Rushduino OSD - Multiwii based open software osd

shufflez · Post by **shufflez** » Wed May 16, 2012 6:58 pm

Hi katana, great to see another octo-builder! Maybe I'll PM you some experience-info, if that is OK?

Regarding the issue with video noise, what happens if you remove the GND contact on the VOLTAGE-TX and VOLTAGE-CAM?
Do you have ferite-filters and twisted/braided cables?
I have a somewhat similar setup, but my TX and Rushduino are powered directly from separate LiPo.
So one LiPo for motors and one LiPo for OSD system (BTW I only have 3S LiPo's yet). BEC from 'OSD-LiPo' powers my Flyduino FC.
Currently my CAM is still powered by flight LiPo.

katana · Post by **katana** » Thu May 17, 2012 9:32 am

Maybe I'll PM you some experience-info, if that is OK?

OK!!

if you remove the GND contact on the VOLTAGE-TX and VOLTAGE-CAM?

i try only to remove ground from 3 pin video tx=NOISE

Do you have ferite-filters and twisted/braided cables?

yes, all cable are twisted and there is 2 lc filter.

today i will make more test.

shufflez · Post by **shufflez** » Thu May 17, 2012 12:37 pm

On the topic of the new Serial protocol, Bledi teased us/me with upcoming (test) version to support this.
Any news on that?

I had too many issues with MWC v2.0, so I have updated my octo to the last dev version, which works like a charm!
But now I can't use the OSD of course

Not that I'll be needing it any time soon, but still, would be nice to have everything set-up!

I keep F5-ing the Rushduino Google code page

Bledi · Post by **Bledi** » Thu May 17, 2012 1:28 pm

yes sorry, we have problems with the new serial link but we are working on it.
For the moment all the protocol is integrate, but we have a important ghost bug and we have to find it.

Sorry for the delay

katana · Post by **katana** » Fri May 18, 2012 9:48 am

Another test, unmout osd from my octo and test only with a 3s lipo, = noise... and noise... :cry:

I have made test, with 3 cam, 2 bec 5v and 2 tx video.

The osd is broken?

Bledi · Post by **Bledi** » Fri May 18, 2012 1:12 pm

no I think your cables are not enought protect and too long.

katana · Post by **katana** » Fri May 18, 2012 3:02 pm

no I think your cables are not enought protect and too long.

But in various test i have reduced the cable, 10 cm from tx to osd and 10cm from osd to Cam. It remained exactly the same noise.

The strange: in each test that I have done, if i bypass the osd , and I connecting directly tx and cam, the image was always perfect, with cables long 10 cm to 1meter...

shufflez · Post by **shufflez** » Fri May 18, 2012 8:00 pm

Do you have a buzzer on FC board? Maybe the magnet in that causes issues?
I assume you've checked all the solder points? Does anything change with FC connected or not?
Does the board actually get enough voltage (either via serial or directly from BEC)?

Just guessing, but hoping to help out!

katana · Post by **katana** » Mon May 21, 2012 5:47 pm

shufflez wrote:Do you have a buzzer on FC board? Maybe the magnet in that causes issues?
I assume you've checked all the solder points? Does anything change with FC connected or not?
Does the board actually get enough voltage (either via serial or directly from BEC)?

Just guessing, but hoping to help out!

1) no, i have made the test on table, no magnet.
2)solder point ok
3)change nothing if Fc conected or not
4)voltage 5.0v ok

I have found the cause of the problem. 8-)

If I attach the cam the problem presents, otherwise the video is perfect.
The cam is the problem? no!

i have test various cam the problem persist.
I have done a test with the analogic exit of a compact camera, i have test pal and ntsc . The problem remain.

the video test:

https://vimeo.com/42560008

Thanks for the help!

djrm · Post by **djrm** » Mon May 21, 2012 9:36 pm

Regarding the video display problem.
It looks like the video signal is overloading the OSD somehow, perhaps you shouold attenuate the input a little. Does the OSD board have/need a termination resistor, if it were missing this could cause a signal overload.
hth David.

