Low cost PCR(under development)

[Subin]

Dear friends,

                     I have already published the arduino code for the "Low cost PCR" machine. Now i have made a small design how that machine look like. The wiring is not completed in that picture. The top there is a lid that can handle 25  samples. The arduino thermistor and fan is used to create the thermo cycler. The bulb and the 12v fan will be controlled by a relay circuit.

" I have added the arduino code, the picture of the machine as an attachment with this mail"

I want suggestions from your side to improve the design...

And i as thinking to have a hacker space in Bangalore or in my college. I think Bangalore is better because many interesting persons are their in Bangalore

[Enrico]

is the amount of heat equal for every tube? i mean, the central tubes may be heated more because of the short distance from the bulb, or is it totally indifferent?

[Subin e k]

Enrico,

                 It will be different if we are using a bulb.But it can be solved by using a heating bed(used n 3d printers).

[Michael Shamberger]

I have seen several DIY PCR designs and I was wondering why those are not able to hold standard PCR plates? 

For example, there are 24,48,96 plates available that are a few dollars each and certified sterile.  

[Cathal Garvey]

Because nobody who made one wanted high throughput. If you do, then why
not make a project of it, and adapt one to handle plates?

--
T: @onetruecathal, @IndieBBDNA
P: +353876363185
W: http://indiebiotech.com

[John Griessen]

On 06/29/2014 10:37 AM, Michael Shamberger wrote:
> I have seen several DIY PCR designs and I was wondering why those are not able to hold standard PCR plates?

Instead of DIY, there could easily be low cost for sale with such if enough line up to buy.
I'll make it.

John Griessen

[Marc Dusseiller]

hoi zäme,

urs has been actively reverse engineered and improved the openPCR:

http://hackteria.org/wiki/Wild_OpenPCR

a good thermal contact is crucial. just an eppi holder usually doesnt really do the trick, but it's fun to make these kinda Jugaad-PCR machines:

http://hackteria.org/wiki/File:Hairdryer_PCR_lab2011.jpg

and about bangalore... go and see yashas and the peeps at sristhi/NCBS out in yelahanka town. they are doing bio-stuff since 2008 and have an opportunity to create a larger space now.

all the best,

marc

[Michael Shamberger]

OpenPcr has a ramp rate of 1 degree C/s.  I checked specs for some commercial versions and they have ramp rates of 3 or 4 degrees C/s.   What are they doing differently?

[Cathal Garvey]

There are two ramp rates; ramp while increasing, and ramp while
decreasing. Most machines can achieve 3-4C increasing because they just
use heating coils, but their decreasing temperature (accomplished with a
peltier, cooling fluid, etc.) may be lower.

IIRC, OpenPCR has a climbing ramp rate of 2-3C, and a reducing ramp rate
of 1C. Larger machines with fundamentally the same architecture (coil +
peltier) achieve higher rates merely by adding extra wattage.

Remember though, an additional consideration is how homogenous the
temperature change is through the block, and at what rate the
temperature change in the block is reflected in the tubes. There would
be little point doubling the ramp rate of the block if the tubes cooled
at an only marginally faster rate. Worse, doubling the ramp rate as
measured by the thermometer may look great on the charts and benchmarks,
but the block may be unevenly cooling and this can have real
consequences when you're expecting your samples to experience identical
conditions.

It's rare that you'll need decreasing ramp rates faster than 1C, in any
case. Have you a special experiment in mind?

[John Griessen]

On 06/30/2014 06:05 AM, Cathal Garvey wrote:
> t's rare that you'll need decreasing ramp rates faster than 1C, in any
> case. Have you a special experiment in mind?

Are there some good research reasons to go for more delta T rate?
My stirred air cooled PCR cycler concept is likely to do more
when I get around to doing it and testing it. And for evenness,
it will be top performing inherently.

[Dakota Hamill]

Well the Idaho Tech halogen light / vortex air thermal cycler which was the first one i got off ebay has one major advantage over any peltier device I've used - it saves a boatload of time.  The ramp rates are much faster, especially the decreasing, as a vent is just flips open and hot air pours out.   It really cuts the PCR program almost in half on time, though I should actually measure it, but it generally finishes a good 30-45 minutes before a peltier device doing a 30x program, which normally takes ~1.5 hours.

It can't hold 96 samples, and requires oil due to no heated lid, but it's worked before and the time it saves more than makes up for those other disadvantages for the uses at the moment.  Obviously being able to run a lot of samples in parallel with a temperature gradient for annealing temp without oil is where the peltier machine is better but, it's also a new machine.  There are plenty of old peltier based devices without heated lids with crappy ramp rates, and for that I prefer the lightbulb machine.

