DRV8256: Outputs are not being activated

Daniel Luis Martins

Prodigy 230 points

Part Number: DRV8256

Dear Sirs,

We assembled some prototypes for using the DRV8256PWPR to drive motors.

It turns out that now, we are having difficulty activating the outputs of the integrated circuit.

Even switching the IN1 and IN2 pins to 1 and 0, the outputs remain off.

Furthermore, whenever one of the inputs is set to 1, the nFAULT pin is set to 0.

We tested with the engine connected, and also without the engine connected. And the result is always the same. We also have the DRV8256PEVM board for signal comparison, and from what we saw, the control pins are the same as on our board.

I would like some help in this case if possible.

Attached is the schematic of the board. Some considerations about this scheme:

- originally, the nSLEEP pin was set to +5V directly. We have already removed it from this connection and placed it for control on the microcontroller. When starting the application, we leave it at 0. Once started, we set it to 1. And when any change occurs in nFAULT, an nSLEEP pulse is given.

- the DVDD pin was connected to the +5V bus on the PCB. But as it is an output, we removed this connection, and left it only connected to the 1uF and 100nF capacitors.

- for the VREF, we leave it connected directly to +5V;

Have a great day everyone,

Kindly,

over 1 year ago

0 Daniel Luis Martins over 1 year ago

Prodigy 230 points

0 Jacob Thompson over 1 year ago in reply to Daniel Luis Martins

TI__Mastermind 23381 points

Hey Daniel,

Given the various changes you mentioned, have you tried a new DRV8256 device to make sure the original configuration didn't damage it? DVDD straight to 5V maybe could damage the device with reverse voltage.

Beyond that, I see you're using the P version of the device, so try the two highlighted rows and make sure these don't work:

Can you measure the voltage directly at the IN1, IN2, and nFAULT pins? Make sure this voltage is above the Input logic-high voltage. And/or try replacing the 22 ohm resistors with a 0-ohm resistor. Typically current limiting resistors used inline with a pin are 100 ohm to 1kohm.

Best,

Jacob

0 Daniel Luis Martins over 1 year ago in reply to Jacob Thompson

Prodigy 230 points

Hello Jacob, how are you?

Thanks for your reply.

I'm working again in our board.

I use one new PCB to test, to avoid any problem.

Yes, we using a P version, for PWM. And yes, we use your table (7.3) to drive a motor (IN1=1 / IN2=0 and IN1=0 / IN2=1), with nSLEEP=1.

But it still doesn't work perfectly.

The voltage measured at the IN1, IN2 and nSLEEP pins when activated, the voltage is 4.9V, above the logic high voltage.

Now, I can only activate one of the two ICs, just IC5 - and only for one of the sides (only when I activate IN1=1 / IN2=0).

When I activate it in reverse (IN2=1 / IN1=0)., the nFAULT pin remains at 0 and does not activate.

Kindly,

0 Daniel Luis Martins over 1 year ago in reply to Daniel Luis Martins

Prodigy 230 points

PCB Project

Hello Jacob, how are you?

Thanks for your reply.

I'm working again in our board.

I use one new PCB to test, to avoid any problem.

Yes, we using a P version, for PWM. And yes, we use your table (7.3) to drive a motor (IN1=1 / IN2=0 and IN1=0 / IN2=1), with nSLEEP=1.

But it still doesn't work perfectly.

The voltage measured at the IN1, IN2 and nSLEEP pins when activated, the voltage is 4.9V, above the logic high voltage.

Now, I can only activate one of the two ICs, just IC5 - and only for one of the sides (only when I activate IN1=1 / IN2=0).

When I activate it in reverse (IN2=1 / IN1=0)., the nFAULT pin remains at 0 and does not activate.

Attached, have a link with the project.

Kindly,

0 Jacob Thompson over 1 year ago in reply to Daniel Luis Martins

TI__Mastermind 23381 points

Hey Daniel,

Can you post a picture of your PCB layout? I want to make sure the thermal pad is properly connected for heat dissipation and distribution.

Can you double check that the charge pump capacitor was correctly populated with 22nf? I've seen similar issues to this when it is accidentally populated with ex. 220nF or 2nf instead.

Lastly, can you post an oscilloscope capture of it trying to drive the motor? include IN1, IN2, nFAULT, and ideally a current waveform of the output.

Best,

Jacob

0 Daniel Luis Martins over 1 year ago in reply to Jacob Thompson

Prodigy 230 points

Hello JAcob,

Thanks again for your help in this case.