LenzGr · Post by **LenzGr** » Mon May 21, 2012 9:47 pm

Hard to tell from the video, but this does not look like "noise" to me, but rather a synchronization problem. Could you create a video that directly captures the OSD output?

katana · Post by **katana** » Thu May 24, 2012 9:39 am

Finally I have found the hardware to record the video.

https://vimeo.com/42752848

In the first part there is the video with the osd, from the minute (1.47) the video bypassing the osd.
To notice also the values of position and baro height...

skytor · Post by **skytor** » Tue May 29, 2012 5:27 pm

hi @all hi katana

i have exact the same vido problem

the camera is a GoPro Hero HD ... i fixed my problem on the V1.2 Board :-)

i lower the resolution ..

the problem i have only @ 1080p/30fps 960p/30fps 720p/30fps

no problem i have @ 720p/60fps an WVGA/60fps

i think the MAX have problems to Handle highres video

see Picture .. it works :-)

(yes i know the time on the right is lalalala)

Bye

katana · Post by **katana** » Wed May 30, 2012 10:06 am

Hi

skytor wrote:hi @all hi katana

i have exact the same vido problem

the camera is a GoPro Hero HD ... i fixed my problem on the V1.2 Board i lower the resolution ..

the problem i have only @ 1080p/30fps 960p/30fps 720p/30fps

no problem i have @ 720p/60fps an WVGA/60fps

i think the MAX have problems to Handle highres video

see Picture .. it works (yes i know the time on the right is lalalala)

Bye

Hi skytor!

I am not the only to have this issue...

The strange is that: the output from gopro video signal is pal or ntsc , I do not believe the record format can be affect the video processing of the MAX... :?:

Am not expert, this ask is for JohnyGab.

In any case I do not know cam where it is possible to modify the resolution in output. I use this (http://www.securitycamera2000.com/produ ... u-DNR.html)

I want to ask JohnyGab if this is hardware or software problem ? at present the osd is unusable.

regards Katana

skytor · Post by **skytor** » Wed May 30, 2012 7:36 pm

Hi Katana

i use PAL video standard ... also in software

Schachti · Post by **Schachti** » Sat Jun 02, 2012 8:52 pm

Hello, I'm from Germany and bought myself an OSD. Unfortunately I can not find a description there.

Katch · Post by **Katch** » Sat Jun 02, 2012 10:39 pm

Just ordered my GoPro HD, Fatsharks and one of these so gonna join the party.

Looking forward to quad FPV fun

kataventos · Post by **kataventos** » Sun Jun 03, 2012 3:44 am

Hi Gab,
today I tested V8 as you asked me, and as you know I use my drone to work. So... I think that V8 is not stable:

1-when I fire up with MWii2.0 it goes strait to PID adjustment; (annoying)
2-when I save settings like wide screen they just stay until I get it off
ETC... the rest is all here in previous post´s...

V7 does not have GPS feed, altitude and fancy stuff but I trust the basic few info like my BattVolt, RSSI and AH to do some verticals

to the rest I use my eyes and hands!
So... my question is, to when a stable dev? I´m ok with the multiwiidev I use, but please let me upgrade sometime soon

Cheers,
H

PS: MAX does work with 50hz and 60hz in Full HD feed! :geek:

palton · Post by **palton** » Sun Jun 03, 2012 4:26 pm

I also get pretty annoyed of the PID adjustment screen at the startup.I sure want it, but not every time i power on..

I did some initial ground testing with my v8 today. Calibration of voltage went fine, pretty near real voltage. I also hooked up an HK OSD gps (at 38400 baud), to the flyduino board i got. Running multiwii 2.0.

Artifical horizon, Flightmodes, GPS sat fix worked fine for me in the rushduino v8. It looks pretty good. I got gps fixed at 8-9 sat and i did some ground checking. I walked 100m away from starting point, where home was set. Distance from where home was set, updated all the time when i was walking. The distance seems to be correct, but the arrow didnt't move much at all. It was pointing the same direction i was walking (away from home), should have pointed in the opposit way, back to home. I guess i don't get a direction when moving so slow, and hence no working arrow.

Altitude went from -25 m to -18m, as i was walking up a small hill. A negative value. Would be great to have a altitude that can be used. a negative value is not very useful..

I will test fly my OSD this week or next weekend.

Schachti · Post by **Schachti** » Sun Jun 03, 2012 6:25 pm

how do I get the MultiWii_dev_20120528 run with the V8 for?

