng background and not Electronics. I was recommended to use one of your products by the University's robotic's team.
Which product of yours do you believe is suitable for such a task? And how should I go about implementing it? I have a good background in programming, so I am not worried about that, but more about the electronics bit. What wires go where and that sort of stuff.
It's actually our final year design project. Didn't have much say in the topic given to us, and it's the first time our Uni has offered it. The Prof in charge doesn't know a lot about the topic, didn't realise that it would be so electronics intensive.
He's suggested we buy the amplification circuit, microprocessor and LED from outside and just set it up ourselves. Which is what we're looking into right now.
The amplification circuit didn't turn out to be as easy as we thought it would be - we tried building our own following instructions on the Mad Scientist Hut blog, but it didn't work. We couldn't get the right transistors, and we don't know anything about them and ended up waiting weeks for some that we ordered to arrive, which ended up being wrong, and then spent days wondering why our circuit wouldn't work. With time running out, we're reluctant to try building our own circuit again; we're looking to find someone who knows this stuff and is willing to make one for us, or we'll have to buy one from outside.
The thing with the chamber is that varying radiation levels create varying levels of current. So in that case, I believe we need to amplify and read these currents with the microprocessor, but I have been told this can't be done? Apparently voltages are what can be amplified, not current?
And with the microprocessor, how do we make it work with an LED display. Like, how can we control the display with the output from the microprocessor? Is it very simple and can be done with coding (sort of like a "cout << "radiation level" << endl;" thing), or does it depend on the display in question and not straightforward at all?
Need more data. What voltage does the ionization chamber work at, and how much current does it pass? Is the current continuous, or in a short pulse whenever an alpha particle is detected?
A text LCD display is easier to drive than an LED display.
EDIT: I did some research and found http://www.inin.gob.mx/mini_sitios/documentos/MRNI-503D0.pdf which suggests that the current is extremely low, between 1fA and 1nA. You will need to measure the voltage dropped across a very high value resistor (e.g. 1G ohm) using an amplifier with very low input bias current. The usual instrumentation amplifiers are unsuitable for this because they have input bias currents of a few nA. You could try one of these http://uk.farnell.com/microchip/mcp603-e-p/ic-op-amp-single-2-7v-8dip/dp/1852052 op amps configured in the non-inverting configuration.
dc42:
Need more data. What voltage does the ionization chamber work at, and how much current does it pass? Is the current continuous, or in a short pulse whenever an alpha particle is detected?
A text LCD display is easier to drive than an LED display.
9V for each chamber, though we might increase that.
Two chambers, one that measures alpha particles, and one that doesn't.
Current is expected to be continuous DC, in the range of 10^-12 to 10^-15 amps. The air is constantly being ionised due to the presence of radon and radon daughters in the air -these particles decay to give off alpha particles, which then ionise the air to produce a +ve ion and an electron.
The electric field present attracts these ions to the cathode and anode, and creates a current flowing in the circuit between them.
Current should be higher in the one measuring alpha particles, as there will be more ionisation overall in that one.
dc42:
EDIT: I did some research and found http://www.inin.gob.mx/mini_sitios/documentos/MRNI-503D0.pdf which suggests that the current is extremely low, between 1fA and 1nA. You will need to measure the voltage dropped across a very high value resistor (e.g. 1G ohm) using an amplifier with very low input bias current. The usual instrumentation amplifiers are unsuitable for this because they have input bias currents of a few nA. You could try one of these http://uk.farnell.com/microchip/mcp603-e-p/ic-op-amp-single-2-7v-8dip/dp/1852052 op amps configured in the non-inverting configuration.
Brilliant, thanks.
Will look to see if I can find that one locally, or something close. Sounds like a hard to get component, would explain why the other sites making ionisation chambers all had circuits that were designed by the creator.
OK, then I suggest you connect each ionization chamber to ground through a 1G ohm resistor, and use a FET input op amp such as the MCP603E/P configured as a unity-gain buffer on each channel. Then feed the outputs of those buffers into a standard instrumentation amplifier such as INA122, and the output of that into the Arduino.
You'll probably need to order the components from a component distributor, you won't find them in the hobby shops.
[EDIT: a single MCP602 op amp will do both channels.]
To drive an LED display, you generally need some external electronics unless you only want to drive a single digit. Unless you need large or bright digits, I suggest you look for an LCD display. 16 characters by 2 lines is a popular size. You can pick them up from eBay or the usual component distributors.
Given all the other issues you're dealing with to get the sensors working I suggest you keep the Arduino part as simple as possible. There are lots of LCD shields which would give you an easy off-the-shelf solution, and you can even get them with an integral keypad if you anticipate needing to control this contraption.
I must say this sounds like a very unfair project to give MechE students - it has a very large physics and electronics component and hardly anything to do with mechanical engineering.
Do you have a source of Radon gas? Do you know whether / how the resulting device will be tested? Building a slowly-varying random number generator would be much easier than building a real Radon detector.
Hahahaha, that's a good idea, will keep that in mind
We've got some granite aggregate, and the mountains around the Uni are a rich source of Radon. We'll probably need a more intensive source though, we'll look into procuring some after we get this working.
dc42:
OK, then I suggest you connect each ionization chamber to ground through a 1G ohm resistor, and use a FET input op amp such as the MCP603E/P configured as a unity-gain buffer on each channel. Then feed the outputs of those buffers into a standard instrumentation amplifier such as INA122, and the output of that into the Arduino.
You'll probably need to order the components from a component distributor, you won't find them in the hobby shops.
I've checked with the distributor here, and the MCP603E is available, but the INA122 is not. What particular features/values am I looking for in this instrumentation amplifer, so that I can look for a replacement?
You could manage without the instrumentation amplifier, if you configure both the op amps to have some gain (the same for both), and read their outputs using 2 analog input pins. If you do use an instrumentation amplifier, then look for one that can run from a single +5V supply, whose common mode input voltage range includes ground, and with rail-to-rail output swing.
nchemlani:
For the LCD Shield, are there any that are compatible with the Arduino Nano?
I haven't noticed any, but it's not something I've ever looked for. I don't recall seeing any Nano shields at all. I'm not sure why you'd want to use a Nano in particular (the genuine Arduinos are much better supported) but I seem to remember seeing a Nano 'prototyping' board which was something the size of a UNO that you could plug a Nano into, which would then accept standard Arduino shields. It seems like a lot of bother and expense, though, compared with just using a UNO.
dc42:
You could manage without the instrumentation amplifier, if you configure both the op amps to have some gain (the same for both), and read their outputs using 2 analog input pins. If you do use an instrumentation amplifier, then look for one that can run from a single +5V supply, whose common mode input voltage range includes ground, and with rail-to-rail output swing.
Most of that has gone over my head
I tried to filter through the available instrumentation amplifiers available on our supplier's website, and I narrowed it down to 27. Assuming I used the correct search terms, will any of these 27 work, or is there something more I'm looking for?
I'm looking into which LCD display to get now - we decided against an LCD Shield because that would need to be plugged into the board, and will thus give us less flexibility when it comes to designing the case/housing of our detector in terms of how we package and set everything up. Would be better to have a display that just needs to be connected by wire and can be placed wherever we want it.
According to this link, The LiquidCrystal library works with all LCD displays that are compatible with the Hitachi HD44780 driver. So I need to find one such display...sigh.
I believe I should be able to download the LiquidCrystal library, plug in it as instructed in the various tutorials, and this face no difficulties. Is this true?
Also, an Elec PG student here said that the INA321 should do the be fine for this job, although that isn't his field of expertise. What are your thoughts?
