You are correct, I forgot I don't have the pin8 of arduino connected.

Pics replaced.

Jalves.

Tried today a bit, and got the same picture like your snapshots.

BUT, look what picture i have with the diyosd. I have the same issue...
I'll ask dennis what could be the problem

Some dimming is seen on your (on jalves's pictures no dimming is seen)...I guess you have to adjust the potentiometer value to get more dimming (or use a pot with a different value)..Do you require more dimming? Does it have to be completly black?
I know that the telemetry part of osd won't be seen, but will the other part of the osd (which we can see) have to be completely dimmed to black? I'm asking this because I don't like very much the completely dimmed look

Some dimming is seen on your (on jalves's pictures no dimming is seen)...I guess you have to adjust the potentiometer value to get more dimming (or use a pot with a different value)

i have a 150Ohm resistor on PIN8 which is responsible for dimming.

Do you require more dimming? Does it have to be completly black?

for debugging: YES. For the finalversion: NO

but will the other part of the osd (which we can see) have to be completely dimmed to black

Of course NOT. That whould be bad.

I'm asking this because I don't like very much the completely dimmed look

Me too.

msev: can you imagine why the chars of the diyosd, and the "bars" of my code are so ugly and not stable?

Jalves: can you please post a picture of a diyosd screenshot? Would be nice to compare.

Maybe its because of innapropriate resistor values..When using the pots there is a sweet spot that gets the most stable letters.

Originally Posted by muerzi

Jalves: can you please post a picture of a diyosd screenshot? Would be nice to compare.

My diyosd (V017):

Thanks. Looks better than mine...
Will try to adjust the poti.

Today i will try to build the OSD like the SimpleOSD by Flytron.
Maybe with the LM1881 the detection of the newline sync is better than with the diyosd osd method with the analog comparator, and i hope the letters will be more stable.

Nope,

I bought the SimpleOSD from flytron (with the video telemetry on mind...) and try the diyosd code with it.

Surprise, surprise, the characters on screen are much more unstable (almost unreadable) than with the analogue comparator of the atmel.

Well I diy-ed the version with lm1881 and in my case they are not unstable. Can you make a video of how unstable yours are?

Jalves:

Have you edited the settings in the code correctly for the diyosd?
What voltagesource are you using?
A lipo, a BEC or a stable voltagesource?

Are you sure there are no other misstakes you made?

You can also try to replace the 680k resistor with a 710k resistor on the rst pin of the lm1881

Originally Posted by muerzi

Jalves,
About the circuit you are trying.
Do you think the LM393 is fast enough to do that?
At 500kHz it should be no problem, but with e.g. 4MHz?

With a typical response time of 1.3us I think it is too slow even for 500KHz.


The the last image looks as dimming does not work (not connected or disabled by software).

1.3 uS are 769230.769230769Hz, so 769kHz.
But than there will will be a problem with the timing.
Maybe we should go to 250 kHz and use more lines...

Originally Posted by muerzi

Maybe we should go to 250 kHz and use more lines...

Or find a faster comparator. On a quick search I found LT1011 which should be good to about 2MHz.

Hey guys.

From now on, i'll be 1 week on holiday (in croatia).
I will be take my ipad with me, but can only read the post on eveninng.

Cheers

Any progress so far from anybody?

My video experimenter arrived, but i'm in croatia till saturday