DIY Electroporator (was: New ideas)
Nathan McCorkle
>
> On Sep 21, 11:43 pm, Nathan McCorkle <nmz...@gmail.com> wrote:
> > Is that a commercial set up? Pics?
>
> My Baekon unit is an oddball. It doesn't use the paddles or wires.
> It has a round cuvette with what looks like a metal BB embedded in the
> bottom which makes contact with a contact in the bottom of the base.
> The top of the base unit has a needle electrode which comes straight
> down the center. On the side is a dial and mechanical height
> indicator which raises and lowers the needle.
>
Very interesting set-up, much different from the BioRad MicroPulser
that I have used.
>
> The BioRad electroporator cuvettes are available in 5 packs if I
> remember right. The dealer called them "qickpacks" or something
> similar. Still, they're way too expensive for what they are. I would
> look into another supplier if you can find one. I think there are
> several that make compatible cuvettes, although you'd probably have to
> call or research a bit to find out what ones would fit.
>
The professor who owns the BioRad electroporator has disposable
cuvettes and will give me one to use for the control experiment. He
also gave me one a few minutes ago so I can design a holder to connect
the piezo pulser to, or for that matter, a personal electroporator
design (once that comes about).
>
> I doubt the cuvettes are autoclavable. But I don't see why you
> couldn't sterilize with another method such as gas, IPA, EtOH, bleach,
> iodine, or even just boiling water. Most of those are probably not
> very good for the electrode material, but I don't see why it wouldn't
> work at least several times each.
Searching online, it seems they are made of polycarbonate, and don't
hold up in the autoclave. There were a few google hits that mentioned
using 3 washes of dH2O, followed by 3 washes of 70% ethanol, then a
wash of 50% ethanol and left under UV
for a while.
Here are two pages form the BioRad manual with some relevant info...
here it says E. coli get 1.8kV, but I swear I read a patent online
last Friday that said they were pulsing with 17-19kV... so maybe you
can vary the current a lot, or maybe their electrode width was much
larger. The table's "mnemonic" lists EC1 and EC2, which are respective
of a 1mm cuvette and a 2mm cuvette.
http://diybioforum.org/projects/biorad1.JPG
http://diybioforum.org/projects/biorad2.JPG
--
Nathan McCorkle
Rochester Institute of Technology
College of Science, Biotechnology/Bioinformatics
Meredith L. Patterson
> Here are two pages form the BioRad manual with some relevant info...
> here it says E. coli get 1.8kV, but I swear I read a patent online
> last Friday that said they were pulsing with 17-19kV... so maybe you
> can vary the current a lot, or maybe their electrode width was much
> larger.
1.8kV across a .1cm gap would be a field strength of 18kV/cm, which is
a measurement I've seen in patents and several papers -- was it that?
Cheers,
--mlp
Nathan McCorkle
>
> Cheers,
> --mlp
Nathan McCorkle
http://pages.towson.edu/jsaunder/Saunders%20Publications/46.Pulse%20Generators%20for%20Electrofusion%20and%20Electroporation.pdf
I have been designing an electroporator circuit with a new mentor who
is a physics and high voltage guy. With the above PDF I have
determined, I believe, how to accomplish power modulation without
having a variable transformer. Have banks of different sized
capacitors to control the pulse width, as the cells need only a
specific amount of power/electric-field-effect (area under the curve
of voltage vs pulse length). These will be TTL controlled, and
probably feed into a multiplexer for control by uProcessor.
I am still working on the piezoelectric el-cheapo setup, but I am not
sure how to bring down the 18kV coming out of it to something that
aligns with the power requirements. (Which I believe is in the Lurquin
paper that so many people generously responded to and forwarded to
me). The oscilloscope readings Simon sent me of the piezo show a pulse
width of 40uS. Which I can't quite calculate based on the equations
lurquin and another paper give. (I wish they had V/cm by pulse width
vs bacterial transformation efficiency).
Today I took notes on the BioRad cuvette I had, using a nice digital
caliper. I will try to fab the cuvetter holder within the next couple
of days.
-Nathan
attached are some PDFs with math in them (lurquin is the most intense
and descriptive.
ben lipkowitz
even possible?
there are numerous ways to host a PDF file, for example filebin.ca or
filebin.org or scribd.com, though scribd sucks (why would anyone use it?)
just imagine i'm downloading my mailbox over a packet radio connection
through a string with tin cans and onto a grooved wax cylinder, OK?
and no, forcing me to switch to gmail is not the answer!
