Making my own incubator

588 views
Skip to first unread message

Mega

unread,
Feb 3, 2012, 2:09:59 PM2/3/12
to DIYbio
Hey,

As I need someting to heat my e.coli cells above 30 degrees Celsius to
make them grow exponentially, I'm building my own heater.

It's made of a very big christmas coffe cup (some 15cm in diameter),
controlled by an attiny 13 and lm335 (temperature sensor).

It's heated by 6 Ohm resisstances (3 parallel -> 3 of parallels in
serial)

It's powered 3,7Volts, 900 mA. Resistances heat up to ~ 64 °C. When
sensor reaches 35°C it will stop heating.

leaking pen

unread,
Feb 3, 2012, 2:26:03 PM2/3/12
to diy...@googlegroups.com
are you heating with the elements directly in your culture, or are you
heating a water bath?

how far from the elements is the temp sensor?

> --
> You received this message because you are subscribed to the Google Groups "DIYbio" group.
> To post to this group, send email to diy...@googlegroups.com.
> To unsubscribe from this group, send email to diybio+un...@googlegroups.com.
> For more options, visit this group at http://groups.google.com/group/diybio?hl=en.
>

Mega

unread,
Feb 3, 2012, 3:04:27 PM2/3/12
to DIYbio
The air is heated, some (few) centimeters above is the petri dish.

There is a 'lattice' (I hope this is the correct english word) that
keeps the petri dish away from the bottom of the cup where the
resistors are, but lets the hot air go through.

The temperature sensor is not near the ressistances (64°C) but above
the petri dish to get somewhat accurate 35°C.

Nathan McCorkle

unread,
Feb 3, 2012, 5:16:40 PM2/3/12
to diy...@googlegroups.com

Sounds pretty good, if you're using arduino environment for programming you can easily add PID control which should stabilize the temp around the set point pretty well.

If you temp sensor is above your petri dish, I would worry the petri is blocking heat, and that the underside of the petri is actually warmer than you want. If you have a second temp sensor, it'd be worth it to put that below the petril, and see if the two sensors report different results.

leaking pen

unread,
Feb 3, 2012, 5:35:10 PM2/3/12
to diy...@googlegroups.com
Sounds good! let us know how it works.

Cathal Garvey

unread,
Feb 3, 2012, 5:40:46 PM2/3/12
to diy...@googlegroups.com
I love this idea! It's really minimalist and simple. A cool hack. :)

I have code for a water bath that uses an arduino and an LM35; you might
be able to re-use some of that code if you like?

In particular, there's an LM35 reading function and some basic
temperature controls, and some failsafe code built in.

It's on github here: https://github.com/cathalgarvey/KettleKontroller

And it's attached. Currently set to 55C for the G.stearothermophilus I'm
trying to culture, you'll have to change that if you want to use it as-is.

Also if you're using it as-is, you'll need to play with "HeatPulseDur"
and "RestPulseDur": they cover how long to turn the heater coil on and
off during heating. Kind of like coarse pulse-width-modulation; you can
use different settings to get different rates of heating out of any
given setup.

Right now, it's pin A0 for LM35 input (with the LM35 on +5V), and heater
control is on pin 13 (In my case, controlling a relay which turns a
kettle on and off)


--
www.indiebiotech.com
twitter.com/onetruecathal
joindiaspora.com/u/cathalgarvey
PGP Public Key: http://bit.ly/CathalGKey

BathProto3

Mega

unread,
Feb 4, 2012, 7:12:15 AM2/4/12
to DIYbio
Thank you.... I have already written the code (I modified the code I
used for measuring the temperature many months ago)

I could post some pictures here if there is an online storage where to
put them in...

Pulse-with modulation would work well with transistors/MOSFETs but as
I have only electronical transistors (maybe can't stand the amperage)
at home I'm using a relais.

Yeah and the temperature I'm measuring is displayed by two LEDs :
First LED shows 'tens', second one 'ones'. So if it was 34 °C (Komma
is not displayed) it would be -> 3times first led blinking , 4 times
second led.
(To avoid using LCD which would mean you'd need a bigger chip like
ATmega8)
> > It's powered 3,7Volts, 900 mA. Resistances heat up to ~ 64 �C. When
> > sensor reaches 35�C it will stop heating.
>  BathProto3
> 6KAnzeigenHerunterladen

CodonAUG

unread,
Feb 5, 2012, 2:00:37 AM2/5/12
to DIYbio
Have you thought about using a styrofoam cooler as the chamber instead
of a cup?

Mega

unread,
Feb 6, 2012, 12:56:36 PM2/6/12
to DIYbio
Yes I have.

But the heating elements may burn the styrofoam (polystyrene).
Although they have just a bit more than 60°C I want to be sure noone
is hurt ;)

My idea is to heat the cup up using a hair-dryer before using it... So
it has already 30°C heat and the circuit doesn't need to heat so much
up (20->30°C takes several (very many!) minutes!!)

Mega

unread,
Feb 7, 2012, 8:16:07 AM2/7/12
to DIYbio
I have bought an aquarium heater :)

The device I built makes some troubles as it 'faints' sometimes and
you have to reset it...

Cathal Garvey

unread,
Feb 7, 2012, 8:58:07 AM2/7/12
to diy...@googlegroups.com
Small aquarium heaters are great for getting 30C. If you want to "hack"
one, they usually allow you to tune the thermostat, so you could
probably tune it to 37C instead of 30C, too?

Great call! :)


--
www.indiebiotech.com

Dakota Hamill

unread,
Feb 7, 2012, 9:35:43 AM2/7/12
to diy...@googlegroups.com
an aquarium heater is what I had in mind for mine as well, inspired by this post I saw a long time ago


I have the styrofoam box I made and glued together and sealed all the cracks, but I havn't put in the effort to finish a relay circuit.  The arduino + 10kOhm thermistor circuit works fine for temperature readings on the serial monitor, but coupling that to being able to turn the heater on and off is the part I'm at now.

