Cheaper OpenPCR

808 views
Skip to first unread message

Jeswin

unread,
Oct 1, 2012, 11:41:48 AM10/1/12
to diy...@googlegroups.com
OpenPCR is a cool thing. The only thing is, it's still quiet expensive
for a kit. Looking at the BOM Josh posted:
https://docs.google.com/spreadsheet/ccc?key=0AiaWH3PL_9CndG1GbmtSZ1I3aGdkQkE4YVhwZ3o5c1E#gid=0

I see that there are lots of parts. Can we build a simpler version,
just like we have simpler Arduinos available? I assume the most
expensive parts are the heat-block, PSU, microcontroller.

Any ideas for simpler lid unit, cheaper heat-block, PSU, and MCU?

Cathal Garvey

unread,
Oct 1, 2012, 5:00:37 PM10/1/12
to diy...@googlegroups.com
I suspect a cheaper heatsink and PSU are low-hanging fruit, here.
I would stick with a standard Arduino; as someone pointed out earlier,
in the �600 pricetag, �20 for an extremely well-supported and
easy-to-use chip isn't a big priority.

Removing the display on the front and relying entirely on faux serial
output would save some money, too. It's a nice feature, and I like being
able to plug out my laptop and have the program run regardless, but it's
still non-essential.

Heat block is probably not disposable. To get the right bores for PCR
tubes you practically have to get special drillbits machined. I recall
Josh or Tito wrote a blogpost on the OpenPCR.org blog about how you
don't screw around with your machinists, because heatblocks etc. are the
kind of component that you can't just knock together at home if you want
decent results.

That's especially true of the OpenPCR heatblock, which is (I think)
chrome-plated copper; that's pretty much as good as it reasonably gets.

Unless you want to use your own non-standard but
more-convenient-to-design-for tubes, I think that'll remain true;
heatblocks will have to be machined, and that'll cost moneys. Unless you
go with heated fluid/gas chambers, that is. I think that's still the aim
for the Amplino team?
--
www.indiebiotech.com
twitter.com/onetruecathal
joindiaspora.com/u/cathalgarvey
PGP Public Key: http://bit.ly/CathalGKey

Conner Berthold

unread,
Oct 1, 2012, 9:33:44 PM10/1/12
to diy...@googlegroups.com

This link is a good diy approach as well and could be built for less than $85 depending on what you have lying around.

Simon Quellen Field

unread,
Oct 1, 2012, 9:38:12 PM10/1/12
to diy...@googlegroups.com
Suppose you took your PCR tubes (that you want to fit into a homebrew
heatblock) and make a mold of them in plaster or one of the rubber mold
making products. Then make a positive of it in plaster.

Now melt a bunch of tin in a cup, and insert your plaster images of the
PCR tubes. Let the tin cool, and then etch out the plaster with some hydrochloric
acid.

No machining, and you now have a big tin heat block. Make as many holes as
you like for tubes, and make as many shapes as there are types of tubes you
might want to use.

At $23 per pound, tin is cheap. It melts at 232 Celsius, so it is easy to melt on the
kitchen stove, but won't melt in the PCR device.

You can insert fins made from tin plate from a 'tin' can while the metal is liquid
if you want to make cooling fins.

Copper has a thermal conductivity of 400 w/(m.K) compared to tin's 67, so you
might prefer zinc at 116. But Silver is even better at 429, and is also something
you can melt in your kitchen easily. It's $20 a pound, but you won't need more
than about $30 worth.

Of course, as someone who used to do bronze casting as a hobby, I know it is
easy to melt copper and pour it into a mold as well, should you want to go that
way. Skip the chrome plating -- just paint it with lead-free solder while it is still
hot.

But I'd go with zinc. It's cheap ($20 per pound, or just melt U.S. pennies), very
easy to melt and work with, needs no plating to resist oxidation, and even if
copper is four times better at heat conductivity, the zinc should work just fine.


