Continue to Site

Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

Register Log in

H brigde for control speed and direction 180VDC brushed motor

Status
Not open for further replies.
Y

yophi

Newbie level 6
Newbie level 6
Joined
Apr 30, 2019
Messages
12
Helped
0
Reputation
0
Reaction score
0
Trophy points
1
Activity points
112
I have problem with my project made H brigde for control speed and direction 180VDC brushed motor. I have used Mr. Thamid method with IC IR2110 and N channel mosfet. I waste many mosfet and IR 2110 to build my project. please help me to solve my problem. I have success for a while but after time to rotating the direction often my mosfet was burn.

Thank you for your attention

regards,
Yophi
From Indonesian student.
 

Attachments

  • IR2110pwm.PDF
    89.8 KB · Views: 170

Hi,

Most probably the main problem is the PCB layout.
--> show your PCB layout

Additionally it could be a problem of your control signal timing.
--> show scope pictures of your control signals HIN and LIN. Dual channel. Different pictures for: low duty cycle, medium duty cycle, maximum duty cycle.

* I strongly recommend to add capacitors at the DC power path

***
Not that problematic issues. Just check if you like to modify:
* I see no benefit in serially connect the voltage regulators
* I see no benefit in the 1k gate-source resistors. Use at least 10k.
* I see no benefit in using optocouplers. Optocouplers are for galvanic isolation between input and output...what you don't have.

Klaus
 
  • Like
Reactions: yophi

    Y

    yophi

    Points: 2
    Helpful Answer Positive Rating
KlausST said:
Hi,

Most probably the main problem is the PCB layout.
--> show your PCB layout

Additionally it could be a problem of your control signal timing.
--> show scope pictures of your control signals HIN and LIN. Dual channel. Different pictures for: low duty cycle, medium duty cycle, maximum duty cycle.

* I strongly recommend to add capacitors at the DC power path

***
Not that problematic issues. Just check if you like to modify:
* I see no benefit in serially connect the voltage regulators
* I see no benefit in the 1k gate-source resistors. Use at least 10k.
* I see no benefit in using optocouplers. Optocouplers are for galvanic isolation between input and output...what you don't have.

Klaus
Click to expand...
bottom view.jpgTop View brd.jpg
Hi Klaus,

Thank you for your reply,

--for PCB lay out I just made my self and you can see at attachment.

--for want to see the signal, i don't have Oscilloscope to inspect it. so just check the signal at LIN and HIN only( the PWM signal from arduino )?

--which DC power path need add a capacitors. you mean on high DC voltages? what value of capacitor need add on there? because I have tried add a electrolyte capacitor with 330uf/ 400v but have burn on IR2110 and mosfet after start to run the motor

* I see no benefit in serially connect the voltage regulators
-- so can you suggest me how to get 12v ( VC IR2110 ), 9v ( supply arduino ), 5v ( for VDD IR2110 )

* I see no benefit in the 1k gate-source resistors. Use at least 10k.
-- so i need replace 1k gate - source resistor with 10k because I just made according Mr. Thamid design.

https://tahmidmc.blogspot.com/2013/01/using-high-low-side-driver-ir2110-with.html

* I see no benefit in using optocouplers. Optocouplers are for galvanic isolation between input and output...what you don't have.
-- if no benefit in using optocouplers for insulating the arduino signal, please suggest me must be better.

Yophi.
 

Hi,

* PCB Layout: Without proper GND plane it´s not suitable for switching power applications. No wonder parts get killed. --> Use one layer as GND plane only, wihtout any other traces.

* buy/lend an oscilloscope

* DC power path capacitors. Value is not that important - then need to be fast to supress spikes. Thus use foil or ceramics. Start with 1uF. But they are useless without GND plane.
The same "fast" applies for all power supply capacitors. A good rule of thumb: place a 100nF ceramics capacitor close to each supply pin of each IC.
C9..C12 are slow ar eelectrolytics which are slow. --> the correct way is like C1 and C2. = A fast cramics in parallel to an electrolytics.

* voltage regulators: --> connect them in parallel.

* 1k / 10k. Nobody can explain why there is so often 1k in schematics. But 1k is a dawback: The gates of the MOSFETS are high resistance = about no DC current. But 1k generates DC current, thus the bootstrap circuit needs to supply this current to maintain bootstrap voltage.
--> 10k allowes smaller bootstrap capacitors, lower PWM frequency, less current peaks to charge the bootstrap circuit.

* optocouplers: They are useless, thus I recommend not to use them. --> Just directly connect the arduino_out to HIN /LIN

Klaus
 
  • Like
Reactions: yophi

    Y

    yophi

    Points: 2
    Helpful Answer Positive Rating
Your problem may be "Shoot-Through" = both MOSFETS on one side of the H-Bridge are ON simultaneously.
What is your Dead-Time?
Usually, hardware is used will guarantee minimum Dead-Time, not software.

Are the MOSFETS getting warm?

How many amps is flowing through the MOSFETS ( min / typical / max ) ?
What is the Stall Amps for the Motor?

What is the DC Supply Voltage for the Motor?
What is the Max Reverse Voltage across any MOSFET?

Are you slowly ramping the RPM's down to 0 RPM before reversing or are you stopping suddenly & quickly reversing?
 
  • Like
Reactions: yophi

    Y

    yophi

    Points: 2
    Helpful Answer Positive Rating
I have success for a while but after time to rotating the direction often my mosfet was burn.
Click to expand...

