Building a non-lethal insect shock mat for bio research, a few questions / General Science and Electronics / Forums

Forums

4hv.org :: Forums :: General Science and Electronics

« Previous topic | Next topic »

Building a non-lethal insect shock mat for bio research, a few questions

1 2 next

Move Thread

LAN_403

aonomus

Wed Sept 24 2008, 04:55PM

Registered Member #1497 Joined: Thu May 22 2008, 05:24AM
Location: Toronto, Ontario, Canada
Posts: 801

Let me cover this formality first: this isn't a thread about ethics of research, etc. I already have approval from my uni, this is a thread about the electronics behind it. If you feel so inclined, PM me with your rant/whine.

Ok. So for a animal behaviour research project I have decided to study the factors that contribute or affect the ability of an insect to learn a behaviour either through positive or negative reinforcement. I have decided on using an electric shock as the negative reinforcement and I'm thinking about the method of implementation.

A few papers have shown that a resistive mat, typically plastic coated with a carbonaceous material (graphite, asphalt, etc) with a current flowing from one end to the other which acts as a resistor, with any insect ontop acting as a object of lower resistance (think: voltage divider) causing current to flow through the insect instead. A recent paper discussed using antistatic foam as a substitute for the plastic mat which would be more readily available, however this is where the first question comes into play.

There are two types of antistatic foam that I have seen before, the pink closed cell foam, and the black/grey open cell foam. I'm wondering if the pink foam is simply a plastic foam with a anti-static agent mixed into the plastic whereas the black/grey foam contains carbon/metallic particles to dissipate static charge.

The protocol listed in the two papers I have read describe using a square mat with 2 electrodes per side to form a criss-cross of potential circuits, and using phase-shifted 60Hz AC to create a rotating electrical current (ie: the 'connected' electrodes form circuits that rotate like a impeller). The rotating electrical current is to compensate for the specimens longitudinal axis being parallel or perpendicular to the current and receiving varying doses (whereas with a rotating field, the RMS dose is identical regardless of orientation).

The phase shifting is accomplished with isolating transformers, 200H chokes, capacitors, etc. My question is whether or not the circuitry is dependent on voltage? (See fig below)

Edit: Journal article:

wrote ...
The Resistive Sheet: A Gridless and Wireless Shocking Technique
Author(s): Nicholas Longo, Lawrence R. Holland, M. E. Bitterman
Source: The American Journal of Psychology, Vol. 74, No. 4 (Dec., 1961), pp. 612-618
Published by: University of Illinois Press

[Edit: Your picture was too large]

DrZoidberg

Thu Sept 25 2008, 03:36PM

Registered Member #350 Joined: Mon Mar 27 2006, 05:14PM
Location:
Posts: 106

You could use graphite spray. But you need several thousand volts because the chitin exosceleton of insects is insulating them and only a fraction of the voltage actually drops between the legs of the bug. If the "mat" is e.g. 10cm wide and you apply 10,000 V then that would be 1000V per cm.
If the legs of the insect are 5mm apart it would get shocked with only 500V.
I don't know if that's enough. I guess it depends on how thick the exoceleton is.
If you used snails you'd need much less voltage. Snails have no insulation at all. Even 1V should shock them.
Does the shock really have to be exactly the same each time? There are simpler ways to shock bugs e.g. with a wire grid but then the shock would have a different strength each time.

Backyard Skunkworks

Thu Sept 25 2008, 04:15PM

Registered Member #1262 Joined: Fri Jan 25 2008, 05:22AM
Location: Maryland, USA
Posts: 451

You really better keep a careful eye on how much current and power the bugs get. I know bugs tend to die quite easily when shocked. I've had a few wood lice crawl into a live disposable camera inverter (for some odd reason) and the 330V at 5mA kills them instantly. The lethal threshold for a lot of bugs might be well below 1mA, so my guess is your gonna have a hard time giving bugs a reasonable shock without damaging/paralyzing/killing them.

guitardude012

Thu Sept 25 2008, 05:10PM

Registered Member #968 Joined: Fri Aug 24 2007, 04:54PM
Location: Norman, Oklahoma
Posts: 104

I would look into an extremely sensitive Ground Fault Interrupter(GFI) setup. It will have to be custom in this current/voltage range I think. You will need a torroid, some op amps, a FET, and a logic circuit most likely.

