DRV8833 and DRV8836

koho

Other Parts Discussed in Thread: DRV8833, DRV8836, DRV8836EVM, DRV8839

Hi,

I am using DRV8833 as LED switching using PWM. My test input is PWM which runs with 20% duty cycle and 250hz to 1khz. I used this pulse as an input-1 for DRV8833. Then the output-1 shows the pulse same as the input.

I changed DRV8833 to DRV8836 due to the size issue. Then, the output of DRV8836 is different, which means the falling time is too slow. What is the difference between?

I couldn't see any difference between them in the datasheet. I used DRV8836EVM and DRV8833polu.

I attach the wave here. The top is the DRV8833 output1 and the bottom is the DRV8836 output1. Both uses the same input file, which is similar to DRV8833 output1.

Thanks for your feedback.

over 5 years ago

0 Rick Duncan over 5 years ago

TI__Guru** 114905 points

Hi user5354909,

There should be no difference in operation between the DRV8833 and DRV8836 assuming:

1) The MODE pin of the DRV8836 is set to 0
2) The xOUT1 and xOUT2 outputs of the H bridges are controlled using the same sequence of xIN1 and xIN2 inputs

Although this appears to be the case in the images, please confirm the assumptions above.

The outputs appear to be entering Hi-z at different times.

Would you please provide scope captures of both outputs (xOUT1 and xOUT2) zoomed in when the xOUT? in the images is being disabled?

0 koho over 5 years ago in reply to Rick Duncan

Prodigy 220 points

Hi Rick,

Both assumption is applied already. I will show the AIN2/AIN1/OUT2/OUT1 in the scope, respectively.

If it goes to Hi-Z, then both should show the same timing? right? If it goes to Hi-Z, then I am okay. But, I want to double-check whether the functions of both are same.

Thanks

0 koho over 5 years ago in reply to koho

Prodigy 220 points

Hi Rick,

I am using DRV8836 EVM. Does it have any capacitor at the output? I couldn't find anything in the schematics and the EVM board.

I also have DRV8839 EVM, The pulse is clear as DRV8833. Let me know if DRV8836EVM has problems, then I have to switch to DRV8839? (I have to change my PCB then..)

Thanks

0 Rick Duncan over 5 years ago in reply to koho

TI__Guru** 114905 points

Hi Koho,

Thank you for the images.

The DRV8836 does not have a capacitor at the output.

I am trying to determine the difference, but will not be in the lab until tomorrow.

From the images, it looks like the transition from outputs enabled to Hi-Z occurs faster on the DRV8836 vs the DRV8833. That was the main reason I requested zoomed in images. Zoming in further (1us/div) should further confirm this.

What load do you have on the outputs when you took the scope shot?

Tomorrow, I plan to connect the outputs to a VM/2 (maybe a 100k/100k divider) voltage first to show when the outputs are disabled.

Then I will add a large pulldown (1M) to GND) on the output to determine if there is some additional leakage.

0 koho over 5 years ago in reply to Rick Duncan

Prodigy 220 points

Hi Rick,

I didn't load anything when I took the scope shot, so it is an opened circuit for the output. I know there is an internal pull down for the outputs. But, I also tested it with LED, which shows the same waveform.

Thank you! I am looking forward to seeing your results.

0 Rick Duncan over 5 years ago in reply to koho

TI__Guru** 114905 points

Hi koho,

I have some more information although I took the data differently than I outlined above. It is important to look at the differential voltage across the outputs to determine if there is enough voltage and current to drive the load. Below I measured each output and used the math function to determine the voltage across the outputs.

I ran the DRV8833 and DRV8836 at 4V with a 1MOhm resistor between AOUT1 and AOUT2.

The scope captures show AIN1 going low. AOUT1, AOUT2, and AOUT1 - AOUT2 is shown indicating when the device entered Hi-Z and the time constant for each device.

FYI -- There is no internal pulldown for the outputs. There is some leakage paths (+/-1uA max for the DRV8833 and +/-200nA max for the DRV8836).

The first two images are the DRV8833 zoomed out and in.

In the first image below you can see the outputs quickly enters Hi-Z and reaches 1V differential voltage in ~10us, 0V in >90us.

The next three images are the DRV8836 zoomed out and in

In the first image below you can see the outputs quickly enters Hi-Z and reaches 1V differential voltage in ~80us, 0V in ~200us. If you more load current between the pins, the voltage should collapse faster. I don't think there is enough charge to continue to drive the LED when disabled.