We are electrically inducing heart attacks in fruit flies to test the effects of various treatments on survival. We are shooting for roughly a 50% survival recovery rate in our control group.

I work with a lab that is studying cardiac function in insects. We have been inducing cardiac arrest using electrodes applied to the outside of the fruit fly. This has a lot of problems to it, attachment of the electrodes is labor intensive, the carapace is a pretty good electrical insulator , Etc

I'm new to the lab and was asked to look over the situation to see if I had any ideas on how they could improve fly handling and throughput. I got the idea of maybe putting the Flies inside a solenoid and inducing Eddy currents to shock them.

I have very limited experience in building electronic devices though I have a decent undergraduate level understanding of physics including electromagnetism.

I don't really know where to start here. I thought of just looking for solenoids online or maybe just an automotive coil. Putting the fruit fly inside a little plastic tube. Sticking it in the core of the coil after the coil is already running current from a car battery. And then just suddenly disconnecting the terminals of the battery and letting the magnetic-field collapse to see what happens to the fruit fly.

I would consider it a victory at this point if I could just show that we could get the fruit fly to die without actually cooking it. We can systematically lower the the magnetic field intensity that we collapse using lower voltages through the coil until we get what our Target is, a shock level that creates about a 50% survival rate.

I'd appreciate any advice on how to approach this or even criticism regarding whether or not it would work. I'm in electronics hobbyist but haven't really done that much period just some basic stuff.

This is an example of the physical principle I am talking about, though out problem is simpler since we want a single disruptive shock and have no portability issues.


  • 2
    \$\begingroup\$ Not sure about induction-coupling to fruit flies, but if you could generate a pulse of RF in a frequency range where the average body-length of a fruit fly was near 1/2 wavelength, you could zap them with RF induced voltage. I've personally been toying with designs (haven't built any prototypes yet...still working out the tracking part) for shooting mosquitoes out of the air using a parabolic-reflector RF setup at about 18Ghz. The problem becomes generating the ultra-high frequency for such applications though...especially with a fruit fly's length being several times shorter yet. :S \$\endgroup\$ Apr 13 '17 at 18:26
  • 2
    \$\begingroup\$ Fruit flies have cardiac function? \$\endgroup\$
    – Trevor_G
    Apr 13 '17 at 18:36
  • 1
    \$\begingroup\$ @Trevor Technically, yes, they have a dorsal vessel that functions similarly to, and is often called, a heart. \$\endgroup\$ Apr 13 '17 at 18:40
  • 2
    \$\begingroup\$ Just show them the bill for all this research.. that should do it... \$\endgroup\$
    – Trevor_G
    Apr 13 '17 at 18:47
  • 1
    \$\begingroup\$ I believe I should call SPCFF. Edit your question and add your diagram. \$\endgroup\$ Apr 13 '17 at 22:40

Electrocardiac functions are pure chemical potential activity with no magnetic field.This is because the material is mainly an insulator which descibes all dielectrics, with some dielectric constant, effective series resistance, (ESR), dielectric constant time that of and thus requires a certain Energy level to activate without damage.

The ESR is responsible for all the heat in capacitors as well as skin burns from the ESR between the electrodes and the cardiac function.

So the objective must be to minimize the ESR which takes perhaps 1000 times (guesstimate) the energy on the chest of a human compared directly on the heart muscle in surgery. Thus attenuation is unavoidable in humans even with large paddles and high k dielectric grease. A similar problem will exist for you on insects and made much harder by the tiny surface area.

Immersion in the dielectric would help greatly but possibly drown the insect. The surge is a high amplitude ,very fast decay due to the low ESR of the interfacial ionization and ESR*Ceq. for the equiv capacitance of the discharge unit and target combined in parallel.

The desired solution is to use properly sized storage capacitance and voltage including cables at 50pf/m typ for twisted pair but use a higher voltage that will exceed the breakdown voltage (BDV) of the carapace.

The energy in any capacitance including the 300pF human finger model is E=1/2CV^2 in Joules , Farads and Volts. Thus a finger at 10kV from a nylon carpet with neoprene shoes will arc 10mm from a thin wire with 300pF or so is E=1/2 300e-12* 1e4^2= 15 mJ

You might only need 1kV with 1mJ to exceed BDV, trigger a cardiac function or failure, and not make fried protein.

Microwave pulses are not the same as a capacitance discharge, as the storage capacitance tends to be too high in order to get sufficient high BDV. Although 1kW microwave oven can generate 10kV it is too much energy.

I suspect the insect BDV is perhaps 1 to 10 kV/mm (guesstimate) like damp wood (not dry) but not mica or kapton which has a much higher kV/mm. I believe the pH level will determine most likely, the BDV/mm carapace material properties as this promotes partial discharge, an early precursor of of BDV.

So get a friction charge generator and make some small film cap or mica range of small caps xxx pF to charge upto 1kV then test on damp wood of same size as fly body. If arc works ~1mm then observé response and adjust C or V to fine tune the forces of a toothpick and electrode needle heads to apply to side to see if it tries to turn with the impulse current in 1 us. Current sense R's may be used in series with 10:1 probe across 1kV rated R with 0.1 ohm but probe ground and tip must be removed and wired directly to probe tip and barrel. Otherwise it will ring badly.

Now that's what I would do. The charge process could be done safely with some tech advisor.

Otherwise get a non ESD proof Nylon carpet and Neoprene shoes and do it the old fashioned way. hahaha. Don't forget the needle heads and suitable grease.

