Continue to Site

Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

PWM MOSFET Solenoid Drive. Flyback Diode gets HOT to the touch. Why is this?

Status
Not open for further replies.

777funk

Newbie level 6
Joined
May 21, 2012
Messages
12
Helped
0
Reputation
0
Reaction score
0
Trophy points
1,281
Activity points
1,392
I have PWM running a solenoid at approx 3.5V from 10-12V DC as a reference. I have a SMT MOSFET (IRLML6246TRPBF) rated at 20V and 4.1A driving the solenoid (measures 6 ohms).

Here's something I know nothing about and have two questions.

1. If I don't have the flyback diode in circuit (parallel with the solenoid) I get a voltage but as soon as I connect the solenoid load it drops to almost Zero. If I have the diode in circuit it works perfectly and holds proper voltage. Not sure why this is?

2. When I touch the diode after a set time it feels HOT to the touch. I'm guessing around 130F. Why would this be? I'd think at pulsed 3.5V and with a 6ohm load I'd have around 0.5Amps. The diode is a 1N4002 which is rated at 1A. Why is it getting hot? Am I missing something or does the pulsed voltage change the calculations?

thanks for any ideas here.
 

Hi
you should use flyback diode of course , and another thing : how much is the frequency of operation ? has your GC capacitor enough current to drive ?
Best Wishes
Goldsmith
 

Can you post the schematic please, it could be for a number of reasons.

1. why is PWM being used to drive this solenoid and at what frequency is the PWM?
2. The rating of a component is a measure of it's maximum current/voltage/dissipation not an indication of actual power loss within it. A 1N4002 will produce approximately (0.7 * I) Watts of heat so even at relatively low current it's small body will get quite hot. If your PWM is at a frequency higher than a few hundred Hz you will also get significant additional dissipation from the diode's reaction time, a Shottky diode would give better results.

Brian.
 

Schematic Posted. This was my first use of Eagle so I'm not sure everything is "correct" but that should be the idea.

Can you post the schematic please, it could be for a number of reasons.

1. why is PWM being used to drive this solenoid and at what frequency is the PWM?

I'm using PWM because I need to control a valve varied by voltage to the solenoid. I believe the frequency is 1kHz


2. The rating of a component is a measure of it's maximum current/voltage/dissipation not an indication of actual power loss within it. A 1N4002 will produce approximately (0.7 * I) Watts of heat so even at relatively low current it's small body will get quite hot. If your PWM is at a frequency higher than a few hundred Hz you will also get significant additional dissipation from the diode's reaction time, a Shottky diode would give better results.

Brian.

So I'm guessing a Shottky would be up to speed as far as reaction time?
 

Attachments

  • PWM Driver.png
    PWM Driver.png
    3.3 KB · Views: 127

In my other/related thread I had asked if a larger PWM input to Gate Series resistor value would slow down the start current or give a softer startup for the MOSFET. I'm wanting to protect the MOSFET upon initial startups.

Also... in my question regarding the Schottky diode, how would I choose the parameters?
 

No no! don't do that ! the reasonable way is using a ferrite bead in series with gate . 1.5 k can destroy all thing ! ciss need enough current . if you can't supply it , it will work at linear region what means high value of dissipation . you understand ?
 

What goldsmith is pointing out is that in this kind of circuit you want the MOSFET to switch fully on or fully off without spending time in the transition between them. If fully off, no current flows so the the power (V*I) is V * 0 = 0, when fully on the current is high but the voltage is very low so the power is I * 0 = 0. If you slow down the rise and fall time by making the series resistor higher in value, the MOSFET spends longer in an intermediate state where both V and I are high and therefore power is lost as heat.

The diode PIV should be rated at the supply voltage plus a safety overhead, as a rule of thumb, use the highest rated diode you can in the space available. A voltage rating above what is actually there is not a problem but one with too low a voltage will potentially fail. The current can still be 1A. In your circuit the voltage across the diode reverses suddenly at switching rate (1KHz), a normal diode will not be able to start and stop conducting quickly enough and for a brief period as the polarity reverses, it will appear almost as a low value resistor and hence get hot. A Shottky diode (or 'fast recovery' diode) is engineered to minimize the time it takes to start and stop conducting so the 'hot' period is much shorter and overall heating much less.

Brian.
 
thank you goldsmith and Brian. Makes perfect sense.

I chose a logic level MOSFET so it would be in the full on Vgs region most of the time.
I'll lower the gate resistor and get a Schottky in there. The 1N4002 is working but I don't like heat. Do I need a Gate resistor? What is the advantage of having it there between the uC and the MOSFET? I know it functions fine without it.

Is there a way to ease the startup load on the MOSFET? I'm a little nervous that I'm exceeding the I ds for it upon startup. Normal opperation is fine... just startup that worries me.
 

Do I need a Gate resistor? What is the advantage of having it there between the uC and the MOSFET? I know it functions fine without it.
to limit current .
try to build this circuit for your input ( by the way how you made PWM ? do you have any totem pole before your gate ?
see below , please :
mosfet.JPG
Good luck
Goldsmith
 

Status
Not open for further replies.

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top