Welcome to Laser Pointer Forums - discuss green laser pointers, blue laser pointers, and all types of lasers

LPF Donation via Stripe | LPF Donation - Other Methods

Links below open in new window

ArcticMyst Security by Avery

Burning Mosfets, could use advice.

Joined
Aug 13, 2013
Messages
14
Points
0
So I have an S06J 405nm diode and driver from DTR. Works great. I'm trying to hook it up to my CNC machine for engraving. Here's the circuit I'm using.
ieiacQn.jpg

I have both this switch circuit and the diode running off of a 5v wall wart power supply with plenty of power.
When I test this on the bench by jumping the 5v power supply to the Mosfet gate it works great. No problems.
When I install it into my machine and connect the gate to a 5v led output that is tied to the Z axis direction pin it will turn on and then not turn off once the gate is disconnected. The mosfet does not get hot. Basically I've fried the mosfet into a closed state between the drain and the source.

Any suggestions?

Mark
 





Joined
Jul 27, 2007
Messages
3,642
Points
63
You should check to make sure you don't have any strange grounding issues first, but other things to look out for is you might want to put a resistor inline with the mosfet gate along with say a 7.5v zener diode to prevent any voltage spikes from toasting the delicate oxide layer on the gate region.

Another option that would eliminate voltage transients is to use an optocoupler between your CNC system output and the gate on the mosfet to switch your laser on and off. I would still recommend the gate resistor and zener since it would prevent problems with static discharges and other mishaps.
 
Joined
Sep 12, 2007
Messages
9,399
Points
113
Looks like this particular fet already has esd protection built in. Do you have a driver for the diode? What is the expected operating current? Is the fet heat sinked? Do the 5V control and the 5V power have common ground?
 
Joined
Jan 14, 2011
Messages
424
Points
0
"When I install it into my machine and connect the gate to a 5v led output that is tied to the Z axis direction pin it will turn on and then not turn off once the gate is disconnected"

Are you actually disconnecting the Gate, or are you assuming The voltage on that line has gone to zero when the LED goes out? Check the voltage on the Gate. Maybe the line drops below Vf of the LED, but not low enough to turn the FET off?
 
Joined
Aug 13, 2013
Messages
14
Points
0
Do you have a driver for the diode? What is the expected operating current?
Yes, it's a constant current 450ma driver.
Is the fet heat sinked? Do the 5V control and the 5V power have common ground?
No the fet was not heatsinked. The 5v control and the 5v power may not have a common ground. They are both separately powered by the same electical outlet.
 
Joined
Aug 13, 2013
Messages
14
Points
0
"When I install it into my machine and connect the gate to a 5v led output that is tied to the Z axis direction pin it will turn on and then not turn off once the gate is disconnected"

Are you actually disconnecting the Gate, or are you assuming The voltage on that line has gone to zero when the LED goes out? Check the voltage on the Gate. Maybe the line drops below Vf of the LED, but not low enough to turn the FET off?


Yes, I actually disconnected the wire from the control to the gate and the diode remains on. I also tried grounding the gate see if that would turn it off and it does not. Seemingly by that time the mosfet is "stuck" closed.

Mark
 
Joined
Aug 13, 2013
Messages
14
Points
0
You should check to make sure you don't have any strange grounding issues first, but other things to look out for is you might want to put a resistor inline with the mosfet gate along with say a 7.5v zener diode to prevent any voltage spikes from toasting the delicate oxide layer on the gate region.


I was considering adding a resistor between the control and the gate only because I've seen that done in a number of similar circuits. Can you explain what the resistor would do there?

Mark
 
Joined
Sep 12, 2007
Messages
9,399
Points
113
The resistor would limit the input current to the gate. The gate acts as a tiny capacitor, and therefore behaves as a short for a few microseconds which can be hard on the signal source.

The 5v control and the 5v power may not have a common ground.

Tie the grounds together unless there is something you are not mentioning that would make this a problem. What you expect to be 5V may be much more if the grounds are not common.
 
Joined
Aug 14, 2013
Messages
2,655
Points
63
Put an 18V zener diode between gate and source. Another thing: this is not a logic level FET and 5V is
not enough to turn it fully on. Consider either using a logic level FET such as the 3055 or a 12-18V signal
on the gate.
 
Joined
Sep 12, 2007
Messages
9,399
Points
113
Put an 18V zener diode between gate and source...

18V zener would be useless. Absolute max gate voltage is 10V according to the datasheet.

Another thing: this is not a logic level FET and 5V is not enough to turn it fully on.

Why do you say that? The gate threshold is 1-2V. You have clearly not read the datasheet and are making generalization. Please read the datasheet before commenting further so we don't confuse the poor guy.
 
Last edited:
Joined
Aug 14, 2013
Messages
2,655
Points
63
18V zener would be useless. Absolute max gate voltage is 10V according to the datasheet.

That is correct. I've never seen a MOSFET with such a low breakdown voltage. You learn
something new every day I guess.

Why do you say that? The gate threshold is 1-2V. You have clearly not read the datasheet and are making generalization. Please read the datasheet before commenting further so we don't confuse the poor guy.

Don't confuse the threshold voltage with the saturation voltage. Threshold is where the
MOSFET just begins to conduct.

My impression was that it was able to be driven fully from a 5v input. https://www.sparkfun.com/products/10213 At least thats what this implies.

Yes. You are right. It was not labeled as a "Logic Level FET" which was also throwing me
off. 5V should be fine, at least at the currents and temperatures you're running it at. So
just the zener diode should be all you need to get up and running. Something like
1N5239BTR: FAIRCHILD SEMICONDUCTOR: ICs & Semiconductors
 

Things

0
Joined
May 1, 2007
Messages
7,517
Points
0
Keep in mind that by putting a resistor on the gate, you're forming a resistive divider, so don't make the value too low. You definitely need to keep that 10K resistor to ground there to drain the gate.

It's hard to tell why it's doing that, though. Do you have any pics of your setup? Which CNC control board are you using? If you're connecting it to an LED on the Z axis dir pin, it shouldn't be 5V, more like 2.5V, a red LED would burn out at 5V (and with current limiting, the voltage is lower anyway).
 
Last edited:
Joined
Aug 13, 2013
Messages
14
Points
0
I don't know how to calculate the appropriate value for a resistor in series with the control line and gate. Any direction or help there? Also, why a zener diode and not just a regular diode? The interweb tells me that the difference is that a zener diode can conduct current in a reverse direction if the voltage difference is high enough. I'm not sure how that comes into play here.

Also, I may be making an invalid assumption about where my control line comes from. The board is proprietary to EZ-Router. Here's a pic of what I'm working with.
oqXootf.jpg


When I move the Z axis up one of these led's goes out. When I move it down it turns on. The voltage at the screw terminal that I'm assuming is connected to that led goes to 5v when the led goes on and 0v when it turns off.


Mark
 




Top