Post by thread:[PIC] Debonce code for HALL device

Harry H. Arends wrote: > Hi group, > > For a project I am using a 16F685 and two HALL sensors as a indexing > and station sensor device. > I know that the TLE4905L device http://www.micropik.com/PDF/TLE4935L.pdf > has a logical output but would it be > necessary to include some debounce code in the routines. To debounce what ? What would be bouncing ? > On the other hand should I use rather a input with TTL or ST. Note that the HALL sensor has an internal ST stage before the output driver.

>> For a project I am using a 16F685 and two HALL sensors as a indexing >> and station sensor device. >> I know that the TLE4905L device > > http://www.micropik.com/PDF/TLE4935L.pdf > >> has a logical output but would it be >> necessary to include some debounce code in the routines. > To debounce what ? What would be bouncing ? Hall devices do not have bounce behaviour within the IC switch itself *BUT* there may still be mechanical variations in what is being sensed which drive the sensor backwards and forwards past its trigger point and effectively cause "bouncing". This could occur eg on a rotating disk if there is vibration present. Generally magnetic sensors will exhibit a degree of inherent hysteresis due to the properties of the magnetic circuit, and will also often have inbuilt explicit hysteresis as well. As Jan-Erik notes, the IC cited apparently has this feature. > Note that the HALL sensor has an internal ST stage before > the output driver. Accordingly, deboune would be required only in extremely mechanically noisy cases. Try it in a worst case situation, or analyse the field variations that you expect to be present, and see if debounce seems necessary. Russell

Thanks for the prompt reply. After the holiday I will setup a test rig with scope to see the signals. I will report the results on the group. {Quote hidden}> -----Oorspronkelijk bericht----- > Van: spam_OUTpiclist-bouncesTakeThisOuTmit.edu [.....piclist-bouncesKILLspam@spam@mit.edu] > Namens apptech > Verzonden: zondag 28 december 2008 14:42 > Aan: Microcontroller discussion list - Public. > Onderwerp: Re: [PIC] Debonce code for HALL device > > >> For a project I am using a 16F685 and two HALL sensors as > a indexing > >> and station sensor device. > >> I know that the TLE4905L device > > > > http://www.micropik.com/PDF/TLE4935L.pdf > > > >> has a logical output but would it be > >> necessary to include some debounce code in the routines. > > > To debounce what ? What would be bouncing ? > > Hall devices do not have bounce behaviour within the IC > switch itself *BUT* there may still be mechanical variations > in what is being sensed which drive the sensor backwards and > forwards past its trigger point and effectively cause > "bouncing". This could occur eg on a rotating disk if there > is vibration present. > > Generally magnetic sensors will exhibit a degree of inherent > hysteresis due to the properties of the magnetic circuit, and > will also often have inbuilt explicit hysteresis as well. As > Jan-Erik notes, the IC cited apparently has this feature. > > > Note that the HALL sensor has an internal ST stage before > the output > > driver. > > Accordingly, deboune would be required only in extremely > mechanically noisy cases. Try it in a worst case situation, > or analyse the field variations that you expect to be > present, and see if debounce seems necessary. > > > > Russell > > -