Bledi · Post by **Bledi** » Sun Jun 03, 2012 6:29 pm

the V8 is not compatible with the last dev version. The serial link is not the same and we try to intergrate it but we have difficulties.

palton · Post by **palton** » Sun Jun 03, 2012 6:46 pm

DId another test now, using my car. Multirotor on top of the car. Satelites fixed are between 7-9.I get speed working when driving with my car, distance is also working as it should. Compass doesn't match the direction I'm driving (updates randomly, sometimes hangs for a while), and neither the arrow pointing home. After 1km, it should have calculated correct way home, even with a low refresh rate gps. It certainly don't point home (about everywhere, except home).

Does this work for others, is the code broken or me having some problem with my setup?

kataventos · Post by **kataventos** » Sun Jun 03, 2012 10:54 pm

Bledi wrote:the V8 is not compatible with the last dev version. The serial link is not the same and we try to intergrate it but we have difficulties.

it´s not easy... you two will have to find a way with Alex, command 'O' etc became obsolete! no more 'O's

Cheers,
H

copterrichie · Post by **copterrichie** » Sun Jun 03, 2012 10:56 pm

Actually, the new protocol is Waaaaayyy better then the old one. Much less traffic and only required information is transmitted.

kataventos · Post by **kataventos** » Mon Jun 04, 2012 2:05 am

@Gab

do tou think that is possible for "us" to make some diy on the RushOSD (not used digital port´s) to communicate with a Frsky D8RII(other´s with telemetry support)? I know that we can bypass the (leveler 3.3v R232) part.

Cheers,
H

Edit: should be easy to code that ( Ithink)

and the best part is that you only have to code it once! there are lots of Frsky out there, maybe is a good Idea...

Bledi · Post by **Bledi** » Mon Jun 04, 2012 8:17 am

copterrichie wrote:Actually, the new protocol is Waaaaayyy better then the old one. Much less traffic and only required information is transmitted.

yes it's true the new protocol is really better but to intergrate it's not so easy

copterrichie · Post by **copterrichie** » Mon Jun 04, 2012 2:09 pm

This is the power behind Team Effort.

JohnyGab · Post by **JohnyGab** » Mon Jun 04, 2012 6:20 pm

Hi everyone.
I saw that many of you have problem or question, i'll try to answer to them the better I can shortly.

For those you still have question about how to power that, here is the explanation :http://vimeo.com/40696974

For the rest of you that have a problem with V8, I know about the PID at start bug, its pissing me off too!

Since the last Dev of the Multiwii, the way to get data over the serial link have changed a lot, and its requesting MAJOR modification to the OSD itself. Problem is : space, i'm currently working on that.

Before : OSD was sending O, and the mutlwii was replying a big bunch of information, the OSD take them and analyse them....and start over.
It was like the OSD sending : "Hey Multiwii board, tell me every information you have "

Now : OSD have to ask one question to get one information , and the multiwii reply the resquested information
its like the OSD saying : " Hey Multiwii, what is you roll and pitch angle" the mutlwii reply : " its 12deg, and 3 deg"
...and next is : "hey multiwii, what are Rc data , stick possition" and the multiwii reply : "1200,1400,1600.......ect"

Its far better because the OSD dont have to process unwanted information, BUT, its now 15 individual request that OSD have to manage instead of 2.

BUT, here come the best advantage : No mode modification to the multiwii code will be required!

I know that some of you are really waiting for that, and its normal. I will not leave it like that, i'm working seriously on that, But I want the next release to be STABLE

copterrichie · Post by **copterrichie** » Mon Jun 04, 2012 6:26 pm

Fantastic Team Work there Johny, keep up the good work.

skytor · Post by **skytor** » Tue Jun 05, 2012 5:07 pm

Hi Johny

i say only THUMBS UP ... great work ... and the ""PID at start bug"" yes i see it .. but .. one short arming the copter and i am on ""normal"" screen mode ... PROBLEM ? no .. it works

bye Skytor

palton · Post by **palton** » Wed Jun 06, 2012 9:36 am

My problem I was writing about earlier,it's not related to the OSD. I think the OSD works just fine with v8. It presents everything fine on screen. It sure looks real nice. I have to say that.

It's my magnetometer that's not working in my case. I haven't really used it before and tested it. It is calibrated. First i thought it was just to close cables, creating magnetic fields, but it's not. I have created very good separation and I have checkd with a compass. Momentarily the magnetometer works fine, and then goes crazy again.

