AO3400 MOSFET problem

[d4v1dcn]

Hi! This is my first post here and sorry if not is the correct topic..

I'm designing a circuit to control a DC motor using DRV8837 H-Bridge driver. This circuit also have a LED lamp with 3x 3528 SMD LED connected in parallel.

The problem here is that if i meassure between +9V pad and -VCC_LEDS terminal i get 9v but mosfet's gate is not receiving any voltage from SW1 or SW2. How it's possible that mosfet are flowing current between source and drain without apply any voltage to the gate?

Hope that you can help me.

Thanks in advance!

[Kurets]

Are the leds actually in circuit and lit up? If so, the MOSFET is either on, shorted, or connected in reverse (so that current flows through the body diode).

1. What is the actual gate to source voltage?
2. If this is zero, what is the resistance drain to source?
3. What readings to you get with diode mode on the dmm if yoi measure from source to drain and from drain to source respectively?

[d4v1dcn]

Are the leds actually in circuit and lit up? If so, the MOSFET is either on, shorted, or connected in reverse (so that current flows through the body diode).

1. What is the actual gate to source voltage?
2. If this is zero, what is the resistance drain to source?
3. What readings to you get with diode mode on the dmm if yoi measure from source to drain and from drain to source respectively?

Hi,

LED's aren't connected to PCB but if i measure in PCB pads (+VCC_LEDS (9v) and -VCC_LEDS) i got 9v

Replying to your questions:
1: 0v
2: 342.2 Ohms
3: From source to drain: 0.45v and from drain to source: 0.49v

Thanks!

[BadeBhaiya]

MOSFETs even in their off state are basically very high value resistors. It is totally possible that your multi-meter may show 9V but if you try to pass even very low current, the voltage will drop to 0V. The burden of the multimeter is usually enough to do this, but in some cases may not be

[wraper]

The circuit shown makes no sense. What are these SW1, SW2? Is the voltage driving these inputs supposed to come from somewhere that is not shown in the circuit as only inputs connected together are shown? Not to say SW1 and SW2 are almost shorted together through R9, R10. DRV8837 has pull-down built in.

[Kurets]

Okay, yes what the third poster says. Although based on your DMM readings for resistance and diode check it seems like the transistor is conducting more than reasonable. Either the gate voltage is not zero or the device is damaged.

The only MOSFET I have ever killed by (suspected) ESD was a 2N7002, which ended up a "weak short" kind of like your example.

Another question, the gate pull down and the H-bridge driver are both mounted? One common thing seen when starting to experiment with MOSFETs is to forget that if you charge the gate capacitance without a discharge resistor attached, the transistor will stay on for a long time.

[d4v1dcn]

The circuit shown makes no sense. What are these SW1, SW2? Is the voltage driving these inputs supposed to come from somewhere that is not shown in the circuit as only inputs connected together are shown? Not to say SW1 and SW2 are almost shorted together through R9, R10. DRV8837 has pull-down built in.

Hi,
SW1 and SW2 are tactile swithes connected to +9v. They are supposed to apply +9v on mosfet's gate on pressed

[d4v1dcn]

Okay, yes what the third poster says. Although based on your DMM readings for resistance and diode check it seems like the transistor is conducting more than reasonable. Either the gate voltage is not zero or the device is damaged.

The only MOSFET I have ever killed by (suspected) ESD was a 2N7002, which ended up a "weak short" kind of like your example.

Another question, the gate pull down and the H-bridge driver are both mounted? One common thing seen when starting to experiment with MOSFETs is to forget that if you charge the gate capacitance without a discharge resistor attached, the transistor will stay on for a long time.

Yes, pull-down and driver are mounted on PCB. I attach the PCB image (note: The pull-down resistor is mounted on the actual PCB even though it does not appear in this design.)

[BadeBhaiya]

Seems like the FET is blown. Getting a source to drain forward voltage of 0.4V is okay, but a similar drain to source voltage even when the Vgs is 0V means you MOSFET is gone. Or could it be a PFET instead of a NFET?

Going by the image you posted, it is apparent that the drain of the FET is in fact not connected to anything, so its the conclusion that I came to.

Hi,
SW1 and SW2 are tactile swithes connected to +9v. They are supposed to apply +9v on mosfet's gate on pressed

Although quite unlikely, it is possible to have killed the FET because of ESD from the pushbutton. Very unlikely though.

[wraper]

The circuit shown makes no sense. What are these SW1, SW2? Is the voltage driving these inputs supposed to come from somewhere that is not shown in the circuit as only inputs connected together are shown? Not to say SW1 and SW2 are almost shorted together through R9, R10. DRV8837 has pull-down built in.

Hi,
SW1 and SW2 are tactile swithes connected to +9v. They are supposed to apply +9v on mosfet's gate on pressed

Then this circuit has no chance to work properly because regardless of which switch is pressed, the result will be the same as either switch will pull up both inputs of DRV8837.
Another error is that absolute maximum Vcc and logic input voltage for DRV8837 is 7V, yet you apply 9V to them. It's probably blown already, and unless you actually measured gate voltage to be 0V, MOSFET is likely activated through blown DRV8837 pushing voltage though its inputs.

[d4v1dcn]

Seems like the FET is blown. Getting a source to drain forward voltage of 0.4V is okay, but a similar drain to source voltage even when the Vgs is 0V means you MOSFET is gone. Or could it be a PFET instead of a NFET?

