Re: [DIYbio] What's wrong with open source PCRs

That is how the first PCR machines worked, as I understand it Simon. Basically big hot tubs with the kind of baskets you see fries cooking in at a fast food joint robotically switching samples back and forth. Or, gasp, a poor grad student doing it manually.

I want to riff on your design, Simon.

Consider two water reservoirs: an insulated hot water carafe that keeps water near boiling (for tea etc) and a plastic cooler filled with ice water. 

The tanks are on a raised shelf. Each drains into an insulated silicone tube, say 1" inner diameter. The tubes connected with a Y joint. The base of the Y connects to a garden pump that pumps water back up to the hot water tank, which can overflow into the cold water tank.

Disconnect the tube from the base of the Y-joint and drop a ring of ~6 PCR tubes in, like shells into a revolver. Follow with thermocouple.

Fashion two servo-actuated hose clamps for the hot and cold tube. Differentially open/close clamps to set water temperature.

Beer-cave style PCR.

And a variant: if you're house's water heater is set to a dangerously high temperature, fashion a similar tube harness for a faucet and sink. Instead of hose clamps, use servos to control the faucet handles.

I wonder what the ramp rates would be.

Mac

On Tue, Jan 26, 2016 at 2:45 PM Simon Quellen Field <sfield@scitoys.com> wrote:

Just so I understand the problem a little better (I have never used a thermal cycler before), let me ask some questions about the specs I have found for one.

The BIO-RAD C1000 Touch Thermal Cycler 96-Well Fast uses 850 watts of input power.

It has a maximum ramp rate of 5 degrees Celsius per second, and an average ramp rate of 3.3 degrees C per second. It holds 96 tubes, each with 0.2 milliliters of liquid in them.

Two other specs are a little confusing to me: the Gradient range is 30-100 degrees C,

and the temperature differential range is 1-24 degrees C.

So we are talking about heating and cooling 19.2 milliliters of water, and the mass of the plastic tube. From the graph on that page, it seems like a typical use case is cycling between 60 C and 90 C, at about 30 to 40 second intervals.

To be on the conservative side, I will assume that the plastic holder takes as much power to heat as the water, and I'll round up to heating and cooling 40 ml of material.

Raising 40 ml of water by 30 degrees C takes 5,020.8 joules.

Doing that at a rate of 5 degrees per second takes 6 seconds.

5,020.8 joules in 6 seconds is 836.8 watts.

Let's call that the brute force method.

Now suppose I have two water baths, one at 60 degrees, and another at 90 degrees.

I use a small motor to move the 96 tubes from one bath to the other.

I can have a small fan that blows over the tubes when they go from the hot bath to the cold bath to speed the cooling, so that when the tubes hit the 60 degree bath they are already at 60 degrees.

Now my heating problem becomes one of maintaining the 60 degree bath at 60 degrees (making up for the losses through the Styrofoam insulation, and the evaporative losses), and maintaining the 90 degrees bath at 90 degrees, making up for the losses in heating the tubes, insulation, and evaporation.

The 90 degree bath loses 5,020.8 joules every minute (cycling 30 seconds hot and 30 seconds cold). That's 83.68 watts. Plus a small amount needed to make up for evaporative and insulation losses. And a bit to run the fan. Call it 90 watts.

At 12 volts, that's 7.5 amperes. At 20 volts, it's 4.5 amperes.

The reason the USB3.1 specification calls for 20 volts at 5 amps is so it can be used to charge laptops and power monitors. I actually own a desktop computer with USB3.1 ports and Type-C connectors, so I may be able to give actual numbers, once I get a device that can charge through a Type-C port. But the manufacturer claims it can do 20 volts at 5 amps. You can get a PCIe card with a Type-C connector for less that $20 and try it out for yourself.

So, it should be just barely possible to build a thermocycler that runs off a Type-C connector. You don't even need to go the water bath route. You can put a laptop battery in it, and get the whole 850 watts while it is running, and charge it from the USB port when it is idle.

But realistically, how much more convenient is plugging it into a USB port compared to plugging it into the wall? Just use house wiring and throw power at the problem.

Having said all that, I think a cool project would be a single well cycler that heats and cools 0.2 ml of solution, powered by USB. That should be doable in 10 watts. :-)

-----

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

On Tue, Jan 26, 2016 at 12:07 PM, Jonathan Cline <jcline@ieee.org> wrote:

The open lab devices don't use 2012 technologies.. they still use 1990's technologies, that's the thing that bugs many of us.  It's not only that they use an Arduino board and then have to employ an external Microchip ADC to do the actual analog work that bugs me.

Laws of thermodynamics means it is physically impossible to use normal 5V USB power for a typical thermocycler.  5W or 10W is not enough power.  1st law: "Can't get something for nothing."  Companies have produced "USB powered coffee mug warmers" before - these are scam products.  Regardless of ramp time, there must be enough energy (power) put into the system so the heat does not dissipate.  Buy one of these USB coffee warmer products and try it out for yourself if in doubt.  60W is possible for an efficient thermocycler but not sufficient for a large mass (heat block) like regardless of ramp time.  And you need more power to suck the heat back out in cooling.  Note: OpenPCR uses a 200W supply and the design uses a traditional large-mass block and still that block is smaller than most thermocyclers.  Two Hundred Watts, yes, capitalized.  First calculate out how to do proper heat transfer to/from PCR tubes within 60W of energy then design a device which satisfies that math (also accounting for losses).  Don't forget the anti-condensation lid.  Rather than reverse engineering the existing devices, start fresh with the hard math, and design something fresh which applies today's available technology.  

