HobbyKing.com New Products Flash Sale
Reply
Thread Tools
Old Sep 12, 2006, 10:23 AM
Registered User
Ralph Weaver's Avatar
Indianapolis, IN USA
Joined Nov 2000
2,305 Posts
Question
ESC Throttle Auto Calibration??

OK, I've searched for a long time and couldn't find anything... Does anyone know where I can get some info or examples about throttle auto calibration? You know, where the controller figures out which way is low and adjusts the range to match the recv pulses... Thanks!
Ralph Weaver is offline Find More Posts by Ralph Weaver
Reply With Quote
Sign up now
to remove ads between posts
Old Sep 12, 2006, 03:15 PM
Registered User
UK
Joined Sep 2005
298 Posts
So you want the esc to calibrate off/on based on the min and max width of incoming pulses? Or the controller figures out which is low and which is high, ie whether to reverse or not?
There generally isn't much need... pulses range from 1 to 2 ms wide, so the esc uses a bit in the middle and expands it a bit... roughly 1.2 to 1.8ms. I believe "low" is always 1ms (roughly).
What's your application/problem ?
MatC is offline Find More Posts by MatC
Reply With Quote
Old Sep 12, 2006, 05:03 PM
Registered User
Ralph Weaver's Avatar
Indianapolis, IN USA
Joined Nov 2000
2,305 Posts
I've done several speed controls. They all are 1.2ms off to 1.8ms on. I want to do one similar to commercial units where the maximum & minimum throttle positions are constantly monitored to adjust the throttle response. For instance… if the radio goes from 1.1 to 1.7, then the controller will use that as the full range.

Also, if the initial throttle pulses are above 1.5, say 1.8, then I know the throttle channel is reversed and use 1.8 as low and 1.2 as high.

I'm a professional embedded programmer with 20+ years experience including commercially available RC products. I don't anticipate any trouble figuring it out, but there are plenty of other smart people out there - maybe I could get the benefit of their experience.

"Everyone learns from their own mistakes, but a wise man learns from the mistakes of others"
Ralph Weaver is offline Find More Posts by Ralph Weaver
Reply With Quote
Old Sep 12, 2006, 06:20 PM
Dimension Engineering
Akron, Ohio
Joined Jan 2002
1,705 Posts
Here's the gist of how we do it.

Have a function called scale. This takes three inputs, the current, max and min. It returns the duty cycle, or whatever intermediate value you want. You call this function once per update cycle.

When you get a new servo pulse, ask "is this valid and bigger than max or less than min" if so, update max and min

Start with max and min being 1300 and 1700 or so.

The trick is ignoring noise thats within the valid range for servo wveforms but outside the valid range for this particular transmitter. I leave that as an excercise for the reader.
Comatose is offline Find More Posts by Comatose
Reply With Quote
Old Sep 12, 2006, 07:13 PM
Registered User
E_ferret's Avatar
Launceston Arpt, Tasmania, Australia
Joined Jan 2004
564 Posts
Quote:
Originally Posted by Ralph Weaver
OK, I've searched for a long time and couldn't find anything... Does anyone know where I can get some info or examples about throttle auto calibration? You know, where the controller figures out which way is low and adjusts the range to match the recv pulses... Thanks!
Hi Ralph, I made a few controllers for sale about 6 years ago, where i allowed for full 100% calibration. Here is roughly how i did it; had a couple of variables ,low-water and high-water mark, preset at 1.5ms on power-on.
incoming pulses were validated to be between 0.8ms and 2.2ms, also a perfect train of 25 valid pulses was required before calibration started. You moved the throttle to full stick and the high-water mark followed. Each time a differential check was made to look for at least 0.6ms between the two markers. You then moved the throttle to idle and the low-water marker followed. when the differential was greater than 0.6ms, the calibration was partially ended, upon the throttle moving back up by at least 5 counts, the 2 markers were then fixed solid to indicate 0% and 100%. each pulse width from then on was normalized between these two markers. I did have a few problems with that, were if you waited too long to lock-in the variables a 'too low' or 'too high' glitch could move the markers outside the range of you throttle stick, only happened a couple of times but enough to 'P-M-O'. I incorporated lots of other pulse conditioning software to prevent glitches etc... Only problem was that i had to run the chip at 16MHz. I used the ISR to handle the incoming pulse width counting and also for the mark/space modulation of the motor pulse. Ran modulation at 2400Hz, with 50 throttle steps between 0% and 100%. Hope this helps

Regards, jacques
E_ferret is offline Find More Posts by E_ferret
Reply With Quote
Old Sep 13, 2006, 07:33 AM
Registered User
UK
Joined Sep 2005
298 Posts
I'd be _extremely_ wary of throttle reverse based on initial position!

Power up system with throttle accidentally set to medium... prop spins... oops, cut the throttle... prop now spins at full tilt.

Best make this one a manual option. Most transmitters can reverse ranges easily enough.
MatC is offline Find More Posts by MatC
Reply With Quote
Old Sep 13, 2006, 08:17 AM
Registered User
Ralph Weaver's Avatar
Indianapolis, IN USA
Joined Nov 2000
2,305 Posts
My personal preference is to use fixed end points and adjust the transmitter, but in this case, there is no way to modify the signal - I get what I get from the TX, no reversing. Also, in this particular case, there is no real potential for injury.

I know I've a bit vague with the actual function of my speed control, but it's a potential comercial product and I can't say too much. I can say it's not intended for the electric plane or robotics market.

The current product also has manual adjustment to adapt to the endpoints and direction of the throttle travel - I'd like to eliminate that, both for ease of use and cost.

My current product uses an MC68CH705J1A and I need to upgrade. Am going to use something from the '09QG family. Will also be able to eliminate the serial EE and the oscilator. Almost went with Atmel, but they've been having some corporate issues, made me nervous. Not a fan of the architecture of the PIC.

Quote:
Originally Posted by MatC
I'd be _extremely_ wary of throttle reverse based on initial position!

Power up system with throttle accidentally set to medium... prop spins... oops, cut the throttle... prop now spins at full tilt.

Best make this one a manual option. Most transmitters can reverse ranges easily enough.
Ralph Weaver is offline Find More Posts by Ralph Weaver
Reply With Quote
Reply


Thread Tools

Similar Threads
Category Thread Thread Starter Forum Replies Last Post
Jeti ESC and Throttle management rcFp 3D Flying 1 Aug 13, 2004 03:23 PM
Jeti BL ESC throttle input query Discharger Power Systems 5 May 28, 2004 11:53 AM
corona esc/throttle question rcaddition Electric Heli Talk 3 May 12, 2004 06:02 PM
ESC Throttle % in Gov Mode - The Big Question K.C. Electric Heli Talk 4 Jul 14, 2003 02:43 PM
9-CAP Tx setup for FMA 20 amp esc...throttle issue e-sailpilot86 Power Systems 0 Feb 19, 2002 09:51 PM