Need help suggestions with circuit

[jrobin9]

Hey guys. I'm trying to build a worm/fish shocker for use in my uncles pond. I found a schematic off the internet and was looking for some input. I have a basic background in electronics but it has been a while and I am rusty. This circuit is designed to produce 600 volts output, 30 cycles per sec @50% duty cycle. I have attached the schematic with notes I made to myself. The cycles need to be adjustable from 30-60 while maintaining 50% duty cycle. I was planning on replacing the R4 trimpot with a regular pot (wired as a trimpot) for ease of in the field adjustment. Any issues with that?

I would also like to bump the volts up to about 1000 with a multi-output transformer but don't know how that would impact the rest of the circuit...

Any design flaws seen or improvements you think could be made?

 

[(*steve*)]

Your description of R4 and R5 should be reviewed.

I presume the capacitors are in uF, but you have omitted the "u".

R6 has nothing to do with timing.

The symbol for Q1 is incorrect and a part number for this is required.

D3 seems to be in a position calculated to do the most damage to your mosfet should there be no load on the output of the transformer.

Can you suggest the voltage rating for C4, and the reason why R7 and the unnamed 1.8k resistor are in series?

 

[Mitchekj]

You're trying to make a flyback converter with a 555, yes?

D3, C4, R7 are a snubber, there to protect the FET when the coil collapses each cycle. (The polarity across the coil changes when the FET switches off.) I haven't checked your math on the values, but ok.

And if I'm thinking right, you're going to blow your whole project up anyway without a catch diode on the output. Though with the voltages involved on the output, you may need one hell of a high reverse voltage rated diode. (Thinking 2kV or so?) The idea being only one coil is conducting at any time. The primary charges while the FET is on, inducing it into the secondary when the FET is off. You need the catch diode to reverse bias the secondary whilst the primary is charging.

AND... those coils are going to be HUGE at 30-60Hz. Have you calculated what the primary inductance needs to be?

Anyway, I guess it could work in theory, though all the flybacks I'm familiar with have been of a step-down nature.

 

[jrobin9]

ok. I am teetering on the outermost edge of my knowledge now so forgive me if I appear ignorant.

steve
Yes the capacitors are uF.
Q1 is an IRF450
D1 yes, no load would not be good at all. That ?SHOULD? never be the case.
C4 should be 3.9uF 350V
Remember I got this schematic from another source but R7 was originally (one) 3.6k resistor. At some point in time it was changed to (two) 1.8k 3W resistors. I am not sure of the reason why.

Mitchekj, yes a flyback converter. I thought a/the snubber diode D3 and a flyback/catch diode were basically the same thing? D3 was being used as the snubber diode fr307 1kv 3 amp.

The coil suggested for use was a 24v/240vR 4 amp (with air gap cut into it). I have not calculated the primary inductance and don't know or can't remember how. I was depending on the original designer's values but my knowledge is hit or miss at this point and I tend to question everything

 

[(*steve*)]

C4 should be 3.9uF 350V
Remember I got this schematic from another source but R7 was originally (one) 3.6k resistor. At some point in time it was changed to (two) 1.8k 3W resistors. I am not sure of the reason why.

The why probably has something to do with the voltage rating of that capacitor.

Resistors have voltage ratings, and 350V probably exceeds the voltage rating for most common resistors. Therefore you use 2 to lower the voltage across them.

What is the power rating of those resistors, by the way?

 

[jrobin9]

3 Watts each. Several parts values were changed after the original design. Thats probably a good thing. It leads me to believe that the design wasn't done and dropped but that improvements were are being made.

 

[Mitchekj]

The catch diode I was referring to would be on the secondary (output) side of your transformer. (Don't think of it as a transformer, it's actually acting like two coupled inductors.) In the flyback, you don't want the primary and secondary to both be conducting at the same time. The catch diode on the output would need to be reverse biased when the FET is ON; and forward biased when the primary is not conducting, FET off. Keep in mind, secondary and primary will be 180 out of phase due to the way inductors work (polarity swaps when collapsing, with a large voltage spike at the start of the transition, all to keep current flowing)... so your positive pulses will actually be coming from the "bottom" side of the output unless you wound the transformer accordingly (the dots on each coil would be opposite eachother schematically,) that's where you'd need to put the catch diode, on the positive rail. (Anode toward the coil, cathode toward the output.)

The snubber (D3, et al) is there simply to protect the FET from the spike induced when the coil collapses. And yeah, they went to two resistors to deal with all the power being dissipated in the snubber. A transil setup may work much more efficiently, depending on the voltage levels involved there.

 

[Mitchekj]

Well, just ran some quick numbers for your coils. (No accuracy guaranteed.)

You'd need about 18mH of primary inductance, at ~18A peak. Turns ratio of 44.6. Something like 18 to 16 gauge wire on the primary? You can use much smaller gauge wire on the secondary. Not to mention the core, which to avoid saturation at that kind of current would be large in its own right. The insulation between windings will need to be formidable, as well.

