Warning: This device is not Underwriters Laboratory approved and, beyond that, it is just downright dangerous. Don't even think about building one—unless you live in a third world country.

The Problem

Some neighborhood mice have decided to upgrade their lifestyle at the expense of the humans who live here. That is, they have discovered that bread crumbs, bits of cheese, peanuts and even some fruit in the kitchen of a house are more desirable that their previous diet.

When Carlos the Wonder Dog was still with us, the mice discovered it was too dangerous to go food browsing in the kitchen. While he was far from a violent dog, he was very inquisitive and when something came running out from under a counter or other location, he would usually drop his built for snow-sized paw to slow it down. More times than not, at least part of the running animal would end up under that pow resulting in at least the loss of limb and many times loss of life.

We had decided a cat was an alternative. Unfortunately, unlike Carlos who seldom stole food from the kitchen counter (well, one freshly baked loaf of bread once), the cat thought it was her private, dog-free, play area. So, she moved to a new home and we went to the more traditional methods of control.

Rat poison was first. Unfortunately, it seemed to lose its value a day after opening the package. Yes, it eliminated a few mice but we wanted an on-going, low-maintenance solution.

The next idea was a mouse trap. We found some that were from China. Serious jaws that could probably trap and possibly kill a fair-sized dog. I first baited one with cheese. The mouse managed to remove the cheese without tripping the trap. My next idea was peanut butter which would stick to the trip mechanism better. Same thing. Yes, it might kill a cat but a rat was just too light, smart or careful to trip it.

One other type of trap is a cage with door. The idea is the mouse enters for the food and trips the door. The first problem is, of course, will the trip mechanism actually work with our smart mice? Beyond that, we then have a live mouse in a cage to deal with. It is a long walk to the nearest neighbor I don't like so we scratched that off as impractical.

The Better Mousetrap

The old story is that success comes from inventing a better mousetrap. But, I am in rural Nicaragua which presents two problems:

  1. Most people don't have this problem because, unlike us, they don't have more food in their house then they can eat today.
  2. Many of the materials an inventor would want are just not easy to come by.

My goal was to come up with not necessarily a better mousetrap but a mousetrap more appropriate for where I live and the available materials. Ultimately, having it be inexpensive to make is a big plus. The version you see here is a prototype. That is, its manufacturing cost is much higher than what it would be when built to local standards and put into production. But, I wanted the flexibility to make changes so I raided my junk box as well as the trash can to produce it.

My Functional Specification calls for a low-cost space where a mouse can enter, discover some tasty food, start eating and get electrocuted. It should consume virtually no energy except when performing electrocutions and be made from locally available materials.

My Design Specification goes on to detail some of the pieces to be used. They are:

  • Mounting base such that the whole unit is stable when placed on the floor of a house.
  • A plastic tunnel which the mouse will enter for his last supper. This tunnel will insulate the area and protect adjacent areas from being contaminated with burned mouse pieces.
  • Aluminum contact plate that the mouse will stand on.
  • A feeding dish of appropriate size for a mouse which will also be electrically conductive.
  • 120VAC operation with DC used for electrocution.

Here is the materials list. I have put in parenthesis what the alternative would be for a Nicaraguan production model.

  • A scrap of wood about 10cm by 20cm. (This is actually optional. A rock could be used to keep a completed unit in place.
  • An old coke can or, more likely, beer can. (It needs to be cut into a flat square which you could do with a pocket knife and the plastic coating removed from the inside which can be done with a dull machete. Personally, I used sand paper.)
  • A large plastic coke bottle.
  • An AC cord with plug. (Just wire direct would be an alternative.)
  • A 1.5K, 10W resistor. (The alternative is to not use one. The only real downside it that you might burn your house down or electrocute a horse or cow.)
  • A 400PIV, 1A silicon diode. (The alternative is not to use one. I prefer DC to AC because, unlike what Edison said, it is much more dangerous than AC because of its holding power but AC will probably work.)
  • A 100uf or larger, 200V capacitor. (Optional: no DC, no capacitor needed.)
  • Some way to connect all this. I used soldered terminal lugs, screw terminals and such. (For the minimalist system, nothing is needed other than twisting wires to the contacts—coke can pieces—as you have eliminated all the other components.)
  • Assorted screws. (One needs to be insulated. I expect a plastic bag can be adapted to be an alternative. The rest can be replaced with nails.)
  • Contact cement. (Well, however you will make the aluminum coke can stay in place inside the plastic coke bottle. Be creative.
  • A mouse or rat. (Not optional if you want to know if this actually works.)


I could take a lot of photos of the construction process but I suggest you go with the concept and be creative. The right approach will depend on what local materials you have.

