4-tube incubator and optical density meter arduino shield
Mackenzie Cowell
ByoWired
On Oct 7, 5:46 pm, Mackenzie Cowell <m...@diybio.org> wrote:
> I'd like to build a simple incubator+OD meter for up to 4 microcentrifuge
My first quick comment is that you might have a look at the following:
http://soslab.ee.washington.edu/mw/images/e/ef/EE449_SP10_Group6_MS5_Report.pdf
In that, they used LEDs and a Light-to-Frequency Converter called a
TSL230 which I think they interfaced with the Arduino. They seem
confident in their results.
I'm in the process now of testing an OD system using a similar set-up,
but it's taking OD readings and some other optical properties through
a syringe, not microcentrifuge tubes. Also, I'm not familiar with the
Arduino and have built my system around the Propeller microcontroller
- sorry. I have an incubator built around the syringes using non-
inductive resistors, but I haven't yet tested the incubator portion
yet. I have digital thermometers using DS18B20's. Although I've
heard of people using PCB traces for heating, it seems to me it might
be easier to use resistors because they can be positioned around the
microtubes without dramas.
Anyway that's it for now.
hope that helps get you started,
Mark
Nathan McCorkle
> I'd like to build a simple incubator+OD meter for up to 4 microcentrifuge
> tubes. For added cuteness, perhaps it could all be built as an add-on
> shield for an arduino. The heating could be accomplished with simple
> resistive traces on the board (right?).
> Questions:
> I've never used eagle cad before to lay out a board or had custom boards
> fabbed, but I have done some breadboard-style prototyping. I'd love help on
> that.
its not too hard, its fundamentally a drawing program, and for digital
electronics you just need to get the wires routed around the board
without crossing... if they have to cross, then you need another board
layer. analog and RF is pretty tough, but there isn't too much to
worry about with digital.
> Can 600nm ODs be taken with simple LEDs and photodiodes?
its in the visible, so cheap detectors will work, a single point is
all you need, so even an sensitive single point detector will still be
relatively cheap.
> Would typical microcentrifuge tubes mess up the OD measurements?
if they don't fluoresce they will only attenuate and diffuse the
signal, so a simple blanking with an empty, or empty with buffer in it
should work.
> How quickly do 1 mL cultures of e.coli k12 reach stationary phase?
> Does anyone want to help me with this?
i hope i can!
questions for you guys:
why 600nm? i suspect its only because its commonly referred to in the
literature, and i imagine it is that way because its a peak in the
spectrum of some easily acquired light source.
--
Nathan McCorkle
Rochester Institute of Technology
College of Science, Biotechnology/Bioinformatics
John Griessen
> questions for you guys:
> why 600nm? i suspect its only because its commonly referred to in the
> literature, and i imagine it is that way because its a peak in the
> spectrum of some easily acquired light source.
The Wash U write up mentioned bacteria they were interested in absorbed that,
while nutrient mix did not. There design is not much help since their prof
made them use a PI controller built into a black box product, (their liquid
handler bot), as part of the system. The Arduino code will not be similar,
just the LED driver and photodiode circuits. I dont' really like the use of many
sets of light sources and detectors. I'd rather see one source, one detector
and a turntable for moving samples and the blank control between the det. and source.
John
John Griessen
> I'd rather see one source, one detector
> and a turntable for moving samples and the blank control between the det. and source.
Another reason to use a turntable is it can scale up to more wells, and you
can add liquid handling later, as the Wash U folks had, but without their
constraints. Since you already have LEDs and photo-detectors on the parts
list, the same ones can be used to do the turntable stop locations.
John Griessen
ByoWired
> fabbed, but I have done some breadboard-style prototyping. I'd love help on
> that.
>
and who will be using it, etc. For example, at what temperature will
the incubator be needed to run? Will the entire unit be sitting at
room temperature, or is this something to be stored in a refrigerator
and yet have the incubator automatically ramping and controlling the
temperature to somewhere above freezing, etc. Does it need to be
portable? Must it fit in your armpit? Is there a cost constraint?
When it comes to PCBs, I have made only one-of-a-kind units, nothing
mass produced, so I've used www.expressPCB.com, which allows you to
make a few boards fairly cheap and very fast. It's not for mass
consumption, though, because the software to draw up the PCBs, though
free, is not useable by other companies (unless you pay a fee and get
Gerber files made from your design.) So in that sense, I suppose it's
not "open." But it's a cheap, fast way to just test ideas and get the
ball rolling, in my humble opinion. After you get all the bugs worked
out, you can always switch to other, open software or whatever.
Also, as for using resistors for heating elements: I suppose you could
even use regular resistors (instead of non-inductive ones) so long as
you keep your PWM frequency fairly low, maybe about 100 Hz, so long as
your power supply doesn't get weird.
To keep it simple and cheap, I wouldn't bother with turntables -
whether manual or servo actuated. It's much easier just to support
more LED/detectors. Plus it allows you to record your data on EEPROMs
or SD cards, etc. whilst you are not around to turn the table, etc..
:-)
JonathanCline
> The heating could be accomplished with simple
> resistive traces on the board (right?).
equations and it can be shown to be more difficult than it's worth.
I'd have to look up the equation but the variable is a function of
height, width, and length. Since height (thickness of copper) is a
constant for the PCB and very small, the length has to be quite large,
while fitting the length in a small space for micro tubes is a
challenge (minimizing width of course) if not impossible. You're
better off using a heating brick (basically a really thick trace slash
resistor) or nichrome wire or a light bulb or a huge resistor (but a
huge resistor won't have the power requirement most likely). It seems
many low-tech/old-school incubators do in fact use light bulbs.
Perhaps you could try a 12V hair dryer, their temperature ranges are
good for this (i.e. not dangerous even if using max current).
## Jonathan Cline
## jcl...@ieee.org
## Mobile: +1-805-617-0223
########################
ByoWired
On Oct 7, 11:12 pm, JonathanCline <jncl...@gmail.com> wrote:
>You're
> better off using a heating brick (basically a really thick trace slash
> resistor) or nichrome wire or a light bulb or a huge resistor (but a
I very much agree that using PCB traces is probably not a good idea.
But if you try to use nichrome wire, consider that it's very difficult
to solder such wire to anything ( I think you need special solder, or
at least a special nasty kind of flux), and to get any decent heat
distribution out of the nichrome wire, you would need to coil the
wire, which then requires you to electrically insulate it from its
neighboring coils - that takes super careful winding of the coil and/
or something like lots of heat shrink tubing - and when you're done,
it's still a fire hazard, IMHO. A coil also creates inductance, which
might be a consideration if you're using PWM controls. I think it's
much more straightforward to use resistors. They're small, cheap,
easy to solder, and can probably give you the power density you need
without tons of other hassles. Heck, to keep one little ole tube
warmed up at room temperature probably wouldn't take more than, I'm
guessing, a half watt or so. Think of your Spongebob night light -
about 5 watts - and if you (carefully) touch it and wrap your fingers
around it so all of its energy is now warming your hand, you can feel
that 5 watts is probably way too much for a tiny tube.
