Servo Pulse - Weatronics
11-07-2012, 02:40 PM
#1
Servo Pulse - Weatronics
I'm using Hitec servos, which have a refresh pulse rate of 20ms (50 Hz). I'm using Weatronics (Giga control version 2.3, Rx/Tx firmware 2.33), which allow setting the specific refresh rate for each servo. However the closest options are 18ms or 21ms. Which on is better for a servo that expects 20ms refresh?
Is 21ms better because at 18ms the servo may have issues keeping up? Or is 21ms worse because the servo is not updated frequently enough?
Thanks!
Is 21ms better because at 18ms the servo may have issues keeping up? Or is 21ms worse because the servo is not updated frequently enough?
Thanks!
11-07-2012, 11:56 PM
#3
My Feedback: (1)
Join Date: Feb 2002
Location: private, UNITED KINGDOM
Posts: 3,672
Likes: 0
Received 26 Likes
on
16 Posts
RE: Servo Pulse - Weatronics
ORIGINAL: fw190pilot
18mz would be the best 21mz is acually faster than the servos.
18mz would be the best 21mz is acually faster than the servos.
Edgar, either 18 or 21 will work and neither you or the servos will notice any difference. iirc the factory default is 18, I leave all mine at the default and all servos work fine.
11-08-2012, 04:25 AM
#5
My Feedback: (1)
Join Date: Feb 2002
Location: private, UNITED KINGDOM
Posts: 3,672
Likes: 0
Received 26 Likes
on
16 Posts
RE: Servo Pulse - Weatronics
ORIGINAL: proline8000
Do not use 21 . Use 18 using a higher refresh rate you could get some intermitting or what looks like uneven movement of the servo. The servo can take up to 20ms refresh rate not more.
Do not use 21 . Use 18 using a higher refresh rate you could get some intermitting or what looks like uneven movement of the servo. The servo can take up to 20ms refresh rate not more.
11-08-2012, 07:18 AM
#6
My Feedback: (1)
Join Date: Feb 2002
Location: private, UNITED KINGDOM
Posts: 3,672
Likes: 0
Received 26 Likes
on
16 Posts
RE: Servo Pulse - Weatronics
Let's put some context on these numbers and explain the misunderstandings by proline and fw190.
ms refers to the number of milli seconds between each frame of data. The higher the number, the longer the time between frames. 21ms is a longer time to wait than 18ms. So 21ms is a slower refresh rate than 18ms.
There is a common myth that good old PPM ran at 50Hz or 20ms between frames. But that's only an approximation to make for easy explanations of how radio control worked. Different tx and different brands ran at slightly different rates and servos were designed to cope with a range of refresh rates around 20ms, but they were not fixed to work at 20ms and nothing else - if they were, they would work with very very few tx! For example, Multiplex and Graupner Tx in 12 channel mode were generating PPM pulses well in excess of 20ms, the data log in my Weatronic module shows the Mpx tx is sending a frame to the module every 25ms, and every servo I used with it on ppm, whether cheap analogue or expensive digital likeJR8511 all worked perfectly at 40Hz, 25ms. Some tx such as 4 channel didn't need to wait the whole 20ms for a frame and could repeat a bit faster and servos had to work with those as well, so they can cope with refresh a bit faster than 20ms. Not much faster, but a bit faster.
Hence 18 or 21ms is fine and all my servos attached to Wea work perfectly on either.
ms refers to the number of milli seconds between each frame of data. The higher the number, the longer the time between frames. 21ms is a longer time to wait than 18ms. So 21ms is a slower refresh rate than 18ms.