But if aquarium heaters already come preset to shut off at certain temperatures, or are adjustable, I get that solves that problem.

The relay circuit would be a nice thing though, and I think Cathal you mentioned you made one, because it could be used for an incubator + a coffee can thermal cycler.


It'd be sweet to have a basic arduino shield / external relay circuit breadboard with some relays that'd be rated for common devices at the correct power/voltage like a 60W light bulb, heat gun, heating pad, blender...I don't know what else...

Also, has anyone gotten Pachube to broadcast their info?  Do you need a piece of hardware besides just the arduino?  Maybe I read their FAQ wrong but I felt like I needed an MS in computer science to make the arduino update it's info to the web

CodonAUG

unread,
Feb 7, 2012, 9:43:18 AM2/7/12
to DIYbio
Depending on how much effort you want to put into it, you could just
buy a temperature controller. The plus side is that they sell them
for outdoors purposes so they'll be less risk from humidity. Downside
is the cheapest one is 130.
I bought one for a microscope incubator (that used a hairdryer) and it
works great.

http://www.aqualogicinc.com/products/temp_controllers/index.htm

Cathal Garvey

unread,
Feb 7, 2012, 10:33:29 AM2/7/12
to diy...@googlegroups.com
Temperature controllers that offer a probe-and-plug-socket level of
simplicity would be great, if you could afford them. However, if you're
willing to put effort in anyway, an arduino, solid-state relay and LM35
will give you a very effective temperature controller for about �25, or
�15 if you go for a cheap arduino clone (though you'd pay the difference
buying the programming cable..).

Darrell Montana

unread,
Feb 7, 2012, 10:16:03 AM2/7/12
to DIYbio
Do you guys worry about using a fan? I would worry about cross-
contamination issues. Have you not noticed it being a probelm? All
the incubators I have used have used convection heaters

David Brunell

unread,
Feb 7, 2012, 3:49:10 PM2/7/12
to diy...@googlegroups.com
For a controller, why not just use an inexpensive PID like this one:

http://www.auberins.com/index.php?main_page=product_info&cPath=1&products_id=14

You connect it to a solid-state-relay like this one:

http://www.auberins.com/index.php?main_page=product_info&cPath=2_30&products_id=9

I've built a lot of temperature controllers this way.  I can post photos, schematics, etc. if you are interested.

David


Mega

unread,
Feb 7, 2012, 3:52:17 PM2/7/12
to DIYbio
Well,

I would have taken a water bath for the petri dish....

A box, where the aquarium heater heats the air is even better because
your bacteria won't surely be flooded by water....

My aquarium heater makes (measured) 30,5-32 °C. (it should've been
able 34°C...)

>30°C is enough to make exponential growth? Or do I have to add hot water from the tea cooker every 5 minutes ;) ?

Before I hack the heater I want one sucessful transformation ;)

Chris

unread,
Feb 7, 2012, 5:54:05 PM2/7/12
to DIYbio
@CodonAUG...nice find on the heater controller, seems like a neat and
well designed general purpose controller

@Cathal...I agree with you that it can be done on the cheap...

@Dakota...here are some other temperature controller kits to purchase
or use as a reference design that might be along what you said would
be nice:

http://www.sparkfun.com/products/81

http://lowereastkitchen.com/wp/

these might fill a middle ground between an expensive pro module with
no hacking and full hacking with bare components. I am a fan of kits
to get you up and running quickly but they are not as difficult as
starting from basic components.

Chris

Tom Randall

unread,
Feb 7, 2012, 11:04:22 PM2/7/12
to DIYbio
I looked at aquarium heaters once, but they dont go much above 30C
(you probably dont want to slow cook the fish) so that makes sense,
but one really needs a heating element that at least goes to 37C for
most restriction enzyme digestions. Here is an alternative that goes
up to at least 60C (http://www.brewhaus.com/Fermenters-and-Parts-
C97.aspx) see the item "110V Submersible Heater with thermostat". I
have used one of these in a styrofoam water bath for years without any
issues, but I dont think you would want the water bath to dry out,
that would likely cause a potential melting/fire problem. This site
(www.brewhaus.com) has a lot of useful equipment that can be used for
purposes other than fermenting/distilling and will ship to residential
addresses. There is a European (Swedish, I think) website that sells
these heating elements also, dont remember the link offhand.

Cathal Garvey

unread,
Feb 8, 2012, 3:58:36 AM2/8/12
to diy...@googlegroups.com
With the TSS method (PEG/MgSO4), 30C is enough.
Indeed, I *always* culture my bacteria (E.coli, B.subtilis, G.xylinus)
at 30C. The only exception is G.stearothermophilus, because it doesn't
grow at less than 40C!

Mega

unread,
Feb 16, 2012, 2:33:16 AM2/16/12
to DIYbio
Just wanted to say my incubator works...

Soon I'll be testing it with bacteria.


I've made some pictures and a video, don't know where I can post
these.
If anyone wants to build his own, write me and I'll send you the
pictures & programming code for the ATtiny & the design of the circuit
via e-mail...


The total costs were quite neglegible... 2-3 € for the ATtiny 13 and
socket, wires didn't cost anything because I ripped them off out of
old devices. Ressistances, which I had at home, may cost some 2 euros
alltogether... Two Leds I used were in a multi-package (20 Leds
for ... don't remember... 2€ ... it was a bargain). The platin costed
5-7€ althouhtt You only need a piece (say 1/10.... So about half a
euro)
The oversized coffee cup was a gift, so it was free. Yeah, the relais
may have costed between 1,5 and 5 euros, really don't remember that...
If necessary I can look it up...

Mega

unread,
Apr 21, 2012, 12:12:17 PM4/21/12
to diy...@googlegroups.com
I soldered the entire board newly, and now it works fine.

Today or tomorrow, I'll be testing it with the heating resistances and an empty petri dish.



Questions: Do you think, I could sell them? (Of course, it will need a testing period prior to selling, to fix bugs and make sure it's safe)
To make a few bucks profit, I'd need to sell a construction kit for a bit more than 20€. Say 25. (I keep looking for cheaper parts, although here they are already quite as cheap as possible) . Soldered versions may cost additional 10€, but they are 'plug-and-play' things.