One important issue is to make sure the heat gets distributed so that
you don't have a burning hot zone in one place on the microtube while
the thermometer is reading a colder part of the tube. To help prevent
this sort of hot spot scenario, you might want to consider jacketing
the tube in a "thermal mass", which could mean that the thing in which
the tube is inserted is made of a metal that helps spread the heat
around - maybe copper tubing from the hardware store would work for
that. Holes could be put in the copper for your LEDs, etc. Resistors
perhaps covered in heat shrink tubing are then taped on with
electrical tape wrapping the whole incubator jacket - or use a larger
piece of heat shrink tubing to wrap up everything but the top access
hole. Maybe a little bit of silicone oil or gel placed on the
microtube before sticking it into the incubator/optical densitometer
would not only help heat conduction but would also perhaps improve the
light transmission characteristics of your LED/OD system.
Of course, depending on how accurate your system needs to be, you
could thermally gang all of your 4 tubes together in one metal block
so that only one PWM control would need to power one strand of heating
resistors, then hope everything averages out okay.
For LED/OD, you could have a single LED in the center, with your 4
microtubes surrounding the single LED, and then have 4 separate
TSL230's reading each of their own microtubes. At least that would
cut down on the number of LEDs you need. And if one of your
microtubes is a blank, then maybe it could serve as your control or
reference.
Cheers.
John Griessen
> For LED/OD, you could have a single LED in the center, with your 4
> microtubes surrounding the single LED, and then have 4 separate
> TSL230's reading each of their own microtubes.
When I suggested a turntable, I was imagining an arduino controlled mover
for it with well locations positively ID'd Turning the table would stir air and
equilibrate temperature well...
JG
ByoWired
On Oct 8, 12:57 am, John Griessen <j...@industromatic.com> wrote:
> LEDs aren't completely uniform in all directions, so that could have wide variations.
an LED that diffused its light fairly well. Of course each microtube
slot would have to have its own baseline measured because of (I hope)
fairly minor variations in the light from the central LED. I think
somewhat diffused light shining through a sample is better than trying
to make a brilliant narrow beam because very narrow beams might be
overly susceptible to tiny bubbles, flecks of crap on the tube, etc.
A more diffused beam will average out minor poop that might otherwise
throw your measurements way off. This is probably especially true if
you are shaking the microtubes occasionally for whatever reason, when
bubbles or froth might get in the light bath... or you happen to have
them in your armpits when jogging through the lobby of Nature
magazine.
Andrew Barney
far I'm imagining a little shield with four small copper tubes with 4
small light bulbs connected to relays. and little tubes inside the
copper being regulated. I have no idea if this is feasible, but it's
an interesting visual.
Nichrome wire probably could be used with some sort of screw clamp
instead of soldering, but i think it's probably overkill and not worth
the effort for this small of a scale. Doesn't the reprap use nicrome
wire to melt the plastic?
I don't really understand your LED measuring device your talking
about, but perhaps this sensor would be helpful.
http://www.sparkfun.com/commerce/product_info.php?products_id=8940
> --
> 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.
>
>
Andrew Barney
(plywood chamber?) where you could put several petri dishes or test
tubes inside, and the temp would be regulated? Perhaps with a
lightbulb and a fan system.
Or individual incubators for single test tubes or single petri dishes?
I'm thinking a box would be able to evenly control the temp better.
Cathal Garvey
I think it would be best to create a temperature controlling board first, which can take accurate input from several sensors and apply accurate outputs through mosfets (whether to fans, heaters or whatever).
On top of this you could then have one of several shields if you aren't using it to manage a custom build. After seeing the excellent success of the Makerbot heated build platform, I like the idea of a trace-powered heater board for space heating. If that's too awkward, a few power resistors.
For the tube-incubator, you could have a board with a central servo holding a light-metal multi-tube carousel. The carousel could be used later for automated liquid handling, but out of the box it's there for agitation and to note samples through a light gate. The carousel is heated by small resistors that are powered by inner and outer rails, which are in turn powered by brushes underneath. The metal spreads heat evenly.
Picture to come. It's just my musing upon reading this excellent list of ideas!
---
Twitter: @onetruecathal
Sent from my beloved Android phone.
On 8 Oct 2010 08:39, "Andrew Barney" <kee...@gmail.com> wrote:
Okay, so are you guy's looking to creating a DIY incubator box
(plywood chamber?) where you could put several petri dishes or test
tubes inside, and the temp would be regulated? Perhaps with a
lightbulb and a fan system.
Or individual incubators for single test tubes or single petri dishes?
I'm thinking a box would be able to evenly control the temp better.
On Fri, Oct 8, 2010 at 12:57 AM, Andrew Barney <kee...@gmail.com> wrote:
> haha, this thread has a...
Cathal Garvey
Attached picture of the carousel board. By separating power and sensor management, people can use the essence of the incubator board to make space heating incubators or to manage other sensor/power requirements easily.
If they weren't monopolised by makerbots, The extruder controller boards are basically an arduino with the sensor/power bits on-board, but it's not shield-compatible.
As for the carousel itself, you could probably use a plastic printed part wrapped in tin foil, experiment with the 'alumide' material from shapeways, or manufacture them yourself. But to get accurate Od readings, cultures will have to be agitated somehow for homogeneity. Using a heat-spreading carousel addresses many problems at once, even if it's awkward to get.
---
Twitter: @onetruecathal
Sent from my beloved Android phone.
On 8 Oct 2010 12:01, "Cathal Garvey" <cathal...@gmail.com> wrote:
I think it would be best to create a temperature controlling board first, which can take accurate input from several sensors and apply accurate outputs through mosfets (whether to fans, heaters or whatever).
On top of this you could then have one of several shields if you aren't using it to manage a custom build. After seeing the excellent success of the Makerbot heated build platform, I like the idea of a trace-powered heater board for space heating. If that's too awkward, a few power resistors.
For the tube-incubator, you could have a board with a central servo holding a light-metal multi-tube carousel. The carousel could be used later for automated liquid handling, but out of the box it's there for agitation and to note samples through a light gate. The carousel is heated by small resistors that are powered by inner and outer rails, which are in turn powered by brushes underneath. The metal spreads heat evenly.
