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.

Why doesn't this H-bridge work?

Status
Not open for further replies.

NFI

New Member
Hello,

I'm trying to design an H-bridge to control a bank of three motors on a 6-wheel robot vehicle. The direction of the motors is controlled by a signal line which is high for forward and low for reverse. In the attached schematic, the direction signal line is represented via two-position switch, which connects to both a pnp and an npn transistor. The idea (at least in my head) is that when the signal line is high, the npn will conduct, turning on two of the n-channel MOSFETS in the bridge and, when the signal line is low, the pnp will conduct, turning on the other two MOSFEST. In the attached schematic, the motor is represented by two LEDs and a resistor (so I can see which way current is flowing for testing).

What happens is this;
Signal line high: npn conducts, pnp doesn't conduct, the MOSFETS seem to partially switch, but the LEDs don't turn on
Signal line low: npn and pnp both don't conduct

I've clearly fundementally misunderstood this circuit, but I can't see why :( Any wiser minds prepared to explain what I'm doing wrong would be greatly appreciated...

Thanks,

NFI

p.s. I forgot to say - I tried to build this circuit as well as simulate it and I get the same result: components used were: npn: ZTX302
pnp: ZTX500
MOSFETS: IRF520A
 

Attachments

  • h-bridge - not working.JPG
    h-bridge - not working.JPG
    30.1 KB · Views: 377
Last edited:
It doesn't work because the base of the second transistor (PNP) is taken to 0v rail when the switch is trying to activate the transistor.
Put a 1k resistor in the base.
What is the 10k and 220R doing in parallel?
You have to look at the circuit and see what voltage will appear on the gate of each MOSFET when it is not turned on and when it is turned on.
Look at the data sheet and find out what voltage needs to be on the gate to turn it on.
 
Last edited:
Hi Colin,

thanks for having a look...

The base of the pnp needs to be taken to zero volts to turn it on, as the bias needs to be in excess of 0.7V collector to base - that's the principle of my circuit - Vcc will turn on the npn, 0V will turn on the pnp.

The idea of the 10K's is to pull the gates of the MOSFETs to 0V when they're not turned on, but to offer a higher resistance than the gate resistance when either the npn or pnp transistors are conducting, hence turning the MOSFETs on. The 220R's effectively in parallel with the 10Ks are then acting as a voltage divider with the other 220Rs, reducing the gate voltage to that required by the MOSFETs - or at least, that's what I think is going on...

Adding a 1K to the base of the npn doesn't seem to make any difference...
 
Try removing the 220 ohm resistors on the mosfet gates.
 
Hi Mikebits,

thanks for the suggestion - looking at the circuit, I can see now that the voltage divider isn't going to work...

removing the 220R's from the gates hasn't made the circuit work, though :(
 
Hello,

I'm trying to design an H-bridge to control a bank of three motors on a 6-wheel robot vehicle. The direction of the motors is controlled by a signal line which is high for forward and low for reverse. In the attached schematic, the direction signal line is represented via two-position switch, which connects to both a pnp and an npn transistor. The idea (at least in my head) is that when the signal line is high, the npn will conduct, turning on two of the n-channel MOSFETS in the bridge and, when the signal line is low, the pnp will conduct, turning on the other two MOSFEST. In the attached schematic, the motor is represented by two LEDs and a resistor (so I can see which way current is flowing for testing).

What happens is this;
Signal line high: npn conducts, pnp doesn't conduct, the MOSFETS seem to partially switch, but the LEDs don't turn on
Signal line low: npn and pnp both don't conduct

I've clearly fundementally misunderstood this circuit, but I can't see why :( Any wiser minds prepared to explain what I'm doing wrong would be greatly appreciated...

Thanks,

NFI

p.s. I forgot to say - I tried to build this circuit as well as simulate it and I get the same result: components used were: npn: ZTX302
pnp: ZTX500
MOSFETS: IRF520A

Your problem is that your using all NFETS, the top two need to be PFETS and then you may have to properly bias them.
Kinarfi
try this
 

Attachments

  • untitled.PNG
    untitled.PNG
    52.2 KB · Views: 263
Last edited:
Can't you see that the top N-channel Mosfets are followers? Their gates need to be at 22V (10V more positive than the source voltages) for them to fully conduct.