-fenn
Meredith L. Patterson
> I am still working on the piezoelectric el-cheapo setup, but I am not
> sure how to bring down the 18kV coming out of it to something that
> aligns with the power requirements. (Which I believe is in the Lurquin
> paper that so many people generously responded to and forwarded to
> me). The oscilloscope readings Simon sent me of the piezo show a pulse
> width of 40uS. Which I can't quite calculate based on the equations
> lurquin and another paper give. (I wish they had V/cm by pulse width
> vs bacterial transformation efficiency).
Since the piezo circuit is rapidly switching on and off, you have a de
facto PWM (pulse width modulation) circuit. Figure out the duty cycle
(percentage of time on vs. percentage of time off) and you can compute
the effective voltage. If the pulses you have are "on for 40 uS, off
for 40 uS" then that's a duty cycle of 50% and an effective voltage of
9kV, assuming that the peak is 18kV.
Have you tried adding resistors in series on either side of the piezo
terminals to induce a voltage drop?
Cheers,
--mlp
AARON LEWIS DINKIN
ben lipkowitz
> Packet radio? What's your call?this is KC2YAN 73
no i'm not actually that cool (yet), i just have a very slow email server
with a 100MB disk quota. it only becomes a problem when i start getting
unsolicited attachments.
J. S. John
>> I am still working on the piezoelectric el-cheapo setup, but I am not
>> sure how to bring down the 18kV coming out of it to something that
>> aligns with the power requirements. (Which I believe is in the Lurquin
>> paper that so many people generously responded to and forwarded to
>> me). The oscilloscope readings Simon sent me of the piezo show a pulse
>> width of 40uS. Which I can't quite calculate based on the equations
>> lurquin and another paper give. (I wish they had V/cm by pulse width
>> vs bacterial transformation efficiency).
>
> Since the piezo circuit is rapidly switching on and off, you have a de
> facto PWM (pulse width modulation) circuit. Figure out the duty cycle
> (percentage of time on vs. percentage of time off) and you can compute
> the effective voltage. If the pulses you have are "on for 40 uS, off
> for 40 uS" then that's a duty cycle of 50% and an effective voltage of
> 9kV, assuming that the peak is 18kV.
I been doing some reading on this subject. Given the fact that an RC
circuit is quiet cheap considering the cost of a capacitor/resistor,
why use a piezoelectric? The data for RC electroporators is already
out there but a piezos would need more testing, correct?
But the thing I haven't been able to learn about yet is how to control
the time of electroporation in an RC circuit? The times I read about
are in the milliseconds so an on/off switch is not the best to control
that? Any ideas not including Arduinos, as I don't have any nor too
much experience with them. Any possible ICs that could do the job? My
electrical experience is low but still learning.
Thank you for any advice.
--
Folding@Home
http://folding.stanford.edu/
Team Maximum PC Magazine (Team ID: 11108)
Rank 3
Meredith L. Patterson
> I been doing some reading on this subject. Given the fact that an RC
> circuit is quiet cheap considering the cost of a capacitor/resistor,
> why use a piezoelectric? The data for RC electroporators is already
> out there but a piezos would need more testing, correct?
An RC electroporator needs HV capacitors of sufficient capacitance. I
have not been able to find a 5 uF capacitor rated for 3kV apart from
expensive (~$150 or more) gas-can-sized things. If you can track down
conveniently sized, inexpensive ones, you'd be doing us all a favour.
Piezos will certainly need extensive testing; the advantage here is
that the core component costs 89 cents in the checkout line at the
supermarket.
> But the thing I haven't been able to learn about yet is how to control
> the time of electroporation in an RC circuit? The times I read about
> are in the milliseconds so an on/off switch is not the best to control
> that? Any ideas not including Arduinos, as I don't have any nor too
> much experience with them. Any possible ICs that could do the job? My
> electrical experience is low but still learning.
I used a 555 timer and a whole bunch of 74xx logic to build out
something that would do a variable number of variable-width pulses at
a variable duty cycle -- and 7-segment LEDs to display all of this
stuff -- and when the cost of all the binary counters, 7-segment
drivers, hex buffers and crap added up to be over $20 and I still
didn't have it right yet, I said "fuck it" and bought an Arduino NG
off a guy who wasn't using his anymore. I took it home, and a few
hours later I had the ability to fire off a variable number of
(fixed-width, fixed-duty-cycle) pulses, with the number set by turning
a potentiometer.