Picture to come. It's just my musing upon reading this excellent list of ideas!
---
Twitter: @onetruecathal
Sent from my beloved Android phone.
>
> On 8 Oct 2010 08:39, "Andrew Barney" <kee...@gmail.com> wrote:
>
> Okay, so are you guy's look...
Jonathan Street
It's been a long time since I've done any bacterial work but could this device replace a conventional incubator? Or would you still need to have a larger incubator for larger volumes and petri dishes? If it isn't going to replace a conventional incubator I would be inclined to rely on that, as Andrew Barney suggests, for temperature regulation and focus on measuring the optical density.
Even then I would probably avoid an arduino shield unless I could find a splash resistant box to put the whole device in.
--
John Griessen
The carousel is heated by small resistors that are powered by inner and outer rails, which are in turn powered by brushes
> underneath. The metal spreads heat evenly.
The brushes are a complexity that can be skipped with no big loss -- unless you want to try
PCR-like speedy temperature changes. Heating the box, and stirring by rotating the carousel is
a good way to equalize temp.
John G
John Griessen
> If it isn't going to replace a conventional incubator I would be inclined to
rely on that, as Andrew Barney suggests, for
> temperature regulation and focus on measuring the optical density.
>
> Even then I would probably avoid an arduino shield unless I could find a
splash resistant box to put the whole device in.
Arduinos are not the least expensive route. I could help develop
a tool like this and be a supplier, and it could be Arduino if there's
real demand. I would also make a version that uses smaller form factors for electronics
and flat flex connectors between modules, since price always matters.
It would program just like an Arduino and be compatible, but need
an adapter from flat flex connectors to modularly connect to
standard Arduino(c) parts.
One way to get away from splash problems is put the controls on top,
with only the detector and light source poking down. All else could be at
ambient above some foam insulation. analog to digital conversion accuracy
suffers with temperature changes of the measuring amplifiers. By using one
source-detector pair, and keeping it at the same temp, and designing it
with temperature stable components yo get real accuracy over time. You
get an instrument instead of a kludge. You get something that can be calibrated
easily by placing known density sample objects in it and running a cal routine.
The known density well sized objects could be sent for NIST cal and rented or
shared like a library book to cal your gear. Why aim lower?
John Griessen
systems engineer, manufacturer of open hardware
ByoWired
Maybe I'm misunderstanding the original poster's question, but I
thought this was for 4 microtubes, in which case a carousel seems like
overkill to me. Personally I hate labware that is difficult to clean,
and all this biological gooey nasty stuff seems to require endless
cleaning. So if you're talking about physically mounting something
wet above your electronic boards, you're asking for trouble. Sooner
or later, somebody is going to drip goo down into your boards and it
will be a freaking nightmare to clean out. Having servo rotators is
also problematic for the same reason. You get gunk dripping down the
shaft and how the heck do you clean that out? Similarly, I would
advise against using anything made out of plywood. How do you clean
it? So if we're trying to keep to Mac's plea for something simple, I
would rule out as many moving parts as possible.
As for controlling the heat, I suggest Mosfets instead of relays. You
can easily use PWM control with Mosfets, but with relays the control
would necessarily be "bang bang", (totally on, totally off) which
might work okay, but it might also wear out the relays faster.
I'm not an Arduino guy, but this shield concept seems to get taken a
little too far by some people. As I said earlier, IMHO, you might
want your electronics well separated from your wet stuff. Also, it's
nice to have the circuit readily accessible to oscilloscope probes,
etc. in case you have to debug things. Having everything packed away
in a brick of PCB boards makes it difficult to troubleshoot or modify
things on the fly.
Another note, generally speaking about incubators whether large or
small: if your control system goes haywire for whatever reason, even
when small voltages are involved, there's always the possibility of a
heat source starting a fire. That's another reason I would steer away
from things made of wood or cardboard. It's safer to use metal, some
types of plastic, and use fiberglass for insulation instead of cheap
foam. That might be showing an abundance of caution, but since these
are DIY systems, you really need to think about what happens when
things go wrong. Even just a few watts of energy, when blanketed by
insulation, can eventually lead to temperatures high enough to ignite
a material.
Brody
> How quickly do 1 mL cultures of e.coli k12 reach stationary phase?
lab strains of E. coli double once per 20min in completely ideal
conditions (temperature, aeration, nutrient levels, cell density,
etc). In practice, I'd say that doubling times around 30-40min are
more of the norm. So, it really depends on the volume of your
culture, the number of cells in your inoculum, and your growth
conditions.
Also don't forget that the cells will go through a "lag" phase before
starting exponential growth. The length of the lag will depend on
what conditions the bacteria were experiencing before you inoculated
your culture. If you had them on plates in the fridge, lag will be
relatively long. If inoculate your culture from another culture
actively growing under similar conditions, lag will be short.
So, back to the question, how long until stationary phase. You will
need to define stationary phase. Traditionally, the cultures are
thought to reach stationary phase somewhere in the range of OD600 =
0.6 to 1.0. A publication from the Journal of Bacteriology (http://
jb.asm.org/cgi/content/full/189/23/8746) shows that the cells change
their physiology around OD600 = 0.3, suggesting that the "exponential"
growth phase ends there. Let's, for the sake of simplicity, define
stationary as OD600 = 1.0. If you inoculate at 0.5, it will take
about a half an hour to get to saturation. I'm sure you can do the
math to adjust your inoculation amount to fit the time frame you are
interested in. That's theory. In practice, the best would be to plot
the growth of your cultures on a semi-log scale (OD on the log axis,
time on the linear axis) and see how the bugs grow in your specific
conditions.
ByoWired
> The known density well sized objects could be sent for NIST cal and rented or
> shared like a library book to cal your gear. Why aim lower?
>
somewhat relative anyway, just to get a growth curve, so plus or minus
a few perfect would be good enough? At least that was the gist I got
out of the UW paper I referenced earlier. Are there applications
where absolute measurements are essential? And, if so, does Mac need
that for this project?
ByoWired
On Oct 8, 2:57 am, Andrew Barney <keen...@gmail.com> wrote:
> I don't really understand your LED measuring device your talking
> about, but perhaps this sensor would be helpful.http://www.sparkfun.com/commerce/product_info.php?products_id=8940
>
it's the same exact thing. If you look near the bottom of the
sparkfun page
http://www.sparkfun.com/commerce/product_info.php?products_id=8940
you'll see it's a TSL230.
:-)
John Griessen
> On Oct 8, 10:40 am, John Griessen<j...@industromatic.com> wrote:
Why aim lower?
>>
>
> Does OD data really need to be that accurate? I thought it was all
> somewhat relative anyway, just to get a growth curve, so plus or minus
> a few perfect would be good enough?