Going by the image you posted, it is apparent that the drain of the FET is in fact not connected to anything, so its the conclusion that I came to.

Hi,

Mosfet is N channel (A09T) from AliExpress. I've tried with 2 different units, same result.

[d4v1dcn]

The circuit shown makes no sense. What are these SW1, SW2? Is the voltage driving these inputs supposed to come from somewhere that is not shown in the circuit as only inputs connected together are shown? Not to say SW1 and SW2 are almost shorted together through R9, R10. DRV8837 has pull-down built in.

Hi,
SW1 and SW2 are tactile swithes connected to +9v. They are supposed to apply +9v on mosfet's gate on pressed

Then this circuit has no chance to work properly because regardless of which switch is pressed, the result will be the same as either switch will pull up both inputs of DRV8837.
Another error is that absolute maximum Vcc and logic input voltage for DRV8837 is 7V, yet you apply 9V to them. It's probably blown already, and unless you actually measured gate voltage to be 0V, MOSFET is likely activated through blown DRV8837 pushing voltage though its inputs.

Hi,

Then this circuit has no chance to work properly because regardless of which switch is pressed, the result will be the same as either switch will pull up both inputs of DRV8837

The circuit works correctly now except the mosfet issue. If i press SW1, motor turns in one direction. If i press SW2, motor turns in the other direction.

The idea is to turn on LEDs when SW1 or SW2 are pressed.

Another error is that absolute maximum Vcc and logic input voltage for DRV8837 is 7V, yet you apply 9V to them

Yes, you are right, this is a first version of PCB. VCC of DRV8834 is connected to VBAT+ (3.7-4.2v) now

[wraper]

Mosfet is N channel (A09T) from AliExpress. I've tried with 2 different units, same result.

It would not surprise me if it's not even N-channel mosfet. Measure direction of body diode (if there is any) between the drain and source.

[wraper]

Then this circuit has no chance to work properly because regardless of which switch is pressed, the result will be the same as either switch will pull up both inputs of DRV8837

The circuit works correctly now except the mosfet issue. If i press SW1, motor turns in one direction. If i press SW2, motor turns in the other direction.

The idea is to turn on LEDs when SW1 or SW2 are pressed.

Another error is that absolute maximum Vcc and logic input voltage for DRV8837 is 7V, yet you apply 9V to them

Yes, you are right, this is a first version of PCB. VCC of DRV8834 is connected to VBAT+ (3.7-4.2v) now

Show actual circuit, how are we supposed to help you when circuit you provide has no bearing with reality?

[d4v1dcn]

Then this circuit has no chance to work properly because regardless of which switch is pressed, the result will be the same as either switch will pull up both inputs of DRV8837

The circuit works correctly now except the mosfet issue. If i press SW1, motor turns in one direction. If i press SW2, motor turns in the other direction.

The idea is to turn on LEDs when SW1 or SW2 are pressed.

Another error is that absolute maximum Vcc and logic input voltage for DRV8837 is 7V, yet you apply 9V to them

Yes, you are right, this is a first version of PCB. VCC of DRV8834 is connected to VBAT+ (3.7-4.2v) now

Show actual circuit, how are we supposed to help you when circuit you provide has no bearing with reality?

Hi, full circuit schematic is attached.

Only for testing purposes, i've replaced A09T (AO3400) by a IRLR024N and LEDs works as expected (only when i press SW1 or SW2)

Regards!

[wraper]

^still no switches shown, if they're connected to 9V, it's not OK. You still have nSleep input connected to 9V. R9, R10 are still 330R, I don't see how U5 IN1, IN2 inputs can possibly operate separately, they're basically shorted together FWIW.

[d4v1dcn]

^still no switches shown, if they're connected to 9V, it's not OK. You still have nSleep input connected to 9V. R9, R10 are still 330R, I don't see how U5 IN1, IN2 inputs can possibly operate separately, they're basically shorted together FWIW.

Hi, switches are connected to VBAT+. Updated schematic are attached 

I don't see how U5 IN1, IN2 inputs can possibly operate separately, they're basically shorted together FWIW.

. What will be te correct method to achieve this?. I want to use SW1 and SW2 to activate H-Bridge and at same time, activate mosfet for LED light.

Thanks!

[d4v1dcn]

Hi,

I've created a new schematic with some upgrades and changes to make a new PCB. Please i would like to know if you see any problems with this circuit and if you could give me your advice on how to improve it.

This circuit is to make a custom electronic screwdriver.

Regards!

[wraper]

What will be te correct method to achieve this?. I want to use SW1 and SW2 to activate H-Bridge and at same time, activate mosfet for LED light.

Use diodes instead of those resistors as you already did in updated circuit. Although use low power Schottky like BAT54 or regular silicon diodes instead of SS14, they have a lot of reverse leakage current and built in pull-down of DRV8837 may be not enough to deal with that.

[d4v1dcn]

Hi guys!

I have now another "problem" with section 6 'front lights'. LED's 61-63 are always on and i don't know why. If i measure between GND and anode of D61 or D62 i got 0v, but if i measure between GND and mosfet's gate or drain, i got around 4v (+VBAT). I'm using IRLR024N mosfet now. Maybe the mosfet is damaged or there is any problem in the circuit design?

New schematic is attached.

Thanks in advance!