3
\$\begingroup\$

I would like to make a suit with 100 or more individually controllable Linear Resonant Actuators. Texas Instruments makes this http://www.ti.com/lit/ds/symlink/drv2603.pdf which take a PWM as an input to control the vibration intensity. I'm wondering if instead of lighting LEDs with http://www.adafruit.com/datasheets/WS2811.pdf could one of these be wired to 3 DRV2603s?

Looking at the data sheets, it seems the DRV2603 accepts a PWM between 10 and 250 khz. I'm not clear on what the WS2811 outputs. The datasheet reads 400 or 800 khz but that is operation freq. Here www.world-semi.com/en/Driver/Lighting_LED_driver_chip/WS2811/ I see 2.5khz scanning freq, so I'll assume that.

The other hurdle I see is digital input high current max of 3uA on the DRV2603 while the WS2811 has a constant current output of 18.5mA. I assume that would fry the DRV2603.

Are these totally incompatible? How would the DRV2603 treat a PWM of 2.5 khz? Would I get 25 levels of intensity instead of 255?

How does constant current work? Is there a way to make this chip output less current? Could I maybe drive an led and then throw a resistor in parallel with the LED to drive the LRA as a side-effect of lighting the LED? I've seen Running Man and I know that would look good at least.

I'm certainly open to more proper solutions but this seems appealing for the price of these components. I am planning on using an Arduino with adafruit's neopixel library if this approach proves feasible. The arduino will be used to communicate with a computer over USB serial, so if an alternate solution involves some kind of USB to lots of PWM outputs, that would be great too.

Thanks for any pointers.

\$\endgroup\$
14
  • 1
    \$\begingroup\$ Input currents are usually requirements, not limits. \$\endgroup\$ Nov 7, 2014 at 5:42
  • \$\begingroup\$ The ws2811 has open collector outputs. You can convert them by using a inverter or a npn transistor. The ws2811 also has selectable current control I believe \$\endgroup\$
    – Passerby
    Nov 7, 2014 at 17:20
  • \$\begingroup\$ Do you want to individually control the strength of each LRA, or just be able to turn it on and off? \$\endgroup\$
    – bigjosh
    May 6, 2015 at 0:38
  • \$\begingroup\$ @bigjosh I want to individually control the strength. \$\endgroup\$
    – meta-meta
    May 7, 2015 at 2:13
  • \$\begingroup\$ I do not think using WS2811s is a good solution for this project. You could instead use something like an ATTINY25 to control the LRAs. Each ATTINY could control several LRAs and also have a connection link back to a central controller. While using a DRV2603 for each LRA may make controlling them easier, I do not think it is necessary and adds to space and cost requirements. Instead, I think you could drive the LRAs from the ATTINYs using some power transistors. LMK if you are still actively working on this project and more details. \$\endgroup\$
    – bigjosh
    May 7, 2015 at 19:16

1 Answer 1

1
\$\begingroup\$

Not sure if you have found answers to all your questions. You could connect the DRV2603 to a Lithium battery supply. You need to maintain the voltage rating of the datasheet (2.5-5V) rather than connecting to a current source.

When the LRAs are vibrating you will see the DRV2603 will consume around 30-90mA to vibrate the LRA.

For vibrating multiple LRAs, you can hookup say OUT+ of 10LRAs to 10 DRV2603s and connect the rest of OUT-'es to a switch. There is an example of this one in the EVM schematic. I would get a digital switch for that one so that you can communicate with a micro-controller.

You can also post your question on the E2E TI Haptics forum too for an in-depth Haptic questions.

\$\endgroup\$
1
  • \$\begingroup\$ Thanks, I was able to get some good info from the TI Haptics forum! \$\endgroup\$
    – meta-meta
    May 13, 2015 at 20:16

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.