I cut the top off of a 2 liter plastic coke bottle. I further cut a slot on one side about 2/3 of the distance down it. This is to 1) make assembly and cleaning easier and 2) prevent the poor mouse from getting claustrophobia. On the side opposite the slot I cut holes for one screw to mount the unit to be base as far as possible away from the bottle bottom and another about 2.5cm from the bottle bottom for mounting the feeding bowl. I also cut a small hole in the bottle bottom to run the wires from the contacts to the electronics section

A single 354ml aluminum beverage can is used to form both the contact plate and the feeding dish. Using a utility knife (or whatever), cut the top and bottom off the can and make a cut such that the side of the can can be rolled flat. The can bottom becomes the feeding disk, the side for the contact place. The top part is discarded. Don't forget to clean off the plastic coating on the contact plate and feed plate.

We are now ready for the mechanical assembly. Note that you will want a bit of space on the other side of the bottle bottom for the electrical parts. Three to five cm will be sufficient.

If you have a nylon screw to attach the food dish to the base, great. If not, be creative. The goal is to connect one wire to the bottom side of the feed dish and run it outside the bottle while fixing the dish and bottle to the base. You also need to put another screw through the other end of the bottle into the base. As this should be beyond the aluminum contact, there is no need for anything insulated.

At this point, you should have the bottle mounted horizontally on the base with the food dish inside. You now want to glue the can inside the bottle. Give it a test fit and, if necessary, trim it so that it does not contact the feed plate or the mounting screw and is smaller than the plastic so you have no exposed edges. (This is still far from safe but that safety step costs nothing.)

You now need to connect a second wire to the contact plate and run it through the hole in the bottom of the bottle to connect it to the electronics. I used a 6-32 screw and terminal lug. This leaves an exposed screw head on the outside of the bottle. While if wired right, this should be on the neutral side of the AC line, it is still worth covering with RTV or tape.

This completes the mechanical construction. We now need to do the electronics. While the minimum would be to just connect the contact plate and feed plate to the AC line (making sure the contact plate goes to the neutral side, that is just a bit too scary for me and also would likely be less effective. It should now be time for a schematic. I have, however, decided to just describe the circuit. If it makes no sense, you probably shouldn't be building it.

The circuit is a simple half-wave rectifier. The diode is a garden-variety silicon with a PIV of at least 400 volts and current capacity of 1 amp. It should cost $.05. The filter capacitor is 100uf, more or less, rated at 200 volts or more. I used a 1.6K, 10W resistor in series with the diode. You can pick what you like.

The resistor limits the maximum current in case you have a short circuit mouse and reduces the fire danger. While it might prevent a dog, cat or person from getting fried, don't bet on it.

With no load (no mouse) the capacitor will charge up to about 170 volts (120 volts times the square root of 2). With the 1.6k resistor in the circuit a short circuit mouse will result in about 4.5 watts of dissipation in the resistor (one-half of 120V RMS squared divided by 1500 ohms). The one-half is in there because it is a half-wave rectifier reducing the effective power to half. I picked a 10 watt resistor for two reasons. First, it is good design practice to have a resistor of twice the power dissipation capacity as is actually needed. Second, if one built the circuit without the rectifier, the power dissipation would be twice as much.

I wired the unit such that the positive voltage will be delivered to the mouse's nose. I know neither whether this is his preference nor is it more effective. The only thing important in the circuit is to make sure that the diode direction is correct for the capacitor direction. That is, the cathode (end with the stripe) of the diode must be connected to the positive (+) side of the capacitor. Not doing this will result in an explosion (of the capacitor).

You may have noticed I used a three-wire cord but did nothing with the ground wire. This is because, assuming that the outlet being used is wired correctly (a big assumption where I live but the electrical wiring in my house was done by me), this will make sure the big contact will be connected to the neutral.

Putting the Unit into Operation

Warning: There is no bleeder resistor in the circuit. This means that even after the unit is unplugged the voltage difference between the contacts could remain close to 170 volts for hours and maybe days. Use something metal with an insulated handle to discharge the capacitor before touching the unit. Or, as an alternative, add a bleeder resistor. A 100,000 ohm, 1/2 watt resistor would up the idle state power consumption to almost 0.3 watts but could save your life.

I recommend you give it a test to make sure the right things are grounded as well as see if there is voltage on the plates. If you have a voltmeter, great. If not, a neon test lamp will suffice. What you want to know is:

  • There is voltage between the two contacts.
  • The big contact (can side) is connected to the neutral side of the line.

If the first test isn't obvious, you have no business making this unit. The second requires more creativity. If you are using a neon test lamp, this is relatively easy. Hold one wire of the test lamp in your fingers. Touch the other one to the feed plate and then the contact plate. If it lights when you touch it to the feed dish and not the contact plate, all is well. If it is the other way around, you need to reverse the wires. (Don't forget to disconnect the unit and discharge the capacitor before making this change.)

