Quatos for 234mm Quad X? Thrust curve measurement?

The new control algorithm for AutoQuad flight controllers.

Re: Quatos for 234mm Quad X? Thrust curve measurement?

Postby Mullet » Wed Jul 22, 2015 11:12 pm

Thanks!

So to clear up confusion... I will need to use the ESC that I'm intending to make measurements with. That is understood. What I don't get is that I don't need to measure real RPM. I thought the whole point of the Thrust Curve Measurement is to do so at specified RPM intervals. Does KISS ESC allow you to specify these intervals via firmware programming tool?

I know there are hints here and there. While that is useful it, would be great to have a full how-to. I have no issue publishing this info even if it is unsupported.
Mullet
 
Posts: 181
Joined: Sun Dec 28, 2014 5:56 am
Location: Fairfield County Connecticut

Re: Quatos for 234mm Quad X? Thrust curve measurement?

Postby aBUGSworstnightmare » Fri Jul 24, 2015 6:41 am

Memos Firmware should have a function you can use for thrust measurement. It is activated by setting MOT_ESC_CALI = 2
(by the way: MOT_ESC_CALI = 1 will calibrate the endpoints for the ESCs).
It steps through the RPM range (MOT_MIN to MOT_MAX) PWM_STEPs of 100. The console window shows you some instructions --> be sure to have the M4 connected to QGC.
Measure the thrust at this points, use the spreadsheet to calculate A1 and A2 and your good to go.
aBUGSworstnightmare
 
Posts: 1460
Joined: Fri Jun 22, 2012 5:24 pm

Re: Quatos for 234mm Quad X? Thrust curve measurement?

Postby Astudillo » Sat Aug 15, 2015 8:33 pm

aBUGSworstnightmare wrote:Memos Firmware should have a function you can use for thrust measurement. It is activated by setting MOT_ESC_CALI = 2
(by the way: MOT_ESC_CALI = 1 will calibrate the endpoints for the ESCs).
It steps through the RPM range (MOT_MIN to MOT_MAX) PWM_STEPs of 100. The console window shows you some instructions --> be sure to have the M4 connected to QGC.
Measure the thrust at this points, use the spreadsheet to calculate A1 and A2 and your good to go.


Hi guys,

I just did my thrust/pwm calibration on my mn1806 and i was wondering where in the spreadsheet to put the numbers as i didn't measure the RPM?

my second question about the calibration procedure is the first measurement at 1000 µs the motor doesn't start spinning, with 1100 it does. And going from 1900 µs to 2000 µs the calibration procedure stops, so the last measurement isn't possible as well. Should i alter my PWM settings, they are now the default settings of :

Arm PWM: 975 µs
Min. PWM 1000 µs
Starting PWM: 1125 µs
Man. PWM 1950 µs

Cheers,
Ray
Astudillo
 
Posts: 139
Joined: Thu Apr 25, 2013 8:23 am
Location: Amsterdam, The Netherlands

Re: Quatos for 234mm Quad X? Thrust curve measurement?

Postby Mullet » Tue Aug 18, 2015 3:35 am

I'm sorry I can't answer your questions. I'm about to go down the rabbit hole here as well.

I haven't tried anything yet, but thought about it for a bit. I guess this is the same thing you're wondering... How does RPM correlate to PWM steps of 100µs?

I bought a tachometer to go along with the Turnigy Thrust tester to be able to try and correlate them. I also bought a simple servo tester so I can manually change the PWM signal.

I'm not sure about your settings, but this is what I'll be starting with...

MOT_ARM: 1050
MOT_START: 1065
Starting PWM: 1110
MOT_MAX: 1950

I'm in the midst of attempting to deal with measuring mass and have questions about that, which will be saved for another thread.

Also, I'll be doing a build log, which tries to illustrate a method of going the Quatos route with KISS ESCs in mind. Thus far, I've taken some basic pictures of my frame, etc. and now have to start formulating my thread.
Mullet
 
Posts: 181
Joined: Sun Dec 28, 2014 5:56 am
Location: Fairfield County Connecticut

Re: Quatos for 234mm Quad X? Thrust curve measurement?

Postby aBUGSworstnightmare » Tue Aug 18, 2015 6:23 am

Astudillo wrote:
aBUGSworstnightmare wrote:Memos Firmware should have a function you can use for thrust measurement. It is activated by setting MOT_ESC_CALI = 2
(by the way: MOT_ESC_CALI = 1 will calibrate the endpoints for the ESCs).
It steps through the RPM range (MOT_MIN to MOT_MAX) PWM_STEPs of 100. The console window shows you some instructions --> be sure to have the M4 connected to QGC.
Measure the thrust at this points, use the spreadsheet to calculate A1 and A2 and your good to go.