I really encourage people to try out the Arduino if they're at all
programming-inclined -- it's a terrific prototyping board for the
ATmega series of microcontrollers, and I think you can use the Arduino
bootloader to program ATmega uCs for use in dedicated circuits (you
just have to burn the bootloader onto the uC first).
Hmm. It's been a while since I've played with mine; I think I'll come
up with a way for it to take serial input for the number of pulses,
pulse width and duty cycle. That sounds like a fun weekend project.
Cheers,
--mlp
Nathan McCorkle
>
>> On Sat, Oct 3, 2009 at 12:23 PM, Nathan McCorkle <nmz...@gmail.com> wrote:
>>> I am still working on the piezoelectric el-cheapo setup, but I am not
>>> sure how to bring down the 18kV coming out of it to something that
>>> aligns with the power requirements. (Which I believe is in the Lurquin
>>> paper that so many people generously responded to and forwarded to
>>> me). The oscilloscope readings Simon sent me of the piezo show a pulse
>>> width of 40uS. Which I can't quite calculate based on the equations
>>> lurquin and another paper give. (I wish they had V/cm by pulse width
>>> vs bacterial transformation efficiency).
>>
>> Since the piezo circuit is rapidly switching on and off, you have a de
>> facto PWM (pulse width modulation) circuit. Figure out the duty cycle
>> (percentage of time on vs. percentage of time off) and you can compute
>> the effective voltage. If the pulses you have are "on for 40 uS, off
>> for 40 uS" then that's a duty cycle of 50% and an effective voltage of
>> 9kV, assuming that the peak is 18kV.
>
> I been doing some reading on this subject. Given the fact that an RC
> circuit is quiet cheap considering the cost of a capacitor/resistor,
> why use a piezoelectric? The data for RC electroporators is already
> out there but a piezos would need more testing, correct?
>
> But the thing I haven't been able to learn about yet is how to control
> the time of electroporation in an RC circuit? The times I read about
> are in the milliseconds so an on/off switch is not the best to control
> that?
uSeconds.... you can change the frequency of pulse and pulse width by
increasing or decreasing the capacitor's uF rating... smaller = faster
charge and discharge, bigger = longer charge, longer width
Any ideas not including Arduinos, as I don't have any nor too
> much experience with them. Any possible ICs that could do the job?
showed that after 200mhz, the electroporation efficiency decreases
(this isn't neccesarily bad, I'm sure it could be used at some point
to control electroporation operation)... control the discharge start
with an SCR that is controlled by TTL.
My
> electrical experience is low but still learning.
>
> Thank you for any advice.
>
>
>
> --
> Folding@Home
> http://folding.stanford.edu/
> Team Maximum PC Magazine (Team ID: 11108)
> Rank 3
>
> >
>
Nathan McCorkle
<clon...@gmail.com> wrote:
>
> On Sat, Oct 3, 2009 at 12:23 PM, Nathan McCorkle <nmz...@gmail.com> wrote:
>> I am still working on the piezoelectric el-cheapo setup, but I am not
>> sure how to bring down the 18kV coming out of it to something that
>> aligns with the power requirements. (Which I believe is in the Lurquin
>> paper that so many people generously responded to and forwarded to
>> me). The oscilloscope readings Simon sent me of the piezo show a pulse
>> width of 40uS. Which I can't quite calculate based on the equations
>> lurquin and another paper give. (I wish they had V/cm by pulse width
>> vs bacterial transformation efficiency).
>
> Since the piezo circuit is rapidly switching on and off, you have a de
> facto PWM (pulse width modulation) circuit. Figure out the duty cycle
> (percentage of time on vs. percentage of time off) and you can compute
> the effective voltage. If the pulses you have are "on for 40 uS, off
> for 40 uS" then that's a duty cycle of 50% and an effective voltage of
> 9kV, assuming that the peak is 18kV.
really matters, as well as the overall effect, but I think its each
field induction effect, and then how many times you induce it that
matters.
>
> Have you tried adding resistors in series on either side of the piezo
> terminals to induce a voltage drop?
Nathan McCorkle
<clon...@gmail.com> wrote:
>
> On Sat, Oct 3, 2009 at 3:14 PM, J. S. John <phill...@gmail.com> wrote:
>> I been doing some reading on this subject. Given the fact that an RC
>> circuit is quiet cheap considering the cost of a capacitor/resistor,
>> why use a piezoelectric? The data for RC electroporators is already
>> out there but a piezos would need more testing, correct?