That means one LARGE, expensive, very custom transformer. Winding it yourself may prove... tricky. I have to admit I'm not up to snuff on the magnetics design required here, skin depth, etc... so you will need to consult someone who is. The bad thing about SMPS design, when something's wrong with the magnetics, it usually destroys the whole shebang. Do not pass go, re-build the whole thing. Trial and error will prove to be frustrating.

I understand the idea of such low frequencies being used based on the heartbeat frequency, yeah? The switching frequency needs to go way up in order to pull the size (to include cost, complexity) of the magnetics down.

On the plus side, it sounds like it may be doable at least. Be prepared to spend some cash, and heartache, though.

(As an aside, that comes out to something like 600ft. of copper wire, just for the primary.)

 

[55pilot]

Hey guys. I'm trying to build a worm/fish shocker for use in my uncles pond.

I do not know about worm and fish, but it will make a very effective people killer. What are your plans to make sure you do not kill yourself, a neighborhood kid or even a hobo that decides to go dipping in the pond? The idea that it is locked, on private property.. etc will not help you. I do not know where you are located, but in the USA the doctrine of "attractive nuisance" makes you liable if someone climbs a fence, breaks a lock, ignores a dozen warning signs and does something really stupid and gets hurt.

That is why, my advice is DO NOT DO IT.

---55p

 

[55pilot]

I understand the idea of such low frequencies being used based on the heartbeat frequency, yeah?

60Hz inducing arrhythmia is very particular to the size/shape of the human heart.

Lower frequencies do penetrate deeper into the body and transmit better in water so they are more lethal than very high frequencies.

---55p

 

[55pilot]

Several parts values were changed after the original design. Thats probably a good thing. It leads me to believe that the design wasn't done and dropped but that improvements were are being made.

Or that the original design did not work and people who know even less have tinkered with it subsequently. Either could be true.

---55p

 

[jrobin9]

Something doesn't add up with the transformer...I may be confusing myself and leading you guys in the wrong direction. Traditionally these type "shockers" put out 100-600 pk pulse volts at around 1/2 amp. The power disipation (in the water, 500 ohm resistance give or take) is normally around 10-50 watts which gives you a shock effect of 10-20 feet.

Let me go back to the basic circuit. In the original design, R5 is set to fcw value 10k. R4 is adjusted to allow for a max input current of 3A (and not touched again). Whatever the resistance value of R4 is set to in order to lock in that 3Amp input value, it effectively locks in the cycles.
Based on the fact the circuit was designed for 30 cycles, lets assume that this puts R4s value locked in at 10k ohm which gives me 30 cycles @ 50% duty cycle. R5 is adjustable externally and would effectively lower the duty cycle from 50% while maintaining the same cycles (since it starts out at the maximum resistance value). So max current on primary is 3A with max duty cycle of 50% @ 30 cycles.

Mitchekj, How does the "air gap" in the transformer effect the inductance of the primary coil? I know it reduces it but don't know what distance of the gap you used in your calculations. I actually dug around the net and found the assembly instructions for this circuit. It's 10 or so pdf pages but does give me much more detail including a wave shape drain on Q1 w 2-3 amps and 500 ohm load. It also stated that the output with 500 ohms is over 300 volts and over 1/2 amp. Here is the circuit description from the instructions.

"The circuit utilizes inductive charging similar to that used in the ignition systems of the older automobiles. The primary of transformer (TI) current charges through mosfet switch (Q1) The current ramps up to a value determined by i=Et/L where E equals the 12vdc input, The inductance of T1 primary and “t” the on time of switch Q1. The pulse energy is now equal to Li2/2 being controlled by the on time as determined by pot R5.
Transformer T1 requires cutting in an air gap necessary to store the inductive energy as the core itself would saturate making T1 useless. The cutting rework done to T1 is shown on fig 3..Timer (I1) is wired up as astable pulse generator with a fixed frequency determined by the total value of pot R5 and trimpot (R4) along with timing capacitor (C2). Trimpot R4 is used to set the maximum core charging time range of R5. Resistor
(R6) and capacitor (C5) decouple the timer I1 Vc from the main 12vdc. LED (D4) along with current limit resistor indicate when power switch (S1) in on."

Safety is a concern but isn't much of one. If this works as intended, I'm not worried about a neighborhood kids finding a 12v battery, breaking into my garage, knowing how to hook it up and saying to his friend "hey you hold these while I flip the switch". As I said earlier the effective range should only be 10-20 feet in the water. "attractive nuisance" applies to everything. Should I not have electricity at my house either? These units only stun the fish, if you remove the power they will swim away after 10-30 seconds. This is the same type thing that the DNR uses to do fish studies in lakes only this is on a much smaller scale.