Need MOSFET driving advice

[brianmc]

I'm designing an IR transmitter for a college project based on binary fsk. I'm using an astable 555 timer running at 126KHz to directly drive a N-Channel mosfet. Vs in this project is 9v. I have 3 IR diodes in series with a resister connected to a Vs. The mosfet is on the low side of the circuit so that when the 555 timer output goes high it activates the fet and and allows the IR circuit to ground. 100mA +- should be flowing.

THE PROBLEM: The IR diodes are on constantly. Used an scope and found that there was a dc level across each of them but the resistor voltage showing a perfect square wave as if the circuit is working correctly. Since I have never used MOSFET before I am at a loss as to what is going on. Looked around online and I believe the problem is ringing but am not sure and need some advice/help. I tried pulling the gate of the fet low with a 1K ohm resister to ground but it had no effect. There is no series resistor between the gate and 555 timer output. This is the fet I'm using: https://ie.farnell.com/stmicroelectronics/stp55nf06/mosfet-n-to-220/dp/9803211?Ntt=980-3211

sorry for the short story

 

[barry]

If you are seeing a square wave across the resistor (0-9V, I presume) but you have a constant voltage across your diodes, you don't have it wired correctly. What you observe is impossible if it's a series circuit.

 

[brianmc]

It is a series circuit, 3 series diodes and a 59 ohm resistor in series as well. Is it possible that ringing on the gate could cause the gate to stay open enough to allow a few mA to flow through and develop voltage across the diodes (1v) with only a few mV across the resistor?

 

[barry]

Forget about the gate. The same current that flows through that resistor flows through the diodes. If you see zero volts across that resistor, then no current flows through those diodes and they should be off. What is the exact voltage you are seeing across that resistor (and what's the voltage at the drain of the mosfet?)

 

[brianmc]

I will get a picture of the scope traces and post them up tomorrow. I fully understand what you are saying about the current being the same through the diodes as the resistor and that's why this circuit has me confused because the scope showed a voltage square wave, 0-9v, across the resistor with the same frequency as the 555 timer output but the diodes had a constant voltage across them. The voltage at the drain is a distorted square wave roughly 6 volts. Thanks.

 

[FvM]

Looks like normal operation. The voltage drop across the 59 ohms resistor will tell you about the diode current. I don't expect that a noticeable current flows in transistor off-state.

 

[D.A.(Tony)Stewart]

The LED's like most diodes will reach the threshold voltage with a tiny current. Since IR LED's have very low ESR, (~ <1 Ohm) the voltage change will be very small.

 

[barry]

The LED's like most diodes will reach the threshold voltage with a tiny current. Since IR LED's have very low ESR, (~ <1 Ohm) the voltage change will be very small.

OP said "the diodes are constantly on" and "there is a DC level" across them. The LEDs should have 1-2 volts across them when they are conducting.

 

[laser-speed]

Excuse me guys ... you may ground the gate with a 1k resistor ... so it keeps off while off state instead of the floating condition ... the MOSFET Gate charge causes that error .

 

[D.A.(Tony)Stewart]

The LED's like most diodes will reach the threshold voltage with a tiny current. Since IR LED's have very low ESR, (~ <1 Ohm) the voltage change will be very small.

The 555 output is low impedance complementary driver, so no issue, but the IR LED's have no pull up so there will always be "some " leakage to maintain a DC voltage across them even tho the series resistor shows near zero voltage drop during Off state.

To prove my point use a 10K pull-up resistor the junction between R and LED cathode. Although you do not need to do this it will prove your assumption that the IR LED's are ON all the time May be incorrect.

If you are trying to send FSK with a pulsed IR LED, you are doing it very inefficiently.
I would use 3V supply with 2 LED's and pulse 150mA per string or 3 LED's with a 4V LiPo with 1 Ohm instead of 9V with 3 LEDS pulsing 100mA and consuming more battery power with less output.

 

[barry]

I guess someone should ask the question: "How do you KNOW the IR diodes are on constantly?". Do you have an infrared detector? Can you SEE infrared?