>
> An RC electroporator needs HV capacitors of sufficient capacitance. I
> have not been able to find a 5 uF capacitor rated for 3kV apart from
> expensive (~$150 or more) gas-can-sized things. If you can track down
> conveniently sized, inexpensive ones, you'd be doing us all a favour.
charge a 400v capacitor, which is controlled by an SCR (400 piv @
1A)... the cap dumps into a flash transformer (300 v in, 4000 v
out)... the pulse width is changed by switching uF values of the cap.
the pulses are triggered by opto-coupled TTL using the SCR (off till
TTL pulse, then on til source voltage drops)
Simon Quellen Field
What you want is a simple voltage divider:
"http://en.wikipedia.org/wiki/Voltage_divider"
A potentiometer is a variable resistor that has three terminals.
One terminal at each end of the resistor, and a third that is connected
to the wiper that sweeps across the resistor. This is perfect for use as
a voltage divider -- you connect the two ends of the resistor to the positive
and the ground, and the wiper lets you select any voltage from zero to
the maximum available from the power supply.
My latest science fiction novel A Twisted Garden is now available in bookstores.
Simon Quellen Field
1. Pulse frequency seems to have no effect on results.
2. Pulse amplitude and number of pulses are the most important.
3. Saturation occurs as the amplitude and number of pulses reach a limit,
and that happens (in pig heart valve cells) at 1.3 kV/cm and 5 pulses total
using 100 microsecond pulses with a 50% duty cycle.
4. Efficiencies of 88% and above were reported at that limit (although the graph has a point at 100%).
5. Bacteria and yeast require higher field strengths than mammalian cells, with
some researchers using as much as 18kV/cm.
6. Exponential decay (as opposed to square waves) is used for bacterial electroporation.
These are the easiest to build, and are what you get with the piezoelectric igniter.
7. Electroporation efficiency can be measured using simply dyes instead of plasmids.
Calcein-AM (Sigma catalog #C0875) is used to tell live cells from dead cells, since it only
stains live cells, and propidium iodide is used since it cannot cross cell membranes
without electroporation. Under the microscope, cells that glow in both colors are live
cells that incorporated the dyes via electroporation. Counting those and the cells that only
absorbed the propidium iodide and calculating the ratio gives you the efficiency.
8. The "microcell" used in one study is something better described as a "millicell", and is
something we could hack up fairly easily. Where they used gold and lithography to lay
down electrodes with millimeter spacing, we could do it by laying down wire by hand under
a magnifying glass or low power microscope. They used 70 micron trace widths -- that is
so close to 40 gauge wire (78.74 micron) that it probably makes no difference. One foot
of 51 micron diameter platinum wire is $40 at Surepure Chemetals:
"http://www.surepure.com/products.php?ID=5&meas1_ID=2&subCat=15&units=Metric"
They have gold wire in 76 micron, 51 micron, 25 micron, and 12.7 micron diameters
for about the same price.
Lay down the wire, glue it in place, then use a razor blade to cut a 1 mm section out, and
you have your microcell electrodes.
(I did not see any mention of 200 megahertz -- that seems like an awfully high frequency
for a high voltage apparatus, especially when frequency seems not to be important to
electroporation.)
My latest science fiction novel A Twisted Garden is now available in bookstores.
Cathal Garvey
This may be just my opinion on the matter. Consider it an apology in advance if I send any unsolicited attachments in the future without giving forewarning.
-Cathal
--
letters.cunningprojects.com
twitter.com/onetruecathal
Kiva.org - Loans That Change Lives
JonathanCline
> Hi Ben.. As much as I empathise with your bandwidth issues
Notebook page on OpenWetWare. This is how the other projects have
proceeded. It is possible to post up results and track project status
that way, including files or pictures. Web 2.0 collaboration!
## Jonathan Cline
## jcl...@ieee.org
## Mobile: +1-805-617-0223
########################
Jake
square or exponential will both work well enough. They're both used
on high-end devices and I haven't seen that much that either is that
important.
With the dye test you won't be actually testing for transformation
efficiency, so I don't think it's that great of a measure. It's easy
enough to test for actual transformation efficiency, and probably
cheaper.
One thing I was thinking on this pizeo electroporator is that it's not
going to work very well with a fixed distance between the electrodes.
You're going to need to vary the distance between them since you can't
really vary the voltage.
I have a Baekon electroporator with plenty of cuvettes. The unit has
an adjustable gap so it should be easy to tune it to find the right
length. Whoever can donate the time and aquire the cells and plasmids
to test transformation efficiency let me know. I'll be happy to loan
out the Baekon unit for as long as it takes to do the experiments.
Once we know how well the lighter works, the correct gap, and the
number of pulses we might just end up with a dirt cheap
electroporator.
