h bridge problem (for brushless motor)

[jayanthyk192]

hi,

I building a brushless motor controller using h bridge. the upper side has irf9540 and the lower side has irfz44n. i connected their sources to 5thv and 0v resp. using logic 5v as gate voltage for both I started the brushless motor. it works but the current is not going beyond 200mA.why is this happening. the current at 5v for both is beyond 3A for both.

thank you

 

[alexan_e]

Have you measured the voltage across the motor?
What is the specification of the motor?

Alex

 

[jayanthyk192]

The motor is 4000kv, internal resistance is 20m ohms.the current should be atleast 3A.

 

[alexan_e]

I haven't heard of a 20m ohm motor, it seems very strange.
Have you got any datasheet?

Did you measure the voltage it gets when it is connected on the bridge?
How does the motor behave directly to 5v supply?

Alex

---------- Post added at 19:14 ---------- Previous post was at 18:51 ----------

You are correct, brushless motors have a very low resistance and they also shouldn't be connected directly to the power supply so I assume you have some kind of controller/current limiter that is responsible for the limited current.

Alex

 

[prmurthy]

Hi,

You cannot control brushless motor with logic. You need PWM to control. As the speed increases the current will increase.

Thankyou

PS: Brushless motors are phase controlled. So you should be having number of phases for your brushless motor.

 

[jayanthyk192]

I haven't heard of a 20m ohm motor, it seems very strange.
Have you got any datasheet?

Did you measure the voltage it gets when it is connected on the bridge?
How does the motor behave directly to 5v supply?

Alex

---------- Post added at 19:14 ---------- Previous post was at 18:51 ----------

You are correct, brushless motors have a very low resistance and they also shouldn't be connected directly to the power supply so I assume you have some kind of controller/current limiter that is responsible for the limited current.

Alex

I'm powering the circuit with smps and it can only give 4-6A max.but the H bridge supplies only 200mA of current.the motor does not even turn.

---------- Post added at 19:41 ---------- Previous post was at 19:38 ----------

Hi,

You cannot control brushless motor with logic. You need PWM to control. As the speed increases the current will increase.

Thankyou

PS: Brushless motors are phase controlled. So you should be having number of phases for your brushless motor.

right now i'm just moving it in steps.so now PWM just simple step by step movement

 

[alexan_e]

I'm powering the circuit with smps and it can only give 4-6A max.but the H bridge supplies only 200mA of current.the motor does not even turn.

So there is nothing to limit the current of the motor?
Are you applying short pulsed or a constant voltage?
And how much is that voltage(measured on the motor), you never answered my question about the voltage that the motor gets, you only mention the 200mA
If you are using the bridge to supply the 5v to the motor then I think it pulls an excessive amount of current that the psu sees as short circuit and it limits the output.

Alex

 

[jayanthyk192]

there is nothing that limits the current.
i'm applying pulses.
The voltage across two terminals during the pulses is 4.5~5mV.
My smps does not limit current because it gave 4A when I held the multimeter across the terminals.

 

[alexan_e]

My smps does not limit current because it gave 4A when I held the multimeter across the terminals.

what exactly did you do , did you short circuit the psu output with the ammeter?
Since you apply pulses you have to give more info on the duration of the pulses and the driver that you use for the mosfets, you need a high current driver for fast switching.

Alex

 

[jayanthyk192]

what exactly did you do , did you short circuit the psu output with the ammeter?

Alex

yes i just put the multimeter pins into the smps and measured the current.

about the timing:the pulses are around 1 second away,i.e change over between each phase.

the circuit is the one attached,high side is URF9540,low side IRF44n

 

[alexan_e]

 

[jayanthyk192]

:shock: this is one of the best ways to destroy your ammeter or damage the power supply



And how much is the duration of the pulses?

Also take a look at https://www.edaboard.com/threads/196377/ and note that you are stressing the mcu outputs and you have a very slow switching speed for the mosfets because the mcu can give a few mA at best.

Alex

the pulses are large almost 1 second,then the next phase is active.switching at this rate should'nt cause problems.the link to your older post is really good.i never thought of the capaitance.i neglected that.but here as the switching speed is like a stepper motor the MC current (gives 25mA) is not being stressed.

i also observed that the p mos are getting heated up considerably and the lower side stays cool.does this have to do something with the above discussion?

---------- Post added at 21:10 ---------- Previous post was at 21:05 ----------

i try to do it really fast.i hear the smps fan reduce noise.it starts lacking power,all of which is being sucked by the meter.

