Control brushed esc (electronic speed controller)

Botdocter · 2011-01-22 07:40

I am using esc's for the first time and am not entirely sure how to drive them.

I forum members told me that i can drive them with servo commands, but that doesn't seem to work.
With my servos i have to use a 4k7 resistor on the signal pin to make them work. Could it be the same thing with the esc's?

[url]Http://aesior.wordpress.com[/url]

Botdocter · 2011-01-22 08:34

Ok i have tried with resistors and the motors turn. But very weak. I can stop them by pinky.
Also they don't seem to react on changing the pulse width. In some cases they don't turn at all, in some cases it stays the same.

Anyone have any ideas?

Pics can be found on the above link!

JonnyMac · 2011-01-22 08:38

You probably don't need the resistors. It would be helpful if you posted a link to the ESC specs.

Botdocter · 2011-01-22 09:29

Descriptions:
Forward with Reverse/Brake (6-12V,2-3CELLS*LIPO.)*
Size 43 x 24 X 21(mm)
Weight* 65(g)
Operating Voltage 6-12(V) or 2-3cells Lipo
BEC 1(A)
Switching Rate 1.7(KHz)
Low Voltage Cut-off 5(V)
Suitable for most controller - Futaba/Acoms/Tamiya/JR/Hitec/GWS receiver

Suitable for most*>23T Brushed Motor*- Tamiya 540 motor.....

Specifications:
MCU Controlled. No stepped operation.
BEC (Battery Eliminator Circuit)
Low torque "Soft Start"
Brake. Reverse.
Extremely smooth throttle step
Safe “power on”
Auto shut down when signal is lost
Low Voltage Cut-off. High temperature Cut-off. High current Cut-off.
Small dimensions and weight
Includes:

50A Brushed ESC
ESC English Instruction
100% Brand New in Bag

Graham Stabler · 2011-01-22 10:04

A lot of speed controls have a start up sequence where the signal sent must be low for a given time (throttle in low position) so that may be an issue.

The fact that it turns but not much is strange, either you are not sending long enough pulses or there is some issue in the wiring whereby the current is being limited.

I assume you are driving the motor from a battery pack (it is charged isn't it!), does the BEC supply the propeller or does it have it's own supply? If you have a battery and a separate supply have you coupled the grounds? They should be coupled so a logic high from the prop translates to a logic high on the controller.

Graham

Botdocter · 2011-01-22 11:21

I believe the startup sequence is indeed the problem. Sometimes i hear a little beep.

So what does "low" throttle mean?

Is that pulsewidth 900 or 1500 as beeing the middle(0)

Graham Stabler · 2011-01-22 11:26

A beep is good as it means it is armed. The pulse widths vary between 1 and 2ms, as you say 1.5ms is the middle, for a servo that means in the neutral position but for an ESC that means half throttle, most probably you need to set it to around 1ms and wait until you hear the arming beep. BUT read the esc manual as it should detail this.

Graham

Botdocter · 2011-01-22 12:22

Yeah that's the problem! I did not come with a manual...

But this sounds quite clear. I think i know what to do now.

Thanx!

Graham Stabler · 2011-01-22 12:25

here you go:

http://www.xaic.cn/english/zzcp/13.pdf

Botdocter · 2011-01-22 14:21

Graham Stabler wrote: »

here you go:

http://www.xaic.cn/english/zzcp/13.pdf

Thanks! allthough it did not gave me a correct sollution.
This is the code that works to initialize but not correctly.

CON
_clkmode = xtal1 + pll16x
_xinfreq = 5_000_000 'Note Clock Speed for your setup!!

ServoCh1 = 6 'Select DEMO servo

VAR

OBJ
SERVO : "Servo32v7.spin"

PUB Servo32_DEMO | temp

SERVO.Start 'Start Servo handler
SERVO.Ramp '<-OPTIONAL 'Start Background Ramping

'Note: Ramping requires another COG
' If ramping is not started, then
' 'SetRamp' commands within the
' program are ignored
'
'Note: At ANY time, the 'Set' command overides
' the servo position. To 'Ramp' from the
' current position to the next position,
' you must use the 'SetRamp' command

SERVO.SetRamp(ServoCh1,1500,100) 'Pan Servo

repeat 1500000 'Do nothing here just to wait for
'background ramping to complete

SERVO.SetRamp(ServoCh1,1400,100) 'Pan Servo

repeat 1500000 'Do nothing here just to wait for
'background ramping to complete

SERVO.Set(ServoCh1,1515) 'actual speed goes from 1450 to 1550

The last pulse width is accually the one that makes the wheel turn.(1515)

The other ones, i can't change otherwise the beep won't sound.

The strange thing is that the speed variations go from 1450(= -100%) to 1550 (+100%)

any ideas??

