Nick Gammon

Member Since: March 11, 2009

Country: United States

  • Product BOB-10859 | about a year ago

    Ditto for DecCalc of course. It uses the same figure.

  • Product BOB-10859 | about a year ago

    Oh, yeah….you do need to have main motor power (VS) to the board. It is not enough to supply 5V to the 5V terminal. If you don’t have motor power to the board, all you’ll read is zeros….(ask me how I know this).

    I agree with the above comment. It’s no good trying to talk to the board “and apply motor power later” because it doesn’t leave reset until VS is above around 8 or 9 V (the datasheet does mention this if you hunt for it).

  • Product BOB-10859 | about a year ago

    In your dSPIN_example sketch I see an anomaly. In the file dSPIN_support.ino the function AccCalc claims to need to multiply stepsPerSecPerSec by 0.137438 but that figure is not supported by the maths:

    // The value in the ACC register is [(steps/s/s)*(tick^2)]/(2^-40) where tick is 
    //  250ns (datasheet value)- 0x08A on boot.
    // Multiply desired steps/s/s by .137438 to get an appropriate value for this register.
    // This is a 12-bit value, so we need to make sure the value is at or below 0xFFF.
    unsigned long AccCalc(float stepsPerSecPerSec)
    {
      float temp = stepsPerSecPerSec * 0.137438;
      if( (unsigned long) long(temp) > 0x00000FFF) return 0x00000FFF;
      else return (unsigned long) long(temp);
    }
    

    According to my calculations the required value would be:

    1 / (2^-40 / (250e-9)^2) = 0.068719476736
    

    Not 0.137438.

    The inverse of 0.068719476736 is 14.551915228367.

    That agrees with the datasheet (page 40) which says:

    The available range is from 14.55 to 59590 step/s2 with a resolution of 14.55 step/s2.

    And multiplying 14.551915228367 by 0x8a gives 2008.1643015146 which is what they say the default setting for the ACC and DEC registers are.

    Am I missing something? Is the datasheet wrong? Or the code wrong?

    The other calculations in that file check out, it’s just this one that doesn’t match up.

  • Product KIT-10956 | about 2 years ago

    This page does not have a link to the source code, but the retired version does. Please add the link to the source here, as it took me a while to track it down.

    Also the supplied (banana) leads don’t exactly go into the holes if you mount the board over the battery pack as shown in the photo. They just limply sit there and flop out again. A solution isn’t particularly obvious, but I mounted mine off-center so that the leads could go through the holes.

  • Product TOL-10806 | about 3 years ago

    Can you please confirm that product TOL-10806 is definitely for the 1102CML scope? The photos show the 1102C and look somewhat different to the photos on the Atten website (knobs in different places etc.).
    Compare to:
    http://www.attenelectronics.com/Uploadfile/image/customer/1400.jpg

No public wish lists :(