--
-- 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 http://groups.google.com/group/diybio.

To view this discussion on the web visit https://groups.google.com/d/msgid/diybio/53B186B8.40606%40industromatic.com.

For more options, visit https://groups.google.com/d/optout.

[Cory Tobin]

> What are they doing differently?

In addition to what others have said, the tube block has a lower mass
than the OpenPCR block. The OpenPCR block is a solid block with holes
drilled in it. If you look at newer machines from BioRad, Thermo,
etc, they have designed the block to have a much smaller mass while
having the same surface area in contact with the tubes. See this
picture: http://www.bio-rad.com/webroot/web/images/lsr/products/amplification_pcr/product_overlay_content/global/lsr_cheese_block_expand.jpg
But as far as I know, shorter ramp rates just means shorter run time.

-cory

[Nathan McCorkle]

On Mon, Jun 30, 2014 at 4:05 AM, Cathal Garvey
<cathal...@cathalgarvey.me> wrote:
> Remember though, an additional consideration is how homogenous the
> temperature change is through the block, and at what rate the
> temperature change in the block is reflected in the tubes. There would
> be little point doubling the ramp rate of the block if the tubes cooled
> at an only marginally faster rate. Worse, doubling the ramp rate as
> measured by the thermometer may look great on the charts and benchmarks,
> but the block may be unevenly cooling and this can have real
> consequences when you're expecting your samples to experience identical
> conditions.

I'd say that uneven heating or cooling will produce a thermocycler
that will be quite hard to use for normalizing experiments... you'd
never be able to run things in parallel.

The fastest ramp rates I've seen is 15C/sec heating, 12C/sec
cooling... see the pic of an advertisement here:
https://groups.google.com/d/msg/diybio/yttYHgz0gFE/o4Icu_3--lwJ

[Nathan McCorkle]

On Mon, Jun 30, 2014 at 8:48 AM, John Griessen <jo...@industromatic.com> wrote:
> On 06/30/2014 06:05 AM, Cathal Garvey wrote:
>>
>> t's rare that you'll need decreasing ramp rates faster than 1C, in any
>> case. Have you a special experiment in mind?

I'd say that you always want faster, unless you want your PCR to be
sloppier. Sloppiness depends on a lot of factors, ramp rate is one,
but a properly designed primer is another.

>
> Are there some good research reasons to go for more delta T rate?

Yes, it gives the molecules less time to be random and jiggle around.
You transition through all the other possible molecular states faster,
all the possible base-pairing of a primer to a target sequence
(avoiding sequences that don't match the best)... the math is probably
something about the speed of the temperature delta being closer/faster
than than the speed of diffusion... thought there are the electronics
of basepairing too, but that'd be where I'd generally start to think
about a model of this.

> My stirred air cooled PCR cycler concept is likely to do more
> when I get around to doing it and testing it. And for evenness,
> it will be top performing inherently.

Yep, instantaneous with nice power metering, and you can probably
start to do some neat target detection stuff with something like
differential scanning calorimetry (maybe pair this with qPCR?)
http://en.wikipedia.org/wiki/Differential_scanning_calorimetry

[Nathan McCorkle]

On Mon, Jun 30, 2014 at 4:35 PM, Nathan McCorkle <nmz...@gmail.com> wrote:
> You transition through all the other possible molecular states faster,
> all the possible base-pairing of a primer to a target sequence
> (avoiding sequences that don't match the best)...

(for the specified temperature end-point)

[John Griessen]

On 06/30/2014 11:04 AM, Dakota Hamill wrote:
> It can't hold 96 samples, and requires oil due to no heated lid,

Heated lids are not essential. Absence of unwanted heat flow is.
My concept will do fast heat flow and even heat distribution without
a separate hater in the lid.

The "lid" won't look like a lid, and will not be near the sample vials.
The swirling air will be near the sample vials.

On 06/30/2014 01:26 PM, Cory Tobin wrote:> the tube block has a lower mass
> than the OpenPCR block.

Having a non-block with zero mass is even better.

On 06/30/2014 06:35 PM, Nathan McCorkle wrote:> do some neat target detection stuff with something like
> differential scanning calorimetry

OK... melting and glass transition phase changes...

Are you meaning measuring temperatures with fine precision inside or next to the
sample vial and comparing heat flow from the heater source by Watts delivered?

The usual plastic vial might get in the way -- maybe it should be silicon glass,
and the sample size would have to have more mass than usual, or the machine would have to have
super insulation. Sounds non-low-cost at first glance.

