Shop our Airplanes Products Drone Products Sales
Thread Tools
May 22, 2013, 10:43 PM
Registered User
I decided to try the Arduino Uno board and got better results with the compiler. The version 18 code compiled as-is but I cannot seem to get the sync to stabilize. The camera is a Sony 600TVL Super HAD that outputs NTSC video. The OSD display looks very similar to that in post #2271. I've tried 10K, 50K and 100K pots with similar results. The USB is being used for power source (5V).

TIA for suggestions,

--Van
Last edited by van-c; May 23, 2013 at 12:49 AM.
Sign up now
to remove ads between posts
May 23, 2013, 06:25 PM
Registered User
Hey guys is there any simple way in the code to remove\disable the GPS and related text? Just point me in the right direction so I don't have to reverse engineer all the code here. I'm looking to just show pack v and mah at this time and add the GPS later if needed.
May 23, 2013, 09:34 PM
Registered User

NEO-6M GPS from Hobby King. How to make it work?


I just received the inexpensive NEO-6M module from Hobbyking- $21.99. I had the OSD working with an Adafruit breakout board, but when I attached the NEO-6M instead, I got nothing. The LED on the GPS flashes at about 1 second intervals. Testing with a simple Arduino Adafruit program shows the GPS putting out something, but not intelligible. The unit comes pre-set to 38400 baud. Can someone tell me how to make this GPS work?

Thanks

Dick
Jun 05, 2013, 08:58 AM
Antonov

Remove or disable the GPS related text


Quote:
Originally Posted by joesa3rd
Hey guys is there any simple way in the code to remove\disable the GPS and related text? Just point me in the right direction so I don't have to reverse engineer all the code here. I'm looking to just show pack v and mah at this time and add the GPS later if needed.
Hi

This is not a pro solution but it worked for me (until someone tells me how to do it properly)

I just moved the text away from the visible area. In the config tab. Change the values as follows (maybe other values for NTSC).

#if (video_system == 0)
// PAL
#define toplinetext -100
#define toplinenumbers -100
#define butlinenumbers 271
#define timer_line -100
#define gps_nmea_line -100
#define summaryline -100
#define current_calc_line 292

Now I only see volts, amp and mah.

//Antonov
www.fpvteamsweden.com
Jun 12, 2013, 10:07 PM
New Hampshire
Anyone add Sonar to this?
For Multicopter low level flying, baro or altitude is not very useful I think, but Sonar should be very useful and fun!
Jun 12, 2013, 10:21 PM
Build to Fly? FLY to BUILD!
Legot's Avatar
Quote:
Originally Posted by NHmoose
Anyone add Sonar to this?
For Multicopter low level flying, baro or altitude is not very useful I think, but Sonar should be very useful and fun!
Thats hard.
Jun 13, 2013, 07:04 AM
No fun above tree level
comode's Avatar
Well, maybe it's because i'm just a noob in electronic and C programming (actually, i wrote my first blink sketch on arduino 1 month ago ), but i tried to write my own OSD prog based on the atmega328/LM1881, and it's really a pain in the ass ! All what i can do is some flickering lines. I tried with timers, SPI and asm delay (i used your delayX functions Dennis, as i don't know asm), and i can't make a single clean lettre that doesn't flicker like hell !
So Dennis, please allow me to express my respect for your work and my thanking to share it (even if i still don't understand half of your code). u rox.
Jun 13, 2013, 05:30 PM
Build to Fly? FLY to BUILD!
Legot's Avatar
I've been thinking, and I don't think its possible with this particular hardware just because of the limits on timing that were mentioned.

On the implementation end of it you would need a gyro stabilized gimbal that's capable of 360 degree continuous motion (with continuously varying speeds) so that you can have an ultrasonic sensor or the like spinning at a constant speed on a level plane.

In software you would have to take the continuously changing range data from the Ultrasonic sensor and plot it as polar coordinates around a little stationary image of the multirotor. This polar graph would then have to spin in relation to the multirotor's orientation (from a magnetometer, which would also determine the speed of rotation for the sensor).

Overall, the hardware is a little tricky but pretty simple. The software is where it gets difficult, you would definitely need allot of power and speed to make it useable. Maybe in some other graphical OSD designed by a legend?
Jul 01, 2013, 09:02 AM
Registered User
bparini's Avatar
Quote:
Originally Posted by rbcodell
I just received the inexpensive NEO-6M module from Hobbyking- $21.99. I had the OSD working with an Adafruit breakout board, but when I attached the NEO-6M instead, I got nothing. The LED on the GPS flashes at about 1 second intervals. Testing with a simple Arduino Adafruit program shows the GPS putting out something, but not intelligible. The unit comes pre-set to 38400 baud. Can someone tell me how to make this GPS work?

Thanks

Dick
In config.h here:

// Baud-rate for GPS
#define BAUD 38400
Jul 03, 2013, 02:19 PM
Registered User
My new Flytron SimpleOSD X2 gives me a hard time to get it running. I posted my problems on the flytron forum but got no response here... I hope that someone here can give me some hints how to improve the performance of the osd.

My camera is the new fatshark cmos 600tvl camera. Summary of my problems:
  • With dim_on set to 1 no information of the osd is readable, it looks like the picture in post 2271
  • With dim_on set to 0 the image is still not very stable but readable, the problems start if there are some darker areas in the image. Then the lines start to move over the whole screen again. When the light situation changes everything is readable again.