Professional incubators cost around (at least, I think) 1000€. On ebay you will get one for >50. (I wanted to buy one for 20€ and was the highest bidder for the longest time, but in the last seconds, one threw in 85€! and there was a fight and it came to ~120€s!)

My incubator has just 1 Watt for heating ( I could produce 2W 'quite' easily - and with more expensive relaises, 10W and more.) , but for small volumes, that should be enough to provide enough heat if you don't open the incubator so that warm air can escape.
When you pre-heat it with a hair-dryer to working temperature, you can cut pre-heating phase to some seconds (vs. many minutes without). 

It has to be said that this is only the electronic board for an incubator, you need to get a polystyrene box or an insulated cooling box or a big coffee cup.



What do you think?? Would you buy such a thing?
My motives are that I can help others buy providing them with as-cheap-as-possible-incubators, and get some funding for my own biology work/research...
The code is open source, because I like the idea of open source ;)


Be honest :D


Nathan McCorkle

unread,
Apr 21, 2012, 6:14:52 PM4/21/12
to diy...@googlegroups.com
I'd say release the design open-source

> --
> You received this message because you are subscribed to the Google Groups
> "DIYbio" group.

> To view this discussion on the web visit
> https://groups.google.com/d/msg/diybio/-/Pxm2yHBkTjUJ.


>
> To post to this group, send email to diy...@googlegroups.com.
> To unsubscribe from this group, send email to
> diybio+un...@googlegroups.com.
> For more options, visit this group at
> http://groups.google.com/group/diybio?hl=en.

--
Nathan McCorkle
Rochester Institute of Technology
College of Science, Biotechnology/Bioinformatics

ByoWired

unread,
Apr 22, 2012, 12:25:40 AM4/22/12
to diy...@googlegroups.com


On Friday, February 3, 2012 2:09:59 PM UTC-5, Mega wrote:


...It's made of a very big christmas coffe cup ...

I cringe every time somebody tells me they are making their own incubator out of things like styrofoam, cardboard, etc.  Sure, it works and it's cheap, but when designing something like this, you have to ask yourself what will happen if a component fails or you have some bizarre software bug that will allow the heater to run non-stop.  It's possible this sort of design could catch on fire, easily setting afire anything nearby.  I would suggest people consider using a metal container of some sort wrapped in fiberglass insulation.  A small metal garbage can, for example, wrapped in the kind of fiberglass insulating blanket they sell for hot water heaters, will work pretty well.  Of course it's not as cheap as a cardboard box, etc. but if the software freaks out or a relay gets stuck, etc. and your heater-resistors run till they ignite something inside, then at least the fire is contained in a non-flammable container that might smother the fire in its own smoke.  Using a metal container, you must take precautions keeping it properly grounded so you don't electrocute yourself, but I think it's worth not risking a fire by going with that design.

My two cents worth.  

Mega

unread,
Apr 22, 2012, 4:07:36 AM4/22/12
to diy...@googlegroups.com
>component fails or you have some bizarre software bug that will allow the heater to run non-stop.

The circuit is designed to be as secure as possible. And it is getting tested.

E.g. the relais is always of if there's no voltage coming from the chip. Most likely, a software error (never had one in the last years) will cause the chip to restart (heating is off).


What I would always do and recommend to do, of course:

Don't place the incubator on wooden floor or carpets. Use a fire alarm device in the same room.
Don't go out of the house when it's running. Place water near it. (It uses a mobile phone transformator with a long cable, so water would just hit 3Volts DC - absolutely no danger, not 230V AC!

I, too, wouldn't like to use polystyrene because it burns very well once in flames. (but seen it many times on the internet)
Maybe, some shop sells cheap cooling boxes?

ByoWired

unread,
Apr 22, 2012, 10:40:12 AM4/22/12
to diy...@googlegroups.com


On Sunday, April 22, 2012 4:07:36 AM UTC-4, Mega wrote:
 

...The circuit is designed to be as secure as possible. And it is getting tested. ...

The general lack of safety considerations in DIY equipment designs is the main reason my attitude has turned against DIY biology.   If DIYers are unwilling to follow what are basic, simple design lessons that have been learned the hard way over many decades of engineering practice, then it's easy for me to imagine that similar attitudes will exist when it comes to performing the experiments themselves.  Probably what will eventually turn public opinion against DIY biology is not the release of "frankenstein bugs" that the press sensationalizes but instead will be the little fires and electrocutions that will occur here and there as a result of the proliferation of ignorant equipment designs.  

John Griessen

unread,
Apr 22, 2012, 1:17:36 PM4/22/12
to diy...@googlegroups.com
On 04/21/2012 05:14 PM, Nathan McCorkle wrote:
> What do you think?? Would you buy such a thing?

The lack of a durable surround for the heater is going to cut sales
down to a very few.

On 04/22/2012 03:07 AM, Mega wrote:
> Maybe, some shop sells cheap cooling boxes?

There are no retailers for partial products, just warehouse companies.
This kind of thing needs some design time and testing and thought.
You're not done yet.

On 04/21/2012 Mega wrote:
> 20€. Say 25. (I keep looking for cheaper parts, although here they
>> are already quite as cheap as possible)

Oh, you can do much better. A 1 Watt heater system can be run by a $0.5 microcontroller
and a $1 DC power supply, and $0.30 worth of passive parts, $0.75 for PCB,
and case parts of decent non-flammable materials for $1 and kit purchasing
time of $0.25 per each ==> $2.80 USD per kit. You could profit at a selling price of
$5 if industrious, profit less at a selling price of $10 if lazy. 20 eur is a
very low volume selling price for what you offer.

I'm not as pessimistic about diybio equipment evolution as byowired, but
to sum it up, you haven't mentioned anything novel or high value that
I'd design into a product yet. And your price ... ehh?

