MultiWii OSD - MWOSD

[Aftershocker]

Hi!
Think this is not right topic to write my problem, but i am using this firmware on Minim osd 1.1. It is working well, still need more free time to play with it. First i started to use it, i was thinking it is compatible with Multiwii 2.3, but seems not, works ok with 2.2 version. When will be update for 2.3?
Ok i was using one of these chinese 5.8g transmitters and sony 700tvl (also 520tvl) cameras. In the Gui had to put or leave Pal mode on.
Then i connected minim with the transmitter and camera set that came with Fatshark predator v2 goggles. I got only grey screen. Everything wired correctly, i just unplugged video in/out from minim, put the jumper wire, and got the video, but when plugged thru minim, only grey screen. Tried to put on and off the video mode jumper on camera, nothing helps. Only thing what i understand, that minimosd just do not like fatshark camera video signal. Seems very strange. Could not find in google similar problems.

[Aftershocker]

Reply to my question! There is not problem with Fatshark camera, but with transmitter! Connected Fatshark camera - minimOsd - simple Chinese transmitter and everything works fine. So there Fatshark transmitter does not like MinimOsd. Been reading there are something to do with white balance in the minim code. But not sure where and what to do.

[kataventos]

On Max sketch there are some defines you can use to lower the white level:

Defines you can use:

Code: Select all

#define WHITE_level_80 0x03
#define WHITE_level_90 0x02
#define WHITE_level_100 0x01
#define WHITE_level_120 0x00

Where to change:

Code: Select all

// set all rows to same charactor white level, 90%
  digitalWrite(MAX7456SELECT,LOW);
  
  uint8_t x;
  for(x = 0; x < MAX_screen_rows; x++) {
    MAX7456_Send(MAX7456ADD_RB0+x, WHITE_level_120); // Here
  }

Cheers,
-KV

[Phrank916]

Correction option for baseflight added to config.h

I'm new around here but I've been following this thread because I love this fork of minimOSD and KVTeam, the GUI is especially nice. Thank you, shikra, for the hard work and effort that goes in to making it awesome.

I'm using the OSD with a naze 32 full and the baseflight standard firmware. I've been watching for the fix mentioned in the quote above to be made public, but it was wondering if I could possibly get the information for the config.h or help out with chasing down the issue. The "DISARMED" in the compass isn't a deal breaker, just a minor annoyance, but as far as I can tell so far, it's almost perfect!

Let me know how I can help. I'm pretty familiar with AVR and Arduino, so I hope I can be of assistance. Thanks!

Phrank

[Phrank916]

FYI. FatShark PilotHD camera also doens't like MAX chip. I have white lines behind text with pilothd camera and when I've add throttle...text was flickering. Now I'm using pz0420H camera and everything is ok.

I was getting the flickering on throttle up until I added the ground line from the 12V camera power source to the ground pin on the video in for the OSD board. The flickering is almost always a ground issue. The max chip is very sensitive to having a common ground with the camera and VTx.

Phrank

[shikra]

@phrank

if (cmdMSP==MSP_ATTITUDE)
Find following in serial.ino and try adding the -180 bit:
{
for(uint8_t i=0;i<2;i++)
MwAngle[i] = read16();
MwHeading = read16();
MwHeading = MwHeading - 180;
read16();
}

[shikra]

@kv

thanks - I have just bought same cam to test and its very dull. I assumed at quick look had to increase for brighter. I'll try reducing.
I have changed soem of the other max chip registers and it definitely improves font sharpness and brightness for the hi res cams. It seems usable to me
pending flight test...

[shikra]

@aftershocker - for your issue - have you checked pinouts on your chinese cam is same as IRC ones ... Mine are not same.... especially the power pins

[Aftershocker]

@aftershocker - for your issue - have you checked pinouts on your chinese cam is same as IRC ones ... Mine are not same.... especially the power pins

Thanks for reply! With pinouts everything is ok, chinese cam have 12v supply and fatsharks have 5v. Otherwise i had something burned . Also i power everything from one battery, all the grounding are triple checked and ok. Just chinese transmitter works fine both cameras and minimosd, fatshark transmitter also works fine with both cameras, but not with minimosd As soon as i unplug video in/out from minimosd and connect together i have video feed, this is simply to check.
Need to play with those white balance settings.

[shikra]

@aftershocker... my only thoughts are - overlooked something in wiring (doesn't sound like it) or maybe if using the 5v to power the output stage of the minim its too much power for that + 5v cam?

[shikra]

For any testers, the repository has been updated with some of the recent changes from feedback here. The GUI remain the same so no need to download and compile that.
I have test flown and it looks OK, but not extensive testing - so take care...
RSSI and current only tested in simulator... really want some feedback here.