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




On Mon, Oct 1, 2012 at 2:00 PM, Cathal Garvey <cathal...@gmail.com> wrote:
I suspect a cheaper heatsink and PSU are low-hanging fruit, here.
I would stick with a standard Arduino; as someone pointed out earlier,
in the €600 pricetag, €20 for an extremely well-supported and

--
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 https://groups.google.com/groups/opt_out.



John Griessen

unread,
Oct 1, 2012, 11:03:31 PM10/1/12
to diy...@googlegroups.com
On 10/01/12 10:41, Jeswin wrote:
> Any ideas for simpler lid unit, cheaper heat-block, PSU, and MCU?

I'll be making one that has a much cheaper heat block.
It doesn't use one, while still doing fast temperature ramp ups.
Can;t work on it next week though -- maybe March.

Phil

unread,
Oct 2, 2012, 9:41:57 AM10/2/12
to diy...@googlegroups.com
Why are people so obsessed with building PCR machines?  Will someone explain the reason?

It is unhelpful.  There is no need to build cheap PCR machines.  You can buy used PCR machines on ebay for $100-$300.
I advise people with those skills to use them designing or building something DIYers need and can't get, like a backup power supply that will keep a small freezer and a small incubator running for three days, or a small, lightweight, laminar flow hood.

Cathal Garvey

unread,
Oct 2, 2012, 9:59:20 AM10/2/12
to diy...@googlegroups.com
Hey, solid points; getting some silver and making a miniature heatblock
by casting is certainly a good possibility for PCR block manufacture.

For a while, you could get sintered silver models from shapeways. Should
have made the most of it while it lasted! But, costly. Probably more
than double its value due to the tech requirements and margins alone.

On 02/10/12 02:38, Simon Quellen Field wrote:
> Suppose you took your PCR tubes (that you want to fit into a homebrew
> heatblock) and make a mold of them in plaster or one of the rubber mold
> making products. Then make a positive of it in plaster.
>
> Now melt a bunch of tin in a cup, and insert your plaster images of the
> PCR tubes. Let the tin cool, and then etch out the plaster with some
> hydrochloric
> acid.
>
> No machining, and you now have a big tin heat block. Make as many holes as
> you like for tubes, and make as many shapes as there are types of tubes you
> might want to use.
>
> At $23 per pound <http://stellartechnical.com/puretinbar.aspx>, tin is
>> in the �600 pricetag, �20 for an extremely well-supported and

Bryan Bishop

unread,
Oct 2, 2012, 11:23:10 AM10/2/12
to diy...@googlegroups.com, Bryan Bishop, Phil Goetz
On Tue, Oct 2, 2012 at 8:41 AM, Phil Goetz <phil...@gmail.com> wrote:
> There is no need to build cheap PCR machines. You can buy used PCR machines
> on ebay for $100-$300.

Those don't always/ever work. The other advantage of using a
community-vetted thermocycler is that it's much easier to get the
community able to doll out support and develop with each other. But if
you know a reputable ebay seller that sells working/tested
thermocyclers, let me know.

- Bryan
http://heybryan.org/
1 512 203 0507

Sebastian S. Cocioba

unread,
Oct 2, 2012, 3:24:22 PM10/2/12
to diy...@googlegroups.com
You could ask the ebay seller to run through the self test. Just write very clear step by step directions on how and to post pics of each test result. Thats how i got my working gene amp 9600. Cheap but very large. Works great! Chances are some sellers would go the extra step to sell their stock. Good luck.

Sebastian S Cocioba
CEO & Founder
New York Botanics, LLC

Sent via Mobile E-Mail

Bryan Bishop

unread,
Oct 2, 2012, 3:28:11 PM10/2/12
to diy...@googlegroups.com, Bryan Bishop, Sebastian S. Cocioba
On Tue, Oct 2, 2012 at 2:24 PM, Sebastian S. Cocioba wrote:
You could ask the ebay seller to run through the self test. Just write very clear step by step directions on how and to post pics of each test result.

They aren't going to buy plasmids/DNA/primers, run gels and post results. That just seems highly unlikely to me. I am not sure if anything less than testing some reactions would be enough, really..