[Nathan McCorkle]

On Mon, Jun 30, 2014 at 5:24 PM, John Griessen <jo...@industromatic.com> wrote:
> Are you meaning measuring temperatures with fine precision inside or next to
> the
> sample vial and comparing heat flow from the heater source by Watts
> delivered?

Not too sure... I feel like the sensor would either be inside the
liquid or maybe non-contact IR? I believe the idea is you command
constant power to the heater (or maybe a constant increase in power),
then watch the temperature, and look for when the temperature curve
isn't smooth. Then do the same thing in reverse cooling the sample.

>
> The usual plastic vial might get in the way -- maybe it should be silicon
> glass,
> and the sample size would have to have more mass than usual, or the machine
> would have to have
> super insulation. Sounds non-low-cost at first glance.

Not too sure.

[Cory Tobin]

> Having a non-block with zero mass is even better.

Sure, except that air doesn't transfer heat to your sample nearly as
fast as aluminum. Imagine putting your hand in a 100C oven vs putting
your hand on a 100C frying pan. That's why most thermalcyclers take
into account the volume of the sample and calculate the temperature
instead of just using the temperature of the block.

I'm not saying air won't heat the tubes sufficiently fast. If you are
optimizing for cost then air is the way to go. But heating and
cooling the air rapidly won't necessarily lead to the temperature of
the sample being heated/cooled faster compared to aluminum.

-cory

[Simon Quellen Field]

Maybe I'm missing something, as I haven't been following this thread from the beginning, but if your heater has to heat and cool the aluminum as well as the sample, the problem you are trying to solve is exacerbated, not alleviated. If you are moving the sample from a hot aluminum block to a cold one, then you are fine. Otherwise, you'd want low thermal mass, and air is pretty good. So would be a design that moved hot and cold water into the space around the sample. The water is a good heat conductor, and it is easy to keep large reservoirs of water at a constant temperature, and then move small amounts of water at the proper temperature into a chamber full of samples. Two valves, and gravity can move the water from the reservoirs through the device and into the sink.

Pretty cheap.

-----

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+un...@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+un...@googlegroups.com.

To post to this group, send email to diy...@googlegroups.com.
Visit this group at http://groups.google.com/group/diybio.

To view this discussion on the web visit https://groups.google.com/d/msgid/diybio/CAC46Jss2toCNHgMK-eKtRJf16ZWsH7xAtjVj0Y9eGEWD%2BWw5vQ%40mail.gmail.com.

[Cory Tobin]

Actually, I take back what I said about air not transferring heat to
the sample very quickly. I remember a device a long time ago, I think
it was called the "Indee" or something like that, that used really hot
air and circulated it around the tubes really really fast. I think
the ramp rate was around 8C/s. They basically made up for the poor
thermal conductivity by heating the air up to like 200C and then
bringing it back down as the calculated sample temperature reached the
target temperature. Which made it faster than the peltier devices.
So maybe air wins out in both cost and speed if designed right.

-cory

[Nathan McCorkle]

Yep, think about a hair drier. I also have a hot air PCR
thermocycler... it uses a light bulb for a heater... but didn't seem
to have a great ramp rate.

For modifying a hair drier heater coil to be arduino-controllabe,
these two instructables should be useful:
http://www.instructables.com/id/Extreme-Surface-Mount-Soldering/step1/Order-Parts/
http://www.instructables.com/id/Closing-the-Loop-on-Surface-Mount-Soldering/

they don't include snubbers, so couldn't be used to control the fan
motors, but it will work for the AC heater component which is going to
be higher amperage than the fan(s) anyway.

> --
> -- 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 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+un...@googlegroups.com.

> To post to this group, send email to diy...@googlegroups.com.
> Visit this group at http://groups.google.com/group/diybio.

> To view this discussion on the web visit https://groups.google.com/d/msgid/diybio/CAC46Jsuo%3DsE1dKG8Z263_RaJS5d%3DkWva%2BucJXbwDQOcawxQ3ug%40mail.gmail.com.

> For more options, visit https://groups.google.com/d/optout.

--
-Nathan

[Josh Perfetto]

We have a prototype Open qPCR machine at the moment which is capable of
ramping ramping 5-10 C/s at various parts of the temperature range. Final
controlled ramps in the PCR region are likely to be more like 5-7 C/s. The
main difference over the OpenPCR was much higher power density in the
thermoelectric modules, which necessitates larger power supplies and higher
current power switching components. Another difference was reduction in
thermal mass of the heatblock by removing more excess material. A third
change was a faster heat exchange rate with the air when cooling (i.e.
larger heatsink and faster airflow). As Cathal pointed out this requires
more engineering in heatblock consistency and thermal control.

