Go Down

Topic: Is this project possible? (Read 5115 times) previous topic - next topic



I'm hoping to build an 40*40*40cm "insect chilling chamber" using a couple of Peltier elements, 5 fans (4 plus one to counter condensation on the lens), 4 fins and some sort of XYZ+ABCDE mover inside the airtight container. By the latter I mean a thing, maybe an arm of some sort, that'll let me position a twig with a chilly bug on it quite slowly and accurately, ie 1 degree per second, to get the twig and bug at the right angle to be able to take shots.

I also need to be able to add temperature and CO2 sensors, a digital readout and infra red control over the whole thing. This is obviously quite an ambitious project I believe Arduino with steppers on gears could be the right approach, but I don't really know where to start or if I'm even on the right path with Arduino.

Is this possible?

Many thanks,


Is this possible?


but I don't really know where to start

With some research. Specifically on Peltier coolers. The amount of cooling needed, and the level of control over the temperature that you need will indicate whether Peltier coolers are appropriate, or not, and how long they will last if used inappropriately.

How much control over the fans do you need? On or off or variable speed? That defines what kind of fan you need.

Stepper motors can move quite slowly. They can be geared up or down. A stepper motor driver, sized to match the stepper motor, will be needed.

Why are you shooting the insects?
The art of getting good answers lies in asking good questions.


Thank you paul, I appreciate the handholding.

Regarding Peltiers, judging from info online, two beefy ones with decent fans and sinks *should* be able to cool a small space down to where I need it (1-2 deg C). All the research I've done indicates that the fans and sinks are really the key to making it work.

On or off is fine for the fans driving air to and from the Peltiers, but variable speed would be nice for the fifth. The fifth is smaller and just needs to drive air over the lens, to prevent condensation. Vibration is a concern so ideally as slow as possible. Heck a small hand held fan adaptation will even do for that. For the other 4, driving air, my dream would be to add a temperature sensor inside and set a target temperature and write a proggie that'll cut out the inside fans for me when it's at the right temp, then resample every Xs(?) or so and come back on when it's some variable point up from the desired temperature (Y). Trial and error will determine X and Y.

Could you point me at some good sources of information about designing arm-like structures or XYZ manipulators using steppers and gears? If I know my options I'll at least know where to start!

I'm making this thing to keep insects alive. I'm a photography nut but I don't like the fact that in order to make extreme macro shots of things like moths they need to be dead, because alive they don't sit still. You can't have movement between plane of focus shots when you're stacking a few hundred shots together to make a master shot with everything in focus. A cold environment coupled with CO2 keeps them perfectly still but also leaves them unharmed. For example it is very standard when you catch moths to keep them in a fridge overnight in a tube to later ideantify - because it stops them flying around and damaging their wings (scales coming off).

This is an example with dead insects, it is good photography but I'd like to do better and be the first to come out with stuff like this of living, not dead posed insects!

Master & Servant by Johan J.Ingles-Le Nobel, on Flickr


Fwiw, I decided to buy an Arduino starter kit to get some experience.


Good idea - the starter kit will help you get your feet wet and enable you to ask the exact questions you have as well as to better understand the answers you receive.

I don't have any advice to offer on your build but I think the project is interesting.  Here's hoping you keep us updated :)


An arduino should be perfect for this, you will need interfaces ( or readymade shields ) for drivers, and stepper motors, but a lot of your project is the practical side.
45 years of editing projects with a knife and soldering iron, then I found Arduino !


Great photograph!
The project is possible but, as you have recognised, it is ambitious.
This link gives useful information.

If your primary objective is the photography you might want to consider purchasing some aspects of the system e.g. the motion system and doing other aspects yourself e.g. fans cooling etc.

You can get robotic arms where control via arduino is an existing feature.


Jan 14, 2013, 11:49 pm Last Edit: Jan 14, 2013, 11:59 pm by nass Reason: 1
That societyofrobots site is fabulous, thanks for pointing me that way! Good advice about splitting the parts, in fact I was already looking at that tonight. I was thinking maybe do the XYZ manually but go for electronic control over 4 ways to rotate and use geared stepper motors. I already built myself a 'usable' XYZ/PRY+R mechanism so I do know the design works, in fact it works amazingly well as it means I can position something and rotate it to any angle and it'll still be in focus:

... but this is an re-engineering bodgejob with araldite and a hacksaw blade, what I'm after is this sort of thing but mechanised and controllable, inside a cooled chamber... so erm, a bit more complex =). The problem is that proper sensitive industry 'universal stages' like this cost a four figure sum so I had to build my own, ie above. I did find some very affordable yellow robotic arm controllable by Arduino but on the vids (and according to a friend who I subsequently discovered has one) it isn't at all 'accurate'. Unfortunately I need something vaguely accurate (1 degree granularity on each of the 4 rotation axis). The feeling I very much have that it's a tradeoff between accuracy and cost... not entirely surprising really =)

Anyhow I'll putz around with this starter kit, then build a peltier container thingy, test that, then try and work out some sort of 4 axis rotation thing. Plenty of time to impress my wife and children with fetching blinking LEDs.


