Originally Posted by Dennis Frie

The transmitting-end is the "easy part". Get a quick sketch up and running and check how much data you can get through, even with bad video-signal.

I would personally do something like this to get started:
- Make a complete white and a complete black character (to indicate 1 and 0)
- Make a "test-sketch" with characters like this:

White white white
black black black
White white white
black black black
etc.

I used something similar as you can see in the video for example at 1:16.

The normal digits receive additional 4 lines, two above and two below the readable digit. The "lines" represent 4 bits, and this is decoded on the ground by a computer with video card. But it could be as well decoded by a fast micro controller.

OSD Cockpit for FPV : 3D cockpit + HUD + 3D radar + antenna tracker + google earth (6 min 0 sec)

I know you don't plan to use a computer, but I can confirm this is very reliable as we get at least 5 consecutive images with same information if using a 5 Hz GPS (PAL=25 full images/s)

After some experimentation, the bits were encoded as bicolor bars. I used small bar with white on the left and black on the right for a 0 bit, and black-white for a 1 bit. This is more reliable than just black or just white, it allows especially to be sure that a valid bit has been read (especially in case of poor video, with black lines or interferences).

I used 4 bits to encode the 10 digits + several signs (such as minus).

Originally Posted by ifrkey

Dennis,

needless to say, bravo!

That CPU has most certainly met its menace. Could CloudBuster be linked to this?

http://www.holmea.demon.co.uk/GPS/Main.htm

Thanks
In regard to the GPS-project, with the amount of GPS-chipset being produced for car-navigation, mobile-phones etc. the cost is really low and performance pretty good.

As Legot pretty much mentioned, price, size, and performance will probably all suffer. With the amount of work required, itīs not really worth the effort.

Originally Posted by Legot

With some work it could, but it would be a huge waste of time and resources.
That project was more of a proof of concept rather than a viable alternative to integrated solutions. Its really expensive too.

Legot and Dennis you are right.

After reading from start to finish your thread Dennis, yesterday, and being fortunate enough to be spared all the agonizing intervening time, that fellows here have suffered, I felt the urge to say something, albeit, pardon my enthusiasm, stupid.

I love homemade projects, and the less custom hardware I see the better, but that's me.

CloudBuster is nearing completion and if I remember correctly you still have 25% of free underutilized space. Before I make your thread below

//DIY Headtracker (Easy build, No drift, OpenSource)//

my bedtime reading for tonight, I would like to ask if support for 3 axis accelerometer, gyro and magnetometer I think, baring in mind the similarities of the two platforms would be easy, time, effort and resources considered, to implement to CloudBuster.

Wouldn't that constitute an IGS whereas by feeding data from a computer via the serial link, could simulate a virtual flight! and test the platform, software wise, to exhaustion?

I am filed with admiration and I thank you, immensely for your gifts

Originally Posted by ifrkey

Legot and Dennis you are right.

After reading from start to finish your thread Dennis, yesterday, and being fortunate enough to be spared all the agonizing intervening time, that fellows here have suffered, I felt the urge to say something, albeit, pardon my enthusiasm, stupid.

I love homemade projects, and the less custom hardware I see the better, but that's me.

CloudBuster is nearing completion and if I remember correctly you still have 25% of free underutilized space. Before I make your thread below

//DIY Headtracker (Easy build, No drift, OpenSource)//

my bedtime reading for tonight, I would like to ask if support for 3 axis accelerometer, gyro and magnetometer I think, baring in mind the similarities of the two platforms would be easy, time, effort and resources considered, to implement to CloudBuster.

Wouldn't that constitute an IGS whereas by feeding data from a computer via the serial link, could simulate a virtual flight! and test the platform, software wise, to exhaustion?

I am filed with admiration and I thank you, immensely for your gifts

No worries, all crazy ideas are welcome.

Yup, still around 25% memory left. That should be more then plenty for a few extra ideas and features. But the microcontroller is still a bit limited.

The headtracker-project is already implemented in a small version, just for tilt/roll (artificial horizon). It will need some work, higher sampling-rate etc. but its shouldn't take up much extra space, as the primary functions are already included in the code.

@Dennis

Why do you include an external 27MHz crystal in your board? I read the spec sheet and the IC has an internal 27MHz crystal.

Head Tracker ? You have the OSD with your HT infused? When? Where? I just ordered parts last night for the Head Tracker. I got a camera in the mail today. Servos are to the ready.....

And ---- Anyone have an OSD board for sale in US - - - bare is fine.

Originally Posted by bstott2

@Dennis

Why do you include an external 27MHz crystal in your board? I read the spec sheet and the IC has an internal 27MHz crystal..

It doesn't...

However, it does have internal capacitors for the pierce oscillator...

Originally Posted by bstott2

@Dennis

Why do you include an external 27MHz crystal in your board? I read the spec sheet and the IC has an internal 27MHz crystal.

Head Tracker ? You have the OSD with your HT infused? When? Where? I just ordered parts last night for the Head Tracker. I got a camera in the mail today. Servos are to the ready.....

And ---- Anyone have an OSD board for sale in US - - - bare is fine.

Where did you see that in the datasheet? I think you have missed something

While the OSD is not suppose to act as a headtracker, a lot of the functions for filtering sensor-data etc. have been implemented in the OSD as well. Itīs used for artificial horizon atm.

Originally Posted by Dennis Frie

The headtracker-project is already implemented in a small version, just for tilt/roll (artificial horizon). It will need some work, higher sampling-rate etc. but its shouldn't take up much extra space, as the primary functions are already included in the code.

Any chance you would implement the mpu6050 stuff from the osd to the headtracker ? Or if you could guide me which parts I have to copy-paste

Originally Posted by msev

Any chance you would implement the mpu6050 stuff from the osd to the headtracker ? Or if you could guide me which parts I have to copy-paste

I will not write further support for the headtracker, but you can pretty much just copy-paste the data-reading from the OSD-project as that have been written for MPU6050. Just be aware that the sample-rate are different on MAX7456 and the headtracker-project

Dennis is your BOM from farnell still legit for the last version of pcb (which i ordered): http://www.rcgroups.com/forums/showp...&postcount=103

Originally Posted by msev

Dennis is your BOM from farnell still legit for the last version of pcb (which i ordered): http://www.rcgroups.com/forums/showp...&postcount=103

Use this one instead (see post #1)
http://www.bobsquads.com/MAX7456-DIY-OSD.html

Thatīs the components I'm using with the latest design.

Originally Posted by Dennis Frie

Use this one instead (see post #1)
http://www.bobsquads.com/MAX7456-DIY-OSD.html

Thatīs the components I'm using with the latest design.

That crystal is available just from USA sick

only 5x3.2mm available in EU.

Originally Posted by msev

That crystal is available just from USA sick

only 5x3.2mm available in EU.

Iīve bought from Denmark where I live without any problems.

Originally Posted by Dennis Frie

Iīve bought from Denmark where I live without any problems.

Where in Farnell? 25 euro shipping for 1 euro part if I order from USA warehouse...And in EU warehouse they don't have it.

Would the smaller part be still ok?

This is what I use and have designed for:

http://dk.farnell.com/jsp/search/pro...sp?SKU=1841955

http://dk.farnell.com/jsp/search/pro...sp?SKU=1841946

Hmm, seems like itīs another crystal Bob have listed.. Not sure how that will fit.