Jeswin

unread,
Oct 2, 2012, 6:17:42 PM10/2/12
to diy...@googlegroups.com
That's right. If the machine runs the protocol, we only know that the
MCU is working. We have no idea if the heat block is working properly
or the temperatures are displayed accurately. The machine could have
moisture damage or a cracked heat-block, for all we know.

$600 is a lot for a single instrument. The coffee cup PCR is a good
one, at half the price. If we use an innovative alternative to
machined heat-blocks, like Simon's suggestion, then the price is
better. Maybe the OpenPCR folks should build a budget version, based
on that?

Sebastian S. Cocioba

unread,
Oct 2, 2012, 6:23:38 PM10/2/12
to diy...@googlegroups.com
Very true. Do you know of any modern Peltier pcr machine that has a user serviceable or hackable Peltier plate? Those things tend to die in 4-5 yrs. my old giant pcr machine runs off of fridge coils and compressors so all i have to do is refill the coolant with prestone 150. Anyone have the same luck with other old pcr machines?

Sebastian S Cocioba
CEO & Founder
New York Botanics, LLC

Sent via Mobile E-Mail

Jeswin

unread,
Oct 2, 2012, 6:38:05 PM10/2/12
to diy...@googlegroups.com
On Tue, Oct 2, 2012 at 9:41 AM, Phil <phil...@gmail.com> wrote:
> Why are people so obsessed with building PCR machines? Will someone explain
> the reason?
>

Because they're a core Molecular Bio equipment and still expensive for
many people. We're talking in dollars. Convert the OpenPCR cost to the
currency in South American, Asian, or African countries and you can
see how expensive it is. How about the curious teenager or a High
School teacher whose budget (or School's budget) won't let them buy
such an expensive machine? If my budget is $1000, how much is left for
the other lab equipment?

> I advise people with those skills to use them designing or building
> something DIYers need and can't get, like a backup power supply that will
> keep a small freezer and a small incubator running for three days, or a
> small, lightweight, laminar flow hood.
>
No idea about the back-up power but as to Laminar Flow (a must have in
any lab). Go to Wal-mart and get a 19 gallon (or less, not sure how
big these are) clear storage container. Cut up the side with a Dremel
to get the opening. Add a fan to the top to blow air in and out in
front of you. Don't forget to add a HEPA filter on the fan. I am
assuming you are not working with dangerous chemicals or bio-hazards.
The will prevent contamination of samples. Might cost you less than
$70, depending on the items you buy.

Lots of examples from fungus growers:
http://www.fungifun.org/English/Flowhood

> --
> 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.
> To view this discussion on the web visit
> https://groups.google.com/d/msg/diybio/-/okgWcVfT8GkJ.

Cathal Garvey

unread,
Oct 6, 2012, 5:27:57 PM10/6/12
to diy...@googlegroups.com
The peltier used in the OpenPCR is readily accessible and is a
relatively easily bought/replaced part; peltiers of that type are
popular among computer modders as an extreme processor-cooling gadget.

Indeed, most of the components in the OpenPCR are off the shelf computer
parts; the ready repair-ability of the device is one of the unique
selling points.

Josiah Zayner

unread,
Oct 8, 2012, 10:30:48 AM10/8/12
to diy...@googlegroups.com
I don't understand why it costs greater than $500? I mean a decent peltier, a temperature sensor, an aluminum block, a heat sink/cooling unit, a power supply and an arduino can all be had on ebay for less than $60. Is it the body?

Jeswin

unread,
Oct 8, 2012, 11:41:07 AM10/8/12
to diy...@googlegroups.com
My guess is that the body, custom parts (mainly the heat-block, also
the screws, whatnot) and probably for developing the software. As
discussed in another thread, OSS != free.

But I had the same thinking as you do. The main parts should not cost
around $500 except the expensive machined heat-block. I'm thinking of
other ways to build a PCR on the cheap and in kit form. Not everyone
wants to scrounge around for parts when you can get it all in one box.
> To view this discussion on the web visit
> https://groups.google.com/d/msg/diybio/-/H6rlv856f3kJ.