All of this increases cost relative to the conservatively engineered
OpenPCR. The increased cost may not be worth it for everyone, but I think
there is certainly benefit to PCR cycles that take 30 minutes instead of 2
hours. Anyways hope this answers your question about how faster ramp rates
are possible in heat block based thermocyclers. There are some more exotic
designs that achieve even faster ramp rates with high speed airjets.

-Josh

On Mon, Jun 30, 2014 at 7:30 PM, Josh Perfetto <jo...@chaibio.com> wrote:

We have a prototype Open qPCR machine at the moment which is capable of ramping ramping 5-10 C/s at various parts of the temperature range. Final controlled ramps in the PCR region are likely to be more like 5-7 C/s. The main difference over the OpenPCR was much higher power density in the thermoelectric modules, which necessitates larger power supplies and higher current power switching components. Another difference was reduction in thermal mass of the heatblock by removing more excess material. A third change was a faster heat exchange rate with the air when cooling (i.e. larger heatsink and faster airflow). As Cathal pointed out this requires more engineering in heatblock consistency and thermal control.

All of this increases cost relative to the conservatively engineered OpenPCR. The increased cost may not be worth it for everyone, but I think there is certainly benefit to PCR cycles that take 30 minutes instead of 2 hours. Anyways hope this answers your question about how faster ramp rates are possible in heat block based thermocyclers. There are some more exotic designs that achieve even faster ramp rates with high speed airjets.

-Josh

On Mon, Jun 30, 2014 at 4:05 AM, Cathal Garvey <cathal...@cathalgarvey.me> wrote:

[Michael Shamberger]

>> Have you a special experiment in mind?

No, just interested in the iterations that open tools are under going.  I am wondering if the scientific tools market could go the same way that software went.  The first versions of open source software were inferior to the commercial vendors but subsequent iterations eventually outpaced the commercial versions.  After that there was mass acceptance.

Makers each build a specific instrument but there is no standard across these.  Metric vs imperial, part reuse, and software could all be standardized across a set of open tool designs that could be iterated on and eventually overtake the current close commercial offerings.  

[Cathal Garvey]

> Yep, think about a hair drier. I also have a hot air PCR
> thermocycler... it uses a light bulb for a heater... but didn't seem
> to have a great ramp rate.

Yea, this; I was going to compare a heat gun to an oven. Air has very
low heat capacity, so it delivers heat in tiny doses, then tends to form
a "cusion" of cooled air around the object. So, ovens are less burn-ey
than a heat-gun for the same temperature; the latter blasts away the
cooled air-shell and replaces it with high-temperature air constantly.

I knocked together a hot/cold air thermal-cycler using a coffee can,
computer case fan, and an art-grade heat gun once. I used an LM35
temperature sensor and suspended samples within the can. At least as
read by the LM35, which was within a sample vial for "realism", ramp
rates of 3-5C either way were achievable by hot/cold air.

Bonus; when dealing with total-immersion fluids like air, you shouldn't
need to worry much about oil or heated lids, because the whole tube will
be heated/cooled simultaneously. If you found that you did have
problems, putting dollops of insulating material on top would probably
suffice to stop the lids cooling fast enough to deposit vapour; silicone
tube beanies! :)

[David Ng]

If you are only planning on small fragments, what about these approaches:

http://www.biotechniques.com/BiotechniquesJournal/2011/January/Rapid-DNA-amplification-in-a-capillary-tube-by-natural-convection-with-a-single-isothermal-heater/biotechniques-307538.html?pageNum=1

http://www.sciencemag.org/content/298/5594/793.full

http://www.jove.com/video/2366/rapid-pcr-thermocycling-using-microscale-thermal-convection

-Dave

[John Griessen]

On 06/30/2014 07:45 PM, Cory Tobin wrote:
> Imagine putting your hand in a 100C oven vs putting
> your hand on a 100C frying pan.

No, the comparison analogy to make is riding a motorcycle with
no jacket in the desert. You overheat rapidly.
Forced air, not still air -- vigorous flow for high delta T.


On 06/30/2014 08:01 PM, Simon Quellen Field wrote:> if your heater has to heat and cool the aluminum as well as the sample, the

problem you are trying to solve is exacerbated, not
> alleviated

Yes. Since we are dealing with often very small sample sizes, I am planning a machine
for those. It may need better than usual vials to conduct well and work at high delta T
rates.