You can't tell what observation will be important. There's no such
thing as qualitative science -- you need good numbers.
At least that was the gist I got
> out of the UW paper I referenced earlier. Are there applications
> where absolute measurements are essential? And, if so, does Mac need
> that for this project?
Like I've said before, my interest is in being a low cost supplier of open hardware
products, so i want to satisfy more than one person. Each person can take
my kits and documentation and do what they like, but I will be making
tools for groups, not individuals. That will fund my desire to have a variety
of projects to design, redesign, improve with outside help, think of new projects,
repeat.
John Griessen
Mackenzie Cowell
--
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.
ByoWired
On Oct 8, 10:19 pm, Mackenzie Cowell <m...@diybio.org> wrote:
> FTIR theory
Another buzz phrase to consider is "matching the indices of
refraction" of the LED housing and the microtube. I think I already
suggested a dab of silicone oil or grease to the outside of the
microtube might help with that. But it's very possible that no such
optical coupling will be necessary.
>
> My goal is traditional incubation temps: 37 and 42 degrees. UV light or 99
> degree temps would be a plus for sterilization.
C for sterilization? I think I would try to make the tube holder/
incubator so it can be totally separated from the electronic system
and autoclaved, tubes and all, especially since 99 C doesn't, by
itself, guarantee sterilization. Design the tube holder so about 20
units can fit into a single pressure cooker?
As for UV, that's a very good question. I've been yearning for a good
LED-type UV source that has a spectrum that can do some serious
sterilization. Any LEDs out there that can crank out a wavelength of
250nm? Is 240nm really necessary to UV sterilize or can some other
wavelength maybe for longer times be used? Please let me know!
I guess the more I think about it, the more logical it might be to
ditch the single LED concept and just go with 4 separate LEDs. That
would cut down on the mechanical headache of making reflectors,
diffusers, etc. and LEDs are super cheap anyway. If you want to learn
how to multiplex things, you might even be able to use multiple
colored LEDs and do your ODs at a variety of wavelengths. Having it
blink from red, green, yellow, oh and especially blue is really cool.
You might even use the 4 tubes to play some kind of memory game, like
Simon or whatever it's called, while awaiting stationary phase.
For agitation, how about a vibrator motor? No, not that kind of
vibrator! I'm talking about the kind used for pagers, cell phones,
etc... Take a look...
http://www.sparkfun.com/commerce/product_info.php?products_id=8449
Mac, you still haven't told us who the target users are for this. Is
this for school kids? IGEM contests? or are you trying to make armpits
obsolete?
:D
John Griessen
> For agitation, how about a vibrator motor? No, not that kind of
> vibrator! I'm talking about the kind used for pagers, cell phones,
> etc...
It's probably the same kind.
ByoWired
On Oct 8, 7:24 am, Cathal Garvey <cathalgar...@gmail.com> wrote:
>.... But to get accurate Od readings, cultures will
> have to be agitated somehow for homogeneity....
How much agitation is enough? Are we talking about a twitch now and
then or do these things have to be shaken like a cocktail? And,
besides the desire for homogeneity, are there any other reasons for
agitation? For example, is there a need to mix air with the sample?
Will bubbles form on the sides of the tube? Or froth on top? Would a
miniature magnetic stir bar/ball/bead maybe do the trick if an
actuating magnetic field can be set up around it?
Do some cultures ever generate pressure that might build up in the
microtubes?
Cory Tobin
> microtubes?
It sounds like the plan is to grow E. coli in 1.7mL plastic tubes
(correct me if I'm wrong). The typical 1.7mL tube is air tight.
You'll probably want to expose the culture to the air some how.
Poking a hole in the cap may be sufficient, although I've never tried.
If the culture is not exposed to air the growth will be significantly
limited. See here: http://aem.asm.org/cgi/content/short/13/1/109 The
article should be free, if you hit a pay-wall let me know. Anyways,
the typical culture tube has a cap that doesn't seal tightly.
One thing to note when building this thing is that water in a 1.7mL
tube doesn't slosh around. To mix liquid in one of these tubes, the
tube has to either be inverted or shaken fairly violently, like with a
vortexer.
As for Mac's question of how long it takes a 1mL culture to reach
stationary phase, I usually grow 4mL cultures and they reach
stationary between 11.5 and 12 hours. The 4mL cultures will be
different than 1mL, but maybe this info is helpful.
-Cory
Cathal Garvey
--
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.
ByoWired
I'm not familiar with the Arduino, so I'm wondering how many pins will
be available on the Arduino for sensors, incubator control, etc.?
Will you want some kind of data display? LCD? TV screen? mouse menu
control? How about data recording on SD cards, EEPROM, etc.?
John Griessen
> One thing to note when building this thing is that water in a 1.7mL
> tube doesn't slosh around. To mix liquid in one of these tubes, the
> tube has to either be inverted or shaken fairly violently, like with a
> vortexer.
Yes, the meniscus will be as tall as the tube is around and not much will
change its shape with ordinary forces. So designing its carousel with
a rapid moving servo that can be programmed to oscillate at or above
a resonance of the liquid/tube combination would help stir. If there were two small
holes in the tube cap, that same vibration movement would cause air flow in/out.
John Griessen
ByoWired
On Oct 10, 1:34 pm, John Griessen <j...@industromatic.com> wrote:
>
>So designing its carousel with
> a rapid moving servo that can be programmed to oscillate at or above
> a resonance of the liquid/tube combination would help stir. If there were two small
> holes in the tube cap, that same vibration movement would cause air flow in/out.
>
challenge hitting that resonance while driving 4 microtubes and their
incubator heaters (and perhaps their LED/Sensor pairs?) all at the
same time as a single unit. The mass of such a 4-tube unit might be
too great for a reasonably sized servo to jiggle into resonance, but
I'm just sorta guessing about that. If you plan on having the whole
thing rotate like a Ferris wheel and take advantage of gravity, I
suppose that would be another issue.
Of greater concern to me, and of something I know little about, is the
potential for sucking in unwanted microbes from the environment along
with the flow of air. I presume you would want some kind of air
filter over your vent holes? And for bacteria, I presume you would
want to eliminate the possibility of bacterial spores getting in, so
that means using a submicron filter, maybe 0.2 microns? Would that be
a necessity? And I hope somebody will clue me in on this - can such
filters get wet? Can they filter air one moment, then get wet, then
continue filtering air?
Instead of designing vent holes, would it be better to divide the
microtube volume into so many parts liquid, so many parts air, so that
there's enough air available to grow a culture to the desired density,
etc.?