Josh Perfetto

unread,
Oct 8, 2012, 2:05:33 PM10/8/12
to diy...@googlegroups.com
The cost breaks down about three ways:

1. Off the shelf components - there is no single obscenely expensive
component of the OpenPCR, but a lot of moderately expensive and even
inexpensive components that add up. You can find cheap temperature
sensors for $1, but these don't have the accuracy and dynamic response
that are required for OpenPCR's 0.5C accuracy. It's the same for a lot
of the components you mention and the other components in the BOM. For
example the specifications of the peltier are hardly any "decent"
peltier. Most peltiers on ebay are used for CPU cooling which are not
suitable for thermocycling. The PSU is capable of 12 A at 12 V. etc.
Even if you add up the McMaster-Carr components, that produces a hefty
bill.

2. Custom components - there are a handful of custom components in the
BOM, and it is very expensive to get these made in the small
quantities needed for the OpenPCR kits. Getting them machined properly
required a lot of back-and-forth prototyping with different
manufacturers. The block is made from 1100 aluminum which is difficult
to machine and most online machine shops don't even offer it.

3. Fulfillment - the whole purpose of providing OpenPCR kits to the
community is to collectively reduce costs by ordering in bulk and
redistributing to individuals. There is a lot of time and money spent
on finding suppliers, ordering, paying shipping and duty, inventorying
components, packaging the 211 total components (83 separate BOM items)
into individual baggies and boxes, accepting online payments, and
shipping out. Many of those components are actually modified by us
because required tools are not readily available. For example the USB
chip on the Arduino is reprogrammed using a hardware programmer. All
of this fulfillment takes an insane amount of work. Those who ordered
OpenPCRs early on knew we had some significant delays in fulfillment,
which was the result of figuring out how to do all of this and ramping
up our capacity. Plus, being a kit, we have to provide a lot of
post-sales support - people may assemble something wrong and require a
replacement part, etc.

I've thought about producing a cheaper design for a while and have
some ideas of how to do it, starting with removal of the heated lid
(which also, many of these cheap ebay units don't have). I haven't
moved on it mainly because I'm not sure it would be very
transformational. If people want to do PCR on a budget, it is already
possible with 3 pots, 3 burners on a stove, 3 thermometers, and some
time. This is the first way I did PCR at home. A PCR machine brings
automation. The purpose of automation is to trade money for time.
Before OpenPCR there were already a range of automation trade-offs
available, ranging from $300 used ebay units to $10,000 new units. I
didn't like any of those trade-offs however. I was looking for a lower
price, open design, fully functional/accurate unit, with a heated lid
(which further automated by removing mineral oil PCR steps, which
often leads to contamination, especially in some public lab classes I
was running). Thus, OpenPCR.

Now lets say there was a $200-$300 OpenPCR. Sure a cheaper machine is
always more beneficial to the community, but how much more? Would it
be worth the effort? Here is my thinking. For someone who wants to do
a small number of basic PCR experiments on a shoe-string, there are
already free or extremely low cost options (i.e. light bulb PCR
discussed previously on this list), so they have choices if the $600
OpenPCR is an obstacle (they can also go to a community lab/community
college if they have nearby). For someone who wants to do some serious
experimentation on an on-going basis (like the guy who wanted to do
chicken parentage testing I referred to this forum), you are going to
need more than a PCR machine (polymerase, primers, digestion enzymes,
gel/capillary electrophoresis, sequencing services, etc), so a $600
PCR machine is actually just a small fraction of the overall cost.

So rather than just making ever more cheaper PCR machines, I am more
interested in:

1. Automated machines, like a self-contained device that can
completely answer the chicken maternity/paternity question and other
such questions on its own

2. Cheaper/open versions of other lab equipment - so much of this
stuff is ridiculously expensive and no open software for - flow
cytomers, mass spec, -80 storage, fluorescent microscope, capillary
electrophoresis, microfluidic drivers, nanodrop, etc.