[Peter Thielen]

Not sure if others have come across the "open source" PersonalPCR design, but it uses the same ideas you are all discussing here.  Some major drawbacks include poor documentation and improper release of design files, such as the .pdf used for the all surface mount component control board.  I would be very interested in building these if others are able to provide an eagle file for the board, as I have access to a fully outfitted Maker lab that would allow me to produce these for ~$100/ea in components.  Honestly, if anyone comes up with a usable design along similar lines as this, I'd be interested in helping out with production.

Note that they also appear to be starting a company with the intent of taking the design closed source again: miniPCR.  

[Michael Shamberger]

>> prototype Open qPCR machine 

When is this going to be ready?  Are you going to post the plans?

[John Griessen]

On 07/03/2014 08:00 PM, Peter Thielen wrote:
> Not sure if others have come across the "open source" PersonalPCR <http://www.personalpcr.com/pcr/home> design, but it uses the

> same ideas you are all discussing here. Some major drawbacks include poor documentation and improper release of design files,
> such as the .pdf used for the all surface mount component control board. I would be very interested in building these if others
> are able to provide an eagle file for the board, as I have access to a fully outfitted Maker lab that would allow me to produce
> these for ~$100/ea in components. Honestly, if anyone comes up with a usable design along similar lines as this, I'd be
> interested in helping out with production.

This design uses heavy metal blocks. I am planning a vigorously stirred air PCR machine
that will sell for under $125 and have a parts cost of maybe $35 in 100's volume. I can't see using Eagle
when KiCAD and gEDA are available and free open. If interested, my timeline is 2nd half of 2015 for launching.

[Nathan McCorkle]

Yes, Eagle seems ridiculous when a project is branded open-source.

--
-- 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 http://groups.google.com/group/diybio.

To view this discussion on the web visit https://groups.google.com/d/msgid/diybio/53BABFDC.7050405%40industromatic.com.

[Dakota Hamill]

I talked to the guys at miniPCR and they seem very nice.  I had followed pesonal PCR for a while but the website never seemed to change and the product was never for sale.  I guess a few guys from Templeman engineering who were working on it spun off miniPCR.  It's a sleek and small looking design and very hand-held looking.  PersonalPCR was going to be $600 and miniPCR's are $800 according to both websites.   Most of the design files were released I think except for perhaps the most important ones, those of the actual circuit.  I can't really blame someone for putting in a hell of a lot of work on a project and wanting to make some money off of it though or maintain some ownership.

So for $800 for 8 wells with miniPCR, vs 16 wells for $600 from OpenPCR, vs 96 wells from a used machine you can get from somewhere for less than $600 (though it wouldn't be as portable and small) it's a toss up as to what you want and who you want to support I suppose.  I've seen a really nice machine in a lab which was a touch screen interface brand new 96 well machine with all sorts of bells and whistles for only $2,500 I think.  

If someone could come in with a price point of $500 or less I think it'd be a hit, because until then I think Ebay or dumpster diving is still a better bet.

--
-- 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 http://groups.google.com/group/diybio.

To view this discussion on the web visit https://groups.google.com/d/msgid/diybio/53BABFDC.7050405%40industromatic.com.

[John Griessen]

On 07/07/2014 11:39 AM, Dakota Hamill wrote:
> So for $800 for 8 wells with miniPCR, vs 16 wells for $600 from OpenPCR, vs 96 wells from a used machine you can get from
> somewhere for less than $600 (though it wouldn't be as portable and small) it's a toss up

What will make any of them better is some handling automation to load and unload from a row of 96 vial holders.
Another feature most of those lack is an app that lets you easily keep track
of what's been done, print labels for vials, etc. I'd like to develop an app for mine that
shows a spreadsheet-like grid of data that maps to what is lined up in wells, and also
detects every time a well is unloaded by a human hand and notes that on the data and could ask
aloud if the human attached to the hand wants a label printed for the vial they are taking.

[Dakota Hamill]

Any extra perks would be cool, especially label printing, I hate writing on PCR tubes with a passion as they are so small with awkward angles and limited surface area. The fundamental drawback I see to OpenPCR and miniPCR and whatever other "DIY" if you want to call it that thermal cyclers is # of samples.  Time is very valuable and 8 samples just isn't enough in general I find, 16 is workable, but if for example, most high school or college classes in the US are 20+ students, and we assume each wants to run a reaction, that means at least two machines would be needed per class, at a cost of $1,200 or $1,600 respectively (for OpenPCR and miniPCR based on prices on website).

At that point, you're better off (in my opinion) spending another $1,000 and getting a 96 well plate gradient PCR machine with all the bells and whistles from a larger manufacturer, or picking a refurbished one off ebay or even cheaper.