Changelog: R1.a1.b0
+ Filter changed to 8 point ring + optional 2nd stage expo filter
+ All sensors now filtered - Battery / Video / Current / RSSI PWM RSSI
+ MWII sensor data remains as from FC
+ Vid voltage divider moved to "approx" same as main voltage
+ Vidvoltage added to OSD menu - with visual display
+ Zero amps adjust added to OSD menu
+ RSSI adjust max + min added to OSD menu
+ Amps and volts in the OSD menu now in Human readable form
+ Heading correction option for to support Baseflight & others added to config.h file
+ Added the plane support changes into a #define for config.h
+ Freed up memory for future additions
+ I2C GPS distance fix - for OSD with issues at distance > 650m

[disq]

I have the font file from an OSD product. If somebody is interested to get 'inspiration' from their font and make better font/characters for this one, here's a preview.

[shikra]

@ disq - thanks. Usefull actually - it confirms what I was thinking to improve the map modes... Will take a close look for other ideas.


A quick vid from testing this morning. Nothing too exciting...

[Aftershocker]

@aftershocker... my only thoughts are - overlooked something in wiring (doesn't sound like it) or maybe if using the 5v to power the output stage of the minim its too much power for that + 5v cam?

Fatshark cam gets +5v from fatshark transmitter, but tried to power from flight controller Ubec, also works fine. Today tried all 4 possible White Balance levels in the code, nothing changed Chinese transmitter works with all possible white balance levels. So the conclusion is, that stupid Fatshark transmitter does not work with minimosd. Will have to do fpv race with Fathark goggles connected with aux video cable to chinese receiver

[subaru4wd]

For any testers, the repository has been updated with some of the recent changes from feedback here. The GUI remain the same so no need to download and compile that.
I have test flown and it looks OK, but not extensive testing - so take care...
RSSI and current only tested in simulator... really want some feedback here.


Changelog: R1.a1.b0
+ Filter changed to 8 point ring + optional 2nd stage expo filter
+ All sensors now filtered - Battery / Video / Current / RSSI PWM RSSI
+ MWII sensor data remains as from FC
+ Vid voltage divider moved to "approx" same as main voltage
+ Vidvoltage added to OSD menu - with visual display
+ Zero amps adjust added to OSD menu
+ RSSI adjust max + min added to OSD menu
+ Amps and volts in the OSD menu now in Human readable form
+ Heading correction option for to support Baseflight & others added to config.h file
+ Added the plane support changes into a #define for config.h
+ Freed up memory for future additions
+ I2C GPS distance fix - for OSD with issues at distance > 650m

Great news. Thanks for your hard work. I will get to testing this new release on the bench this weekend. I even picked up a current sensor the other day so hopefully I can get that working.

I will post back when I have results. Thanks again Shikra!

[scrat]

FYI. FatShark PilotHD camera also doens't like MAX chip. I have white lines behind text with pilothd camera and when I've add throttle...text was flickering. Now I'm using pz0420H camera and everything is ok.

I was getting the flickering on throttle up until I added the ground line from the 12V camera power source to the ground pin on the video in for the OSD board. The flickering is almost always a ground issue. The max chip is very sensitive to having a common ground with the camera and VTx.

Phrank

Thanks will try, but then I'll have ground loop. I have ground connected on 5v side of the minimosd.

[hinkel]

This OSD is very fine Thanks Shikra !

[ABL]

So the conclusion is, that stupid Fatshark transmitter does not work with minimosd.

Not sure about this. May be cable problem?
I used minim with multiwii mods with IRC600mW TX. No problems, but this was maybe 1 year ago, probably more (sold all <censored> fatshark/irc gear and happy with my decision). I do not see reason why they should not work now.

[HFMan]

Shikra,

I tried your latest firmware this weekend.

FYI- recall that I run 6S, so I still wonder why my main voltage adjust number and video voltage adjust number are not identical- they both have the same resistor voltage dividers on board (Witespy minimOSD).

1) RSSI was better- a bit easier to adjust manually. Thanks for that! I think it could use a bit more smoothing- a bit jumpy.

2) Current Sensor- busted completely. No matter what numbers I put in for zero and current adjust, it simply doesn't seem to vary at all. I have a Flytron 50A current sensor, 100 gain (1V - 10amps current). I've checked the current sensor with a multimeter, it looks good.

3) Video Voltage- I noticed this wasn't working at all- always reads zero volts. I'll have to get out my lab supply and see if something changed here.

I didn't get a chance to thru it a lot as I was having other issues and had to focus on those first. I really WISH I could get the current sensor to behave!

...Keep up the good work, it's getting better every release!