Mine are not rhetorical questions. I'm really hoping for some
experienced biologists, Arduinoologists, etc. to plug some insights
here.
:-)
ByoWired
On Oct 7, 5:46 pm, Mackenzie Cowell <m...@diybio.org> wrote:
>... For added cuteness, perhaps it could all be built as an add-on
> shield for an arduino.
> I've never used eagle cad before to lay out a board...
I, too, have never used Eagle before, so I looked up the freebie
version and it says:
"The useable board area is limited to 100 x 80 mm (4 x 3.2 inches)."
http://www.cadsoft.de/freeware.htm
I suppose if you want to go the freebie Eagle route, you'd have to
take into account the board size limitations.
I just thought I'd point that out.
John Griessen
> It might be a
> challenge hitting that resonance while driving 4 microtubes and their
> incubator heaters (and perhaps their LED/Sensor pairs?) all at the
> same time as a single unit.
Engineering is the balance of price and performance.
To the engineer anything is possible, few are feasible, and
odd combinations sometimes win out.
To the businessman all the is "possible" is buyable off the shelf.
You can go above resonance and do a move then wait a while
and get stirring for cheap. Resonance peaking is not necessary.
"rotate like a Ferris wheel" --> not in my wildest dreams.
" air filter over your vent holes?"--> goretex sonic bonded.
"hoping for some experienced biologists, Arduinoologists"--> zero Arduinology here...
I think they're vastly overrated.
JG
ByoWired
On Oct 10, 10:32 pm, John Griessen <j...@industromatic.com> wrote:
>
> "rotate like a Ferris wheel" --> not in my wildest dreams.
Personally I think a carousel is unnecessarily complex. How about
achieving agitation with a single solenoid that can periodically
"kick" the rack of microtubes? If the rack is somewhat springy, it
will even wiggle around a little after mixing some air into the
fluid.
> " air filter over your vent holes?"--> goretex sonic bonded.
to the entire unit? And how will this goretex filter be cleaned? or
would it be disposable? How would it be sterilized?
> "hoping for some experienced biologists, Arduinoologists"--> zero Arduinology here...
> I think they're vastly overrated.
>
like a toy to me. But I bet that's because it's so popular and so a
lot of what people make with it are toys. I think I've heard people
say the same thing about the Propeller chip, that it's a great one-
trick pony, etc. and I can't understand that opinion. I suppose
everyone is a xenophobe when it comes to other people's
microprocessors. Ha! So tell us what you're using so we can all
publicly fear and loathe it. :-)
John Griessen
How about
> achieving agitation with a single solenoid that can periodically
> "kick" the rack of microtubes? If the rack is somewhat springy, it
> will even wiggle around a little after mixing some air into the
> fluid.
Sure, that will work and a carousel can be kicked also and use only
one set of light source and detector and a blank tube can be
moved in front of it for a zero reference cal.
>> "hoping for some experienced biologists, Arduinoologists"--> zero Arduinology here...
>> I think they're vastly overrated.
>>
>
> What do you prefer to use?
MSP430 is good. Arduino chips, (Atmel), are good, and the programming
environment is a plus by way of popularity. Lilypad is something to
look at also -- same programming environment, different form factor.
From sparkfun site
==========================
http://www.sparkfun.com/commerce/product_info.php?products_id=9266
http://www.sparkfun.com/commerce/product_info.php?products_id=9218
* The recommended serial connection board is the FTDI Basic. If
you have multiple Arduino Pro boards (many people do) you will only
need to buy one FTDI Basic board ($13.95)
* A big difference between the Pro boards and the Duemilanove is
voltage. Duemilanove runs at 5V, the Pro comes in either 3.3V or 5V.
Select the correct voltage for your project. Many sensors operate at
3.3V so you will need to decide which voltage will work best for
your specific projects.
=====================
The problem I have with modular Arduino for instrumentation is
they're expensive and don't have mounting holes compared to
a custom board. Their use of through hole
parts is for experimentation with microcontrollers, and once you get
enough experience there is little need for that. I see people refining
and adding to a specific instrument design much more if it is inexpensive.
Low cost will attract users and developers.
Not being constrained to the Arduino connectors and through hole parts
simplifies and cost reduces further.
I use flat flex connectors between modules. They hand solder
well and cost only $.25 US in quantities.
That kind of slim connector is a natural for separating the light source
and detector from the main microcontroller and user interface related parts
so the detector hangs down next to a tube and the main board is up from
leaks and sealed away from splashes.
The detector is quick change then,
which is good since it will likely evolve/mutate more often than the main board.
John
John Griessen
http://ecosensory.com/diybio/carousel_greycoded-1.jpg
and have come to wanting to layout
a carousel size and want to know what is the right well size?
Is 7.95mm, (as 90's PCR machines had), good for a well size,
or are there smaller capped tubes now?
I'm going to layout 4 and 8 and 16 tubes along with absolute encoder marks
on the plastic carousel so it can be pushed on the shaft of a simple DC motor
that drives it directly. The encoder marks will have LED/phototransistor
black/white detectors above them that the microcontroller can use to move to
the right spot so the optical density emitter/detector pair is lined up with
any of the wells. The same movement can be used to do stirring.
Here: http://en.wikipedia.org/wiki/Rotary_encoder you can see
two versions of code wheels that can be laser printed to do the move-to-a-well function.
I'm thinking of a hinged or somehow accurately located lid that has the main
controller board, optical density emitter/detector pair, and encoder detectors
in it aiming down. The motor can be below the carousel -- with a splash containing shape
surrounding the motor shaft, and the carousel center being like a roof
with a drip edge for the motor shaft area so no goo gets in it. That lower deck
below the carousel will need drain holes also to be goo proof.
John Griessen
Jonathan Street
http://www.coleparmer.com/1/1/50833-eppendorf-microcentrifuge-tubes-capacity-1-5-ml-500-pack.html
ByoWired
On Oct 11, 11:27 pm, John Griessen <j...@industromatic.com> wrote:
>...
> controller board, optical density emitter/detector pair, and encoder detectors
> in it aiming down.
Cathal Garvey
--
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.
--
letters.cunningprojects.com
twitter.com/onetruecathal
twitter.com/labsfromfabs
ByoWired
> two versions of code wheels that can be laser printed to do the move-to-a-well function.
optical properties and geometry of the sample container are not
normally an issue in making measurements. Measuring ODs through the
sides of a microtube will be a totally different story. The geometry
of the microtube will cause the tube to have various lensing effects
that can converge light or diverge light, maybe even do both if your
source light beam is fat enough. These effects can probably be
accounted for empirically, BUT the incidence angles of the optics must
be highly repeatable. If your carousel system can't bring your
microtubes back to the same point, then it's possible that even the
slightest non-repeatability will screw up your measurements. Before
you decide on a carousel system, you had better figure out how
sensitive your microtubes are to deviations in the light path. And
then compare that sensitivity to the repeatability and accuracy of
your carousel driving system. It's possible that just a degree or so
in system slop will render your design useless. I'm guessing this
will especially be the case if you are beaming light down through the
tops of the tubes or up from the bottom. Along that direction, beam
deflection will be subject to the thickest cross sections of plastic,
and therefore the lensing effects will be greatest. This is not to
say that the alternative (beaming light through the sides of the
microtube) will be inherently repeatable: you will have to be careful
with that design, too.
John Griessen
> 1.5 ml eppendorf tubes which are slightly larger. A quick google search suggests 11 mm.
Thanks.
On 10/12/2010 08:16 AM, ByoWired wrote:
> How do you plan on incorporating the heater/incubation aspect to this?
Keeping it simple, where heat up time in the range of 30 minutes is OK, I am
thinking of heating the air in the enclosure, and making the polypropylene
carousel wheel have low mass near the tubes by CNC carving out zones for
air to flow in. The tubes don't need to be held in a hole made by one drilling,
they can be held in a hexagon drill hole pattern with a hole out of it so the
side contact is tiny and has plenty of air flow space.
http://ecosensory.com/diybio/carousel_tube_wells-1.jpg
On 10/12/2010 08:35 AM, Cathal Garvey wrote:
> For bonus points, a carousel or shaker with separated metal pieces to hold the samples, each with a small heating resistor, so
> samples can be kept at different temperatures (for really fast growth-curve analysis of growth vs. temperature).
I'm planning a small inexpensive machine with USB and an ethernet option and a web server interface,
from which you can aggregate data readings with RSS or some such, so your table of data can evolve in
front of your eyes. 4 tubes is a small enough number to heat all at the same temp and have another unit going
at a different temp.
If there's really a demand for one tube controllability, the thing to do is put serial or IIC (I2C)
ports on the regular 4 unit machine so it's controller and USB can be used to relay data from
one tube (non-shaker) units. A one tube unit is not my idea of a
first machine to make, since I cant imagine that it would sell the best. 4 isn't many.
I bet 8 tube units will sound good to some folks.
The price is going to something like $80 for 4 tubes.
You can't make a shaker for a one tube, plus a controller and a comm port to sell for $20.
So a one tuber (non-shaker) would sell for $40. Would that matter compared to
4 8 and 16 tube agitated machines? Costing $80 $110 and $130? Really?
On 10/12/2010 10:41 AM, ByoWired wrote:
> If your carousel system can't bring your
> micro-tubes back to the same point, then it's possible that even the
> slightest non-repeatability will screw up your measurements.
I dreamed up a nice code wheel design last night with some greyscale at the boundaries
of the line up marks for wells. The code wheel parts for telling which well won't need that
since that is a big coarse step from well to well. A band of greyscale accurately positioned
relatively to the wells can be done by making the wells accurately spaced to each other, and
then making the code wheel pattern accurate wrt itself. The pattern can be made by laser printer
toner transfer and not accurate relative to the wells, except for decent concentricity.
Mismatch of the code wheel pattern and wells can then be accounted for by moving to a level
along the greyscale and controlling the position to that level. I am sure it will be able to do
+/- .076mm (3 mils) repeatability.
John Griessen
PS any more votes on what a good well size is?
Ross Durland
Nathan McCorkle asked a while back, "Why 600nm?" I agree there's
nothing magic about 600 nm. In this application, you're not even
looking at true absorbance. Instead, you're looking at light
scattering by the individual cells. (That's why OD600 depends not just
on the number or total mass of bacteria in a sample, but also on their
size & shape.) I've seen people use OD550, OD590, & OD660.
I agree with ByoWired's concerns about accurate OD measurements. I
think it will be extremely difficult to do in the proposed set-up.
Microfuge tubes aren't flat, aren't optically transparent, and aren't
uniform. On top of that, accurate OD requires a homogenous sample, so
good mixing is necessary.
As an alternative to microfuge tubes, you should know that there are
companies that sell plastic disposable cuvettes for visible wavelength
OD measurements. These typically hold a few mL total. Perhaps they
would work? They're rectangular and transparent, so they'd eliminate
some of the above concerns. I don't know if anyone makes them with
tight-fitting lids, though.
The other question is whether OD measurement is needed at all for an
instrument like this. I can imagine why someone might want automated,
periodic or continuous monitoring of OD in multiple cultures. But I
can't see why they'd want to do it in microfuge tubes. I've seen
people grow E. coli in microfuge tubes before, but only for simple
things like isolating plasmids from a cloning experiment. That doesn't
require any OD measurement or careful monitoring. Just inoculate,
close the lid, and put in a 37C oven. If available, you can put the
tubes on a rocker or rotator for better mixing. You don't get a really
high cell density, mainly due to limited O2, but it's usually good
enough.
You may well be able to build something that does what you're
describing, despite the identified problems, but are you sure this
isn't a solution in search of a problem? If you're just doing it for
your own purposes and/or enjoyment, by all means go for it. But it
sounds like you hope to sell these, so you might want to take a closer
look at your intended market.
John Griessen
> I can offer a few comments from a PhD molecular biologist.
OD requires a homogenous sample, so
> good mixing is necessary.
>
> As an alternative to microfuge tubes, you should know that there are
> companies that sell plastic disposable cuvettes for visible wavelength
> OD measurements. These typically hold a few mL total. Perhaps they
> would work? They're rectangular and transparent, so they'd eliminate
> some of the above concerns.
Sounds good. I've not put time into a layout yet, so your comment
is very welcome and helpful. We've been thinking of Makerbot
output as an affordable way to design and make parts that keep the parts
count low and assembly easy, and rectangular openings are easy for
a 3D printer if tolerances are loose.
I don't know if anyone makes them with
> tight-fitting lids, though.
That might not be important. If it turned out to be a want,
I could make some silicone plugs for rectangular shapes.
> The other question is whether OD measurement is needed at all for an
> instrument like this. I can imagine why someone might want automated,
> periodic or continuous monitoring of OD in multiple cultures.
Yes. Adding automation to also add or remove material as well as do
OD measurement is why I like a carousel based system.
I can see wanting your microbes to grow fast, and not reach
a high density where they are living in their own pollution,
since that might affect their behavior and slow their reproduction.
So adding automation to siphon off cells, add nutrients without
adding contamination would get you rapid production of generations
of microbes sexually reproducing, right?
What is used for automated pipetting for purpose of avoiding contamination
when moving volumes like .1 to .5 ml? What kind of plastic disposable
pipettes are available?
Just inoculate,
> close the lid, and put in a 37C oven. If available, you can put the
> tubes on a rocker or rotator for better mixing. You don't get a really
> high cell density, mainly due to limited O2, but it's usually good
> enough.
With the agitation capability I'm planning, good mixing and good
O2 would get good densities, and rapid growth. Selling points?
> You may well be able to build something that does what you're
> describing, despite the identified problems, but are you sure this
> isn't a solution in search of a problem?
No. I might make an agitated incubator without OD at all first.
The carousel agitation will get even temperature, and good air flow
compared to tubes in an oven on a rocker. The carousel will have
absolute positions encoded on it, so it can twirl to fan the air inside
and stop at a location of any well reliably. I can imagine the OD measure
would be valuable if the growth rate was optimal because it might be possible
to get completion in the same work day. The machine will have ability to
communicate, so alarms can go off when an OD level is reached. You could program
it to "cook" and when the alarm goes off, you have enough cells for the next step.
> But it
> sounds like you hope to sell these, so you might want to take a closer
> look at your intended market.
Yes, thanks. Where else is a good place to ask?
John Griessen
Jonathan Street
You seem to be massively expanding the scope of this project. Laboratory automation is an interesting project but it goes way beyond a simple incubator or spectrophotometer.
I've already expressed my reservations over a combined incubator/spectrophotometer. I think it is important to consider what experiments would benefit from this combination. If you want to grow lots of cells however carefully you optimise the conditions in an eppendorf you won't even begin to approach the yield from a 500 ml flask. If your interest is specifically in studying growth rates then how will you keep the conditions in the eppendorfs optimal.
I think there would be markets for a vortexer, an incubator and a simple spectrophotometer but not for this strange cross-breed.
ByoWired
On Oct 13, 1:53 pm, Ross Durland <rhdurl...@yahoo.com> wrote:
> but are you sure this
> isn't a solution in search of a problem?
poster had a specific user in mind when requesting a simple 4
microtube device and I've been waiting for more inputs on who exactly
wants this and what are the design/cost constraints, etc. but there
have been few inputs since the idea got tossed out here. I'm guessing
maybe Mac got busy with IGEM and so forth. Due to a lack of inputs, I
boldly presumed this device was perhaps meant to compete with the all-
too-popular armpit incubators we've heard so much about and therefore
I've suggested keeping it simple and avoiding the temptation to go to
Full Metal Robot on this one.
As far as measuring OD is concerned, I'm guessing that it's possible
to do ODs in such a tiny tube (or something similar to ODs that will
give indication of growth phases, etc. ) but I'm also guessing it will
require great care with optical alignments and much experimentation to
get results that are meaningful. It's also very likely it wouldn't
work with all types of microbes.
Some details on the intended user/application would help define the
problems and therefore the solutions at this point. Otherwise this
has been just a fun five day brainstorming session.
John Griessen
> Otherwise this
> has been just a fun five day brainstorming session.
It *has* been fun.
Hopefully I'll get more
interested responses as I look elsewhere. So far the directions of wants are all over the place,
so it's hard to tell what would sell enough to cover costs. I'll work up some kickstarter
campaigns and see if I get any ideas to pre-sell.
JG
Russell Whitaker
>
> On Oct 11, 11:27 pm, John Griessen <j...@industromatic.com> wrote:
>
>>
>> Here: http://en.wikipedia.org/wiki/Rotary_encoder you can see
>> two versions of code wheels that can be laser printed to do the move-to-a-well function.
>
> In everyday OD measurements, the sides of the cuvette are flat, so the
> optical properties and geometry of the sample container are not
> normally an issue in making measurements.
I don't know about your lab, but many labs have long gone over to
NanoDrop measurement without cuvettes; a few like the 2000c
model have an _optional_ cuvette well, but for "good enough"
quantitation of dsDNA extractions, dilutions, & PCR product
quantitation, the non-cuvette retention system is the way to go.
http://www.nanodrop.com/HowItWorks.aspx
If you need "oh my god" spectacularly sensitive, reproducible
quantitation, the Invitrogen PicoGreen kit is the way to go
(I have lots of experience with both).
> Measuring ODs through the
> sides of a microtube will be a totally different story.
Don't. even. attempt. this.
Russell
--
Russell Whitaker
http://twitter.com/OrthoNormalRuss
http://orthonormalruss.blogspot.com/
Russell Whitaker
>
> I agree with ByoWired's concerns about accurate OD measurements. I
> think it will be extremely difficult to do in the proposed set-up.
Let's call it "practically impossible," because it's likely the
reproducibility between tubes and between lots of tubes will result
in variances so large as to render the results statistically useless.
> Microfuge tubes aren't flat, aren't optically transparent, and aren't
> uniform.
And that's reflected in their low, low prices.
>
> As an alternative to microfuge tubes, you should know that there are
> companies that sell plastic disposable cuvettes for visible wavelength
> OD measurements. These typically hold a few mL total. Perhaps they
> would work? They're rectangular and transparent, so they'd eliminate
> some of the above concerns. I don't know if anyone makes them with
> tight-fitting lids, though.
>
Dr. Durland, I've had to order these on a few occasions during my
undergrad days at Columbia. They range anywhere from $1-$3 dollars
*per cuvette* because of an exacting same-lot optical absorbance
uniformity requirement. For the benefit of others who don't know: the
very reason they're square is to provide an effectively-normal angle of
incidence for the light beam emitted down the very
narrow groove at the bottom of the well. Everyone who uses cuvettes
uses *two* at minimum: at least one sample and - always! - one
blanking standard with DI water or TE buffer or whatever the
assayed sample happens to be diluted into, in order to determine
the baseline absorbance which the instrument can use to subtract
from the sample measurement.
>
> You may well be able to build something that does what you're
> describing, despite the identified problems, but are you sure this
> isn't a solution in search of a problem? If you're just doing it for
> your own purposes and/or enjoyment, by all means go for it. But it
> sounds like you hope to sell these, so you might want to take a closer
> look at your intended market.
>
This last point needs to be said more often.
Russell Whitaker
>
> As far as measuring OD is concerned, I'm guessing that it's possible
> to do ODs in such a tiny tube
You need to specify what you're trying to measure. 1.5ml Eppendorf
tubes are not "tiny" for most OD absorbance assays, at least in respect
of dsDNA concentrations: 1ul (1 microliter) is routine. Look up "NanoDrop."
