HSR-8498-HB operational behaviour
HSR-8498-HB operational behaviour
by robots42
Mon Aug 25, 2008 11:36 pm
I have just tested the operation of a HSR-8498-HB servo.
There has been some mention on the RoboSavvey Forum that because the HSR-8498-HB servos are digital they only need one pulse to go to a position.
My tests indicate that this is not the case.
On test the servo was run from a Parallax BS2 which allowed complete control of the pulse length, frame rate and number of pulses.
When run on intermittent pulses these servos, unlike analog servos, can get confused and lock-up against an end stop.
Only the right ankle servo was tested.
RN right ankle
forward =2190uS
vertical =1080uS
back = 520uS
if any pulse is <440uS>=460uS and the pulse length rapidly changes to 460uS then servo switches off until >~570uS.
if pulses are less than ~30 then sometimes the servo starts up and goes beyond its set point and locks-up hard against an end stop.
Analog servos operate fine with infrequent and/or erratic frame times as long as they are >~10mS,
all that happens is that the torque drops off as the frame time increases beyond ~20mS.
The HSR-8498-HB can not be safely operated like that.
While I have been trying to make sense of the Hitec documentation and determining what is up-to-date information regarding RN I have been writing some pages and the other day I decided to put them on the web. They are on my website at
http://www.davidbuckley.net/DB/RoboNova/RoboNova.htm
My rewrite of the Template program to use paged routines is a work in progress and will allow any number or routines to be called from the Remocon controller.
David
There has been some mention on the RoboSavvey Forum that because the HSR-8498-HB servos are digital they only need one pulse to go to a position.
My tests indicate that this is not the case.
On test the servo was run from a Parallax BS2 which allowed complete control of the pulse length, frame rate and number of pulses.
When run on intermittent pulses these servos, unlike analog servos, can get confused and lock-up against an end stop.
Only the right ankle servo was tested.
RN right ankle
forward =2190uS
vertical =1080uS
back = 520uS
if any pulse is <440uS>=460uS and the pulse length rapidly changes to 460uS then servo switches off until >~570uS.
if pulses are less than ~30 then sometimes the servo starts up and goes beyond its set point and locks-up hard against an end stop.
Analog servos operate fine with infrequent and/or erratic frame times as long as they are >~10mS,
all that happens is that the torque drops off as the frame time increases beyond ~20mS.
The HSR-8498-HB can not be safely operated like that.
While I have been trying to make sense of the Hitec documentation and determining what is up-to-date information regarding RN I have been writing some pages and the other day I decided to put them on the web. They are on my website at
http://www.davidbuckley.net/DB/RoboNova/RoboNova.htm
My rewrite of the Template program to use paged routines is a work in progress and will allow any number or routines to be called from the Remocon controller.
David
