Simonk ESC vs standard (my opinion)

Post Reply
tovrin
Posts: 705
Joined: Tue Sep 20, 2011 4:08 pm

Simonk ESC vs standard (my opinion)

Post by tovrin »

put some 20a simonk flashed generic ESC's http://witespyquad.gostorego.com/speed-controllers/ready-to-fly-simonk-house-esc-20-amp.html on my quad, it had been sitting for awhile due to a single aeolian ESC flaking out.

the first thing i noticed, was the quad was MUCH quieter in flight. the second thing i noticed is how much more stable my quad is while 'loitering' I used to have a nice 3 foot perimeter around my mpu-6050 500mm quad in all directions where it would 'float', now its down to 6 or 7 inches (though my Z axis seems about the same, given i dont have a barometer i find that fair)

the rate will need tuned now, where I would get some jerky response no matter how soft i push the stick, now i can nudge the stick and get minimal movement out of the quad, but its so soft that i have to really push the stick to get it to move much at all..

I'm very impressed! oh, and this was all done on crappy chipped props, going to pick up some fresh props today.

copterrichie
Posts: 2261
Joined: Sat Feb 19, 2011 8:30 pm

Re: Simonk ESC vs standard (my opinion)

Post by copterrichie »

Where I saw a dramatic differences is with the usage high KV motors. I am using 1700KV motors on Mollie II and there is no way I could get those motors to work with standard ESCs. Just wish the MWC had more than 128 PWM steps.

crashlander
Posts: 506
Joined: Thu May 05, 2011 8:13 am
Location: Slovenia

Re: Simonk ESC vs standard (my opinion)

Post by crashlander »

copterrichie wrote:Where I saw a dramatic differences is with the usage high KV motors. I am using 1700KV motors on Mollie II and there is no way I could get those motors to work with standard ESCs. Just wish the MWC had more than 128 PWM steps.

Use WII-ESC from ziss_dm and add EXTENDED_MOTOR_RANGE and you will get almost full 8 bit (250 steps) on a328.

Regards
Andrej

copterrichie
Posts: 2261
Joined: Sat Feb 19, 2011 8:30 pm

Re: Simonk ESC vs standard (my opinion)

Post by copterrichie »

Thank you Andrej..

User avatar
Hamburger
Posts: 2578
Joined: Tue Mar 01, 2011 2:14 pm
Location: air
Contact:

Re: Simonk ESC vs standard (my opinion)

Post by Hamburger »

... or extend the range in simonk firmware and add EXTENDED_MOTOR_RANGE

copterrichie
Posts: 2261
Joined: Sat Feb 19, 2011 8:30 pm

Re: Simonk ESC vs standard (my opinion)

Post by copterrichie »

I questioned this before on RCG and was told the Simonk ESC has 400 steps default, so if the MWC had 250 steps, that would almost be two time the precision. Just think what that would do for Altitude holding. :)

User avatar
Hamburger
Posts: 2578
Joined: Tue Mar 01, 2011 2:14 pm
Location: air
Contact:

Re: Simonk ESC vs standard (my opinion)

Post by Hamburger »

copterrichie wrote:I questioned this before on RCG and was told the Simonk ESC has 400 steps default, so if the MWC had 250 steps, that would almost be two time the precision.

I am not following the simonK firmware thread at RCG for the most part but I posted over there what to change to make simonK firmware aware of the EXTENDED_MOTOR_RANGE incoming signal range (because my FC was configured that way already). I never went into the simonK code to analyze how exactly it would interpret/convert its input range. But since the input range is user-configurable I expect it would deal with different settings in an intelligent way to make the most out of it. It works good enough for me (but I do not engage alt.hold). If still in doubt, you could always ask SimonK.

copterrichie
Posts: 2261
Joined: Sat Feb 19, 2011 8:30 pm

Re: Simonk ESC vs standard (my opinion)

Post by copterrichie »

SimonK is the one that told me this. From 1ms to 2ms is divide into 400 segments. He said, he did this by using two registers when normal ESCs can only do 200 steps. I have no reason to doubt what he the creator stated. :)

and the added precision would not only benefit Altitude holding but overall stability as well.

tovrin
Posts: 705
Joined: Tue Sep 20, 2011 4:08 pm

Re: Simonk ESC vs standard (my opinion)

Post by tovrin »

Hamburger, to extend the range in the ESC i would have to reflash the ESC with range altered simonk firmware, correct?