I just wanted to report, so people dont think there is a non working OSD software. In my case, it's definitively the setup that isn't working correct. I have to test a bit more, get another IMU maybe.

haydent · Post by **haydent** » Mon Jun 18, 2012 11:46 pm

JohnyGab wrote:Yes and no
Its the same processor, same video driver, so it will display something. But, you sill need to modify hardware to have voltage/rsssi/current mesurement. Also some code modification sill be required to make spi communication between max7456 and atmega328p

So, directly no, but if you ask me "johny, with that hardware ...are you able to make it work" i'll have to say yes. But, shy dont you want to encourage me by buying mine ?

would it work easier with mega 256 ?

skytor · Post by **skytor** » Tue Jun 19, 2012 9:02 pm

Hi Haydent

the ATMEGA328P and the MAX7456 are fix mounted on the rushduino board .. there is nothing that can change on the hardware easy ;-)

the only problems (for me)

1) the ""bug"" on startup -> rushduino goes to the setup menu ( arm the motors an you jump to the ""fly mode" )
2) the temperature dosnt work ( V1.2 Board ) -> on screen always 0°C (- nice to have but not realy needed
3) i dont know how to setup the ampere measurement :-(

to display on the screen

the rest works fine and stable on my ATMEGA 2560 / Flyduino MEGA Flight Controller with Free IMU 0.4.3 + serial FMP04 GPS Modul

by skytor

palton · Post by **palton** » Tue Jun 19, 2012 9:40 pm

Does your arrow point home correct skytor? Curious if my setup got probs, or the OSD isn't working as it should.

My arrow isn't reliable when flying. Doesn't point to home, although home is set when gps finds its satellites, alt and distance from home set is working, locking at 7-11 sat (mostly 9-10). GPS is very slow to find the satellites, much slower than my mediatek 3329 that i use for APM. Magnetometer works and show the compass direction correctly. I haven't tried with a "good" gps yet. I got an HK "apache OSD gps" that i hooked up to MWC. I don't know refresh rate but most likely very low.

Temp works, but not reliable values. Doesn't matter to me. I got the v1.2 board.

I'm using Flyduino MEGA Flight Controller with Free IMU 0.3.5_ms with HK gps from combo kit (Apache OSD) running at 38400bps @Serial 2 on flyduino.

kataventos · Post by **kataventos** » Tue Jun 19, 2012 10:35 pm

For me things work when they work for real 100%, they do not work with this and that bug from version to version things seem to stay just almost...
For example Multiwii, it works from the beginning, the difference is that is getting better and better every day staying with the same type of hardware and architecture, we can still have it running on a 328p.
We have not reached the hardware full capacities if you know what I mean, so please do not say that except this or that it works great! It does not.

Cheers,
H

palton · Post by **palton** » Tue Jun 19, 2012 11:22 pm

Hmm, what did you mean with that? not sure I got it right.

But I cant see anyone saying it works great in this thread, just what different people (like me or skytor) got working and not working, and what matters for each individual, how important it is to get working or not. I'm interested of knowing if it's working or if i got anything incorrect with my setup (which i believe i don't have anymore).

I agree with you,regarding things doesn't work util they really do, stuff that is supposed to be stable. I don't have an expectation that Rushduino would be perfect, as it's not that mature (yet). This is clearly stated for this version. If it was stable then I would expect it to work at 100%. For now, I expect the code will work in the future. I don't expect current development release of MW to be 100% working either, but when it goes into final release as stable. Then I expect it to work fully.

kataventos · Post by **kataventos** » Tue Jun 19, 2012 11:33 pm

I´m on this (rusduino osd) from the beginning, probably one of the first´s out here, gave ideas to like your count down to RSSI min, so, probably you are not figuring out what I said.
I have one piece of hardware just like you, and against many people I always defend that Gab´s work was and is on the right direction, but it is time to fully take the best advantage of what we got...
@palton Question: what do you read on BAT2?

H

Edit: I dont know how to code, but I know how to press who knows

palton · Post by **palton** » Wed Jun 20, 2012 7:54 am

I got your point now, didn't get what you meant from the beginning as I was saying. I agree with you, considering the hw/sw have been around for a while now. I bought my hardware about when ver 8 sw was available. From that perspective it doesn't seam unreasonable, but looking at time frame back, you got a point..