I definitely want to support small projects/companies like that, and I'm fine at times paying a bit more to help a startup type company like that out, but some of the specs need to be at least somewhat comparable to what can be had for cheaper (ebay) or for a little more money from another manufacturer.  I honestly don't care that much about the interface as long as I can program in a cycle, or the ramp time as long as it is within reason, but if I want to mass produce a linear plasmid backbone I'd rather be able to run a lot of PCR samples all at once, not 8 or 16 every 2 hours.  

I've purchased gel boxes and other things from places like IORodeo because they offer products that are cheaper than or equal to (generally cheaper than) other established companies and still match, come close to, or exceed the functionality of the competitor's product.  An 8 well or 16 well thermal cycler for $600/$800 in my eyes does not exceed (I guess it still matches) the functionality of what can be had from competitors or on ebay.  

Not that I've come up with anything better though, so I can't knock them, and I'm not trying to anyway.  I like the people that are behind the projects and hope they continue to create badass stuff like a qPCR machine!  

[Peter Thielen]

They haven't responded to any emails, which I guess is understandable if they are planning on taking it closed source.  Unfortunate, though, as one more closed source idea is one more piece taken from the community as a whole.  They do have the board in the documentation now, it's just not in an easily usable form -- I'm hoping to get something very similar up and running in the near future, and I'll post here if I'm successful.  I've added up all the build costs, and I think it's already south of $100/unit in parts.  

Regarding the use of "vigorously stirred air" and other methods, I think it's likely that you'll run into a lot of issues with transferring protocols from typical blocks if your ramp speed is too fast.  There are a number of other companies that have produced these types of machines over the years, and they never really caught on for whatever reason.  No reason to think it wouldn't work, but there's something to be said for having a certain amount of mass for temperature stability.  The description from the personalPCR site is spot on though if you've looked at new thermocyclers lately -- the blocks themselves seem to be about the only space for innovation while keeping a traditional 96 well format.  Thermocycling for PCR isn't rocket science, yet overthinking things is pretty easy.

On Monday, July 7, 2014 12:39:47 PM UTC-4, Dakota wrote:

I talked to the guys at miniPCR and they seem very nice.  I had followed pesonal PCR for a while but the website never seemed to change and the product was never for sale.  I guess a few guys from Templeman engineering who were working on it spun off miniPCR.  It's a sleek and small looking design and very hand-held looking.  PersonalPCR was going to be $600 and miniPCR's are $800 according to both websites.   Most of the design files were released I think except for perhaps the most important ones, those of the actual circuit.  I can't really blame someone for putting in a hell of a lot of work on a project and wanting to make some money off of it though or maintain some ownership.

So for $800 for 8 wells with miniPCR, vs 16 wells for $600 from OpenPCR, vs 96 wells from a used machine you can get from somewhere for less than $600 (though it wouldn't be as portable and small) it's a toss up as to what you want and who you want to support I suppose.  I've seen a really nice machine in a lab which was a touch screen interface brand new 96 well machine with all sorts of bells and whistles for only $2,500 I think.  

If someone could come in with a price point of $500 or less I think it'd be a hit, because until then I think Ebay or dumpster diving is still a better bet.

On Mon, Jul 7, 2014 at 11:42 AM, John Griessen <jo...@industromatic.com> wrote:

On 07/03/2014 08:00 PM, Peter Thielen wrote:

Not sure if others have come across the "open source" PersonalPCR <http://www.personalpcr.com/pcr/home> design, but it uses the

same ideas you are all discussing here.  Some major drawbacks include poor documentation and improper release of design files,
such as the .pdf used for the all surface mount component control board.  I would be very interested in building these if others
are able to provide an eagle file for the board, as I have access to a fully outfitted Maker lab that would allow me to produce
these for ~$100/ea in components.  Honestly, if anyone comes up with a usable design along similar lines as this, I'd be
interested in helping out with production.

This design uses heavy metal blocks.  I am planning a vigorously stirred air PCR machine
that will sell for under $125 and have a parts cost of maybe $35 in 100's volume.  I can't see using Eagle
when KiCAD and gEDA are available and free open.  If interested, my timeline is 2nd half of 2015 for launching.

--
-- 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 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+un...@googlegroups.com.

To post to this group, send email to diy...@googlegroups.com.
Visit this group at http://groups.google.com/group/diybio.

[Dakota Hamill]

They haven't responded to any emails, which I guess is understandable if they are planning on taking it closed source.  Unfortunate, though, as one more closed source idea is one more piece taken from the community as a whole.  They do have the board in the documentation now, it's just not in an easily usable form -- I'm hoping to get something very similar up and running in the near future, and I'll post here if I'm successful.  I've added up all the build costs, and I think it's already south of $100/unit in parts. 