There is a common myth that good old PPM ran at 50Hz or 20ms between frames. But that's only an approximation to make for easy explanations of how radio control worked. Different tx and different brands ran at slightly different rates and servos were designed to cope with a range of refresh rates around 20ms, but they were not fixed to work at 20ms and nothing else - if they were, they would work with very very few tx! For example, Multiplex and Graupner Tx in 12 channel mode were generating PPM pulses well in excess of 20ms, the data log in my Weatronic module shows the Mpx tx is sending a frame to the module every 25ms, and every servo I used with it on ppm, whether cheap analogue or expensive digital likeJR8511 all worked perfectly at 40Hz, 25ms. Some tx such as 4 channel didn't need to wait the whole 20ms for a frame and could repeat a bit faster and servos had to work with those as well, so they can cope with refresh a bit faster than 20ms. Not much faster, but a bit faster.
Hence 18 or 21ms is fine and all my servos attached to Wea work perfectly on either.
11-08-2012, 07:21 AM
#7
RE: Servo Pulse - Weatronics
Thanks for the feedback so far.
In general I have been using the default (18ms) with no problem. However I have a new setup where the servo is moving agressively back and forth. If I touch the surface, it will stop. This is frecuent, but not all th etime. I'm wondering if the refresh rate may be a factor here.
When using the slower refresh rate, the 1ms gap (21-20) means the servo is "needing" and update and not getting it... ?
When using the faster refresh rate, the 2ms (20-18) means the servo may not yet "done" with the previous move and now getting the new position too early.. ?
So early (18) is always better than late (21)?
Thanks!
In general I have been using the default (18ms) with no problem. However I have a new setup where the servo is moving agressively back and forth. If I touch the surface, it will stop. This is frecuent, but not all th etime. I'm wondering if the refresh rate may be a factor here.
When using the slower refresh rate, the 1ms gap (21-20) means the servo is "needing" and update and not getting it... ?
When using the faster refresh rate, the 2ms (20-18) means the servo may not yet "done" with the previous move and now getting the new position too early.. ?
So early (18) is always better than late (21)?
Thanks!
11-08-2012, 07:29 AM
#8
My Feedback: (1)
Join Date: Feb 2002
Location: private, UNITED KINGDOM
Posts: 3,672
Likes: 0
Received 26 Likes
on
16 Posts
RE: Servo Pulse - Weatronics
Edgar, the servo spends most of its working life "needing" an update and very little of its time getting an update. The servo data pulse will be between 0.9 and 2.1ms long. Then it has to wait approx 18ms with no data until another data pulse of 0.9-2.1ms comes along and then it sits idle for another approx 18ms. Thus changing the refresh rate from 21 ms to 18ms simply means it sits idle for 3ms less but it is still sitting idle for approx 15ms. It makes no practical difference, the servo electronics are designed for a pulse at around 20ms but with latitude either side of that.
Servo behaviour like that is much more suggestive of a faulty servo. You used the magic word "Hitec" and that's the first place I would be looking for any fault. I won't let them into my workshop let alone into my models. But that's a story for another thread.
Servo behaviour like that is much more suggestive of a faulty servo. You used the magic word "Hitec" and that's the first place I would be looking for any fault. I won't let them into my workshop let alone into my models. But that's a story for another thread.
11-08-2012, 08:38 AM
#10
RE: Servo Pulse - Weatronics
ORIGINAL: Edgar Perez
Thanks for the feedback so far.
In general I have been using the default (18ms) with no problem. However I have a new setup where the servo is moving agressively back and forth. If I touch the surface, it will stop. This is frecuent, but not all th etime. I'm wondering if the refresh rate may be a factor here.
When using the slower refresh rate, the 1ms gap (21-20) means the servo is ''needing'' and update and not getting it... ?
When using the faster refresh rate, the 2ms (20-18) means the servo may not yet ''done'' with the previous move and now getting the new position too early.. ?
So early (18) is always better than late (21)?
Thanks!
Thanks for the feedback so far.
In general I have been using the default (18ms) with no problem. However I have a new setup where the servo is moving agressively back and forth. If I touch the surface, it will stop. This is frecuent, but not all th etime. I'm wondering if the refresh rate may be a factor here.
When using the slower refresh rate, the 1ms gap (21-20) means the servo is ''needing'' and update and not getting it... ?
