[Open_electroporator] recommenadtions after John Griessen's lame pulser design ideas

[TQ Hirsch]

Having read a bit further back in the thread (Gmail has some really glaring
flaws WRT threading), I see that the concept of a programmable waveform
generator has been floated, and so far the main concern has been shielding.
The way that I'd lay out the electronics is on three boards: a high-voltage
power supply, a low-voltage control board (with a µC and any computer
interfaces), and the high-speed, high voltage switch. You'd only be worried
about EMF from the switch onward, and that can be solved by wrapping a
Faraday cage around it (and hey, free heatsink!) You'd then run the actual
HV waveform to the wet unit via a coax cable. Everything even remotely
noisy would then be surrounded by a Faraday cage.

That said, this seems like a particularly easy solution, so I'm sure that
I'm missing something. (for starters, I know very little about EM
radiation) If so, I'd appreciate enlightenment.

--tq


On Tue, Mar 18, 2014 at 5:45 PM, TQ Hirsch <thequux at thequux.com> wrote:

> On Fri, Mar 14, 2014 at 10:32 AM, John Griessen <john at industromatic.com>wrote:
>
>> On 03/14/2014 11:39 AM, NeonJohn wrote:
>>
>>>
>>> I think this is a case where a microprocessor is inappropriate.  One can
>>> spend months duplicating in an algorithm what a few discrete components
>>> and a dedicated chip can do.
>>>
>>
>> The problem is the algorithm is subject to change and having a sequence
>> of pulses you can program is better than having a steady ripply output.
>> Experiments are one shot, and will just cook the live critters if left on
>> too long.  So some programability is really desired.
>>
>
>  It is also an option, presuming that ~20ns resolution is acceptable and
> you can afford $3-4 and some board space, to build what is essentially a
> single-bit DDS: hook a microcontroller to an SRAM, have the microcontroller
> write a 1-bit PCM signal to be output to the SRAM and then play it back
> with discrete logic or a CPLD (It would be a few 74-series chips plus the
> µC and SRAM). For bonus points, it doesn't have to be a 1-bit DAC. You can
> build an 8-16 bit DAC out of discretes fairly easily (presuming that you
> have a HV switch at your disposal), and thus have the world's only
> programamble waveform (not just programmable pulse) electroporator.
>
> --tq
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://cibolo.us/pipermail/open_electroporator/attachments/20140318/63fa50fc/attachment.html>
-------------- next part --------------
_______________________________________________
Open_electroporator mailing list
Open_electroporator at lists.cibolo.us
http://lists.cibolo.us/mailman/listinfo/open_electroporator