The chips came in. I ordered the IRL540 and this is what I got:



I didn't see anything online for P30 N06LE. (I messed this up somehow because I see a helpful spec sheet when I look right now) After a few minutes of impatience, I swapped one of the new chips in place of one of the old chips, pin for pin. When I dropped the signal from the Arduino, I expected and hoped that the tube of magnets would jump to the activated coil. Instead, the LED lit for a half second, then the chip exploded in my face with a giant spark. I barely caught myself before falling off the stool.


I'm off to try to decipher this new spec sheet: https://www.sparkfun.com/datasheets/Components/General/RFP30N06LE.pdf

I see a diode in the circuit. Anybody able to explain in layman's terms what the diode does and how it could be involved in explosion prevention?

 

The chips came in. I ordered the IRL540 and this is what I got:

View attachment 152300

I didn't see anything online for P30 N06LE. (I messed this up somehow because I see a helpful spec sheet when I look right now) After a few minutes of impatience, I swapped one of the new chips in place of one of the old chips, pin for pin. When I dropped the signal from the Arduino, I expected and hoped that the tube of magnets would jump to the activated coil. Instead, the LED lit for a half second, then the chip exploded in my face with a giant spark. I barely caught myself before falling off the stool.

View attachment 152299
I'm off to try to decipher this new spec sheet: https://www.sparkfun.com/datasheets/Components/General/RFP30N06LE.pdf

I see a diode in the circuit. Anybody able to explain in layman's terms what the diode does and how it could be involved in explosion prevention?

OUCH! ...

 

Don't worry about the diode. It is a parasitic element though it is helpful in conducting reverse current through the device.

Apparently those plastic breadboards can handle a lot more current than I thought!

 

Apparently those plastic breadboards can handle a lot more current than I thought!

I thought mosfet power type circuits were not to be built on those type of boards. due to all of the "parasitic's" involved with them.

 

When I dropped the signal from the Arduino,

I forget, are you using a gate driver? If you're driving those directly for the Arduino you bought the wrong type of mosfets. To drive them directly from an Arduino, you need "logic level" mosfets.

 

Ben Varvil said: ↑
When I dropped the signal from the Arduino,

If that means you disconnected gate drive and let the gate float, then that cold be reason for the problem. The gate has a very high input resistance and it also gets leakage current from the drain, so it has a tendency to turn itself on and to do so slowly rather than switching quickly. When turning on of off slowly the transistor passes through its analog range during which its power dissipation can become very large.

 

If that means you disconnected gate drive and let the gate float,

That's the reason to use both a driver to gate resistor( to stop ringing) and a gate to source resistor(to keep the mosfet off when it's supposed to be off). Many online circuits now days don't show a gate to source resistor.

 

Don't worry about the diode. It is a parasitic element though it is helpful in conducting reverse current through the device.

Apparently those plastic breadboards can handle a lot more current than I thought!

Hmmm, so me getting diodes, like in this circuit, won't help reduce the explosion part of my problem... damn.

Photo Credit goes to this instructables page: http://www.instructables.com/id/Controlling-solenoids-with-arduino/

 

I thought mosfet power type circuits were not to be built on those type of boards. due to all of the "parasitic's" involved with them.

By those "types of boards", are you talking about the black case the chip is enclosed in, or are you talking about the breadboard I assembled the test circuit on? I never intended to use a breadboard, but my 2 hours of time turned to about 7 minutes... and I was just doing a quick and dirty test to see if something would move. I guess exploding counts as moving, but...

To me, a parasite is something that eats off of a host animal in a destructive way... are you speaking allegorically or is parasite also a circuit term. Either way, I don't follow...

 

I forget, are you using a gate driver? If you're driving those directly for the Arduino you bought the wrong type of mosfets. To drive them directly from an Arduino, you need "logic level" mosfets.

The article I linked earlier said the same thing and suggested the IRL540 chips as logic level mosfets. That's what I ordered. Again, I'm not sure what I got.