 

I guess that's one of the reasons to keep something closed-source?  If you put a lot of blood sweat and tears into designing something and then release the plans on how to build everything, someone else can come in out compete you on production costs, marketing, or price, and you effectively put yourself out of business by trying to "do good".   Unless they feel like giving you something in return, you've effectively done a lot of work for nothing more than a pat on the back.  That doesn't pay the bills, but you might earn karma.  

The question of how "open or closed" I want to be on certain things is one I think of often, and I imagine many other people do to.  I believe in the inherent goodness of people and have a strong desire to want to help people, but I see how quickly people can turn evil or take advantage of kindness.  Reading Childhood's end by Arthur C. Clarke was actually pretty interesting because it gave a glimpse of what society could look like in some "ideal" form.  Maybe it's not always the best idea to give everything away for free.  I don't know, it's a tough question.  Everyone wants credit for their work, and to be able to make a living off it to.  If your work isn't protected, it seems quite easy for you to easily be forgotten about and for anyone who uses it to owe you absolutely nothing.

I think over time it's easy to become jaded, and to give up on outright changing the way the game is played.  In the end, what do we even want the rules to be?  

It seems to me it'd be a lot easier to make a lot of money first, then try to change the rules, or at least you'd be more comfortable giving your time, designs, and ideas away because you're already financially independent.  

[John Griessen]

On 07/07/2014 07:11 PM, Dakota Hamill wrote:
> I'd rather be able to run a lot of PCR samples all at once, not 8 or 16 every 2 hours.

How would 30 in 30 minutes be? That might fit in a footprint of 20cm diameter on the bench. Or maybe a 20 cm square.

On 07/07/2014 08:49 PM, Peter Thielen wrote:> Regarding the use of "vigorously stirred air" and other methods, I think it's likely

that you'll run into a lot of issues with
> transferring protocols from typical blocks if your ramp speed is too fast.

It will be programmable and tightly controlled.

> there's something to be said for having a certain amount of mass for temperature stability.

Active control will give you tighter temperature than a lagging heat sink mass will.

> the blocks themselves seem to be about
> the only space for innovation while keeping a traditional 96 well format.

Why traditional meatl blocks and why use the 96 well format inside a machine?
Why not robot handle the vials and load them into a 96 well holder when done
so as to be ready for human handling between automated stations?

They could even be sorted into different rows based on identity data entered as they
are started. The map to what's what would be on the screen.

On 07/07/2014 09:29 PM, Dakota Hamill wrote:> you effectively put yourself out of business by trying to "do good".

That's a belief.

Here's another belief scenario:
I've seen people in business do fine with no IP protection, just putting things out there
at a good price, by doing good design, and matching to what is wanted. I've done one little test product
that way and it paid back development and made some money before seeming to reach all the available buyers in the word.

Good design is often as not done by one person rather than a group, and so does not
have to cost big money like many investor backed
companies spend. When spending on a product is high, the sale price has to be high, and the volume is lower
(of the people you convince to buy), so spending to close off copying is effective -- for the low volume you
sell at least. When you keep costs down from the start, there's not as much need for defense
against copying -- no one can undercut your price easily, so they don't rush to copy. The product life cycle
is still going to obsolete a thing in a few years because of improvements, so it's not such a big deal
to try to kill of copying.

[Nathan McCorkle]

On Mon, Jul 7, 2014 at 8:37 PM, John Griessen <jo...@industromatic.com> wrote:

>> the blocks themselves seem to be about
>> the only space for innovation while keeping a traditional 96 well format.
>
> Why traditional meatl blocks and why use the 96 well format inside a
> machine?
> Why not robot handle the vials and load them into a 96 well holder when done
> so as to be ready for human handling between automated stations?

For one thing the 96-well plate is going to be much easier for a
non-robot to use for certain experiments/procedures. They also work
for microscopy, since they have two parallel planes, I've not seen any
tubes that would work similarly. A limitation I can think of
immediately is that plates are usually made from harder more brittle
plastic (polycarbonate maybe) which is very very clear. If you were to
use this type of material for a tube, it might crack when
popping/snapping the top off/on. I don't know of any softer plastics
used for tubes that are also clear... not saying they don't exist...
but it's probably easier to adapt to plates than to design/produce a
new tube.

And actually, robots do use plates, I bet they'd prefer them too...
much easier for a robot to open them I think, as the covers are
loose-fitting like petri dishes.