User avatar
Hamburger
Posts: 2578
Joined: Tue Mar 01, 2011 2:14 pm
Location: air
Contact:

Re: Simonk ESC vs standard (my opinion)

Post by Hamburger »

yes, that is my understanding.

sim
Posts: 18
Joined: Tue Oct 18, 2011 11:40 pm

Re: Simonk ESC vs standard (my opinion)

Post by sim »

That's not what I said, I don't think. ;) In the case of my code, I measure at CPU clock speeds and scale to 16-bit PWM steps. So, 1ms typically shows up as 16,000, and 2ms shows up as 32,000, and the 16,000 values between those points is scaled to fit in POWER_RANGE (800 steps, by default, to keep it inaudible). If you calibrate it differently, it still scales with the same precision.

The newer C-based wii-esc (version 2) runs the timer at CLK/8, so it measures in .5µs steps (instead of .0625µs steps), but in theory that is way more than needed, and the use of SDM works in any previous error, so in theory there should be no fixed step limit other than what can be measured.

I've only used naze32, so I have't looked at ATmega-based multiwii stuff. The problem is the hardware PWM output?

copterrichie
Posts: 2261
Joined: Sat Feb 19, 2011 8:30 pm

Re: Simonk ESC vs standard (my opinion)

Post by copterrichie »

sim wrote:That's not what I said, I don't think. ;) In the case of my code, I measure at CPU clock speeds and scale to 16-bit PWM steps. So, 1ms typically shows up as 16,000, and 2ms shows up as 32,000, and the 16,000 values between those points is scaled to fit in POWER_RANGE (800 steps, by default, to keep it inaudible). If you calibrate it differently, it still scales with the same precision.

The newer C-based wii-esc (version 2) runs the timer at CLK/8, so it measures in .5µs steps (instead of .0625µs steps), but in theory that is way more than needed, and the use of SDM works in any previous error, so in theory there should be no fixed step limit other than what can be measured.

I've only used naze32, so I have't looked at ATmega-based multiwii stuff. The problem is the hardware PWM output?


My error, here is the post I mistaken for your answer: http://www.rcgroups.com/forums/showpost ... count=4261

Originally Posted by simonk

; NOTE: We do 16-bit PWM on timer2 at full CPU clock rate resolution, using
; tcnt2h to simulate the high byte. An input FULL to STOP range of 800 plus
; a MIN_DUTY of 63 (a POWER_RANGE of 863) gives 800 unique PWM steps at an
; about 18kHz on a 16MHz CPU clock. The output frequency is slightly lower
; than F_CPU / POWER_RANGE due to cycles used in the interrupt as TCNT2 is
; reloaded.

paddytfm
Posts: 8
Joined: Sun Aug 07, 2011 9:07 pm

Re: Simonk ESC vs standard (my opinion)

Post by paddytfm »

Has anyone observe lost of calibration on ESC after time.
It can happen at anytime and of course when you want to fly far away home.
The fact is the first couple ESC & motor the upper left esc on our quadx does not start at all.
I have changed the ESC and motor but same result.
My setting are 1000 2000 . 1100 min throttle. last 2.2 dev .simonk 30A
Has anyone any idea ,is there any esc initialisation done at beginning of program that could be different including always the first ESC and not the others.
If someone had a solution.
Otherwise great flight when all motors works fine
Paddytfm

tovrin
Posts: 705
Joined: Tue Sep 20, 2011 4:08 pm

Re: Simonk ESC vs standard (my opinion)

Post by tovrin »

my simonk exc have never lost calibration. wish i could help you.

paddytfm
Posts: 8
Joined: Sun Aug 07, 2011 9:07 pm

Re: Simonk ESC vs standard (my opinion)

Post by paddytfm »

Thank's for answer and any help
I'll try to calibrate each time i load a new upgrade.
Paddytfm

Post Reply