John

Avery louie

unread,
Apr 22, 2012, 1:32:22 PM4/22/12
to diy...@googlegroups.com

I dont think that styrofoam is a bad idea.  Plenty of bird-egg incubators use it, and those are considered safe.  If you really wanted to go safer you could use a thermal shutoff switch like those used in mcus to prevent damage.

I think that as in any situation you have a responsibility you not harm others and do your due dilligence, but out seems that you have taken that into consideration.

That said, I do not know what the market for incubators is, but dont be afraid to be entrepreneurial.

--A

--
You received this message because you are subscribed to the Google Groups "DIYbio" group.
To post to this group, send email to diy...@googlegroups.com.
To unsubscribe from this group, send email to diybio+unsubscribe@googlegroups.com.

Andreas Sturm

unread,
Apr 22, 2012, 3:16:55 PM4/22/12
to diy...@googlegroups.com
A microchip for 0.5$ ??? Where do you get it from? China? US?


The cheapest one I use is ATtiny 13 which costs in my country 2.02€ (~3$  I think!) . When I import them from Germany, I get them for 1.80 € (above 10 pieces) and I have to pay for delivery. I've been using them for some years now, and trust me, if they were to get cheaper for me, I wouldn't waste money.

A relaise, that can stand 1A is not as cheap as you said, at least in my country. I looked for the cheapest components.





A passive security system would be worth some thinking: E.g.  A  PTC resistor and a transistor. When heated above, say 70°C, the PTC gets so much resistance that it cuts the relaise off.

John Griessen

unread,
Apr 22, 2012, 3:40:52 PM4/22/12
to diy...@googlegroups.com
On 04/22/2012 02:16 PM, Andreas Sturm wrote:
> A microchip for 0.5$ ??? Where do you get it from? China? US?

US http://search.digikey.com/us/en/products/PIC10F204T-I%2FOT/PIC10F204T-I%2FOTCT-ND/718268

> A relaise, that can stand 1A is not as cheap as you said, at least in my country. I looked for the cheapest components.

Use FETs. They handle that level of power if switched quickly by the 50 cent micro and cost $0.25.

Mega

unread,
Apr 23, 2012, 9:50:54 AM4/23/12
to diy...@googlegroups.com
A FET or MOSFET sounds great... They should be far cheaper than a relaise.

John Griessen

unread,
Apr 23, 2012, 10:11:30 AM4/23/12
to diy...@googlegroups.com
On 04/23/2012 08:50 AM, Mega wrote:
> A FET or MOSFET sounds great... They should be far cheaper than a relaise.

Yes, and they switch very quickly, but may need some careful selection to shut off
completely when using a single DC supply voltage for everything...
Will you show us a sketch or schematic of your circuit? What temperature sensor
do you use? Are you using nichrome wire to heat? How many A2D or comparator inputs
are you using on your microcontroller?

Well commented code examples at http://wiki.sgmk-ssam.ch/index.php/Hands_On_AVR#SGMK_tiny
that have the main code chunks of a heater control done already.

John

Andreas Sturm

unread,
Apr 23, 2012, 3:43:15 PM4/23/12
to diy...@googlegroups.com
Well, 2 Outputs for the LED indication of temperature, one ADC (you called it A2D).
I use LM335 because it's very very easy to use. Of course, I could also use a PTC (about half of the price or even lower, but accuracy?). and one pin  for the heater switch.  Yeah, I'll draw a schematic, and attach it to a message.
Maybe I'll add a PC-fan to distribute the air inside. (But adding things won't make it cheaper, unfortunately)

Thanks for the link!


2012/4/23 John Griessen <jo...@industromatic.com>


John

Mega

unread,
Apr 23, 2012, 3:44:56 PM4/23/12
to diy...@googlegroups.com
Well, 2 Outputs for the LED indication of temperature, one ADC (you called it A2D).
I use LM335 because it's very very easy to use. Of course, I could also use a PTC (about half of the price or even lower, but accuracy?). and one pin  for the heater switch.  Yeah, I'll draw a schematic, and attach it to a message.
Maybe I'll add a PC-fan to distribute the air inside. (But adding things won't make it cheaper, unfortunately)

Thanks for the link!

John Griessen

unread,
Apr 23, 2012, 4:19:00 PM4/23/12
to diy...@googlegroups.com
On 04/23/2012 02:43 PM, Andreas Sturm wrote:
> Well, 2 Outputs for the LED indication of temperature, one ADC (you called it A2D).
> I use LM335 because it's very very easy to use. Of course, I could also use a PTC

That is $1 each in ones qty, so maybe you could use the inherent
temperature dependent volts in circuitry you already have. Often microcontrollers
take a known steady small current diode junction in the chip and arrange to
switch its voltage to the input of
the measuring ADC, with which you can use
a look up table of temperatures. A little more work in code and testing and
a whole dollar off! No big loss in accuracy for the range you care about.
Degrees C +/- 0.5 degree is easy. If there's not an internal convenient diode volts
to measure, you can add a diode and resistor to +DC power supply as your temp sensor
attached to a comparator/ADC input.

John

Cathal Garvey

unread,
Apr 23, 2012, 4:28:08 PM4/23/12
to diy...@googlegroups.com
Hm, I wouldn't skimp on temperature control personally. It's the
critical "input" of the circuit, and LM35s can achieve 0.1C accuracy,
and are very easy to code for. Most other forms of temperature readout
that I've encountered are a bit batch-variable, and I wouldn't be
surprised if the same is true of something cool but hack-ey like your
onboard diode idea (which *is* an awesome hack, I'll grant you!).

For reproducible results, it's nice to offload variables and pay the
extra 50c for a decent IC package.

John Griessen

unread,
Apr 23, 2012, 4:46:52 PM4/23/12
to diy...@googlegroups.com
On 04/23/2012 02:44 PM, Mega wrote:
I use LM335 because it's very very easy to use. Of course, I could also use a PTC (about half of the price or even lower, but
> accuracy?).

