Different ways to discharge circuits?

[bonedoc]

Ok, so when using a logic level or typical NPN to conduct a heavy load (or inductive load), one typically hooks the cathode to the gate, but does the anode go to ground or the source? The reason I ask is that I see both, and in this example it is the ground.

I think I will purchase some IN4746. They are 18V and 1W. any good?

http://www.ebay.com/itm/50x-Zener-D...914?pt=LH_DefaultDomain_2&hash=item4cfa28cb0a


Thanks again!

 

[(*steve*)]

If you're switching an inductive load the diode typically goes across the load. This is because a reverse voltage is generated when power is removed. The diode serves to redirect it away from the transistor (which it would otherwise destroy).

In this case we're limiting the voltage and the diode is reverse biased because that's how zener diodes are used. They break down (non-destructively) at a certain voltage.

In this case you have a device that has a limit of +/-20V (see Vgs) between the source and the gate. So the zener id placed between the source and the gate in such a way that it will break down if the voltage exceeds something less than 20V. The series resistor limits the current and prevents the zener from being destroyed.

Using an 18 volt device is OK. It is probably a 5% part, so the breakdown voltage may be as high as 19V. But it's still less than 20V. If it was a 10% tolerance device then you'd be cutting things very fine.

Resqueline suggested a 16V part because 16 (or 17 or 18) leaves a comfortable margin, and 15V is typically what is suggested to turn these devices hard on.

The 1W rating is more than adequate, you could use a 400mW (or even less) device here.

edit: look here. (same seller) they are 15 volt parts.

 

[bonedoc]

Thanks again! That make a lot of sense. I was saying 18V because that is what his schematic showed. I will lower to 15-16V.

I guess I should not say gate to source and gate to ground, since the source it usually at the ground The schematic on here shows the cathode at gate and the ground/source at the anode.

I learned a lot! Thanks! If you ever do consultations for a fee, let me know, I am happy to help those who help me. Ian

 

[(*steve*)]

I was saying 18V because that is what his schematic showed.

And yes it does.

I misread it as 16V probably because that's what I was expecting to see.

 

[bonedoc]

Well, I am testing this as we speak. I have it exact to the above spec. One problem....works great, until I get to high voltages. At 300V, the 1/2W 60 Ohm resistor catches on fire ;-)

So....I am sure its voltage rating sucks. I may have to get a high voltage resistor, or perhaps a coupe 30Ohm resistors in series?

 

[bonedoc]

Well, I am testing this as we speak. I have it exact to the above spec. One problem....works great, until I get to high voltages. At 300V, the 1/2W 60 Ohm resistor catches on fire ;-)

So....I am sure its voltage rating sucks. I may have to get a high voltage resistor, or perhaps a coupe 30Ohm resistors in series?

 

[bonedoc]

and so does a big a$$ 1W 68 Ohm resistor!

 

[(*steve*)]

1W is tiny.

Try something lie this

If you look a the datasheet you'll find that a 25x overload is permitted for 1 sec or less.

Note that the voltage rating is insufficient. You may be better with three 20W 30 ohm resistors in series.

This should be totally safe, and really we're interested in a resistor that can carry the instantaneous current without acting like a fuse. It *should* be conservative for your circuit.

 

[GonzoEngineer]

If you can tell me the approximate resistance you need, I have a lot of different Kanthal Bulk Ceramic resistors in my junk boxes.
I am sure I have something that would work, but some are rather large.
If I have something, it is yours for free if you pay shipping.

 

[bonedoc]

Hey Gonzo, I think I located some nice to-022 resistors online for cheap. Thanks though! I appreciate it. You guys help me a lot, so anything I can do to help back, let me know. I have a laser cutter. so if anyone needs something strange, let me know ;-)

Can someone tell me what the formula is for calculating discharge times? I saw this posted here:

5 X C X R = T

Where c= charge and r = resistance. I dont know what the 5 is for though....if it is current or what.

 

[BobK]

A capacitor discharging thorugh a resistor goes to 1/e of it's original voltage in RC seconds. The 5 gives a discharge to 1/e^5 or 0.0067 times its original voltage. Throretically, the voltage never reaches zero, so the time to completely discharge a capcitor is infinite.

Bob