 

Ben Varvil said: ↑
When I dropped the signal from the Arduino,

If that means you disconnected gate drive and let the gate float, then that cold be reason for the problem. The gate has a very high input resistance and it also gets leakage current from the drain, so it has a tendency to turn itself on and to do so slowly rather than switching quickly. When turning on of off slowly the transistor passes through its analog range during which its power dissipation can become very large.

Me "dropping the signal from the arduino" means that I sent an output pin to high. The output was connected through a 220ohm, then through an LED. I'm still processing the rest of what you're saying. There is stuff in there I need to understand. So far, I'm getting that when on, or off, the mosfet has zero heat. But, in the transition between those two states (I'm assuming that's what you mean by "analog range") it creates heat. So, if I weren't clearly giving an on or off, i'm cooking the chip... the enclosed chip, which can't expand, so it explodes. I think.

 

That's the reason to use both a driver to gate resistor( to stop ringing) and a gate to source resistor(to keep the mosfet off when it's supposed to be off). Many online circuits now days don't show a gate to source resistor.

No, I don't see resistors between gate and source when I look online. Interesting.

I've always thought that if I have a ton of current flowing and then suddenly stop it, there must be a pile up of current at the stopping point. A train whose cars pile up after the locomotive hits some object, to use a bad analogy. Is that what is happening? Is that the issue your source to gate resistor is addressing?

To speculate further, that resistor doesn't let a bunch of current through, just enough to dampen the massive stopping of current... it gives the train cars a way to dissipate their energy... to stay on the tracks...

Am I in the ballpark?

 

Hi,

To me, a parasite is something that eats off of a host animal in a destructive way... are you speaking allegorically or is parasite also a circuit term. Either way, I don't follow...

It is an 'analogy', but a real thing in circuits. Usually applies to a couple of things, mainly capacitance and inductance, but often include resistance. Like many, many things Google is your friend in understanding electronics. Just Google "circuit parasitic's" will get you better understanding than I can give you.

And the board I was talking about was the bread board, they are well known for parasitic problems. And shouldn't be used for either high voltage or high amperage.

The article I linked earlier said the same thing and suggested the IRL540 chips as logic level mosfets.

After re looking at the Mosfet(which you keep calling a chip, which it's not. A chip usually means a IC or Integrated Circuit, where a mosfet is a type of transistor or component. I know semantics again. ) The data sheet for it is kind of misleading. It calls for the gate to source voltage to be +/-10V but calls it both, 'logic level drive compatible' in the description. And gives the 'continuous drain current @ 5V. So I guess it's a logic level after all. I usualy just look at the VGS to decide. Sorry, my mistake.

To speculate further, that resistor doesn't let a bunch of current through, just enough to dampen the massive stopping of current... it gives the train cars a way to dissipate their energy... to stay on the tracks...
Am I in the ballpark?

Not really. While it may do that, it is a very small amount of the turn off process. What the gate to source resistor does though is gives a high resistance path between the gate and source of the mosfet when it is supposed to be off. This keeps the mosfet from getting turned on by (watch for it) parasitic voltages present in the circuit. The gate in the mosfet is a capacitor in reality, and can/will collect small amounts of voltage from the surrounding things, to the point it will turn on all by itself. And that's not something you want to happen.

The actual turning off of the mosfet takes place in the "drive circuit" either from a driver or in your case a micro. The gate to source resistor is a high value. I use 10K but have seen as little a 2.2K. The small price of that resistor can mean the difference of a circuit working, or even save a higher priced item in a circuit.

I again hope you don't think I'm talking down to you, because that's not my intention. I started learning this stuff in my middle 50's and no one would explain it in easy terms. And many here would say I'm not explaining it correctly. But it is what I have found out on my own and I am trying to do it in a practical not technical way, to try and help you.