Here's a good write up of using a diode for temp sensor.
It needs a volt reference, so pick a micro that has one of those.
Most MSP430 have one. (See below)

http://www.rst-engr.com/rst/articles/KP90APR.pdf


Here is some thermostat write up, and code for MSP430:

http://benlo.com/msp430/

He uses a micro with a $2.50/ qty one price though...so some reading of the
datasheets of the minimal MSP430G2230IDR in 8 pin SOIC pkg for 0.76 USD qty ten, $46/qty 100

That one has internal temp sensor and 4 channel ADC 10 bits.

I need to start something with that one!


John

Nathan McCorkle

unread,
Apr 23, 2012, 4:52:41 PM4/23/12
to diy...@googlegroups.com
On Sun, Apr 22, 2012 at 10:40 AM, ByoWired <byow...@gmail.com> wrote:
> Probably what will eventually turn
> public opinion against DIY biology is not the release of "frankenstein bugs"
> that the press sensationalizes but instead will be the little fires and
> electrocutions that will occur here and there as a result of the
> proliferation of ignorant equipment designs.
>

I generally agree with this. Taking the time to make sure fires don't
happen is worth it to me. Whether that means using a ceramic coffee
cup, or metal box with foil coated fiberglass. You /DO/ leave an
incubator on when you're out of the house, unless you're a shut-in.

I also agree that this is more important in terms of publicity, a fire
from a crappy incubator WILL definitely BE BAD PRESS.

John Griessen

unread,
Apr 23, 2012, 5:20:37 PM4/23/12
to diy...@googlegroups.com
On 04/23/2012 03:28 PM, Cathal Garvey wrote:
> Most other forms of temperature readout
> that I've encountered are a bit batch-variable, and I wouldn't be
> surprised if the same is true of something cool but hack-ey like your
> onboard diode idea

It's the physics behind all the temp sensors, the differences are in the implementation details.
Why would you pay more for a method or mechanism if you can't get any better results
than with the low cost method? You don't just always pay more, you shop for a low accuracy need for low dollars
and hi accuracy for hi dollars. Precision, or fine grained resolution, is inexpensive,
and you can transfer the accuracy from another temp probe to your machine and get repeatable results
without a temp standard built in.

This particular MSP430, for example, MSP430G2230IDR, has a nicely engineered diode temp sensor
inside, that can be switched in to one of its ADC channels to read the temperature of the chip.
The diode will have a small batch to batch random variation, but will always repeat very closely
as temperature differences happen. There is a temperature compensated volt reference for the
ADC so volts measured are truly accurate. Temperature repeatability of .01 deg C is probably
possible, although that's not accuracy, and would only agree that closely on long equilibration
soak times after a temperature change.

All you need is to put the chip in the stirred air you want to heat and you're accurate
to a very reasonable precision, and you could get better accuracy by calibration.
Repeatability and precision are easy to get with silicon microcontrollers.
That one I mention costs $46/qty 100, or 46 cents each and has four channels of 10 bit analog converter.

Is 0.1 degree C inherent reference standard accuracy necessary for incubation? No...
repeatability is nice, an easy cal feature is nice... Your code can have a mode where
when you press a calibrate button, it changes its look up tables to use what your external
fancy temp sensor reference standard says. The code could take input as up/down buttons
to adjust the temperature lookup table while controlling temp at 35 deg C. You would put the tip
of your hi res. platinum temp probe in the incubator for the calibration for a few minutes for each
up/down command until stable at 35.0 deg C. Then it will repeat, and be a transfer standard
for the cal thermometer for many months or years with as good accuracy and precision
as the cal thermometer even though it does not have an internal temp standard that
accurate. It does need a very stable, repeatable volt standard inside or on board, and
to make no changes to the chips used to keep its cal.

John

Nathan McCorkle

unread,
Apr 23, 2012, 6:31:18 PM4/23/12
to diy...@googlegroups.com

I want to watch RH for my incubators, this seems like the best option due to resistance to damage at 100% RH :
http://www.sparkfun.com/products/10167

Simon Quellen Field

unread,
Apr 23, 2012, 7:23:49 PM4/23/12
to diy...@googlegroups.com
This is an incubator, guys.
What organism do you know that maintains its temperature to 0.1C?
Why would you think that was important for anything you are going to incubate?

Let's suppose that you need to incubate something for 10 hours (600 minutes).
Suppose you set one incubator for 310 kelvin, and another for 310.1.
Thus one is to get 186,000 degree minutes, and the other gets 186,060 degree
minutes.

You open them both, and immediately the air temperature changes by 11 degrees
to 26, the temperature of the Petri dishes full of agar.

It takes both incubators 20 minutes to settle down to their set temperature.
You have an average of 5 degrees below the set point during those 20 minutes.
That is 100 degree minutes off already, which is more than the 60 degree minutes
by which the two incubators have been set.

But more to the point, suppose we are incubating E. coli.
Let's look at some of the protocols in use. One says to incubate at 37 Celsius for
18 to 24 hours.  Another of them says to expect a 2-fold loss of transformation for
every 15 minutes that a 1 hour incubation at 37 Celsius. A third calls for incubating
at 37 Celsius for 30 minutes, then incubating 'at room temperature' for 10 minutes.

It does not look to me like anyone is calling for one tenth of a degree of accuracy in
temperature. But more to the point, given that how long the cultures are incubating
is so variable, being off by a couple degrees does not seem like it is going to make
any difference.

If it does make a difference, is temperature the right thing to be measuring?
If the reason we incubate at 37 for an hour is to get a certain density of critters
per milliliter, why not incubate until the optical density is some particular value?

Maybe we shouldn't bother with electronics at all.
Run some hot water from the tap and adjust the temperature to 37.
Get a 5 gallon insulated beer cooler and fill it with water from that tap.
Put your Petri dish in a zip-lock bag, set it in the water, and close the cooler lid.
Take it out an hour later. The water temp is still 37.

But that takes all the fun out of building the device. So let's use the device to
do sous vide cooking, or to incubate chicken eggs (which will hatch in 21 days
even if you are off by 10 degrees). Your sous vide pork chop will come out great
even if you are off by 5 degrees.

