has anyone here tested the gws pg-03 gyros? i am starting to suspect that they glitch a lot... anyone here know anything about them?

Hi Jeremy,

the total weight of my DF is slightly less than 600g. It's not the problem for the Todd's Models' Prop. For my configuration, the DF can lift up at about 50 - 60% PWM duty cycle.

My configuration is as below:

Prop: EPP1045
Motor: GWS 350C Ratio 5.33:1
Microcontroller: 2 pcs of PIC16F876A
MOSFET: IRFZ44N
Gyro: NEC-TOKIN CG-L43
Battery: 1500mAH Li-Po Battery

ADI, I have a question. May I know what is the maximum voltage for the motor i'm using? Currently, I'm using the 11.1v Li-Po battery and I limit the PWM duty cycle to about 75% maximum (75% x 11.1v = 8.33v). Will it damage the motor? Or I should change the battery to 7.4v? My PWM frequency is 20KHz. At first, I'm using 1KHz for my PWM frequency. The motor will run faster but hotter and noisier. So I change the freq to 20KHz. Now the motor is not so hot, but the MOSFET is hot. I have been searching around the net for the optimum PWM frequency for the motor. But I got no luck with it. some say around 1KHz, some say must more than 10KHz, and even some are using >100KHz (Intellicopter)...

i can answer that. that will most definatly damage those motors. those motors are wound for 6v, and because these things sit in ground effect, they get nearly no effective cooling. good news is these motors are cheap, so you could probably put heat sinks on to help them last a bit, and then after maybe a half hour of runtime, replace all four motors and check to see how badly worn the motors are. look for wear on the brushes, and you can gauge about how long you can go before replacing motors. this is important because brushed motors will sometimes wear differently and cause un-balanced responses, one motor to produce more thrust than the other, etc. NEVER REPLACE JUST ONE MOTOR AT A TIME. it will be like just replacing one tire on your car. if you replace one, replace them all.

you may also want to consider re-timing your motors. if i am not mistaken,the 350c motors are being run backwards for our application. i think ADI had an article on how to retime those motors, but if you timed them correctly and tested it to make sure it is good, you may be able to increase your power output and reduce your heat output significantly.

hi ADI,TVE
I'm glad to be wellcome here.I haven't had my pics and also my website yet.
I'll soon make my own website and post some pics when my helicopter have good performances and stable.
May be i haven't calibirated the gyros and tuned the PD constant to the good values yet,so my heli is a bit unstable now.This is my first project about RC model,although i had a glow engine airplane and never fly it and the others before.My experiences in robotic make me confident to do this project and would make it autonomous like TVE did (i hope that),just only move the throtle stick so my fingers is free.
Would like to hear form all of you guys!

Hi TVE,
i would be very appriciated if you could send me your code about your DF
I use powerful DSC dsPIC30F6014 from Microchip and C30 to compile the entire program,it can hande all the maths routines,deal with floating point ,Kalman filters,averaging sampling,PD control algorithm with small amount of time (<1ms).I'm happy with its performance but it rather hot ,may be it's more hot than MOSFETs!,the MOSFETs stay cool (however there is no problem at all)
To ADI:I'm waiting for the day you add my website to your links(ofcourse,my heli must have to be very stable first )
Best regards!

if you cook up a webpage i will host it under my domain if you want until you can get an angelfire account...

A stupid question here. Can I just spray the WD-40 to the motor brush to reduce the wear off? Just for your information, I have attach a small heatsink (the heatsink for PC RAM) to the motor and it's just a little bit warm when running. I still can touch the motor with my hand for a long time..

heat sinks work wonders. i dont know about wd40, but i do know that aero car makes a brush lubricant that is really super conductive. this means less electrical brush noise, less heat, less wear, and better efficiency. http://www2.towerhobbies.com/cgi-bin...?&I=LXBVR8&P=7. i dont know about wd40. take a motor apart, measure the impedance of the brushes with wd40, and measure it without. see what happens.

Kwaiweng Wrote:
May I know what is the maximum voltage for the motor i'm using? Currently, I'm using the 11.1v Li-Po battery ...etc
[/quote]

Kwaiweng ..... I feel that 11.1V is probably too high. I don't need 11.1 V battery. See my #3 video at my website ('More Video' link) I have plenty of power with 7.4V Are you using schottky diodes on your motors for back EMF ?

Here's what I'm using:

Prop: EPP1045
Motor: GWS EPC-300C-C Ratio 5.33:1 (PWM at 1200 Hz)
Microcontroller: Atmel AT90S2313
MOSFET: IRF7456 (10V gate drive)
Gyros: NEC-TOKIN CG-L43
Battery: 7.4V 4400 mAH Li-Pol (Weight = 190 grams)

Anthrax ..... Your DF doesn't need to be perfect for pics and video here. The first video I posted had a catastrophic crash at the end. So don't worry about perfection.
The hardest thing to do is get it off the ground, even for a few seconds. 'Off the ground' is success here.