I'l try to answer your questions:

Jacob Thompson said:
Can you post a picture of your PCB layout? I want to make sure the thermal pad is properly connected for heat dissipation and distribution.

Attached, I put some pictures from our PCB.

/resized-image/__size/320x240/__key/communityserver-discussions-components-files/38/Captura-de-Tela-_2800_916_2900_.png

/resized-image/__size/320x240/__key/communityserver-discussions-components-files/38/Captura-de-Tela-_2800_915_2900_.png

/resized-image/__size/320x240/__key/communityserver-discussions-components-files/38/Captura-de-Tela-_2800_914_2900_.png

/resized-image/__size/320x240/__key/communityserver-discussions-components-files/38/Captura-de-Tela-_2800_907_2900_.png

/resized-image/__size/320x240/__key/communityserver-discussions-components-files/38/Captura-de-Tela-_2800_906_2900_.png

/resized-image/__size/320x240/__key/communityserver-discussions-components-files/38/Captura-de-Tela-_2800_905_2900_.png

/resized-image/__size/320x240/__key/communityserver-discussions-components-files/38/Captura-de-Tela-_2800_902_2900_.png

Jacob Thompson said:
Can you double check that the charge pump capacitor was correctly populated with 22nf? I've seen similar issues to this when it is accidentally populated with ex. 220nF or 2nf instead.

I reviewed and checkedwith our supplier the PN of the capacitor: PN: CL10B223KB8NFNC

They bought from Mouser:

br.mouser.com/.../CL10B223KB8NFNC

Attached, one picture from our DRV8256 IC.

/resized-image/__size/135x240/__key/communityserver-discussions-components-files/38/20231212_5F00_084349.jpg

/resized-image/__size/320x240/__key/communityserver-discussions-components-files/38/20231212_5F00_084415.jpg

Jacob Thompson said:
Lastly, can you post an oscilloscope capture of it trying to drive the motor? include IN1, IN2, nFAULT, and ideally a current waveform of the output.

Sure, I'll turn on the pieces and get the waveforms.

0 Jacob Thompson over 1 year ago in reply to Daniel Luis Martins

TI__Mastermind 23381 points

Hey Daniel,

Thank you for the further information! Give me another day to look into this.

Can you post less blurry pics of the top and bottom layers like this one? You should be able to post them directly on E2E by using Insert, Image/Video File.

Best,

Jacob

0 Daniel Luis Martins over 1 year ago in reply to Jacob Thompson

Prodigy 230 points

Hello Jacob,

Sure ... sorry for the images. I used Insert, but I think that they resize when I transformed to link.

IC1

PCI1

PCI2

PCI3

PCI4

PCI5

PCI6

PCI7

0 Jacob Thompson over 1 year ago in reply to Daniel Luis Martins

TI__Mastermind 23381 points

Hey Daniel,

Ahh much better, thank you for full-res images.

Have you tried this on multiple boards, and done an A/B swap to see if it is the motor driver chip that isn't working or if it is always the driver in IC5 (for example) that isn't working?

And on your layout the ground powerpad on the device does not have a good thermal connection - seehttps://www.ti.com/lit/an/slva959b/slva959b.pdf#page=9

The bottom layer appears to have thermal reliefs on the vias when it should be a solid pour, and the top layer is also broken up for some reason.

Best,

Jacob

0 Daniel Luis Martins over 1 year ago in reply to Jacob Thompson

Prodigy 230 points

Hi Jacob,

Thank you again for your feedback.

In fact, I can improve the bottom layer (in blue), making it completely solid. The top layer (in red) yes, I even tried to make it flow without breaks, but I can do better.

And I hadn't seen this manual you sent, it helped a lot.

But the strangest thing is that the IC doesn't get hot. Because even testing it without any load, it already hits the nFAULT pin and does not activate the output.

Then I press SLEEP and it resets nFAULT. But when I go to activate it, it shows this "error".

0 Jacob Thompson over 1 year ago in reply to Daniel Luis Martins

TI__Mastermind 23381 points

Hey Daniel,

Can you try it without D3 in there, just connect VM directly to power supply? And also the VAR, though from your setup I think you have already removed that from the board for testing.

I did check and confirm that your device is legitimate and not a counterfeit, so that's good.

The TI E2E™ design support forums will undergo maintenance from July 11 to July 13. If you need design support during this time, open a new support request with our customer support center.

Motor drivers

Motor drivers forum

DRV8256: Outputs are not being activated