OSD glitches on throttle

[GoogleCodeExporter]

OSD glitches on throttle

[GoogleCodeExporter]

I should add, the 'glitching' is generally that the overlay text flashes off 
briefly then comes back on, frequency ranging between 2Hz down to every few 
seconds.

The worst case I saw today the overlay was totally disrupted, barely any text 
showing, and that which was visible was out of place.

Seems it might be the video overlay chip.  Perhaps more decoupling caps 
required?

Thanks,

John

Original comment by [email protected] on 8 Apr 2012 at 9:33

[GoogleCodeExporter]

I have a BEC that makes all data disappear, when i move servos.
Not coming back till i reset MinimOSD.

I do not understand what is happening at all.

Original comment by [email protected] on 10 Apr 2012 at 9:37

[GoogleCodeExporter]

It's not a defect. Every OSD out there in the marketing always recommends the 
usage of a dedicated video-battery. That's why MinimOSD has the option of being 
feed externally and also has two stages. That's also why APM has the option of 
a separated servo rail powering (apart from the logic circuitry). Any servo 
introduces noises on the power line. The nature of this noise depends on the 
servos brand and model. It would get worse on analog servos. So, it's difficult 
to filter it by just decoupling caps. It would need filters, fair-rite, 
multiple steps decoupling caps and not necessarily would be successful. So, the 
external small battery is the best, clean and easy way to do it. Some guys have 
done it with a second source and BEC. On my own setup I'm using a 800ma battery 
and a BEC feeding the board instead with APM. The newest MinimOSD hardware 
(v1.5) will turn it even more easy, due it has a switching power instead of a 
regulator, so you can tie the power and use the second stage to feed the board 
with 12V battery, without extra heating.

Original comment by [email protected] on 10 Apr 2012 at 10:26

• Changed state: Invalid

[GoogleCodeExporter]

Thank you Sbenigno,

So after v1.5 the overheating problem is non-existent? 

That i was afraid of.

Was also afraid, that this noise problem also effects APM.
Does it?

Original comment by [email protected] on 10 Apr 2012 at 1:27

[GoogleCodeExporter]

Yes. 1.5 is an improvement of MinimOSD hardware. But I'm using my old version 
pretty fine and I'll not replace mine for now.

Regardless of using or not an OSD: if your servos noises are too high and your 
APM circuitry is getting power from the servo rail, it could interfere on 
analog devices reading. I cannot affirm that with 100% sure for all cases, but 
some interferences would be possible. Like I said, that's why APM has the 
option of isolating the servo rail from APM's circuitry power.

Original comment by [email protected] on 10 Apr 2012 at 1:51

[GoogleCodeExporter]

How do i know if i have the new 1.5 osd?

My osd has a sign v0.1 on it.

Original comment by [email protected] on 10 Apr 2012 at 5:33

[GoogleCodeExporter]

MinimOSD version 1.5 has "v1.5" on it.
This "v0.1" silk on our boards is a production typo. It should be "v1.0" 
instead.

Original comment by [email protected] on 10 Apr 2012 at 6:33

[GoogleCodeExporter]

Thank you,

Than i still have to be careful with overheating. :(

Original comment by [email protected] on 11 Apr 2012 at 8:05

[GoogleCodeExporter]

It would be helpful to capture the essence of this issue on the miniosd wiki 
pages.  As it stands right now, it warns against an independent power supply 
due to the overheating issue on the 2nd stage regulator.

Do I understand correctly that you are recommending that the minimosd power 
should be bridged (eg power both stages from the APM), but that the APM itself 
should be independently powered separately from the servo power rails?

Is there a reference on the APM2 documentation how to achieve this?

Thanks,

John

Original comment by [email protected] on 11 Apr 2012 at 9:53

[GoogleCodeExporter]

>>>>>"Do I understand correctly that you are recommending that the minimosd 
power  
should be bridged (eg power both stages from the APM), but that the APM  
itself should be independently powered separately from the servo power  
rails?"

Yes. That is one line of thought, one way you could chose to go.
There are many other alternatives to try avoiding noises from servos.
That strongly depends on your setup. If you have too many things hooked up on 
your board, you will be in trouble. That's because APM has a thermal protection 
there (on the circuitry isolated from the RC system power) the current cannot 
exceed 500mA.

Please, take a look here on APM wiki at "Powering the Board": 
http://code.google.com/p/ardupilot-mega/wiki/Hardware 

Original comment by [email protected] on 11 Apr 2012 at 11:14

[GoogleCodeExporter]

Thanks for the clarification - I rewired my system in this configuration, and 
still saw glitching on throttle.

However, I have been powering my cam and VTX also from the main battery - as 
soon as I power that from an independent battery the problem disappears.  So, 
it seems that the engine noise was actually being injected into the 2nd stage 
of the minimosd via coupling on the video lines.

I'll run an independent video battery from now on.  Thanks for your input on 
this.

I agree it's not a defect in minimosd as such but a wiki page clarification 
would be really useful I think.

Original comment by [email protected] on 11 Apr 2012 at 11:18

[GoogleCodeExporter]

[deleted comment]

[GoogleCodeExporter]

What is the current draw of a GPS, magnetometer, airspeed sensor, and Minim OSD 
together? Is it too many for APM1? 

Original comment by [email protected] on 11 Apr 2012 at 11:57

[GoogleCodeExporter]

[deleted comment]

[GoogleCodeExporter]

I have a very inaccurate ammeter. It did show that my APM1 is using more than 
500ma!
GPS, magnetometer, airspeed sensor, and Minim OSD was connected.
I have found this impossible.
Some people are using the same sensor set that i do but they also have Xbee 
connected, so their current draw should be even higher than my...


Is there anyone equipped with a accurate ammeter?

Original comment by [email protected] on 12 Apr 2012 at 10:20

[GoogleCodeExporter]

Ok i have actually NEARLY the same issue, but i spotted also:
Arming my board will increase the glitches.. noticable but still readable.
it comes extreme if you power up the motors.

i will go for a secondary source, keep the noisebridge closed since i route 
only V-in and V-out. Cam and reciever are still using the mainbattery. 
hopefully a success..

if not:

secondary source, open bridge 12v from cam V-In V-Out

if not:

secondary source, open bridge, 12 TO cam V-in V.out, BEC from secondary source 
to APM1.

sidenote:
if i disconnect the camera and have grayscreen overlay, i barely see the osd 
glitching. so i thought its a sync failure maybe?

Original comment by [email protected] on 19 Jul 2013 at 3:23

[GoogleCodeExporter]

I have same problem here. When copter is initializing i see the osd ( it 
flickers a bit ) When arming and starting motors the osd telemetry dissapears .

I'm having a walkera x350 pro with ilook+ that i modded. What should i change ? 
I have a bit an electronic background but thats long time ago. Pls use simple 
exlanation pls :)

Original comment by [email protected] on 4 Jan 2015 at 1:26

[GoogleCodeExporter]

I too had terrible video flickering and loss of osd information from the 
minumosd when activating servos. Originally had digital and analog sides of 
minmosd linked and powered from apm. Uncoupling this link and providing the 
minumosd analog side with its own battery did not solve problem. After much 
trial and error I discovered that bridging the output and input sides of the 
apm with the jumper (jm1) was the problem. Removed jumper and supplied separate 
power source to input and output sides of apm solved flickering problem and 
loss of osd information. Hope this helps.

Original comment by [email protected] on 23 Feb 2015 at 6:41