If you already have a workable mechanical design then you are further on than I was thinking.
I was going to mention that accuracy might be a problem with cheap off-the-shelf solutions, the ones I saw did not give a lot of technical information. One other issue will be size. Although you have no load to move the mechanism will still take up a lot of space in your box.

This might be of interest


Thank you Radman, I must be thinking vaguely along the right lines as I had already found and bookmarked that very page! Unfortunately my knowledge is too limited to know whether servos or motors are the answer =(.

Been looking into Peltier more, it appears that maybe I need to use a water cooling system to absorb heat rather than fans driving heat away. It's difficult to find data on what cooling Peltiers can and cannot achieve.


I believe you want to use a stepper motor with a leadscrew (threaded rod) and build or purchase what would be called a "linear stage". If you purchase one you might be lucky and get a used/surplus unit for ~$200; if you buy new expect to pay industrial prices.

A stepper motor is controlled with a fixed number of steps (typically 200, or 1.8 degrees per step) per rotation. A servo, as would be used with model planes/cars/boats (an "RC servo"), moves only within an arc of 60 degrees; the precision is controlled by a potentiometer which can't be counted on as highly precise; expect twitching over time as the potentiometer wears out. A DC motor / rotary encoder combination is also referred to as a "servo". It is capable of both high speed and high accuracy but is by far the most expensive option.   

Peltier coolers will have a "Qcmax" specification which states how many watts of heat the unit can move. You can convert watts to BTU to get a more real world idea of how much cooling that is -- .29 Watts = 1 BTU which can cool 1 pound of water 1 degree F in 1 hour. If you're just cooling tiny insects I can't imagine you'd need more than a few tens of watts of power and a water circulation system shouldn't be necessary.


Jan 16, 2013, 01:06 pm Last Edit: Jan 16, 2013, 11:50 pm by nass Reason: 1
Many thanks Chagrin!

Stages can be purchased a lot cheaper than that online if you know where to look and I already have 6 of them with various levels of sensitivity =). The best one I have can give me 1 micron increments which I might use for 20:1-100:1. You are correct that the most sensitive models do use stepper motors with very fine screws, but those are beyond what I need for this, which is just the ability to move the top four 'rotator' degrees of freedom (like the device on the photo above) and in the X, Y and Z planes. That movement doesn't need to be to 1/100 of a mm though :).

In fact, what's fascinating to me is that every time I mull over this project I am still coming up with design variations. For example I have a whole mechanical setup like this already standing on my tabletop studio using a microscope base and slide positioning table and that combo is actually a perfectly ok XYZ mover. So why not forego the electronic one and just do something with Peltiers and the 4 rotational degrees, and then just have it sitting on my slide table to do the XYZ movement. I have a machine that'll do the stacking increments already so all I'd need is all this in a space with a flexible curtain that'd allow a lens to poke in to take the shots but also keep it airtight. That means I only need to worry about 4 microsteppers which at least keeps it simple. Decisions decisions!

That's really useful info about Qcmax, thank you so much. One question though, when I'm trying to calculate the amount of power needed, shouldn't I also take the amount and weight of of into consideration?

On another note my starter kit is here, courtesy of mouser. Looks nifty, fun fun fun!



I am happy for others to correct me on this, but basically there is a very big overlap between the application of servos and stepper motors and you could use either. In your particular case a servo may be better as I would expect it to generate less heat and, since it is not stepping, to have less vibration.

In general servos only operate through 180 degrees, but that probably meets your needs. One thing I am not sure about is speed, the larger the move you are trying to make the faster the servo will try to go. You could move the servo in small increments, but that might bring back the possibility of vibration.


On or off is fine for the fans driving air to and from the Peltiers, but variable speed would be nice for the fifth. The fifth is smaller and just needs to drive air over the lens, to prevent condensation.

Why not put a small tray of silica gel in the bottom of the box instead? After use, you can dry it out in the oven, wife permitting, and use it again and again.


This week at walmart I saw what appeared to be a Peltier wine cooler that held maybe eight bottles of wine and was priced somewhere around $50-$60. A small dorm refrigerator might also be used as a cooling box.
Google forum search: Use Google Search box in upper right side of this page.
Why I like my 2005 Rio Yellow Honda S2000  https://www.youtube.com/watch?v=pWjMvrkUqX0

Go Up