  • \$\begingroup\$ If I'm reading correctly, this answer describes a more efficient way of killing the fruit flies individually, the way the OP stated they're trying to avoid (having to apply electric shocks to each insect individually). According to the Question post, they've already been accomplishing this part successfully, but are trying for a more Adalf Hitler-esque approach of wiping out large quantities of them in "group showers," while still using the electrically-induced cardiac arrest killing mechanism. \$\endgroup\$ Apr 14 '17 at 0:07
  • \$\begingroup\$ you aren't reading correctly. The energy has nothing to do with voltage alone which is necessary to pass thru the carapace. The energy can be controlled by precise control of the pF storage cap and thus 1/2CV^2 and adjusted to V just above threshold for BDV while current may limited with a Series R. \$\endgroup\$ Apr 14 '17 at 0:15
  • \$\begingroup\$ Roughly correct. We would still probably be shocking the individually, but it would be nice to not have to knock them out with anesthetic, mount them and attack probes individually. If we could just take them. Put them in a small polyethylene tube. Put the tube in a chamber, flip a switch, and then count survivors and monitor recovery time the would be 3 times as fast. \$\endgroup\$
    – user146252
    Apr 14 '17 at 0:23
  • \$\begingroup\$ Also it would not surprise me if we are physically hurting them in the mounting process and giving them burns on their high resisted carapaces when we run current through them. These are confounds. We are not look at thermal injury resistance, insulation properties, infection to damaged exoskeleton or physical toughness to handling. \$\endgroup\$
    – user146252
    Apr 14 '17 at 0:26
  • \$\begingroup\$ If you had a 1M Ohm series R connected to a 100pF cap charged to 10kV , it would be an invisble arc then limit the current to 10uA. How does this damage the subject? Then adjust C,V,R with suitable electrode area and grease to reduce mA/mm2 to make it ideal . Most of the energy thru an excessive large R would be dumped in the R and not the carapace. Now Get'er'dun.! \$\endgroup\$ Apr 14 '17 at 0:28

I'd envision something like this for the "RF generator":


simulate this circuit – Schematic created using CircuitLab

I made it with a voltage-controlled-switch in place of the spark-gap, so it would be simulable in CircuitLab for you. If you open the circuit, click [Simulation], then [Time-Domain]. Set the Start Time to 113.2n, End Time to 113.7n and Time Step to 3p, it will show you some estimation of the output waveform...lot's of unevenly spaced spikes (giving rise to tons of harmonics), followed by a damped sine wave of around 8GHz.

If you attach a feed from the illustrated antenna to a waveguide feedpoint, then any fruit flies (or likely any small insects) placed inside any portion of the qaveguide should be pretty easy to dose with a lethal amount of RF exposure, IMHO!


  • The 5MHz clock signal was chosen rather arbitrarily, other frequencies could be used, you'd just need to adjust the windings on the flyback transformer appropriately.
  • D3 and R1 are both for protection of the clock source (was simulating some rather unfriendly feedback there without them).
  • D1 and D2 turn the transformer into a "flyback" type, which seemed the most appropriate method for this kind of voltage step-up to me for this circuit.
  • \$\begingroup\$ Please comment if you feel an answer deserves a downvote. There's nothing wrong with downvoting if you disagree with an answer, but it's only constructive if you comment to give a reason why you think it's a bad answer (well, at least for answers that aren't obviously in violation of site policies). ... Otherwise, I quess we'll just have to assume the downvoter was an avid SPCFF member ;) \$\endgroup\$ Apr 16 '17 at 0:47

I doubt that you can directly create a 'heart attack' in that structure for a couple of reasons:

  1. The structure is probably too small to sustain an arrhythmia, assuming the cells are vaguely like mammalian cardiac muscle. Ventricular fibrillation depends on the heart muscle recirculating the depolarization wave. That structure is so small it is unlikely to be able to create a re-entrant wave.
  2. Tissue responds to current density. That is what causes muscle depolarization as well as tissue heating. It would be very difficult to get a high gradient in the cardiac muscle to cause depolarization without getting a high gradient everywhere and cooking the fly. That is, it will be tough to get a large difference in gradients to preserve the fly and affect the heart.

Do you know what the specific mechanism of 'heart attack' is for the current process? It would be very helpful to know if you are creating an arrhythmia or cauterizing the muscles.

That said, you might have success applying an micro-electrode very near the cells responsible for initiating contraction, if that's what fruit flies have. The return electrode needs to have a high surface area so the current density decrease rapidly with distance from the small electrode. A conductive liquid might be a good choice. In that sort of setup you could restrict the zone that is cauterized.

In a human heart the sinoatrial node is a small clump of cells that initiates every heart beat. If the fruit fly has a similar structure then a current applied very near it to cauterize those cells might stop the heart action without too much damage to other tissues. (Mammals actually have a hierarchy of pacemaker cells that will progressively takeover if the cells 'up the chain' fail. I would guess that fruit flies do not have that sort of sophistication.)

In any event, you should plan on quite a few cooked flies while you are working out the details. Good luck!

  • \$\begingroup\$ We already do it with electrical current directly apply heady to tail. It causes fibrilation that lasts for up to minutes. that we can detect Magnetic induction has already been used in human and mammal brain, muscle, nerve and cardiac stimulation. \$\endgroup\$
    – user146252
    Apr 20 '17 at 20:16

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.