I can think bugs should be closed so it actually works good, before implementing new serial protocol. So its matching MW 2.0, the current stable version.

I don't measures anything with either the Bat2 or RSSI, as I don't have any use for it. I don't need to measure any low voltage source. Pinout have changed between v1.1 and v1.2. RSSI pin is moved now on the board. I haven't tried any of those (bat2 or RSSI). Does they work?

kataventos · Post by **kataventos** » Wed Jun 20, 2012 10:46 am

palton wrote: I haven't tried any of those (bat2 or RSSI). Does they work?

Just RSSI.

skytor · Post by **skytor** » Wed Jun 20, 2012 7:22 pm

palton wrote:Does your arrow point home correct skytor? Curious if my setup got probs, or the OSD isn't working as it should.

My arrow isn't reliable when flying. Doesn't point to home, although home is set when gps finds its satellites, alt and distance from home set is working, locking at 7-11 sat (mostly 9-10). GPS is very slow to find the satellites, much slower than my mediatek 3329 that i use for APM. Magnetometer works and show the compass direction correctly. I haven't tried with a "good" gps yet. I got an HK "apache OSD gps" that i hooked up to MWC. I don't know refresh rate but most likely very low.

Temp works, but not reliable values. Doesn't matter to me. I got the v1.2 board.

I'm using Flyduino MEGA Flight Controller with Free IMU 0.3.5_ms with HK gps from combo kit (Apache OSD) running at 38400bps @Serial 2 on flyduino.

Hi Palton

YES the arrow work fine .. an YES home is set after i have a satfix to .. BUT the calibration of the mag sensor is littel bit confused ... i try to do it serveral times without reach a propper/stabel working ... after ???? 10 .. 20 ??? calibration it works .. why ?? i dont know :-(

my serial GPS need at cold start ... up to 1 minute to have a satfix .. after 90 seconds i have 7 ~ 9 satelites

kataventos · Post by **kataventos** » Wed Jun 27, 2012 1:47 am

Hi Gab,
any news?

shufflez · Post by **shufflez** » Wed Jul 04, 2012 8:36 am

Hi guys,

After awaiting patiently for support of the new serialprotocol for Rushduino OSD, i'm getting a little anxious now.
Another release candidate (stable version!) has been announced (MWC 2.1) and the progress I've seen on your google code is only for I2C GPS

Can you please let us know when we can expect (even a beta) of the new Rushduino OSD MCE (v9?).

Bledi · Post by **Bledi** » Wed Jul 04, 2012 8:41 am

Yes we are working on a new release, we had wait for a stable serial protocol before finish it.
Stay tune

JohnyGab · Post by **JohnyGab** » Wed Jul 04, 2012 3:29 pm

Some update everyone,

I've succesfully integrated the Rc Chanel and artificial horrizon as well. The major thing now is PID tuning via OSD, in fact, everything exept the display itself as to be re-coded into the osd because everything was relying to that particular old communication protocol.

For now, as I can see, there is still now "Stable" version ( I mean, not dated revision, but true new "version" of the code ) of multiwii, there is no way I want to make something that have bug that can cause problem in the display.

Please, be patient, because it will really worth the wait.

When i'll release that version, you will have NO modification to the multiwii to make, it will just work plug and play.

shufflez · Post by **shufflez** » Thu Jul 05, 2012 6:24 pm

Good to hear maestro, will be eagerly awaiting the new release!
If you need any beta-testers running the (b)leading edge Multiwii versions, please let me know!

Matt. · Post by **Matt.** » Fri Jul 06, 2012 5:38 am

Any chance you can send me a version without the pid tuning.
At least I can test the rest.

JohnyGab · Post by **JohnyGab** » Mon Jul 09, 2012 4:18 am

Here is a little video update !

https://vimeo.com/45417858

shufflez · Post by **shufflez** » Mon Jul 09, 2012 8:43 am

Can't wait Jean!

Bledi · Post by **Bledi** » Sat Jul 14, 2012 1:38 pm

Release is for few days ...
http://youtu.be/CNouIFwaM3Q

We have to test it on fly and it's ok

skytor · Post by **skytor** » Tue Jul 17, 2012 7:43 pm

Hi

MultiWii 2.1 is airborne and works fine ... now i am waiting for OSD *smile :mrgreen: