diy pcr machine temperature switch delay?
Jeswin
the DIYbio community? How does that compare with commercial lab
equipment? Were any design changes made to address this and how much
of an affect is this on the reaction?
Cathal Garvey
temperature?", the OpenPCR is somewhere in the range of 2C/s, the
lightbulb PCR is something like 0.5C upwards and 2-5C downwards, and
hot-air systems like cyclercan can be as high as 5C/s.
I have graphs somewhere documenting sample programs on cyclercan but I'd
have to dig them up.
Heat transfer from hot air to tubes may limit the actual efficacy of
such methods, though, whereas heat transfer is probably most efficient
with a thermal-block system such as OpenPCR.
--
www.indiebiotech.com
twitter.com/onetruecathal
joindiaspora.com/u/cathalgarvey
PGP Public Key: http://bit.ly/CathalGKey
John Griessen
> Heat transfer from hot air to tubes may limit the actual efficacy of
> such methods, though, whereas heat transfer is probably most efficient
> with a thermal-block system such as OpenPCR.
I've done a lot of heat transfer measurements, some designs of
heat sinks and controllers for them, and I just cannot see that theory
you are proposing. Having a metal heat sink can be a good way to extract heat
out of a conductive small space, but, if you have a vial made of
polyethylene, the layers of air, plastic and the sample will dominate
the heat flow.
The electrical analogs may help if you have thought about Ampere and Kirchoff
concepts of volts and amps. A massive heat sink is a reservoir as well as a conductor
of heat, which is working against rapid change -- its electrical analog is a capacitor.
The electrical analog of the air gap and plastic of the vial is resistance to heat change --
temperature is modeled as volts and heat flow as current. When the physical properties
are all arranged as above, either in heat and temperature or volts and amps, the differential
calculus has the same forms for the equations to solve.
An air heat exchange system can be just as fast as one with metal blocks.
By doing a big turbulent stirring -- lots of motion -- you can get rapid
heat flow between two small heat capacity materials, the sample in a vial
and the large quantities of air rushing by, and smoothly controlled by a heater
and/or a vent door to cooler air. Another advantage of moving air stirring
for heat flow in or out of a vial is that the stirring can be had by
moving the vials just as well as moving the air -- and when you move the vials
they shake and stir! So solid blocks sitting still because of their
power wiring, thin film heaters, and thermocouples could never change
a vial temperature at the center of the sample as fast as a shaker system could.
All that would change fast with a stationary block system is vial wall temperature.
The moral of the story is don't copy a 1992 PCR machine for diybio, or
you'll be outdistanced in bang per buck very soon.
I'll be making one of these machines,
but meanwhile it just bugs me to hear bad theory
going out on the list...
John
Jeswin
> An air heat exchange system can be just as fast as one with metal blocks.
> By doing a big turbulent stirring -- lots of motion -- you can get rapid
> heat flow between two small heat capacity materials, the sample in a vial
> and the large quantities of air rushing by, and smoothly controlled by a
> heater
> and/or a vent door to cooler air. Another advantage of moving air stirring
> for heat flow in or out of a vial is that the stirring can be had by
> moving the vials just as well as moving the air -- and when you move the
> vials
> they shake and stir! So solid blocks sitting still because of their
> power wiring, thin film heaters, and thermocouples could never change
> a vial temperature at the center of the sample as fast as a shaker system
> could.
I would have to agree with this. Hot air can be removed much faster
than cooling a metal block. Then in air heat exchange method, you only
have to worry about the conduction properties of the sample vial,
which we can't do much about. Am I understanding correctly?
Nathan McCorkle
I forgot to take a photo of the ad I recently saw in a magazine, but it was advertising 15C/sec ramp rates for super fast PCR. Ramp rate also effects specificity of primers... faster ramp means less time for hybridization at temps otjer than that which was programmed.
Sent from my mobile Android device, please excuse any typographical errors.
--
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.
John Griessen
> you only
> have to worry about the conduction properties of the sample vial,
> which we can't do much about. Am I understanding correctly?
Pretty much. Although, you can always do something -- it just helps to
identify what can be used effectively, so you concentrate
your change efforts on the effective instead of
the myriad dead end paths. When you're able better to see
dead end paths immediately, you will be able to imagine open
paths and stop asking a world wide list vague
questions like, "How much is the delay in the PCR
machines currently being built for the DIYbio community?"
How can I answer that? It's too vague. Are an being built? The jury's out.
The most prolific providers of hardware
are the most tight lipped. I'm only talking so much because I'm time/money limited
and not outflowing hardware fast enough. Otherwise, I'd send you SPAM.
Tasty yummy helpful SPAM though. Kinda like Simon's answers... Sorry Simon :-)
So, my ideas on hearing this are: Thin the sample vial wall. Change the
sample vial wall material to a conductive one, (except when you want to
do light transmission measurements, since most heat conductive materials
are less light transmissive -- I think...). Change the heat transfer by the physical
roughness of the vial wall -- surfaces that disrupt laminar flow
will exchange heat faster, so go for those. Sand paper treatment?
Etch? I don't know -- will take experimenting, but the concept to keep in mind
is increase contact and thus exchange of heat from HOT air molecules.
Increase speedy air contact with vial by upping its speed. Increase
heat transfer at the vial wall to culture media solution interface
by shaking things up. Identify heat conductive material that also
transmits UV to IR light. Schott glass?
On 12/17/2011 04:30 PM, Nathan McCorkle wrote:
> I forgot to take a photo of the ad I recently saw in a magazine,
but it was advertising 15C/sec ramp rates for super fast PCR.
> Ramp rate also effects specificity of primers... faster ramp
means less time for hybridization at temps other than that which was
> programmed.
Hey Nathan, that would be some really tasty SPAM if it had more details.
I can imagine a fast ramp with air heat exchange. I'll be working on showing it with
lab measured data backing it up. Lot's of to do's though -- isn't the holiday
season a chilling effect on progress?
John
keen101
good a conductor of heat as it is a heat sink. The homemade PCR
machine i am currently building and testing is a thermal air system
with a tube holder made from a PCB instead of aluminum. Currently the
problem i am working on with mine is that the mica heater in my
machine currently has a thermal cutoff switch that activates before i
can get the temperature up to 100c.
Here is a link to a nice presentation that i found to be quite helpful
where the topic is rapid cycle PCR.
http://www.arup.utah.edu/media/PCRtalk/PCRtalk.html
-Andrew
Cathal Garvey
the less you expose your DNA to high temperatures in the 90+ range, the
better the outcomes; depurination of some nucleotides can occur after
extended heating. So, the minimum time at 95C needed to denature DNA is
all that's called for. 20s is normal, but you could probably go lower
for "easy" reactions.
Andrew Barney
-Andrew
Nathan McCorkle
Philisa by Streck (streck.com)
15C/sec heating, 12C/sec cooling
--
Nathan McCorkle
Rochester Institute of Technology
College of Science, Biotechnology/Bioinformatics
Chris
We have been building a device we have dubbed
"PersonalPCR" (personalpcr.com for a quick picture) and we are
achieving 0.5C/s ramp up and 0.4C/s ramp down times. The tube holder
is a small aluminum cylinder and heating is achieved using a resistive
heating element (power resistor) and cooling is achieved via a small
cooling fan. Our design diverges from previous designs like the
OpenPCR that uses an aluminum block because we are focusing on using a
minimum amount of aluminum as possible. We are almost thinking of
each tube as a individual element. We are asking the question how
little aluminum do you need...? One answer is of course, none as
evidenced by the lightbulb PCR, but we feel there is value to keeping
some aluminum...
We have been designing the PersonalPCR with cost in mind.
Specifically, we are focusing on designing a device which leverages
circuit board manufacturing as much as possible. I am currently
prototyping the next revision that removes the power resistor and
places the traces directly on the circuit board. Initial tests
suggest that we are getting closer to 1C/s by dropping that huge
resistor from the equation. So we need some aluminum to conduct the
heat to the tube and the aluminum block can start to be thought of a
circuit component that can easily be 'populated' right on the circuit
board. You can begin to see how we can begin to leverage circuit
board manufacturing techniques to build up an entire device. I should
mention I am not interested in making a new type of tube (at least not
right away) since I want to be compatible with existing equipment and
existing pipelines that DIYBio people and scientists already have.
Someone on these groups has mentioned that PCR machines are like the
mousetrap of DIYBio, everyone thinks they can build a better one,
which is great. We are interested in small modular devices that can
be reliably manufactured, reliably work and sold new at a low price so
I think circuit board manufacturing is the way to go.
I have prototype circuit boards so drop me a line if you are
interested in getting your hands on one. BTW the current version is
all wrapped up as an arduino shield and its an open hardware project.
Brian Doom
| Sample capacity | 16 0.2 ml tubes |
| Average ramp rate | 1 C/s |
| Temp range | 10 – 100 C |
| Heated lid range | Ambient – 120 C |
| Temp accuracy | +/- 0.5 C |
| Temp uniformity | +/- 0.3 C |
| Input power | 100 – 120 VAC, 220 – 240 VAC, 50-60 Hz, 200 W max |
| Ports | 1 USB A |
| PC Control Software | Windows, Mac, Linux |
| Display | 20 x 4 backlit LCD |
| Program memory | Unlimited on PC, thermocycler stores last program when disconnected |
| Dimensions (W x D x H) | 13 x 20 x 25 cm |
| Weight | 3.5 kg |
| Software License | GPL |