 

[alexan_e]

so you have 1 sec one polarity on then one second the other polarity on and then then redo, is that right?
You may have cross conducting mosfets if you don't apply a deadtime between pulses, it will take a while for the mosfets to turn off so you have to wait for them to turn completely before turning on the other two mosfets.

To go back in my original assumption you have 1sec wide pulses putting the motor at a 5v supply, I think that this is seen as a shortcircuit for the psu and that is why you measure just a few mV.
A simple test is to put a resistor in series with the motor, a value like 2 ohm (5W if you have or even 1-2W for a short while), then try to see if the motor gets 1.5A

Alex

 

[jayanthyk192]

so you have 1 sec one polarity on then one second the other polarity on and then then redo, is that right?
You may have cross conducting mosfets if you don't apply a deadtime between pulses, it will take a while for the mosfets to turn off so you have to wait for them to turn completely before turning on the other two mosfets.

To go back in my original assumption you have 1sec wide pulses putting the motor at a 5v supply, I think that this is seen as a shortcircuit for the psu and that is why you measure just a few mV.
A simple test is to put a resistor in series with the motor, a value like 2 ohm (5W if you have or even 1-2W for a short while), then try to see if the motor gets 1.5A

Alex

i had 2 10 ohms resistors,i put them in parallel and measured the voltage: 1.8 V then both of them burnt out.2 10 ohms were the least i had.the resistors were 1/4 watt each.

 

[alexan_e]

two 1/4w in parallel become 1/2w but its still too low, anyway you have measured 1.8v across the motor I assume, we can't calculate the current because we need the voltage drop on the resistors but if we assume that the voltage was 5v-1.8v=3.2v
3.2v/5 ohm=0.64A

I think the problem is that you are not using any mean to limit the current on the motor, a resistor will limit it for testing but its inefficient for permanent usage.
You need to use PWM but you also need a proper mosfet driver in that case because you will have very short pulses.

Do you have any kind of load that can draw some power from the 5v so that you can connect it in the bridge to see if you can get a few amperes?

Alex

 

[jayanthyk192]

i just dont know what happened,but after burning the resistors the circuit works.the motor now actually spins.but both the mosfet types are burning hot,i think its well above 100 degress c.even with the heat sink on its burning.is it supposed to be like that?

 

[alexan_e]

Both your mosfets have a gate threshold of 4v max so with a Vgs of 5v as you use now you will not get a very low Rds-on, in addition this on resistance is many times higher than the motor resistance so you are getting more watts on the mosfets than on the motor, it is like a 3 resistor divider.
The irf9540 will have about 0.3 ohm with the Vgs of 5v and irfz44n a resistance of about 50m ohm

Alex

 

[jayanthyk192]

Both your mosfets have a gate threshold of 4v max so with a Vgs of 5v as you use now you will not get a very low Rds-on, in addition this on resistance is many times higher than the motor resistance so you are getting more watts on the mosfets than on the motor, it is like a 3 resistor divider.
The irf9540 will have about 0.3 ohm with the Vgs of 5v and irfz44n a resistance of about 50m ohm

Alex

yes,i saw the datasheet and at 5V the current should reach around 20A,and the resistance should be very small.but now i chnaged the motor and used a hard disk motor.my meter shows 7ohms but its atleast 3ohms(many of them are).but still its getting heated badly.as you said the gate voltage must be low i think.so should i decrease the voltage and then check?

thank you for all the help you're providing.again:thank you.

 

[alexan_e]

you are welcome.

irf9540 is the worse and a calculation from the graph shows a resistance of about 0.3 ohm and less than 4A with your Vgs
irfz44n a resistance of about 50m ohm and can give more than 10A
so the problem in your case is basically the P mosfet and should be getting much warmer than the N mosfet.

which voltage do you want to decrease?

Alex

 

[jayanthyk192]

you are welcome.

irf9540 is the worse and a calculation from the graph shows a resistance of about 0.3 ohm and less than 4A with your Vgs
irfz44n a resistance of about 50m ohm and can give more than 10A
so the problem in your case is basically the P mosfet and should be getting much warmer than the N mosfet.

which voltage do you want to decrease?

Alex

i think i would have to decrease both.i'm thinking of using transistor switches to operate the gates.or maybe i have to use PWM to limit current.

you are 100% right,the p side is hotter than n side.

thank you