So for the past week or two I've been attempting to build one of the ignition coil powered, electric fence controller circuits out there. I picked this one primarily because I have all the parts for it. And it only required 1 555 timer.

Source: https://danyk.cz/ohrad_en.html

So truth be told, my first build I wired the 555 entirely backwards and it's all my fault.. I read the datasheet wrong on this was pin 1. I was a dumb-dumb.. end up popping the 16v 2600mf cap in the process.

Nothing else appears to have been damaged.. the voltage feed was 13vdc. Anyway so last night was my second attempt, and this time (as per the 555 datasheet) I added a .01 cap to ping 5 and ground, and connected pins 4 and 8 to vcc. The power mosfet Im using is an IRF520N, instead of the 830/840. I have seen a YT video on the exact same circuit and that guy used an IRF250, looking at the specs I believe the 520 should be fine.
Also I did not have a 1M ohm resistor.. so I used a Piher PT15 trimpot set to 1M ohms. Oh and the 190 ohm resistor/red led off Pin 3 for activity.

Anyway, last night when power was applied nothing at all happened. If the caps, resistors, fet all check out then.. what are the chances I fried the 555?

Gross soldering pics:

 

Post schematic.

 

Post schematic.

Whoops! It's missing the LED off pin 3. And the first cap (470/16v) is a 6800uf/25v. The original called for a 4700uf I believe.

 

I suggest you run a simulation before implementing the circuit next time.

 

That 'Hokey' looking Earth symbol , not where you would connect to the earth stake, normally with this design.!.

 

Certainly that 0.33 mfd capacitor will smooth out your intended high voltage spike. And the earth connection would be to the battery side positive.
But I would add a resistor in series and the put that 0.33 cap between the top of the primary and the source (negative 12 volts). That will give a spike when the transistor switches on and off. And certainly with a fence charger you want a spike.

 

I designed mine for two short HV shots, every second.

 

I suggest you run a simulation before implementing the circuit next time.

So first time LTSpice user.. the Run process was broken.. took a bit to fix but. I have the entire circuit in there, still entirely unsure how things SHOULD be configured in here but following the advice of a random youtuber I used a pair of inductors with a Spice Directive covering them both. Set the inductance based on the square root of the average ignition coil turn ratio of 100:1. The Sq root ended up being 40000 and 1.

Anyway, using all the same values as I did physically.. even a pot for the 1M resistor. And for analysis I'm assuming transient and picking the various places to see the voltages. I'm not getting the hi/low functions off the 555 so it bothers me that so many people claim to have built this and it worked for them.

 

Hold up.. it appears to be working. I had a 555_any selected instead of an NE555. No idea what the difference is but I'm reading pulsed output on pin 3.

 

I designed mine for two short HV shots, every second.

This is what I'm leaning towards. I found your old posts for others trying to make this circuit but I never could find your circuit posted.

 

Mine was easy if familiar with Picmicro as it is all in the programming.
But there are a few designs out there using a 555.

 

The last batch of fence chargers I looked at were mains powered and they used a binary counter IC to produce the trigger pulse, which fired an SCR to discharge a capacitor into the transformer primary. Each one had the secondary burnt open by a nearby lightning strike. The unwind/ splice properly/rewind task was given to an assistant.

 

Yeah I'm not here trying to reignite the whole diy fence charger thing.. it's been beat to death in a lot of places. I am curious for thoughts on my spice file though. I've made some tweaks and (if Im doing it all right) when I probe pin3 I'm getting almost 3 pulses per second using a 50u cap for C4.
I did not choose specific like mfg or part numbers for the components but the available specs should be accurate.