When using the faster refresh rate, the 2ms (20-18) means the servo may not yet ''done'' with the previous move and now getting the new position too early.. ?
So early (18) is always better than late (21)?
Thanks!
11-08-2012, 03:22 PM
#12
RE: Servo Pulse - Weatronics
ORIGINAL: HarryC
Let's put some context on these numbers and explain the misunderstandings by proline and fw190.
ms refers to the number of milli seconds between each frame of data. The higher the number, the longer the time between frames. 21ms is a longer time to wait than 18ms. So 21ms is a slower refresh rate than 18ms.
There is a common myth that good old PPM ran at 50Hz or 20ms between frames. But that's only an approximation to make for easy explanations of how radio control worked. Different tx and different brands ran at slightly different rates and servos were designed to cope with a range of refresh rates around 20ms, but they were not fixed to work at 20ms and nothing else - if they were, they would work with very very few tx! For example, Multiplex and Graupner Tx in 12 channel mode were generating PPM pulses well in excess of 20ms, the data log in my Weatronic module shows the Mpx tx is sending a frame to the module every 25ms, and every servo I used with it on ppm, whether cheap analogue or expensive digital likeJR8511 all worked perfectly at 40Hz, 25ms. Some tx such as 4 channel didn't need to wait the whole 20ms for a frame and could repeat a bit faster and servos had to work with those as well, so they can cope with refresh a bit faster than 20ms. Not much faster, but a bit faster.
Hence 18 or 21ms is fine and all my servos attached to Wea work perfectly on either.
Let's put some context on these numbers and explain the misunderstandings by proline and fw190.
ms refers to the number of milli seconds between each frame of data. The higher the number, the longer the time between frames. 21ms is a longer time to wait than 18ms. So 21ms is a slower refresh rate than 18ms.
There is a common myth that good old PPM ran at 50Hz or 20ms between frames. But that's only an approximation to make for easy explanations of how radio control worked. Different tx and different brands ran at slightly different rates and servos were designed to cope with a range of refresh rates around 20ms, but they were not fixed to work at 20ms and nothing else - if they were, they would work with very very few tx! For example, Multiplex and Graupner Tx in 12 channel mode were generating PPM pulses well in excess of 20ms, the data log in my Weatronic module shows the Mpx tx is sending a frame to the module every 25ms, and every servo I used with it on ppm, whether cheap analogue or expensive digital likeJR8511 all worked perfectly at 40Hz, 25ms. Some tx such as 4 channel didn't need to wait the whole 20ms for a frame and could repeat a bit faster and servos had to work with those as well, so they can cope with refresh a bit faster than 20ms. Not much faster, but a bit faster.
Hence 18 or 21ms is fine and all my servos attached to Wea work perfectly on either.
Roger
11-09-2012, 04:08 AM
#14
RE: Servo Pulse - Weatronics
Harryc yes you are right I was in the middle of somthing at work and was not going to post but must have ( brain fart) . The only differance you may see if you use 18ms or 21ms is when you use servo reverse the servo center may move. This is the same thing as when you use different mfg's servos on your factory systems
11-09-2012, 04:16 AM
#15
My Feedback: (1)
Join Date: Feb 2002
Location: private, UNITED KINGDOM
Posts: 3,672
Likes: 0
Received 26 Likes
on
16 Posts
RE: Servo Pulse - Weatronics
ORIGINAL: proline8000
Harryc yes you are right I was in the middle of somthing at work and was not going to post but must have ( brain fart) . The only differance you may see if you use 18ms or 21ms is when you use servo reverse the servo center may move. This is the same thing as when you use different mfg's servos on your factory systems
Harryc yes you are right I was in the middle of somthing at work and was not going to post but must have ( brain fart) . The only differance you may see if you use 18ms or 21ms is when you use servo reverse the servo center may move. This is the same thing as when you use different mfg's servos on your factory systems