[John Griessen]

On 07/07/2014 11:11 PM, Nathan McCorkle wrote:
> the 96-well plate is going to be much easier for a
> non-robot to use for certain experiments/procedures. They also work
> for microscopy, since they have two parallel planes, I've not seen any
> tubes that would work similarly. A limitation I can think of
> immediately is that plates are usually made from harder more brittle
> plastic (polycarbonate maybe) which is very very clear.

OK. There are plenty of reasons to like plates with wells, but the ones
aimed at PCR don't seem to be clear, or flat bottomed for microscopy.

http://eshop.eppendorfna.com/products/Eppendorf_twintec_96_PCR_plates
only cost $82 each!

Biorad is much better price:
http://www.bio-rad.com/en-us/sku/hsp-9661-hard-shell-low-profile-thin-wall-96-well-skirted-pcr-plates
$4.30 in qty 50's.

There *are* plates with purposely thin walls that would be good for heat
transfer, and good for air PCR if used with sealing tape or cap strips
over the tops.

I think individual tubes are what I will search for, with the idea of
a handler moving them to a 96 hole rack as they come out of the PCR cycler.

[Nathan McCorkle]

Ahh, I guess I've actually only worked with qpcr plates and cell culture plates... I believe they're all a standard format which makes it easy for robots.

https://www.google.com/search?q=qpcr+plate+96

Those are good for heat, and rather than a loose lid they get a sticky clear transparent film applied on top.

--
-- 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 http://groups.google.com/group/diybio.

To view this discussion on the web visit https://groups.google.com/d/msgid/diybio/53BB9FB0.10003%40industromatic.com.

[Peter Thielen]

The eppendorf item comes in packs of 20, so $4.10 is actually cheaper than the biorad ones.  They are also designed with automation in mind, and as someone who does a whole lot of sample processing I definitely prefer them over individual tubes.  Thin wall tubes are very fragile and tend to break -- I vastly prefer skirted or semi-skirted plates over individual tubes or tube strips.  If I have to go smaller than 96 I typically use these.

Also, Dakota, I totally agree with your thoughts on open vs. closed.  At the same time, I also think your comments on the sub-$500 mark for a thermocycler is spot on.  I've started to think about the market for consumer level biotech products vs. research grade.  I have hundreds of thousands of dollars of equipment at my disposal at work, but if I wanted to do something in my spare time at home I would go for the bare minimum.  No need for fancy robotics to label tubes or move them into new racks -- that's a discussion for much higher throughput work and not necessarily the DIYbio crowd.  

[Dakota Hamill]

Peter, best of luck and I hope you do make something happen, sounds like you have a nice facility or access to one!  The label maker was just a cool wish list item, by no means would it be a deal breaker for me or anyone else I think.  I'm interested to see all the new projects people are working on.

John, yes I agree that a good product, a good design, and a good price will keep a product and company alive and well even with no protection.  Good engineering or good service speaks for itself and draws customer loyalty and customer $$. 

--
-- 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 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+un...@googlegroups.com.

To post to this group, send email to diy...@googlegroups.com.
Visit this group at http://groups.google.com/group/diybio.

To view this discussion on the web visit https://groups.google.com/d/msgid/diybio/b0e2dab4-e381-46d9-aaaa-128151f0ffe5%40googlegroups.com.

[John Griessen]

On 07/08/2014 02:50 AM, Nathan McCorkle wrote:
> Those are good for heat, and rather than a loose lid they get a sticky clear transparent film applied on top.

Hmm... That supplier has $1 each prices for plates and sticky seals -- could be good.

What if my carousel air PCR concept could hold 2 96 well plates for agitation,
air PCR, and handling? With the range of special 96 well plates available, there could be lots of buyers...

The sticky seals don't seem automatable -- some kind of caps might be better for that.

The machine would need a 133 mm radius of two plates twirling, plus some room
for containing everything...30.6 cm diameter lab bench footprint.

On 07/08/2014 06:39 AM, Peter Thielen wrote:> that's a discussion for much higher throughput work and not necessarily the DIYbio
crowd.

Oh, the DIY crowd doesn't seem to buy anything at all, just kludges something. I'm thinking
of the low costs needed as the mechanics of bioengineering gets more popular and out of the
pure research zone, not DIY. But, the DIYers will like my prices too, all 100 of them on the planet...

[John Griessen]

On 07/08/2014 09:31 AM, John Griessen wrote:
> What if my carousel air PCR concept could hold 2 96 well plates?

Or, just make a spindle to hold plates and spin one, and depend on the
inside shape of the machine to create turbulence and so even temperatures
and heat flow for the central wells that would be moving at lower velocities.

Then the machine footprint would be 77mm radius plus 20 -->
194mm diameter lab bench footprint for
holding a 128mm x86mm plate.