[shikra]

@ Hinkel - thats a very cool autoland.. pretty impressive

@ HFman

For RSSI - Will look at alternatives for the smoothing. Problem is if the RSSI values jump a lot, the filter starts to impact real changes.
Any chance you could connect a servo tester to OSD. Just to be sure that its the rx PWM thats varying - and not code or other issue.

For Current sensor - I'll send you some test code to see whats happening on your setup.
Can you advise what OSD hardware your are using.
For hardware current sensor - its working for me on a test bench with simulator here, but appreciate that is not the same as real world.

For Vid voltage,
I would expect divider values shoudl be approx the same now - although note resistor tolerances will mean a +/- difference

[HFMan]

Shikra- I'll put in your debug firmware a bit later today. I'm using Witespy minimOSD board (already has all the signals broken out to pins on the board and 15x dividers on the two voltage pins). Thanks for the prompt response!

[Phrank916]

@shikra

The heading fix for baseflight worked! The word "disarmed" is no longer in the compass between N and W. The problem now is the "minus 180" line added to serial.ino has literally flipped my compass 180. Is there a way to flip it back around without getting "DISARMED" back in the compass?

Phrank

[PatrikE]

In Serial.ino try to Replace

Code: Select all

#ifdef BASEFLIGHT
    MwHeading = MwHeading - 180;
#endif

with

Code: Select all

#ifdef BASEFLIGHT
    if (MwHeading >= + 180) MwHeading -= 360;
#endif

[shikra]

Thanks Patrick. Missed that one - updated code for next sync.....

[Phrank916]

Thanks Patrick. Missed that one - updated code for next sync.....

In Serial.ino try to Replace

Code: Select all

#ifdef BASEFLIGHT
    MwHeading = MwHeading - 180;
#endif

with

Code: Select all

#ifdef BASEFLIGHT
    if (MwHeading >= + 180) MwHeading -= 360;
#endif

Thanks guys! This worked. My OSD looks perfect now!

[bicycle]

Hey guys, I'm setting up this OSD with Naze32, GPS, and current sensor and I'm having the following issues

Amperage and consumed mAh stay at 0, this worked fine in KVTeam and nothing has changed in HW (let me know what you need me to test!) Changing "amps adjust" has no effect and changing "zero adjust" to anything but 0 sets the current meter to 90A all the time (I have set AMEPERAGEMAX=900 as it's a 90A sensor)

RSSI scaling appears to have changed - I am using RSSI thru MSP from naze32, it was fine in KVTeam without touching scaling at all, but not RSSI scaling is way off and I only get 0% (off) and 100% with nothing in between. RSSI calibration simply sets high and low values to 246.

Thanks!!!

[shikra]

pm sent...

[bicycle]

replied

[ToxicToast]

Shikra, thanks for putting this version together. I am running it on a naked minimosd board (no multiwii/naze) in a fixed wing airframe for a bare bones osd. I have it configured to display only the rssi, flight voltage, and the flight timer-which is perfect for my popwing. It fits my needs perfectly-thanks so much for this version!

Have one request from you: is there a way to disable the blinking of the voltage display?

Chris

[shikra]

@toxictoast:
Have you reduced voltage alarm down? 0 should disbale blink

or try changing Blink2hz =! Blink2hz; to Blink2hz =1;

@bicycle/hfman - thanks for teh PM with data. I'll take a look today

[shikra]

Repository updated with bugfix for reported issues + requests for improvements.
RSSI and current sensor should now be working properly again

bugfix: compass heading reversed in baseflight mode
bugfix: now able to disable mode icon if required
bugfix: MWII RSSI bug fixed
bugfix: typo incorrect anal pin handling
+: added ability to hide ARMED/DISARMED status via config.H if required
+: added fastpixel support - *may* improve display for hi-res cameras 650tvl+

[subaru4wd]

Repository updated with bugfix for reported issues + requests for improvements.
RSSI and current sensor should now be working properly again

bugfix: compass heading reversed in baseflight mode
bugfix: now able to disable mode icon if required
bugfix: MWII RSSI bug fixed
bugfix: typo incorrect anal pin handling
+: added ability to hide ARMED/DISARMED status via config.H if required
+: added fastpixel support - *may* improve display for hi-res cameras 650tvl+

Wonderful! I just downloaded and tested the code in the Repo and everything works great!

I am using a HobbyKing 9ch UHF receiver with OpenLRSng 1.6 and PWM RSSI. After performing the calibration, my RSSI seems to be responding well with the Transmitter. However, the Max is set to 5, and min is 0, which only gives me 0-20-40-60-80-100%

*EDIT* fixed my PWM RSSI and now i have 122 steps of resolution.

[ToxicToast]

@toxictoast:
Have you reduced voltage alarm down? 0 should disbale blink
or try changing Blink2hz =! Blink2hz; to Blink2hz =1;

Yes, that resolved it. It hadn't occurred to me that the voltage alarm values were tied to the alarm-thanks for clearing that up.

