13 thoughts on “Adding Quadrature Encoder to the Zumo Chassis”

1. Pingback: Tutorial: Ultra Low Cost 2.4 GHz Wireless Transceiver with the FRDM Board | MCU on Eclipse

2. I like it! Also thanks for posted the signal tweaking and measuring notes. I haven’t done much with encoders, this will be helpful.

   I am attempting this as well, however I intended to include the Zumo Shield and Reflectance sensor. However rather then driven the shield with an Arduino, I am using a Parallax Propeller on the Prop ASC+ board that is compatible with shields but has the 8 parallel core board.

   Biggest issues is the shield’s reflectance female header mounted on the shield is getting in the way of the. I have a couple of ideas to work around this, however not sure yet which route I will take…

   LikeLike

   Reply ↓

   • Hi Matt,
     not sure if you have seen it: I used the Zumo shield as well, see

     The Freedom Zumo Robot

     Zumo Line Following with FRDM-KL25Z

     
     The header itself was not an issue, but the fact that my board has 3.3V signal levels, but the reflectance PCB needs 5V. See above post how I have solved it. Adding the shield on top of the encoder is a problem as this means the shield is on a higher position than originally designed: you would need to extend the reflectance header, and as well think about how to extend the battery contacts so they reach the shield.
     I hope this helps, and happy zumoing 🙂

     LikeLike

     Reply ↓

3. Pingback: A new Freedom Board: FRDM-KL46Z | MCU on Eclipse

4. Pingback: Mini Sumo Robot Competition running with FRDM-KL25Z | MCU on Eclipse

5. Pingback: Mini Sumo Robot with Proximity Sensors | MCU on Eclipse

6. Pingback: Processing the Pololu Motor Shaft Encoders | MCU on Eclipse

7. Pingback: New Zumo Robot PCB’s Arrived! | MCU on Eclipse

8. Hi Erich,
   I am using the Quadrature Decoder of the FTM – however I only use the init_FTM bean and access the registers directly.
   Now I need to also get the value of the two input bits (phaseA and phaseB) for use in auto-detecting presence.
   I tried ‘bits=(GPIOB_PDIR&0x0600)>>17)&0x3;’
   to get PTB17 and PTB18 into the varible ‘bits’, but it doesn’t seem to work right (always 0)
   I can see the bits I want with the EmbSys registers window GPIO->PTB->PDIR , and they change like I want when I move the encoder, but how do I pull them into a varible?

   Brynn

   LikeLike

   Reply ↓

   • Ugg. clearly I need to and with 0x060000 and not 0x0600. but it still doesnt seem to work. How do I pull the value of those bits out without messing with the FTM quadrature setup?

     LikeLike

     Reply ↓

     • I’m using normal sampling to get the bits as GPIO bits: way easier than messing up with the FTM (I understand you have configured it as FTM?).
       Is the PDIR not a 16bit register? so ‘>>17’ would make it zero?

       LikeLike

       Reply ↓

       • The reference manual says it is a 32 bit register. it also says that if it is not configured for digital it will read 0, but I can see the bits I want toggleing like they should with the debugger – but I am having trouble getting the bits in under program control

         this is what it says:
         Port Data Input
         Reads 0 at the unimplemented pins for a particular device. Pins that are not configured for a digital
         function read 0. If the Port Control and Interrupt module is disabled, then the corresponding bit in PDIR
         does not update.

         400F_F050 Port Data Input Register (GPIOB_PDIR) 32 R 0000_0000h 50.2.5/1381

         LikeLike

       • I think you might have a problem how your GPIOB_PDIR is declared (macro)?
         Here is how I read it:
         
#define QuadCounter_GET_C1_PIN_LEFT() (McuGPIO_GetValue(C1_Left))
#define QuadCounter_GET_C1_C2_PINS_LEFT() ((QuadCounter_GET_C1_PIN_LEFT()?2:0)|(QuadCounter_GET_C2_PIN_LEFT()?1:0))


bool McuGPIO_GetValue(McuGPIO_Handle_t gpio) {
return McuGPIO_IsHigh(gpio);
}

         which calls:
         
bool McuGPIO_IsHigh(McuGPIO_Handle_t gpio) {
assert(gpio!=NULL);
McuGPIO_t *pin = (McuGPIO_t*)gpio;

         #if McuLib_CONFIG_CPU_IS_KINETIS
return GPIO_PinRead(pin->hw.gpio, pin->hw.pin)!=0;
#elif McuLib_CONFIG_CPU_IS_LPC
return GPIO_PinRead(pin->hw.gpio, pin->hw.port, pin->hw.pin)!=0;
#elif McuLib_CONFIG_CPU_IS_IMXRT
return GPIO_PinRead(pin->hw.gpio, pin->hw.pin)!=0;
#endif
return false;
}

         which uses the SDK:
         
static inline uint32_t GPIO_PinRead(GPIO_Type *base, uint32_t pin)
{
return (((base->PDIR) >> pin) & 0x01U);
}

         I have setup the pins like this:
         
static void QuadCounter_PinInit(void) {
McuGPIO_Config_t gpioConfig;

         /*
* Left: PTC16, PTC17
* Right: PTC10, PTC11
*/
McuGPIO_GetDefaultConfig(&gpioConfig);
gpioConfig.hw.gpio = GPIOC;
gpioConfig.hw.port = PORTC;
gpioConfig.hw.pin = 16;
gpioConfig.isInput = true;
C1_Left = McuGPIO_InitGPIO(&gpioConfig);
gpioConfig.hw.pin = 17;
C2_Left = McuGPIO_InitGPIO(&gpioConfig);
gpioConfig.hw.pin = 10;
C1_Right = McuGPIO_InitGPIO(&gpioConfig);
gpioConfig.hw.pin = 11;
C2_Right = McuGPIO_InitGPIO(&gpioConfig);
}


         Erich

         LikeLike

This site uses Akismet to reduce spam. Learn how your comment data is processed.