That can be the starting point for the post mortem analysis.

Did you allow the motor to come to rest before you try to reverse the direction?

When you start the motor (or reverse it) do you SLOWLY ramp up the voltage?

Is the motor with load when you try to reverse it?

ALSO

Starting from rest, you have no problem running the motor in one direction, correct?

Did the power mosfets get unduly warm (I mean really hot) when running for a long time?

Starting from rest, can you start the motor in the reverse direction without problem?

Is the dead time under hardware or software control?

Try to use a small ferrite bead with the gate of EACH mosfet.
 

Your problem may be "Shoot-Through" = both MOSFETS on one side of the H-Bridge are ON simultaneously.
-- I think this point its okay. because my code turn on Q1 & Q4 then turn off all mosfet for 5 seconds then turn on mosfet Q2&Q3 for reverse rotating

What is your Dead-Time?
-- i dont know what is Dead-Time, but before I turn on the other mosfet it have 5 second for freewheel ( all mosfet are OFF ).

Usually, hardware is used will guarantee minimum Dead-Time, not software.

Are the MOSFETS getting warm?
--not warm anymore.

How many amps is flowing through the MOSFETS ( min / typical / max ) ?
What is the Stall Amps for the Motor?
-- this brushed motor is 180VDC with 0,3 A

What is the DC Supply Voltage for the Motor?
--just for diode rectifier, i think 310VDC for supply motor

What is the Max Reverse Voltage across any MOSFET?
--I used IRF840 N-Channel mosfet and I set value PWM for max 150-160VDC

Are you slowly ramping the RPM's down to 0 RPM before reversing or are you stopping suddenly & quickly reversing?
--I stopping suddenly with turn off all signal to HI and LI IR2110 and have 5 seconds delay for reversing ( not ramping UP )

I have update my schematic below, please help me to inspect my schematic must be work or still faulty.
 

c_mitra said:
That can be the starting point for the post mortem analysis.

Did you allow the motor to come to rest before you try to reverse the direction?
--yes have 5 seconds to come rest or freewheel and the got o reverse direction

When you start the motor (or reverse it) do you SLOWLY ramp up the voltage?
-- not Slowly ramp up. just according the PWM value from analogRead of Potentiometer.

Is the motor with load when you try to reverse it?
-- nothing load on motor .but have A gearbox on it


ALSO

Starting from rest, you have no problem running the motor in one direction, correct?
-- yes no problem and can running.

Did the power mosfets get unduly warm (I mean really hot) when running for a long time?
--not warm anymore

Starting from rest, can you start the motor in the reverse direction without problem?
-- I am sorry, I have not try before.

Is the dead time under hardware or software control?
I dont know what is dead time.. but my code have 5 seconds for freewheel with millis feature.( turn off all mosfet )

Try to use a small ferrite bead with the gate of EACH mosfet.
you mean use small ferrite to smooth signal to gate of each mosfet? like a inductor filter?

Thank you I have update my schematic below.IR2110.png
Click to expand...
 

Hi,

To your schematic:
* use GND symbols, they improve readability. And they are for free. ;-)
* you need a bulk capacitor and a ceramics capacitor at the input of the voltage regulators
* you need a bulk capacitor and a ceramics capacitor at the bridge supply
* I recommend to replace the 1k resustors with 10k resistors.
* I recommend to replace the UF4007 at the bridge with schottky ones. The lower forward voltage avoids that the slow Mosfet_body_diodes become conductive.
* I recommend to add protection at the signal at X2.

Klaus
 

you mean use small ferrite to smooth signal to gate of each mosfet? like a inductor filter?
Click to expand...

Yes; they will give some protection to noise. It will also reduce the rise time (and the fall time) of the pulse at the gate but will prevent false triggering of the mosfet.

Use filter capacitors and bypass capacitors liberally; they are not expensive. They are essential at the inputs of the three regulators (see post #9).
 

Hy Mr Klaus,

Thank you for your suggest,
I have updated my schematic but, I dont know how much the Value of C6, C9 capacitor at bridge supply and which type of diode schottky at the bridge. I need check at my local market.

what do you think about my schematic after re draw? more better than before.?

Thank you for your sharing.

Yophi.IR2110.png
 

Hi,

C6, C9, C11, C12.
Ceramics: use standard value 100nF (up to 1uF)
Electrolytics: there are many tutirials how to calculate them, even online calculators. Look for "bulk capacitors for power supplies".

The schottky need low enough forward voltage at the expected current to prevent the Moset internal diode from getting conductive. It needs to withstand the peak current, it needs to withstand the bridge voltage. Use online selection guides.

Klaus
 

what do you think about my schematic after re draw?...
Click to expand...

Not related to the present circuit but why you are feeding your arduino with unregulated voltage? It needs regulated supply.
 

c_mitra said:
Not related to the present circuit but why you are feeding your arduino with unregulated voltage? It needs regulated supply.
Click to expand...

I will use 7809 for supply the arduino at other PCB. the other PCB will have 7809 and display ( LCD16x2 or 7 segment ) and some button for control.

thank you.
 

Sorry I cannot find any obvious mistake with the circuit. What is the PWM frequency you are using? I suggest using the lowest frequency compatible with your requirements.

Introduce some dead time in the software.
 

Status
Not open for further replies.

Similar threads

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top