Graham Stabler · 2011-01-22 14:46

Is it a bi directional esc? If so initialise at 1500 otherwise try at 1000

Rather than use a repeat loop for the delay use waitcnt, e.g. waitcnt(cnt+clkfreq*2) gives a 2 second delay

I don't understand this statement: "The strange thing is that the speed variations go from 1450(= -100%) to 1550 (+100%)"

Graham

Graham Stabler · 2011-01-22 14:47

Oh and put your code in code tags when posting to the forum.

[code] your code here [/code ]

(without the space before the last bracket.

Botdocter · 2011-01-22 16:15

Graham Stabler wrote: »

Is it a bi directional esc? If so initialise at 1500 otherwise try at 1000

Rather than use a repeat loop for the delay use waitcnt, e.g. waitcnt(cnt+clkfreq*2) gives a 2 second delay

I don't understand this statement: "The strange thing is that the speed variations go from 1450(= -100%) to 1550 (+100%)"

Graham

Init only works from 1400 to 1470

I meant that servos go from +- 900 to 2100. And these controllers go from 1450 to 1550

Graham Stabler · 2011-01-22 16:18

The range of 1450 to 1550 seems far too small, considering you don't think it is working how did you arrive at the upper limit of 1550?

I would try removing any ramping and then giving the controller plenty of time at each setting.

Graham

Botdocter · 2011-01-22 16:22

What do you mean with arrive? I tried without ramping. It made no difference. Also longer waiting periods did not matter.

Graham Stabler · 2011-01-22 16:39

I mean why do you think that 1550 is the upper limit of the esc.

Graham

p.s. arrive: because you came to a conclusion after a journey of discovery.

Botdocter · 2011-01-22 16:59

above 1550 it doesn't go faster... so that must be the limit. at least with this calibration.

Graham Stabler · 2011-01-22 17:00

And at the point where it does not go any faster is it significantly slower than if you had connected it straight to the battery pack?

Graham

Botdocter · 2011-01-22 17:02

oh no! it goes even faster than i expected.

Graham Stabler · 2011-01-22 18:05

I'm confused, the motor goes faster than expected through the speed control? So what is the problem?

Botdocter · 2011-01-22 18:21

That it is so sensitive. But it works this way so i'll go from here.
Thank you for your help.

Graham Stabler · 2011-01-22 18:25

I think the range over which it appears to operate is also too small. Can you test the escs on an RC transmitter?

And if you run the exact same code on a servo how does it respond?

Graham

Botdocter · 2011-01-22 20:30

With a servo it works normal. Like a servo should.

Accually i believe the range is a bit wider. 1300 to 1500 with 1400 beeing 0. So that's a range of 200. Maybe because of different initialisation.
I was planning on trying to connect it with a receiver. Lets see if that'll work.

I'll get back to you...

JasonDorie · 2011-01-22 23:08

Many speed controls are "adaptive", meaning that they figure out where the max throttle is on takeoff, because most aircraft take off at full throttle. You may be seeing this behavior when the throttle maxes out. That said, most have a range of 1200 to 1800 us for the pulse width, some from 1100 to 1900, and fewer still from 1000 to 2000. Almost all center at 1500. My experience is with brushless controllers for aircraft, so yours may differ, especially if it's for a car and has reverse.

Graham Stabler · 2011-01-23 03:26

Jason,

I've heard of speed controllers with start up sequences that require the user to push to full throttle first and then to zero to initialize but I've never heard of speed controls adapting based on the use of full throttle on take off, do you have an example of such a speed controller? I really can't see how that would work as models requiring full throttle at take off would just get full stick at take off and that would not allow adaptation to anything.

Graham

JasonDorie · 2011-01-23 12:56

Pretty much every controller requires a low throttle condition to initialize. At that point they assume the "low point" they're seeing is zero. They set their internal uppper-end value to something useable, but on the low side, like 1750us. On takeoff the user applies full throttle, and the ESC uses this newly seen high value as the "real" high end of the throttle range.

Most cheap ESCs for aircraft do this, or they'll use fixed points like 1200 and 1800 (I'm making these numbers up). Look for "automatic throttle range" when shopping. If you find one with a helicopter setting, or one that offers "fixed endpoints" or "programmable throttle endpoints", it means you can assume the values won't change, or they can be set by the user. For electronics projects, those tend to be much easier to work with.

Botdocter · 2011-01-23 13:38

Thank you for your reply. Allthough the esc's you're talking about are flight ecs's. I have a esc with brake and reverse. I'm not sure it works the same. Will try with rc transmitter later tonight.

Botdocter · 2011-01-24 08:16

Ok i hooked up the rc transmitter and got exactly the same values.
1400 is middle.

So this is the way to go.

I tried initializing within my main spin file, but that doesn't seem to work. Where in the file should i place the init script?
I placed it right under

Cognew(Drive,@drivest........

Botdocter · 2011-01-24 09:54

Could it be that it has something to do with the esc switching rate of 1.7khz?
I once read something about 2khz that the propeller uses or am i totally off here?

Control brushed esc (electronic speed controller)

Comments