-Josh
> To view this discussion on the web visit
> https://groups.google.com/d/msg/diybio/-/H6rlv856f3kJ.
>

Josh Perfetto

unread,
Oct 8, 2012, 2:10:22 PM10/8/12
to diy...@googlegroups.com
On Mon, Oct 8, 2012 at 11:05 AM, Josh Perfetto <jo...@snowrise.com> wrote:
> 2. Cheaper/open versions of other lab equipment - so much of this
> stuff is ridiculously expensive and no open software for - flow
> cytomers, mass spec, -80 storage, fluorescent microscope, capillary
> electrophoresis, microfluidic drivers, nanodrop, etc.

Oh, and qPCR ;-)

-Josh

Pieter

unread,
Oct 23, 2012, 4:36:55 PM10/23/12
to diy...@googlegroups.com
With the Amplino we started out using a hot air design for our device, but it turns out it was a real pain to calibrate the prototype each time we wanted to test it. Also it's hard to get a homologous temperature in the entire airflow. A 0.5 C resolution, to rival with other PCR machines, did not seem feasible within our technical capabilities. 

Fortunately there are many alternatives to heating, like infrared or microwaves even.

The cheapest way of doing PCR is probably setting three waterbaths at the right temperature, and move around your sample :) 

Simon Quellen Field

unread,
Oct 23, 2012, 7:13:51 PM10/23/12
to diy...@googlegroups.com
How important do people think it is to have 0.5 degree accuracy?
Clearly the samples are going to go from one extreme to the other
over some period of time, and that time has not been mentioned
as all that important (moving samples by hand between three water
baths, for example).

With so many protocols being frozen at 'what worked for Dr. X', and
never being optimized or cost reduced, the idea of giving up on a
solution because it didn't meet a specification that had no justification
seems unfortunate.

One thing the DIYBio community might do to further the art is to
characterize how important (or unimportant) it is to have precise
temperatures and timings. Imagine how much harder soldering would
be if we had to control the temperature to 0.5 degrees, and control
the timing to the millisecond. When 'heat it until it melts, and then
stop' is just fine for most things, unnecessary precision in protocols
can cause harm.

A graph that shows efficiency plotted against temperature extremes
and time at and between those times would let experimenters decide
how much they are willing to pay for that efficiency in equipment costs.
For something like PCR, where you might get double the yield by simply
running another cycle, is efficiency that important?

I am assuming that PCR needs to have the solution heated above some
particular temperature to open the DNA, and then cooled to some other
temperature to allow synthesis, and that these temperatures can be
exceeded by some amount, probably without harm. My bet is that the
degree to which they can be exceeded is quite a bit more than 0.5 degrees.

There is probably also a minimum time that is required at each
temperature extreme, and that not much harm comes from exceeding that
time by a few percent. But it might also be that doubling or tripling the
time also causes no harm, other than the delay in getting the product.

If the protocol works if simplified to 'heat it until most of the DNA splits,
then cool it until most of the DNA is double again', then maybe the timing
and temperature are the wrong things to measure, and we should build a
device that measures percentage of single stranded DNA, perhaps by a
color change or change in resistance of the solution.

-----
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 post to this group, send email to diy...@googlegroups.com.
To unsubscribe from this group, send email to diybio+un...@googlegroups.com.
Visit this group at http://groups.google.com/group/diybio?hl=en.
To view this discussion on the web visit https://groups.google.com/d/msg/diybio/-/8MhX9miw0CkJ.

Josh Perfetto

unread,
Oct 24, 2012, 1:19:27 AM10/24/12
to diy...@googlegroups.com
The main problem is the sensitivity of the annealing temperature, not efficiency. For many primers 5 C may be the difference between non-specific amplification and no amplification at all, with the desired specific amplification somewhere in the middle. How sensitive the annealing temperature is depends on the specific primers. One great way to decrease this dependency is by using touch-down PCR, which we are adding to future versions of the OpenPCR.

If you have to make a trade-off, it is ok to have less accuracy as long as you have repeatability. Though this may mean you have to make more attempts to get the PCR reaction to initially work, at least once you do find an annealing temperature setting that works, it will work for future reactions as well, even if the annealing temperature you set is in fact several C different than what the machine is actually doing.