You should now be ready for your first customer. Just put a bit of food on the feed plate, put the unit where it has easy access for mice and impossible access for kids, dogs and such and plug it in. A zapping noise and the smell of burning flesh will indicate a success.

On the food, it is best if it is in small granules and a bit wet. After all, you don't want to make it difficult for the mouse to eat. And, more important, this will increase the conductivity and ensure that their nose will touch the conductive plate.

Results and on to Version 2

Executive Summary: It didn't work. At about 2AM I heard a mouse enter the unit. There was definitely a zap and the mouse exited. He lived but probably will not want to try the unit again.

I can only speculate on why it didn't work at this time. The most likely situation is that the pads on his (or her) feet are not as conductive as I had hoped. A second possibility is that the food was not wet enough/conductive enough. This has brought about a design modification which I will call Version 2.

It is a simple change. I simply converted the power supply to a voltage doubler. That means one more diode and one more capacitor. The output is now about 330 volts. While I probably should have changed the value of the resistor, I didn't.

I also changed the food composition. This time it is crumbled white bread moistened with salty water. We will wait until tonight to see how this works. It may not get tried unless we get a new mouse as the old one is probably not very excited to give the unit a second try. This is not a serious issue as there are many potential users here.

Making a U.S. Version

If this unit works well, you might be interested in making a U.S. model. That is, one that can legally be sold in the U.S. While I don't have all the details, here are some of the considerations.

  • Warning labels will, of course, be required
  • The power components will need to be in a covered box with, of course, more warning labels.
  • An isolation transformer will be needed as things connected directly to the power line are dangerous. An alternative would be to use a wall wart and DC-DC con verter.
  • It is likely a Ground Fault Interrupter (GFI) would be required.
  • All materials will need to be fire-safe.
  • Materials will need to be non-toxic and probably food-grade as this is a type of feeder.
  • UL testing will be required. (I believe the minimum cost for this is $25,000.)
  • As the device is dangerous for animals, particularly mice, I expect SPCA approval or possibly some other organization will be required.

There are probably many more requirements but should help get you started. That is, if you are crazy enough to think a US-approved version is anything other than a serious joke.


Research results

As the problem has not gone away, I have been doing a lot of research. Yes, there are electric and electronic mouse traps on the market. Reading about them and reading patents has given me new ideas. Here is a bit of data.

A mouse being stuned but not killed seems to be a common problem. I thought this happened with my first day test. This can be addressed.

  • As the nose or mouth contact means the mouse only needs to move its head, it can quickly back off from my trap. The typical electric or electronic trap uses two floor contacts so the current will flow between front and rear paws.
  • It needs to be difficult for the mouse to back up or turn around. This is not the case in my relatively wide bottle trap with no slope or baffles.
  • One patent application showed kill statistics for traps with no, two and three baffles. The three baffle trap was significantly better in their claim. An alternative to baffles was a floor that was initially up hill but would pivot down once the mouse was half way in.
  • It would seem that contact time, not voltage is the issue.

Let me say that if my neighborhood store had electric traps, I would buy one. They are relatively inexpensive ($30 to $50), small, safe and battery operated (30 dead mice per four AA batteries). But, that is not an option. So, back to Nicaraguan construction.

I am thinking the trap chamber could better be built using a 500ml plastic bottle. The bottom (instead of the top) would be cut off, Two strips of aluminum can would be used to create the floor contacts. The wires would be attached on the inside so there would be no exposed high voltage outside the bottle. The bait would be attached to the cap so no open slot for bait access would be needed.

I am not sure about the baffles yet but it would seem that they could be constructed using more bottle pieces glued inside the trap bottle. Adding them, however, would make it harder to remove the dead mouse. (Commercial traps seems to be plastic boxes with an opening lid for mouse disposal. That would be nice but I hate the idea of actually having to buy something.

With the changes for better/longer contact, the idea for some sort of zap limit becomes more important. If I elect to go to a battery-powered trap (or using the 12 or 24VDC we have here) I will probably add timer circuitry to shut off the trap after some sustained zap time. As one trap claims kill in less than five seconds, another in less than 10, a 15 to 30 second timer makes sense.

Day 2 Report

At around 2AM I woke up to the sound of something in the kitchen. Listening carefully, it was clearly our pesky mouse touching the plastic edges of the MataRata Version 2. I waited. Eventually I heard a zap noise and then nothing else. I went back to sleep being confident I would find our friend inside. Well, no such case. At 4AM I got up and the MataRata was empty. I am both disappointed and confused. If I stood on the aluminum plate and touched the food plate with my tongue, I would be no longer. How does a mouse survive? In any case, it is clearly time for a serious redesign. More when I have a clue.