In other words, the temperature sensor in the microcontroller is already more
sensitive and accurate than you need.

-----
Get a free science project every week! "http://scitoys.com/newsletter.html"




Simon Quellen Field

unread,
Apr 23, 2012, 7:46:07 PM4/23/12
to diy...@googlegroups.com
Why spend $8 on a sensor when you can just use a second temperature sensor?
Use the one in the microcontroller for the air temperature, and use a $2 LM34
in a shoelace in front of the fan for the wet bulb temperature. The other end
of the shoelace sits in the water bath that is keeping the temperature constant.

What are you doing to change the humidity, once you decide that you don't like
the sensor reading?

And it just occurred to me that it might be silly to use an LM34 when another
microprocessor is a third of the cost. The two processors can talk to one another
via I2C or SIP. But since the first processor has spare ADC pins and a voltage
You get 2 degree accuracy, which I'll bet is just fine for any incubator.

-----
Get a free science project every week! "http://scitoys.com/newsletter.html"


To unsubscribe from this group, send email to diybio+un...@googlegroups.com.

John Griessen

unread,
Apr 23, 2012, 7:55:50 PM4/23/12
to diy...@googlegroups.com
On 04/23/2012 05:31 PM, Nathan McCorkle wrote:
> I want to watch RH for my incubators, this seems like the best option due to resistance to damage at 100% RH :
> http://www.sparkfun.com/products/10167

OK, that's a $10, $8 sensor. We had a $2.80 bill of materials going for this incubator so far...

You might want to use that if 92% RH was optimum and you needed that. But...
it's a little out of line price wise, so....

With an incubator that stirs air for even temperatures, it will have some flow of air you
can use to advantage to measure RH by definition. The definition is pretty much a differential
temperature between wet and dry with plenty of air flow over the wet especially since
max evaporation, ( as in swinging a wet bulb thermometer around and around for a while),
has some relation to temperature drop in various humidities, and you can create a look up table of
values to do the translation. So, in an incubator, you rig a diode voltage vs. temperature sensor
($0.25 for the first one, $0.11 for the second), for wet and dry using a 1N4007 diode
for each, (and ones from the same manufacturing batch), measured by and ADC channel
all coming from the 46 cent MSP430G2230IDR microcontroller I mentioned earlier, and referencing the same
precision voltage accuracy standard MSP430G2230 has built in, and you get the same result, accurate
rel. humidity, as you get with a $10 sensor.

The $2.80 and climbing bill of materials incubator version may lack some kind
of convenience feature the sensor has, but for incubation purposes, it will hold RH
and temperature rock steady if you have some kind of movable door to let air flow across
a wick into some distilled water, or not, by degrees, in a negative feedback control loop
via the code and the above mentioned measuring sensors and ADCs.

The bill of materials is climbing higher now since we have added more performance to
the requirements, but the system is set up for heaters, and controlling one more is easy,
so we can consider Ni-Ti springs as a wet/dry door mover and be climbing the performance
hill for cheap! All that's left is ways to make 3D shapes in water proof materials
inexpensively. Then you can make the swirl chamber for air flow, and wet dry areas
with a wick and H2O reservoir, and a snazzy outside look in 3D. How much would
that 3D enclosure cost?

For 3D enclosures, I've been thinking the well proven rapid prototype method of
ink jetted water/glue holding together starch could be good for some low volume
first articles -- you can harden starch with cyano-acrylate, (crazy glue). Or 3D
print the negative shape to get a mold, harden with cyano-acrylate, coat with something
for durability and mold release, and use it as a mold till it wears out and print another...

A good material for an incubator would be HDPE. For some shapes with low accuracy requirements,
your incremental cost of producing slow low pressure molded HDPE part of about 250 grams
could be (scrap price sm vol $2/kg) x.25kg + molding effort/costs $0.4 = $0.9.

That boosts the kit BOM total to maybe $4. (with a little mfrg infrastructure put in place)

John

Simon Quellen Field

unread,
Apr 23, 2012, 8:37:32 PM4/23/12
to diy...@googlegroups.com
The enclosure can be vacuum-formed polycarbonate, the same thing they
use for 2 liter soda bottles, and for all those packaged items you see in
stores with clear plastic bubbles around them.

Then the enclosure becomes the packaging, and the cost is very low.
And building the mold to vacuum-form over can be done with Lego, or
Fimo modeling clay., or whittles out of pine or balsa. Making your own
vacuum-forming rig is simple.

Instead of Nitinol hinges, why not run the fan forwards when you want more
humidity, and backwards when you want less? It sits in a window between the
incubator and the water dish. You can add some baffles to keep the air still
when the fan isn't running, to save on heating costs, but if you have a large
enough water bath, the temperature won't change that much.

Connect the fan to two of the microcontroller's output pins. Pin A goes high
while Pin B is low, the fan goes clockwise. Make Pin A low, and Pin B high,
and it goes counterclockwise. Both pins low (or high) and the fan stops.

I knew those extra pins on the 14 pin DIP were going to come in handy...


-----
Get a free science project every week! "http://scitoys.com/newsletter.html"




John

Nathan McCorkle

unread,
Apr 23, 2012, 8:39:55 PM4/23/12
to diy...@googlegroups.com
If I made minimum wage in the U.S., $10 is just about 1.3 hours, its
well worth it to me to get rid of dealing with autoclaving shoelaces
and water storage containers to get rid of scum that builds... and
they're a source of scum/unwanted growth///possible routes for
contamination.

Maybe if you're talking developing world africa where people's time is
worth $10 every 1.3 weeks... but its just not worth it for me to
bother dealing with.

Not to mention the calibration, recalibration, curve fitting, etc...
with a sensor like this its done, and performs pretty well with a
simple buffer to extend the range of the sensor wires to any distance
a lab might see for incubator or environmental monitoring.

