Continue to Site

Welcome to our site!

Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

  • Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

MOSFET Headache

Status
Not open for further replies.

turbobill

New Member
Hello All,

I'm having a real problem trying to get a MOSFET to work like it's suppose to( I think ).

Using an astable 555 multi vibrator output to switch MOSFET on|off is what I want to do. What I'm getting is a good square shape on the scope from the 555 but the MOSFET just stays on all the time.

The 555 output is at 12v 200ma at about 3ms on and 3ms off, there is 12v at the source and a steady 12v at the drain( a straight line on the scope ).

The MOSFET can handle 24V 70+ amps and operate at nano second speeds.

I want the MOSFET drain signal to look like the gate, right? Not happenin'.

Any help is much appreciated,
William
 
You forgot to attach your schematic.
You forgot to say which Mosfet.

Since the source is at +12V then the Mosfet must be a P-channel type. Then it turns on when its gate goes to ground and it turns off when its gate goes to +12V.

But the highest output voltage from an ordinary 555 is only about +10.8V if it doesn't have a load, lower with a load, so some Mosfets will not turn off.

If you use a Cmos 555 then its output goes up to +12V when it doesn't have a load and it will turn off the Mosfet. The Cmos 555 has a low output current so the Mosfet won't switch quickly.
 
Last edited:
A Mosfet with a 24V/70A+ load and without a heatsink will be melted inside. Therefore it is shorted.
 
You forgot to say which Mosfet.
MOSFETs are N-Channel because when I told the Mouser tech guy I wanted to switch the anode he told me I needed N-Channel. Gate Threshold is < 2v .

The timer is an ST SA555 General Purpose Single Bipolar.

Twelve volts at the source and the drain? That makes no sense at all. How is it connected in circuit?
The 12v at the source is from the + bus and the 12v from the drain is because the MOSFET is always conducting. I'm using an automotive tail light as a load.

So, if I'm using an N when it should be a P, then if polarity is switched it should work. Yes/No/Maybe.

Thanks,
William
 
N-channel should have its source grounded and it's drain connected to the load, and be used as a low-side switch. So source = ground, +12V = to load, other side of load = drain. Should work hunky-dory when you rig it up like this, otherwise you might have damaged or gotten a bad FET.

Sounds like you are using a nice FET btw.
 
Last edited:
A schematic?

WHich mosfet? We know its N-channel. How about which one.. as in a part number....
 
WHich mosfet? We know its N-channel. How about which one.. as in a part number....

Have three different ones: ON NTB125N02R, Infineon SPP73N03S2L-08 and Fairchild FDD8896

Have tried each one ... same thing. They are all N-Channel
 
Of course they don't work. You need a P-channel Mosfet if its source is connected to +12V and the load is connected from its collector to ground.

If you use a P-channel Mosfet with a low threshold voltage then the 555 won't turn it off.
 
Yeah, you either need a proper P-channel, or use one of the N-channels you listed above in the way I described above.

The N-channel connected properly is the better solution for the reason AG mentioned above.
 
N-channel should have its source grounded and it's drain connected to the load, and be used as a low-side switch. So source = ground, +12V = to load, other side of load = drain.

Yup, works good! But I really do need to switch the positive instead of the negative.

If I run the output of the 555 straight to the gate w/ no resistor, will it turn off the P-Channel?

Thanks,
William
 
The resistor being in series with the gate or not won't matter, it's just a current limiting device. The gate voltage will quickly become the same as the output voltage of the 555. Just make sure that the output of the 555 reaches as high to the +12V rail as possible, and select a FET with a large-ish Vgs turn on threshold to ensure that it will turn off when the gate voltage gets close to the power rail.

BTW, a question from me to the group: are 555's typically rail to rail or do you lose a volt or so as far as the output voltage is concerned relative to the supply voltage?
 
Last edited:
You do need to use a p-channel MOFET to switch the positive line.

You can use a bootstrap capacitor and an n-channel MOSFET but it's difficult to know whether it's possible without seeing your schematic.

An alternative would be to connect the 555 timer circuit to another 12V supply on top of the 24V supply to give 36V but your 24V supply would need to be able to sink the current.
 
speakerguy79 said:
The resistor being in series with the gate or not won't matter, it's just a current limiting device.
No it is not. The gate doesn't draw current. The series resistor feeding the gate stops the Mosfet from oscillating at a very high frequency.

BTW, a question from me to the group: are 555's typically rail to rail or do you lose a volt or so as far as the output voltage is concerned relative to the supply voltage?
The datasheet shows that its output has two emitter-followers in series to pull the output high. The output high voltage typically has a 1.3V loss with no load and has a max loss of 2.25V when it has a load of 100mA.
The output of a Cmos 555 is rail-to-rail when it has no load.
 
Driving such big N-MOSFET with 555 for high side.
The mosfet need at least 2Amp peak to reach nanosecond switching. And also higher voltage to supply Vgs.
Might be you connect its source to +12V and it on all the time caused by it's parasitic diode.

You can used bootstrap if your 555 provide repetitive switching. Need a capacitor, diode and level shifter.
 
Last edited:
No it is not. The gate doesn't draw current. The series resistor feeding the gate stops the Mosfet from oscillating at a very high frequency.

I was speaking in terms of initial charge injection; the resistor keeps the driving device from seeing the largely capacitive load of the gate and reduces the initial drive current; once gate voltage has stabilized, no current flows. We use the series R to the gate when driving them directly from microcontrollers that have limited Iout capability.

I would be interested to hear more about the oscillation problem, I hadn't considered it before.
 
turbobill said:
I would also like to hear more about that.
In Google, search for Mosfet Gate Resistor. It has articles that describe the Mosfet ocsillating at a very high frequency and a small series gate resistor stops it.
 
The Gate Oscillation

We know that inductor and capacitor doesn't dissipate any power. That's why if we connect charged capacitor and inductor without resistor will cause oscillation at its own frequency.
Mosfet gate is small capacitor and small inductor in series. If we switch Vgs, inductor and capacitor charged, and oscillate at high frequency. Connecting a resistor in series will help to dissipate their energy and stop the oscillation.
 
Status
Not open for further replies.

Latest threads

New Articles From Microcontroller Tips

Back
Top