John Griessen
> On Wed, Oct 13, 2010 at 5:02 PM, ByoWired<byow...@gmail.com> wrote:
>>
>> As far as measuring OD is concerned, I'm guessing that it's possible
>> to do ODs in such a tiny tube
>
> You need to specify what you're trying to measure. 1.5ml Eppendorf
> tubes are not "tiny" for most OD absorbance assays, at least in respect
> of dsDNA concentrations: 1ul (1 microliter) is routine. Look up "NanoDrop."
>
If there are enough people lined up to crowd-fund it, I can make a
full-plastic-robot that will do an OD straight through the middle
of many cylindrical tubes 11mm across with an aiming accuracy of
+/- 0.3mm, compare to low and high OD reference tubes for calibration
purposes, act as an agitating incubator with OD-level-met-alarms
for each well, stir well by rapid and semi random rotary motion,
drive air flow through the oven zone for high O2 concentrations,
and function as a centrifuge.
That will probably happen in stages instead of on the first product iteration.
Later, fluid handling can be added such that fluid could be moved between vials,
or moved to a waste drain -- I just need to figure operate a disposable polypropylene
dropper and move it vertically with low incremental complexity added.
John Griessen
ByoWired
On Oct 14, 12:32 am, Russell Whitaker <russell.whita...@gmail.com>
wrote:
>
> I don't know about your lab, but many labs have long gone over to
>
> http://www.nanodrop.com/HowItWorks.aspx
>
That's certainly a very clever system, but I don't see how it lends
itself to measuring anything, say, overnight. Unless you were careful
to control the humidity to exact conditions, I think nanodrops would
be vulnerable to nanoevaporation, and probably wouldn't be the kind of
DIYer system that was originally asked for.
>Let's call it "practically impossible," because it's likely the
>reproducibility between tubes and between lots of tubes will result
>in variances so large as to render the results statistically useless.
first. After all, as I understand it, OD measurements are more of a
relative measurement than anything else. I've recently had some
involved discussions with a biologist about the meaning of OD, and I'm
told I don't understand the concept. Yet, it seems to me that OD is
relative in the sense that there are no absolute units for it. It's
not defined as "so many photons per cm" or whatever. The reason you
need to "blank" the OD device is so you establish a reference, and
from that blank you then get an OD. To measure OD on one machine and
then take your sample to a different machine will give you no
meaningful ODs because the second machine needs a blank for a
reference. The second machine can't "just tell" what the OD is
without that blank. Wikipedia gives a formula which I believe is
standard, and I don't see any units in there:
http://en.wikipedia.org/wiki/Optical_density
Most important is that your light measuring system needs to provide a
linear response (X amount of light gives Y output of signal, and 2X
light gives an output of 2Y signal, etc.) That's why I suggested the
TSL230. It's pretty darn linear over it's entire range.
Also, OD is not a pure "straight line" measurement when dealing with a
liquid that contains bits and pieces that scatter the light, as cells
do. In a batch of microbes, the OD is actually the sum of light
transmission AND scattering effects, which is a somewhat messy system,
optically speaking. So, given that the system is somewhat messy to
start with, using microtubes might be just another factor that needs
to be accounted for. I'm thinking it might be possible to "subtract"
the microtube's optical messiness so long as the tube itself is not
allowed to wiggle around and the light path is kept at a fixed angle.
The microtube will scatter and transmit light in bizarre ways, but so
long as you keep it fixed and measure the OD across its diameter, I'm
guessing the messiness won't bury the signal in all the noise.
I think it's worth a try if anyone is interested.
John Griessen
> range anywhere from $1-$3 dollars
> *per cuvette* because of an exacting same-lot optical absorbance
> uniformity requirement.
hmmm.... I've never been thinking to get density measurements
you would base dilutions of solutions on, only messy suspensions
of microbes or cells that are reproducing to get a reasonable stop point for
an incubator -- when they look like a grey cloud in the water.
The droplet-on-surface-of-microscope-objective way of
putting the sample to assay between source and detector needs to wait
for another instrument altogether. 1 to 3 cents per container is the target.
The target market is the el-cheap, but neato market. Most serious
researchers will take much more time and absolute proof before they
will buy anything. Meanwhile...
On 10/13/2010 11:32 PM, Russell Whitaker wrote:
>> ODs through the
>> > sides of a microtube will be a totally different story.
> Don't. even. attempt. this.
No need to worry. Just substitute "suspension density" for OD
and call it an incubation completeness sensor is all.
I'd wondered what it meant when I heard a mention of spectrophotometer...
OD was the wrong jargon to use describing an incubator for cheaply contained samples
conveniently grown with agitation at temperature.
John Griessen
Evgeny Vlasov
On Thursday, October 7, 2010 at 4:34:43 PM UTC-6, ByoWired wrote:
On Oct 7, 5:46 pm, Mackenzie Cowell <m...@diybio.org> wrote:
> I'd like to build a simple incubator+OD meter for up to 4 microcentrifuge
> tubes....
My first quick comment is that you might have a look at the following:
http://soslab.ee.washington.edu/mw/images/e/ef/EE449_SP10_Group6_MS5_Report.pdf
In that, they used LEDs and a Light-to-Frequency Converter called a
TSL230 which I think they interfaced with the Arduino. They seem
confident in their results.
I'm in the process now of testing an OD system using a similar set-up,
but it's taking OD readings and some other optical properties through
a syringe, not microcentrifuge tubes. Also, I'm not familiar with the
Arduino and have built my system around the Propeller microcontroller
- sorry. I have an incubator built around the syringes using non-
inductive resistors, but I haven't yet tested the incubator portion
yet. I have digital thermometers using DS18B20's. Although I've
heard of people using PCB traces for heating, it seems to me it might
be easier to use resistors because they can be positioned around the
microtubes without dramas.
Anyway that's it for now.
hope that helps get you started,
Mark
Simon Quellen Field
--
-- 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. For more options, visit this group at https://groups.google.com/d/forum/diybio?hl=en
Learn more at www.diybio.org
---
You received this message because you are subscribed to the Google Groups "DIYbio" group.
To unsubscribe from this group and stop receiving emails from it, send an email to diybio+unsubscribe@googlegroups.com.
To post to this group, send email to diy...@googlegroups.com.
Visit this group at https://groups.google.com/group/diybio.
To view this discussion on the web visit https://groups.google.com/d/msgid/diybio/e3bae866-b90a-4f5d-91ee-f722c644dd34%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.
Brian Degger
https://depts.washington.edu/soslab/turbidostat/pmwiki/pmwiki.php?n=Main.News
To view this discussion on the web visit https://groups.google.com/d/msgid/diybio/CAA0yOM5PTpeM8k7JgsBtRSFF2%3DKba5hb6iFGgV639iPedrPuCA%40mail.gmail.com.
--
Brian Degger