As for the question of how to fix humidity, I'm not sure if piezo
water misters can be autoclaved, but its a candidate choice for me.
>>> diybio+un...@googlegroups.com.
>>> For more options, visit this group at
>>> http://groups.google.com/group/diybio?hl=en.
>>>
>> --
>> You received this message because you are subscribed to the Google Groups
>> "DIYbio" group.
>> To post to this group, send email to diy...@googlegroups.com.
>> To unsubscribe from this group, send email to
>> diybio+un...@googlegroups.com.
>> For more options, visit this group at
>> http://groups.google.com/group/diybio?hl=en.
>
>
> --
> You received this message because you are subscribed to the Google Groups
> "DIYbio" group.
> To post to this group, send email to diy...@googlegroups.com.
> To unsubscribe from this group, send email to
> diybio+un...@googlegroups.com.
> For more options, visit this group at
> http://groups.google.com/group/diybio?hl=en.



Simon Quellen Field

unread,
Apr 23, 2012, 8:48:50 PM4/23/12
to diy...@googlegroups.com
It sounds like we should be designing an incubator that can be autoclaved
as an entire unit. Until you mentioned humidity, I was picturing incubating
covered Petri dishes, so the incubator itself would not need to be hyper-sterile.

Were you going to autoclave the Styrofoam?
Maybe just a soaking in bleach would do.

-----
Get a free science project every week! "http://scitoys.com/newsletter.html"


Jonathan Cline

unread,
Apr 23, 2012, 9:00:07 PM4/23/12
to diy...@googlegroups.com, jcline


On Saturday, April 21, 2012 9:25:40 PM UTC-7, ByoWired wrote:
 
 when designing something like this, you have to ask yourself what will happen if a component fails or you have some bizarre software bug that will allow the heater to run non-stop.  

One good failsafe is to use the laws of physics to eliminate any potential of runaway (I'm not licensed so I can say "eliminate any" instead of "eliminate most except in rare cases", hah).  Use a PTC in the supply.  Or as a cheap (non-approved) substitute, couple the heatsink of the voltage regulator to the incubator heater element, for example by using a machine screw with spacers so that normal operation doesn't affect the voltage regulator, while excessive heat is conducted through.  If the heater goes into runaway, the voltage regulator will shut down, which cuts off voltage to the software-controlled circuit.  With some rudimentary tweaking of the mechanical design, this would (should) provide a failsafe cutoff (about 150C) prior to flash point of other materials.  Also this could tend to brownout the supply near this threshold so both the digital side and software side need to protect against brownouts, which they should be doing anyway.   (For example, PIC chips have a brownout detection bit.)   This is the difference between DIY and professional equip.  Mostly polishing on the design, though very important.  

As for the "software error (never had one in the last years)" .. well there have been a few studies of mission critical software, for example NASA satellite firmware, and a significant yet small part of code on all projects studied found the software to be doing crazy things inside like attempting to write to read-only memory, running code from non-existent locations, wiggling pins that didn't exist, and so on....  Much like biology, software finds a way - it wants to be free!  :-D 



-- 
## Jonathan Cline
## Mobile: +1-805-617-0223
########################

Nathan McCorkle

unread,
Apr 23, 2012, 9:03:59 PM4/23/12
to diy...@googlegroups.com
On Mon, Apr 23, 2012 at 8:48 PM, Simon Quellen Field <sfi...@scitoys.com> wrote:
> It sounds like we should be designing an incubator that can be autoclaved
> as an entire unit. Until you mentioned humidity, I was picturing incubating
> covered Petri dishes, so the incubator itself would not need to be
> hyper-sterile.
>
> Were you going to autoclave the Styrofoam?
> Maybe just a soaking in bleach would do.
>

I don't think needing to autoclave the incubator is really a good
idea, unless you're talking about it being pretty small. Sprayable
with ethanol or bleach without dissolving should be a pretty good rule
otherwise. Or maybe disposable.

Simon Quellen Field

unread,
Apr 23, 2012, 9:11:00 PM4/23/12
to diy...@googlegroups.com
Indeed!

What I like even better than clever hardware design is a design that is safe
due to basic physics. 

If the heater is a thin wire in the water bath, if the bath runs dry, the wire
disintegrates, breaking the circuit.

But even the proposed design, unmodified, might be inherently safe.
The Styrofoam insulator is the only thing allowing the heater to get hot enough
to melt the Styrofoam (let alone set it on fire). If you remove the Styrofoam,
then air convection keeps the heater from getting hot enough to do any damage.
And, of course, melting the Styrofoam reduces the insulation enough to limit
the temperatures the heater can get to, given that we are talking about some
pretty low currents.

-----
Get a free science project every week! "http://scitoys.com/newsletter.html"


--
You received this message because you are subscribed to the Google Groups "DIYbio" group.
To view this discussion on the web visit https://groups.google.com/d/msg/diybio/-/HOhKhZE01mkJ.

Nathan McCorkle

unread,
Apr 23, 2012, 9:23:24 PM4/23/12
to Simon Quellen Field, diybio
Its just not worth it to mess around with this sort of junk tech
unless you're in the developing world and have no money at all... or
you're a 5-10 year old child. When the shoelace gets grimy, I'll be
the one saying "I told you so" when you don't know why you're
continuing to get spore contamination in your incubator.

Anything with sitting water needs autoclaving after time.


On Mon, Apr 23, 2012 at 9:14 PM, Simon Quellen Field <sfi...@scitoys.com> wrote:
> So you can spray the shoelace with bleach.
> :-)
>
>For that matter, filling the humidifying tank with hydrogen peroxide
> instead of water would seem to also work fine. The wet bulb won't care
> if it is wet with peroxide instead of water.
>

That sounds like a good point for keeping the water clean, longer...
but at some point, life finds a way to get grimy

John Griessen

unread,
Apr 23, 2012, 9:40:53 PM4/23/12
to diy...@googlegroups.com
On 04/23/2012 07:37 PM, Simon Quellen Field wrote:
> Instead of Nitinol hinges, why not run the fan forwards when you want more
> humidity, and backwards when you want less?