 

Your exploding FET could perhaps be due to the way your breadboard is constructed and being used. Breadboard sockets aren't designed to accept fat conductors. Pushing the FET pins in broadside-on might bend the socket contacts out far enough for the contacts in adjacent rows to touch each other. If rows 6 and 7 were shorted the FET might turn on partly.

 

Your exploding FET could perhaps be due to the way your breadboard is constructed and being used. Breadboard sockets aren't designed to accept fat conductors. Pushing the FET pins in broadside-on might bend the socket contacts out far enough for the contacts in adjacent rows to touch each other. If rows 6 and 7 were shorted the FET might turn on partly.

Thanks. Good to know. And the "partly" area is when FETs heat up, if I'm comprehending my reading correctly.

Anyway, the 20 FETs that I ordered from China arrived today. They had the correct part number: IRL540N (Amazon's did not). As I only had a few minutes to put towards the project, I swapped our the exploded FET from earlier with one of the new one just to see what would happen.

Interesting that the slow mo shows two explosions. Wonder what that's about...
ShortBus, I'm still processing your reply. Thank you.

 

If I can make a suggestion -get rid of that plastic breadboard and solder the MOSFET into the circuit. That will eliminate one likely cause of your problems.

Edit: Changed "LED" to "MOSFET"

 

The data sheet for it is kind of misleading. It calls for the gate to source voltage to be +/-10V but calls it both, 'logic level drive compatible' in the description. And gives the 'continuous drain current @ 5V.

I found this spec sheet: https://www.vishay.com/docs/91300/91300.pdf
It says at the top the VGS is 5V and that the absolute max is +/- 10V.
If the plastic breadboard did connect the gate to the source, I'm guessing that would put 12V on the VGS... leading to the explosion.

The actual turning off of the mosfet takes place in the "drive circuit" either from a driver or in your case a micro. The gate to source resistor is a high value. I use 10K but have seen as little a 2.2K. The small price of that resistor can mean the difference of a circuit working, or even save a higher priced item in a circuit.

Got it. This helps a lot. Quick clarification when you say "price" are you talking the cents for the actual resistor, or are you talking about loss in performance somehow?

I again hope you don't think I'm talking down to you

No, no, no... I'm not too proud to listen even if you were. Your wording is helpful.

After my reading and processing, I feel confident with this next try. Will post. Thanks again.

 

Hi Ben,

Quick clarification when you say "price" are you talking the cents for the actual resistor, or are you talking about loss in performance somehow?

Yeah, the cents for the resistor verses the price of everything else. I am learning this stuff the hard way like you. And with the circuits out there on the internet(Youtube and instructables especially) that don't show even the basics of how all of the Application Notes right from the manufactures say how to do things, to me is bad. The circuit may work long enough to make a very short video, but will it work over time? Or work at all when some one else builds it? And for some people they won't even try to figure out why it doesn't work and give up on electronics. The companies that make the components do a lot of very expensive research on the parts, so why not use the things they found out?

Your wording is helpful.

I try to put it into words that even I could understand. There are a lot of very knowledgeable people on this site, and I've learned a lot from them. But many of them have done this type of thing for so long that they forget what it's like not to know what their talking about. For some one like us that are just starting out, a more practical, grassroots discussion is better. So that has been what I've tried to do with people, both to help them and to try and give back to the forum for the help I've gotten.

 

I felt confident this one was going to work. Then it exploded.
I got rid of the plastic breadboard. I added a 10K resistor between the gate and the source.
Using a mosfet is electronics 101, but I'm racking up an embarrassing number of failures here.
Anybody see what I'm doing wrong?

 

I felt confident this one was going to work. Then it exploded.
I got rid of the plastic breadboard. I added a 10K resistor between the gate and the source.
Using a mosfet is electronics 101, but I'm racking up an embarrassing number of failures here.
Anybody see what I'm doing wrong?

View attachment 152689 View attachment 152690

I suggest you carefully redraw a diagram of what you've just done and post it here. Draw your diagram according to the way you've actually wired things up, and not the way you thought you did it. Translation: work your way backwards.

Then we might be able to help.