My desire for 0.5 C accuracy was to make it easy to repeat published protocols and get a specific PCR reaction to work with a minimum amount of tries (which have a cost in terms of time plus gel runs). This can of course be compromised on in the interests of costs, though I think accuracy is a relevant consideration for less experienced users who may be struggling to get their first PCR reaction to work -- it is great if you can rule out machine error.

-Josh

Nathan McCorkle

unread,
Oct 24, 2012, 2:19:06 PM10/24/12
to diy...@googlegroups.com
On Tue, Oct 23, 2012 at 7:13 PM, Simon Quellen Field <sfi...@scitoys.com> wrote:
> How important do people think it is to have 0.5 degree accuracy?
> Clearly the samples are going to go from one extreme to the other
> over some period of time, and that time has not been mentioned
> as all that important (moving samples by hand between three water
> baths, for example).

The temperature transition time is also as important as the annealing
temperature, as Josh talked about. The transition temperature matters
because while polymerase is most efficient at higher-than-annealing
temps, it still works, so if the transition is slow, it gives
opportunity for non-specific priming to start extending. A few months
ago I found an article touting 15 degrees C per second for its temp
ramping. This means it will give more repeatable results, and cycle a
bit faster too.

I remember reading about using a laser to heat a droplet of PCR
reaction... I think something like that might work for DIYbio pretty
well. Sit the whole thing on top of a cold peltier, and add heat only
when needed. The small droplet would be small enough that your
off-the-shelf peltier would be sufficient.

--
-Nathan

Chris Templeman

unread,
Oct 24, 2012, 2:20:19 PM10/24/12
to diy...@googlegroups.com
@Pieter does this mean the Amplino is no longer going to be hot air driven?  If so what technique is the Amplino team pursuing?  I really think targeting malaria is a great idea and I hope you guys succeed.  

Josh Perfetto

unread,
Oct 24, 2012, 2:29:50 PM10/24/12
to diy...@googlegroups.com
On Tue, Oct 23, 2012 at 4:13 PM, Simon Quellen Field <sfi...@scitoys.com> wrote:
With so many protocols being frozen at 'what worked for Dr. X', and
never being optimized or cost reduced, the idea of giving up on a
solution because it didn't meet a specification that had no justification
seems unfortunate.

Also wanted to add, the great thing about a 0.5 C accurate PCR machine is that you can pick an arbitrary target, use software like PrimerZ to design primers and calculate an annealing temperature, and often get the PCR to work the first time. I.e. it is a very general purpose instrument.

If the goal though is to get a specific PCR reaction to work cheaply, there are some interesting options. One is isothermal amplification - there are many variants but one interesting one is LAMP http://en.wikipedia.org/wiki/Loop-mediated_isothermal_amplification. Another interesting technique is using convection with a simple heater and described here: 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.

With both of these techniques, it will take more effort (could be 1+ weeks) to initially get your reaction to work, but then once you've identified the exact conditions and primers, you can roll this protocol out much more cheaply - without an accurate PCR machine or any PCR machine for that matter.

-Josh

John Griessen

unread,
Oct 25, 2012, 12:13:56 PM10/25/12
to diy...@googlegroups.com
On 10/24/2012 12:19 AM, Josh Perfetto wrote:
> My desire for 0.5 C accuracy was to make it easy to repeat published protocols and get a specific PCR reaction to work with a
> minimum amount of tries (which have a cost in terms of time plus gel runs).

Yes, accuracy beyond repeatability is a lowest total cost feature, if you can
deliver accuracy inexpensively. I think it is possible with stirred air
and blend doors, but don't have any experiments going just now... still distracted
by other projects that generate money for rent.

On 10/24/2012 01:19 PM, Nathan McCorkle wrote:
> I remember reading about using a laser to heat a droplet of PCR
> reaction... I think something like that might work for DIYbio pretty
> well. Sit the whole thing on top of a cold peltier, and add heat only
> when needed. The small droplet would be small enough that your
> off-the-shelf peltier would be sufficient.

That's a great lateral thought -- work with tiny volumes to reduce costs of equip.
while upping performance of same equip.
Reply all
Reply to author
Forward
0 new messages