Cheers ADI

yup...i did put the schottky diode for the motor. I'm using 1N5819 (1 amp) as I can't get the 1N5822 here. I use 11.1v Li-Po battery for it cause the original DF also use the 11.1v battery and i think the motor is similar right? I have check the voltage across the motor and also the current with multimeter for different PWM frequency.

PWM freq : 1KHz
PWM duty cycle (Full throttle) : 62%
voltage across motor : 7v
current : 7.5A

PWM freq : 10KHz
PWM duty cycle (Full throttle) : 62%
voltage across motor : 6v
current : 5.5A

PWM freq : 20KHz
PWM duty cycle (Full throttle) : 62%
voltage across motor : 5.5v
current : 3.5A


So I choose 20KHz for the freq. the motor is running quite cool, just the MOSFET is hot. Maybe most of the power is dissipated by the MOSFET

I could use something like this too, but they won't ship by air (flamable)

ADI

I'm using SR560 - 60V 5A Schottky diodes. They also sell 1N5822. You can get both of them here:
http://www.sicom.co.nz
This is my local electronics store here in New Zealand and he does ship overseas also.

Cheers ADI

Will it make much different after changing the diode? cause the shipping to my country is not cheap. How long is your flight time? mine one is just a few minutes only, i think is less than 10 minutes and it's not continuous.

It may not make much change, but 1N5819 just seemed a bit low, current wise.
I'm using 4400 mah lipol so will have more flying time than 10 minutes.
I haven't tested total battery flying time yet. Have you checked out the shipping cost at http://www.sicom.co.nz ?

ADI

Hi Kwai Weng,

If you are using Mabuchi motor (RK370SD), below is the datasheet link:
http://www.mabuchi-motor.co.jp/cgi-b...AT_ID=rk_370sd

If I'm didn't remember wrongly, GWS300 or GWS350 is using RK370SD.

You can get 1N5822 easily in Sim Ling Tower (Singapore) or you can order through Farnell (Malaysia).

Below is the link, you guys might find it useful:
http://ww1.microchip.com/downloads/e...tes/00847a.pdf
(don't need to use the brakes)

Read PWM Issues (page 9) - 20khz/10khz/1khz

Keep Going! All the best!

Hi ADI,
just give me some time,i'm busy with my works and programming my heli to be better.Didn't sastify with its performances yet.It could stay at level for long time but when i use rudder stick to command it turn,then it becomes unstable and difficult to make it level again.Need more improvements!.May be Angelfire is a good place to have free website.
To spacelam,thanks for your help to get an angelfire account.
Good luck to everyone!

Rudder can be tricky to use because it's on the same stick as throttle and it's easy to accidently alter throttle.

ADI

i've also had the same experiance with the rudder. i've fond the best solution is to use rudder only as a trim, or for slow adjustments. afterall, other than to keep it straight, what do you need rudder for? this thing will fly in whatever direction you choose anyway...

also, ADI, if you cant find the conducta drops in new zealand, i am sure if you talked to customer support at tower they would ship it by boat for you. they are really nice people.

if not, here are some other retailers that might.

http://www.loystoys.com/aero-car/
http://www.happyhobby.com/hobb_html/aerocar.htm

Hi anthrax,

I will send you the code, but first I have to write it I have not implemented the Kalman filter yet. I believe that Aussie Mechatronics Guy has though...

As soon as I get the new version flying, I will start sharing my source code. I don't think it will be too long from now... maybe a couple of weeks? It will be longer than that before I implement the Kalman filter though..

tve

I'm using the 11.1v battery cause I already own 2 sets of it. One is 1500mAH and one is 1250mAH. Is there any way for me to continue using that battery without damage my motor? Can I do it electronic wise? I don't want to waste the battery and I don't have the extra budget to buy a new 7.2v battery. Or should I separate the 3 cell battery into 2 cells? But one cell will be wasted by this way...

the motors are $6, and if you lube the brushes and keep the heat sink on i am sure it will last a little while. brushed motors are rather cheap anyway, so when they go out it will only cost about $20 us to replace them all, and becuase our flight times are short that could be 4 months from now.

Hi ADI,Spaceclam
Thanks for yours advices about rudder control.But i wonder why danvel did it without problem
(movie Danvel's DSCN1886_z_turn.wmv)
To TVE:waiting four your new video!,about Kalman filter,i think i have successfully deal with it
when modified source code on autopilot webpage( tilt.c) .
Cheers!

i guess he's just got it figured out. remember, i dont know what you can do with all of your guys's fancy programming, i just know what i can do with the components i have available. inherently, a rudder turn would be very unstable, but if you've got powerful stabilization than i guess it does not quite matter...