My bet is that USB 3.1 "20V 100W" won't happen mass market ever - there's no mass market need.  It might be used in a niche market for industrial control (products which will remain expensive because the industrial market is willing to pay).  Even if 100W is in the technical standards, people have to want/need a product for companies to have any drive or success building the product.  So that spec won't get built.  However "12V 60W" I bet could become mass market because of electric cars and smart/autonomous cars (automotive market is 12V).  But this will happen much later than the rest of USB-C rollout, maybe coinciding with broad adoption of smart cars -- which are still currently niche products.  I could be wrong on my bet which would mean it is a great bet to make against me.  You'd still need a 100W USB hub to power that single high-power 12V/60W USB-C port plus the two or three standard-power USB-C ports to round out a typical 4-port hub for a product, and this power has to come from somewhere - for example, automotive batteries if it's in a smart car (12V/15A accessory power = 180W).

The following has been repeated often enough in this group over the years that I call for a chorus  --  say it with me now:  5V/2A USB power is insufficient to power lab devices due to the laws of thermodynamics.  5V USB for a communications connection or small microcontroller power or a few LEDs, sure; heater power: no.

## Jonathan Cline  ## jcline@ieee.org  ## Mobile: +1-805-617-0223  ########################    

On 1/26/16 8:24 AM, Ujjwal Thaakar wrote:

I think you're absolutely right about that. I don't see any reason to use a DC jack anymore. In fact I was thinking about USB-C for power. Haven't actually looked into USB 3.0 and 3.1. I'm not even sure exactly what they are. I am already looking to move beyond the 2012 technologies used in the open machines so far. I'm also thinking about IoT or control using a mobile app. Not sure if that makes much sense going forward. Will be figuring it out as I go.

On Tuesday, January 26, 2016 at 4:21:30 AM UTC+5:30, Jonathan Cline wrote:

Future lab devices will be USB 3.0 powered by 12V, no external power supply.  60W is enough for an efficient PCR machine aka thermocycler.  Build for the future, have it hit the streets in 2017 as fully powered by USB only. [But monitor market acceptance of the standard while developing the design.]  Check out this summary:

USB 3.1: Released in July 26, 2013, USB 3.1 doubles the speed of USB 3.0 to 10Gbps (now called SuperSpeed+ or SuperSpeed USB 10 Gbps), making it as fast as the original Thunderbolt standard. USB 3.1 is backward-compatible with USB 3.0 and USB 2.0. USB 3.1 has three power profiles (according to USB Power Delivery Specification), and allows larger devices to draw power from a host: up to 2A at 5V (for a power consumption of up to 10W), and optionally up to 5A at either 12V (60W) or 20V (100W). The first USB 3.1 products are expected to be available in 2016, and will mostly use USB Type-C design.

On Sunday, January 24, 2016 at 12:40:19 AM UTC-8, Ujjwal Thaakar wrote:

Forget this thread. I just received a message from Josh Preffeto that I am trying to cheat him by subverting Indian customs and making a personal profit. That just makes me sad.

Too frequently in open designs, when a branch occurs (that is what you are trying to do), the project originators get angry.  In some cases the branch far surpasses the original, and in some cases, the original subsumes innovations on the branch.  The originators should not get angry but that's beside the point.  Only by re-engineering human nature will that difficulty be eliminated.   OpenPCR was not cost optimized as concluded in previous discussions (even though it is marketed as a low cost PCR machine).  However consumable costs for experiments will far surpass the cost of the PCR machine so if you can't afford the current price, you're already in trouble aren't you?  This is not to say it's not a good project, this is to say, there may be better projects to focus effort on, especially, reducing the cost of the consumables [reagents] or optimizing a workflow which uses smaller quantities of consumables [such as sub-2 uL].


## Jonathan Cline
## jcl...@ieee.org
## Mobile: +1-805-617-0223
########################

--
-- You received this message because you are subscribed to the Google Groups DIYbio group. To post to this group, send email to diybio@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 diybio@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/56A7D20F.5080509%40ieee.org.

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

--
-- You received this message because you are subscribed to the Google Groups DIYbio group. To post to this group, send email to diybio@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 diybio@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/CAA0yOM5NgEX4Brhsta3jTT21F6uSoYNq4tEUpR01Dd56ZUE7Jw%40mail.gmail.com.
For more options, visit https://groups.google.com/d/optout.

--
-- You received this message because you are subscribed to the Google Groups DIYbio group. To post to this group, send email to diybio@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 diybio@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/CA%2BiLdW0Pgnczex-x7049iswp_-5g9M9pghSeV85rLgf%3D058XpQ%40mail.gmail.com.

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