-Jake
Nathan McCorkle
>
> Just a couple thoughts... Simon mentioned a few points, but I think
> square or exponential will both work well enough. They're both used
> on high-end devices and I haven't seen that much that either is that
> important.
>
> With the dye test you won't be actually testing for transformation
> efficiency, so I don't think it's that great of a measure. It's easy
> enough to test for actual transformation efficiency, and probably
> cheaper.
>
> One thing I was thinking on this pizeo electroporator is that it's not
> going to work very well with a fixed distance between the electrodes.
> You're going to need to vary the distance between them since you can't
> really vary the voltage.
>
> I have a Baekon electroporator with plenty of cuvettes. The unit has
> an adjustable gap so it should be easy to tune it to find the right
> length. Whoever can donate the time and aquire the cells and plasmids
> to test transformation efficiency let me know.
willing and cooperative professor.
> I'll be happy to loan
> out the Baekon unit for as long as it takes to do the experiments.
The commercial electroporator that I will be testing the piezoelectric
sparker against is the property of a professor that doesn't want to
loan it out, even on campus for diagnostic/academic engineering
research.
I feel like loaning equipment is a great idea, as long as the borrower
pays shipping that adequately protects the equipment (including
insurance). I think there would need some sort of publicly viewable
loan check-in / check-out, accountability system to keep track of
loaned valuables and ensure things don't get lost.
>
> Once we know how well the lighter works, the correct gap, and the
> number of pulses we might just end up with a dirt cheap
> electroporator.
>
>
> -Jake
Nathan McCorkle
Eugen Leitl
> Hi Ben.. As much as I empathise with your bandwidth issues, asking everyone
> else to change their habits because you refuse to change yours is a little
Again I notice the attitude that the only way that counts is the way I do
things. Everybody must use my proprietary mail provider, everybody has
free broadband and free storage everywhere, especially on the road,
especially roaming. How could it ever be otherwise?
> odd. Nobody's forcing you to change to Gmail or whatever, but if you don't,
> it's not really fair to ask us to use roundabout ways of doing perfectly
> normal things!
The perfectly normal way to deal with attachements, especially large
attachements, is to post an URL to them. Surely an advanced user like
you knows how to do that? Or do you know how to do that, and just do
not bother, since it's less work for you?
> This may be just my opinion on the matter. Consider it an apology in advance
> if I send any unsolicited attachments in the future without giving
> forewarning.
You know, if more people start using the attitude such as yours in anger,
this list won't be a very enjoyable place. Please consider using more
restraint about what you think is the proper way of doing things.
--
Eugen* Leitl <a href="http://leitl.org">leitl</a> http://leitl.org
______________________________________________________________
ICBM: 48.07100, 11.36820 http://www.ativel.com http://postbiota.org
8B29F6BE: 099D 78BA 2FD3 B014 B08A 7779 75B0 2443 8B29 F6BE
Cathal Garvey
Simon Quellen Field
as a voltage divider. That means that if you set the spacing on the cuvettes to
be close, you can vary the voltage from too much to not enough in a nice smooth
manner (use a 10 turn pot and you will have more control than you'll ever need).
2. It may be easy to test E. coli without the dye, but testing L. acidophilus would
require getting plasmids for it, which is a problem we haven't solved yet. And
spending money solving the plasmid problem before we know we can get them
into the bug doesn't seem like the right sequence. Since we can already make
E. coli competent, my interest in the electroporator was for L. acidophilus. Other
members of the group will have different goals (and just building an electroporator
for its own sake is a fine goal!). So my interest in using the dye is that it looks like
a cheap and easy way to tell if you have succeeded in opening up a bunch of bugs
without killing too many, and without having to actually get plasmids for your bug.
If we did get the plasmids, and they didn't work, we wouldn't know if the problem
was the electroporator or something else. The dye gives us a way to test that has
fewer independent variables.
That said, working with E. coli first seems to be an important first step in testing the
electroporator, dye or not, and I agree that for E. coli using plasmids right away is
probably better than using the dye.
Just a couple thoughts... Simon mentioned a few points...
Jake
electroporate dye into something if you don't have any plasmid
available to do anything with?
I'll be sending out my Baekon (adjustable electrode gap) unit and a
modern electroporator power supply to Nathan McCorkle sometime this
week. He plans to analyze the unit in his electronics lab and
determine how it's working. He'll also be testing electroporation
efficiency of pizeo sparking vs. the modern unit.
This should be interesting!
-Jake