The voltage dividers of the 220 ohm resistors at the transistors limit them to produce gate voltages that are far too low at only 6V. Get rid of the emitter resistors and use series base resistors instead.
 
Many thanks for all the replies I've had helping me on this; in repsonse...

Colin55 : apologies, I meant PNP - my fingers just typed NPN - it unfortunately didn't make any difference either way, though...

Kinarfi : I don't understand why that's a problem? I only have N-Channel MOSFETs available to me, which is why I'm using an NPN BJT to switch one pair and a PNP BJT to switch the other pair. If I were simply connecting the signal line to the gates of the MOSFETs, I'd need P and N channel, but not in this sitation, I don't think? I also don't understand your circuit - when the lower MOSFETs conduct, the upper MOSFETs will have very little voltage at their gates - also, the 10Ks will limit the current to so low a level, the LEDs probably won't conduct, either.

Audioguru : no, I can't see that. All the MOSFETs in this circuit are working as voltage controlled switches, surely? Although the IRF520A has a maximum Vgs rating of +/-20V, they have a threshold Vgs of 2.0V - 4.0V, so 6V should be enough to make them conduct. As it is, I've previously removed the lower 220R's, so the Vgs at all the MOSFETs is closer to 12V in any case - the circuit still doesn't work, though.

Any further thoughts will be much appreciated,

Thanks,

NFI
 
Last edited:
1. First unhook the H- bridge from it's drive circuit.
Then take one Mosfet, and using a voltage divider at it's gate term. determine what input voltage satisfies the mosfet to turn on a motor, when it's source term. is grounded.
Then using the same motor load, hook up the motor to ground and see what input voltage satisfies the mosfet when it's source term. is loaded.

You'll find that the second experiment, will be a little eratic, due to the changing voltages that occur across the motor load.
Because remember the voltage at the gate of a mosfet is refrenced from it's source term. (NOT GROUND alone).

So your top transisters (mosfets), are going to need a higher input voltage to overcome the voltage drop across a active load as a motor, qas well as the drop across the lower Mosfet as well.

2. Once you have established the input voltages needed, put together the H-bridge with the mosfets only, (NOT the drive circuit), with the motor as the load.
Input the required voltages at the gates of the mosfets, to see how well it works. Take voltage measurements across each mosfet, VDS, to determine if they are conducting heavily (low voltage drops), so as to not burn out.

3. When it's working properly and the mosfets are running cool, then redesign your drive circuitry, using one NPN transistor for each mosfet gate, then if you need both a positive and ground signal, then you can add your inverting circuitry, as needed directrly at the input, so it is converted to a positive signal to drive the NPN's which in turn drive there mosfets.
This way all the mosfets gates are driven with a positive signal through seperate transistors, so as to keep any unwanted, interaction between mosfets at a minimum. Which means a sure positive signal at each gate as required.
 
Last edited:
Audioguru : no, I can't see that. All the MOSFETs in this circuit are working as voltage controlled switches, surely? Although the IRF520A has a maximum Vgs rating of +/-20V, they have a threshold Vgs of 2.0V - 4.0V, so 6V should be enough to make them conduct. As it is, I've previously removed the lower 220R's, so the Vgs at all the MOSFETs is closer to 12V in any case - the circuit still doesn't work, though.
Look closely at the rating in the datasheet for the "gate-source threshold voltage". Then the current in the Mosfet is only 0.25mA. How fast does your motor spin with a current of only 0.25mA? The gate must be 10V more positive than the source for the N-channel Mosfet to be fully turned on.

If your gate is only 6V then the source is maybe only 1V or less.
Now that you removed the lower 220 ohm resistors when the gate is 12V then the source is maybe only 6V or less.
For the source to be at +12V then the gate must be at +22V.

An H-bridge driver IC for Mosfets has a bootstrap capacitor that forces the gate of the upper N-channel Mosfet above the power supply voltage to fully turn it on.

If the upper Mosfets are P-channel type then they are turned off when the gate is at +12V and are fully turned on when the gate is at 0V.
 
Status
Not open for further replies.

Latest threads

New Articles From Microcontroller Tips

Back
Top