This is what helps to prevent electrocution in the bathroom and makes sense to try.

I have a hard time believing the 5mA kills the bugs "instantly"... I imagine they are safe for a few ms.

Backyard Skunkworks

Thu Sept 25 2008, 11:37PM

Registered Member #1262 Joined: Fri Jan 25 2008, 05:22AM
Location: Maryland, USA
Posts: 451

guitardude012 wrote ...

I have a hard time believing the 5mA kills the bugs "instantly"... I imagine they are safe for a few ms.

Well I can't be 100% sure they died right at that time, but it made the bugs instantly stop moving, and they never started again. I still have no idea why they "migrated" right into my inverter though.

I guess it's possible that the inductance of the little transformer might provide enough of a jolt to do it, the 5mA is steady state when shorted across a DMM.

aonomus

Fri Sept 26 2008, 04:19AM

Registered Member #1497 Joined: Thu May 22 2008, 05:24AM
Location: Toronto, Ontario, Canada
Posts: 801

I don't think the GFCI would help at all since the current going into the variac is constant, and if there are isolating/step up or stepdown transformers on each 'leg', the GFCI might not work properly... I would have to put a small CT on each hot/neutral leg of the shock mat and use comparators.

Since the research paper made no mention of shock hazard I can only imagine what might happen...

Grant

Fri Sept 26 2008, 11:53AM

Registered Member #7 Joined: Thu Feb 02 2006, 04:32AM
Location: Central Florida
Posts: 18

I believe that Guitardude was suggesting the concept of a ground fault interruption system designed specifically for the shock mat in order to keep the current that the insect receives in check.

A GFCI breaker like that in your lavatory was just an example and would indeed serve no real purpose for the OP's intent.

aonomus wrote ...

I don't think the GFCI would help at all since the current going into the variac is constant, and if there are isolating/step up or stepdown transformers on each 'leg', the GFCI might not work properly... I would have to put a small CT on each hot/neutral leg of the shock mat and use comparators.

Since the research paper made no mention of shock hazard I can only imagine what might happen...

guitardude012

Fri Sept 26 2008, 03:24PM

Registered Member #968 Joined: Fri Aug 24 2007, 04:54PM
Location: Norman, Oklahoma
Posts: 104

Grant wrote ...

I believe that Guitardude was suggesting the concept of a ground fault interruption system designed specifically for the shock mat in order to keep the current that the insect receives in check.

A GFCI breaker like that in your lavatory was just an example and would indeed serve no real purpose for the OP's intent.

Right this would be a GFI you would build yourself. It would go between the Variac and the shock mat.

aonomus

Tue Oct 14 2008, 03:35PM

Registered Member #1497 Joined: Thu May 22 2008, 05:24AM
Location: Toronto, Ontario, Canada
Posts: 801

Ok so a bit of a necro, after looking at the cost of either buying 200H chokes to build a 90deg phase shifting circuit, I'm almost wondering if I could build a square-wave switching circuit to simply send voltage in both directions and perpendicular to each other (ie: 2 full-bridge circuits, one for each axis).

1. Is there any way to smooth a squarewave to approximate sinewave? (Maybe use a LCR network??)
2. Should I set the fullbridges to rotate the E-field, or simply generate 2 90deg phase-shifted squarewaves, smooth to sinewaves, and then input to transformers to step-up?

I have a little more faith in a solid-state setup mainly because the old paper which has protocol on this states that it uses a 1550V transformer and 200H chokes which both seem either impossible to get or expensive (vacuum tube era components I believe?)

I might have more luck trying to feed the solid-state inverter output into a pair of MOT's and just choosing the right voltage to give me approx 1500V.

Bored Chemist

Tue Oct 14 2008, 04:54PM

Registered Member #193 Joined: Fri Feb 17 2006, 07:04AM
Location: sheffield
Posts: 1022

It might be easier to generate the different phase signals at low voltage then amplify them enough to feed the primaries of a quartet of step up transformers.
I'm sure it's within the capability of a tatty computer and a couple of sound cards but I'd have no idea how to implement it that way.

1 2 next

Moderator(s): Chris Russell, Noelle, Alex, Tesladownunder, Dave Marshall, Dave Billington, Bjørn, Steve Conner, Wolfram, Kizmo, Mads Barnkob