I'm just thinking of my combined incubator/centrifuge/liquid handler machine,
where the turntable for holding vials doubles as the fan.

Still, that reversing idea to change function might work in that setting also...

Hmmmm....

John Griessen

unread,
Apr 23, 2012, 9:55:30 PM4/23/12
to diy...@googlegroups.com
On 04/23/2012 07:48 PM, Simon Quellen Field wrote:
> It sounds like we should be designing an incubator that can be autoclaved
> as an entire unit. Until you mentioned humidity, I was picturing incubating
> /covered/ Petri dishes, so the incubator itself would not need to be hyper-sterile.

>
> Were you going to autoclave the Styrofoam?
> Maybe just a soaking in bleach would do.

Yes, RH is hard to benefit from if it has to be sterile.
What need is that?

Instead, could you control RH enough in a vapor permeable vial that is soaking in the desired temperature
so as to decouple temp from humidity?

Otherwise you might as well call it a flow cabinet incubator with UV, HEPA filters, etc.
and it gets large and unwieldy.

Won't mold grow on a RH sensor too? How do you sterilize it?
Toss it out every month at $10 + installation swap out costs? Naahh...

Maybe UV, HEPA filters, etc *is* what you need. They could be small,
and keep the sterile zone separate from the RH sensor and keep it accurate
by maintaining a good flow around the through the filters.

Controlling RH and temp sounds complex. Probably will be starting with
a machine to process vials, rather than open dishes of culture needing
a warm humid breeze.

John Griessen

unread,
Apr 23, 2012, 10:01:54 PM4/23/12
to diy...@googlegroups.com
On 04/23/2012 08:23 PM, Nathan McCorkle wrote:
> Its just not worth it to mess around with this sort of junk tech
Now you're acting surly. I'd appreciate it if you'd stop and
maintain a polite style on this list, whatever mood you're in.
What are you categorizing as junk tech?

You could be pessimistic, imagining what is not.

I'm not thinking of any junk assemblages. Who mentioned any?

Nathan McCorkle

unread,
Apr 23, 2012, 10:20:35 PM4/23/12
to diy...@googlegroups.com
On Mon, Apr 23, 2012 at 10:01 PM, John Griessen <jo...@industromatic.com> wrote:
> On 04/23/2012 08:23 PM, Nathan McCorkle wrote:
>>
>> Its just not worth it to mess around with this sort of junk tech
>
> Now you're acting surly.  I'd appreciate it if you'd stop and
> maintain a polite style on this list, whatever mood you're in.
> What are you categorizing as junk tech?

Sorry, my opinion is just that shoe strings and unnecessary buckets of
water is a bad option to take if you're trying to do biology, not
spend hours on something that will get grimy, or if the wick reservoir
spills.

Like I said, /I want RH control and sense/, this is not necessary for
doing a petri here or there, or an eppendorf of media.

Simon Quellen Field

unread,
Apr 23, 2012, 10:23:12 PM4/23/12
to diy...@googlegroups.com
Nathan can get surly with me -- I usually deserve it.
:-)

Email is a low bandwidth communications medium, and short comments are easily
misinterpreted because we all insert tone when the medium does not communicate
it for us.

His points are completely valid -- he is looking at the problem from a perspective
we have been ignoring. When he brought up humidity as something he wanted to
control, we left our simple low-cost incubator behind, and issues such as sterility
become important.

John's point about the humidity sensor getting damaged by autoclaving is quite
relevant, since even if it isn't damaged outright, it probably needs calibration
afterwards, and his time savings go away.

When two obviously bright people disagree, it is usually because they are not talking
about the same thing.

As for junk tech, I think shoelace qualifies. Why use a new one when that old pair of
sneakers in the closet is just sitting there?
:-)

Now, let's look at ways to make cheap autoclavable humidity sensors.


-----
Get a free science project every week! "http://scitoys.com/newsletter.html"


--
You received this message because you are subscribed to the Google Groups "DIYbio" group.
To post to this group, send email to diy...@googlegroups.com.
To unsubscribe from this group, send email to diybio+unsubscribe@googlegroups.com.

John Griessen

unread,
Apr 23, 2012, 10:25:37 PM4/23/12
to diy...@googlegroups.com
On 04/23/2012 09:20 PM, Nathan McCorkle wrote:
> I said,/I want RH control and sense/, this is not necessary for

> doing a petri here or there, or an eppendorf of media.

OK. It sounds like you're not suggesting how to make an incubator
in general, but a specific research experimental setup.

could you control RH enough in a vapor permeable covered
vial/dish/box that is soaking in the desired temperature
so as to decouple temp & RH from sterility?

John Griessen

unread,
Apr 23, 2012, 10:36:18 PM4/23/12
to diy...@googlegroups.com
On 04/23/2012 09:23 PM, Simon Quellen Field wrote:

> Now, let's look at ways to make cheap autoclavable humidity sensors <http://www.youtube.com/watch?v=Ui300HzoTOQ>.

http://www.johnsondiversey.com/Cultures/en-NZ/OpCo/Products+and+Systems/Categories/Wettex.htm
"Wettex Sponge Cloth is manufactured from a special blend of cotton, cellulose and salt. The cellulose provides excellent
absorbency whilst the salt is dissolved and rinsed away in the manufacturing process, leaving pores or hollows which improve the
high level of absorbency even further. "

So, is there some residual salt in Wettex sponge cloth?
I wonder how long it might stay calibrated?

Calibration is a good thing. You can make it easy if you put some work into
the code and the machine to be calibrated. It doesn't have to be a drag.
Really measuring things is the foundation of science.

Simon Quellen Field

unread,
Apr 23, 2012, 10:49:19 PM4/23/12
to diy...@googlegroups.com
Just because I like junk tech (I make a very good living that way!) I am still
interested in solving Nathan's problem at low cost.

One option is scraping the plastic coating off of a carbon composite resistor.
Now the resistance is a function of humidity, and a simple voltage divider gets
you the value.

For controlling the humidity, maybe all you need is one value.</