I searched this thread for informations and learned that changing the resistor values between the two pins of the atmega and the video signal influence this behavior. Currently both resistors have a value of 100Ohm.

To better understand the problem I hooked the video signal to a scope. The problems are pretty obvious to spot. First a plot without the osd connected:



Note that the peak-to-peak voltage is ~0.7V. The sync signals for the new frame a done correctly (from what I can tell) and the LM1881 should have no troubles to recognize the frame/line starts. This situation changes with the osd connected:



The peak-to-peak voltage is now ~1.5V and the black signal is below the original sync signal level... it is not surprising that the LM1881 gets confused and starts generating spurious interrupts.

Do I have to increase or lower the resistor values to adjust the output of the osd to the camera? I don't have much experience in smd soldering so I don't want to try messing around without knowing that the result will probably be an improvement. Or do I have to do something completely different like capacitors somewhere?

I also did some improvements to the firmware of the SimpleOSD X2. It now supports digital rssi (PWM signals) from receivers like the ezuhf rx. After some short tests the results look quite good but more testing has to be done to make sure everything still works like it should. The source code is in my github repository: https://github.com/schugabe/DIY-OSD
Jul 04, 2013, 10:51 AM
Registered User
bparini's Avatar
You use PAL or NTSC?


Quote:
Originally Posted by schugabe
My new Flytron SimpleOSD X2 gives me a hard time to get it running. I posted my problems on the flytron forum but got no response here... I hope that someone here can give me some hints how to improve the performance of the osd.

My camera is the new fatshark cmos 600tvl camera. Summary of my problems:
  • With dim_on set to 1 no information of the osd is readable, it looks like the picture in post 2271
  • With dim_on set to 0 the image is still not very stable but readable, the problems start if there are some darker areas in the image. Then the lines start to move over the whole screen again. When the light situation changes everything is readable again.

I searched this thread for informations and learned that changing the resistor values between the two pins of the atmega and the video signal influence this behavior. Currently both resistors have a value of 100Ohm.

To better understand the problem I hooked the video signal to a scope. The problems are pretty obvious to spot. First a plot without the osd connected:



Note that the peak-to-peak voltage is ~0.7V. The sync signals for the new frame a done correctly (from what I can tell) and the LM1881 should have no troubles to recognize the frame/line starts. This situation changes with the osd connected:



The peak-to-peak voltage is now ~1.5V and the black signal is below the original sync signal level... it is not surprising that the LM1881 gets confused and starts generating spurious interrupts.

Do I have to increase or lower the resistor values to adjust the output of the osd to the camera? I don't have much experience in smd soldering so I don't want to try messing around without knowing that the result will probably be an improvement. Or do I have to do something completely different like capacitors somewhere?

I also did some improvements to the firmware of the SimpleOSD X2. It now supports digital rssi (PWM signals) from receivers like the ezuhf rx. After some short tests the results look quite good but more testing has to be done to make sure everything still works like it should. The source code is in my github repository: https://github.com/schugabe/DIY-OSD
Jul 04, 2013, 01:07 PM
Registered User
The fatshark camera can switch between PAL and NTSC. I have tried both and the problems where the same in both. Currently run it with NTSC and have configured the OSD accordingly (I changed the define in confdig.h to VIDEO_SYSTEM 1 and fixed the bug where video_system was lower case and not upper case).
Jul 04, 2013, 01:56 PM
Registered User
bparini's Avatar
You have the problem like this?:
When i power-on, the OSD show correctly.
When i run the motor, the OSD show shaky

1. I am use the version 0.18 moddified to support 100% the SimpleOSD X2 board. (Link here: https://www.rcgroups.com/forums/show...postcount=2254)
2. I am use NTSC system.
3. I fix my problem changing the values in the code:

#if (video_system == 0)
// PAL
#define toplinetext
#define toplinenumbers
#define butlinenumbers
#define timer_line
#define gps_nmea_line
#define summaryline
#define current_calc_line

I hope this helps.
Jul 04, 2013, 04:04 PM
Registered User
I did some more experiments and changed the dim pin to another pin and connected this pin over a resistor with the video line... then I tried different resistors and found that a value of 200 Ohm makes the picture as good as without diming enabled. The black background is not very black with this resistor but better than nothing :-D

The image still gets unstable just based on the content of the image. If the brightness is uniform distributed over the image the osd data is displayed readable. But as soon a darker spot is in the image the osd starts going crazy.

@bparini:
Thank you for your suggestions. I run the same firmware as you just with some adjustments I made. But these are not the source of the issues as I can observe the same behavior with the original firmware.
The esc/motor can be ruled out as source for this problem as I don't have a esc/motor connected to the system currently.
What camera do you use?
Jul 04, 2013, 04:15 PM
Registered User
bparini's Avatar
Will be defective the board?...

I use the 808 #16 v2 with video output.


Thread Tools

Similar Threads
Category Thread Thread Starter Forum Replies Last Post
Discussion Diy osd karl k FPV Talk 2 May 03, 2011 12:31 AM
Idea New DIY OSD? atari7202 FPV Talk 1 Mar 25, 2011 12:08 AM
Discussion see my own diy OSD video Passion Aerial Photography 7 May 27, 2010 01:28 AM
Discussion REAL homebrew DIY OSD - Check it out! jafoca FPV Talk 8 Jun 29, 2009 06:52 PM
Discussion picoOSD DIY PIC12F683 based OSD in C kbosak FPV Talk 6 Jan 31, 2008 07:47 AM