Hi guys,

I just did my thrust/pwm calibration on my mn1806 and i was wondering where in the spreadsheet to put the numbers as i didn't measure the RPM?

my second question about the calibration procedure is the first measurement at 1000 µs the motor doesn't start spinning, with 1100 it does. And going from 1900 µs to 2000 µs the calibration procedure stops, so the last measurement isn't possible as well. Should i alter my PWM settings, they are now the default settings of :

Arm PWM: 975 µs
Min. PWM 1000 µs
Starting PWM: 1125 µs
Man. PWM 1950 µs

Cheers,
Ray


Hi Ray,
the answer to your questions is just in front of you!
Your motor will start to spin when the PMW signal reaches the value defined by MOT_START. I'm using the values below with KISS:
MOT_ARM 1055
MOT_MAX 1950
MOT_MIN 1080
MOT_START 1095

In case of thrust measurement - with an increment of 100us - the last value you will measure is 1900, since 2000 is higher than MOT_MAX (1950).
You need to measure the RPM at each PWM step. You have no direct RPM control when using KISS ESC.
aBUGSworstnightmare
 
Posts: 1460
Joined: Fri Jun 22, 2012 5:24 pm

Re: Quatos for 234mm Quad X? Thrust curve measurement?

Postby Astudillo » Tue Aug 18, 2015 6:50 am

Thanks for the answer guys!
Astudillo
 
Posts: 139
Joined: Thu Apr 25, 2013 8:23 am
Location: Amsterdam, The Netherlands

Re: Quatos for 234mm Quad X? Thrust curve measurement?

Postby Mullet » Tue Aug 18, 2015 3:37 pm

So it sounds like the M4 acts as the servo tester with precise intervals of PWM. Then with the tachometer you'd measure the RPM at that given PWM interval. You'd also measure the thrust in grams with your thrust tester. Then you could plug the thrust value with the RPM value in the excel spreadsheet and it'd fit the required formula?
Mullet
 
Posts: 181
Joined: Sun Dec 28, 2014 5:56 am
Location: Fairfield County Connecticut

Re: Quatos for 234mm Quad X? Thrust curve measurement?

Postby okallio » Tue Aug 25, 2015 8:00 pm

I have Arduino based test rig capable of producing 8kHZ PWM pulse or normal 50-500Hz/1000-2000us servo pulse to control ESC. Output can be stepped at wanted levels etc. 10-20-30...% PWM or 1100-1200-1300us servo pulse or whatever wanted. I use BLHeli flashed ESC in closed loop mode.

So if I measure thurst on given steps say for example 10% increments. Then I feed these thrust numbers to Excel. This is clear. But just to be sure. If I don't measure RPM (which is irrelevant cause I will use same ESC in copter than to measure thrust?) is it ok to put those step (say 10,20,30,40...) to rpm column and then calculate A1/2? And then what will I use for DEFAULT_MOT_VALUE_SCAL ? I'm using aqv7.1.1852-hwv8.5-quatos-ext_esc.zip SW.

Or is this all irrelevant because A1/A2 describe some kind of least squares fit of curve that will tell approximated model of thrust data? And DEFAULT_MOT_VALUE_SCALE is intended to be used in conjunction with ESC32 to bundle FC and ESC32 to use same "range"?
okallio
 
Posts: 16
Joined: Sat Nov 29, 2014 9:33 pm

Re: Quatos for 234mm Quad X? Thrust curve measurement?

Postby chschmid » Sat Aug 29, 2015 11:26 pm

For the thrust curve you can use your tool to step up the rpm but you must measure the rpm anf thrust for ea step.
Then use the pair rpm/thrust for the excel to calculate the least square regression and generate the 2nd order function for quatos. Mot_value_scal needs to match the rpm_scale on the ESC32 side.
Quatos is requesting thrust from the motor. With this information (thrust curve & RPM scale) the thrust requested can be calculated to a specific RPM.

Cheers
Christof
chschmid
 
Posts: 1800
Joined: Wed Jul 11, 2012 7:41 pm
Location: Herrliberg, Switzerland

Re: Quatos for 234mm Quad X? Thrust curve measurement?

Postby okallio » Sun Aug 30, 2015 9:27 am

I will be using BLHeli ESCs not ESC32. Sorry I was not clear in my first posting. So there is no rpm setting there.

My actual question was that is rpm used somewhere "inside" Quatos calculations that we must know it? So does rpm range somehow affect calculations that are made in control loop? I understand this in ESC32 case and there yes because speed setting is done in manner of requesting specific rpm from ESC32. But when using some other ESC then we just rely on the fact that some PWM input to ESC produces the same output (thrust) in both test bench and copter.

So when using some other ESC than ESC32 is it necessary to know RPM?

Perhaps I will just add measurement to my rig and be done :)
okallio
 
Posts: 16
Joined: Sat Nov 29, 2014 9:33 pm

PreviousNext

Return to Quatos Adaptive Control

Who is online

Users browsing this forum: No registered users and 1 guest

cron