[shikra]

@subaru
Thanks for testing. All sounds promising.

For RSSI, tHat is a really low value. I'll check I haven't messed that up somehow, but could you ask in the openers support forum what frequency the RSSI PWM runs at?

[subaru4wd]

@subaru
Thanks for testing. All sounds promising.

For RSSI, tHat is a really low value. I'll check I haven't messed that up somehow, but could you ask in the openers support forum what frequency the RSSI PWM runs at?

Sure no problem, i can get ahold of the developer tomorrow. I will check the configuration utility to see if its adjustable. I am also going to put a servo on the rssi output of my RX to see how the servo reacts to RSSI.


On a lighter note, i set up my 4ch RX with analog RSSI out to a seperate OSD and configured it accordingly and the analog RSSI is working great. Calibration gave me a max of 163 and a min of 0, and walking around the house showed remarkibly accurate results.

[subaru4wd]

could you ask in the openers support forum what frequency the RSSI PWM runs at?

Looks like the way I have it setup the PWM RSSI is at 8kHz but there is an alternate way of doing it for 50kHz servo PWM. Im thinking that is my problem, so tomorrow I will reconfigure the RX and test it.

[ABL]

bugfix: typo incorrect anal pin handling

What? 'buttfix' maybe? =)
Feature req: "fat" horizon line. I know, this can be fixed by changing fonts, but why not have this by default? Now horizon line is very thin and hard to see (if you're using "minimal" setup - AHI, voltage, RSSI - no compass, no sidebars, etc)

[subaru4wd]

Shikra,

Changing my RSSI PWM from 8kHz to 50kHz fixed the problem.

Now when I do a RSSI calibration, my max is 244 and the min is 122. Plenty of resolution now, and seems pretty accurate as I walk away with the transmitter. I will provide a better video soon.

[bicycle]

RSSI from openLRSng is 8kHz and 50 ---> Hz <--- (not kHz) servo-pwm. so one 1-2ms pulse per 20ms.

[subaru4wd]

RSSI from openLRSng is 8kHz and 50 ---> Hz <--- (not kHz) servo-pwm. so one 1-2ms pulse per 20ms.

Then somebody needs to tell KHA to update the wiki because this is what it has to say:
RSSI Injection

A RC servo channel can be replaced with RSSI information, and set to an output port to drive a servo or OSD with 50kHz servo-pwm. Note that this injection will replace that servo channel, or you can map this to an additional RC channel when in 8- or 12-channel mode, so RSSI will be on channel 9 or 13 and not overwrite a PPM channel you are using.

[bicycle]

good catch, will let them know

[Muddiman]

Hey Guys,
I need some help, I’ve been pulling my hair out on this one.
I have the following hardware:
MultiWii Pro EZ 3.0 flashed with 2.3
RCTimer 90 amp current sensor
3dr radio telemetry connected via EZ connector serial port 3
Lawmate 1.2 Ghz V2 VTX
1/3-inch Sony CCD camera
LED’s on the arms
MinimOSD V1 flashed with new MWODS code (4-17-14 http://www.flypix.co.uk/ftp ) connected via EZ serial port 0
All neatly installed in a DJI 450 frame.
I have everything working in the MinimOSD but the current sensor. Voltage and current connections from the current sensor are connected to the MultiWii board. Voltage on pin A0 and current sensor on pin A1. No matter what setting I use for calibration the current sensor reading stays static. I tested the current sensor and its functioning as it should. Any ideas?

Thanks

[LVNeptune]

Question regarding the software current emulation. Has anyone actually used it in real world. Interested to know results. How does it calibrate? Obviously you need an external current sensor hooked up to measure the current. Do you configure the current at 10%-20%-30%-etc-etc ? or do you only calibrate it at hover throttle. I would guess it needs to be configured at different %'s because each motor is different and there is no linear similarity between the different motors with amperage draws.

[Deet]

Question regarding the software current emulation. Has anyone actually used it in real world. Interested to know results. How does it calibrate? Obviously you need an external current sensor hooked up to measure the current. Do you configure the current at 10%-20%-30%-etc-etc ? or do you only calibrate it at hover throttle. I would guess it needs to be configured at different %'s because each motor is different and there is no linear similarity between the different motors with amperage draws.

DO a hover flight, take note of duration

Recharge the lipo and note the mAh put back into the pack

Divide charge by time to get current per minute

Dial that in.

Fine tune over 2-3 flights and its VERY accurate provided your hover is close to mid throttle.

[LVNeptune]

You misunderstand. I have a current sensor already so I can measure, accurately, the current at specified intervals. My question is how do you calibrate it? I haven't flashed MWOSD yet. Does anyone have screenshots of the calibration procedure?

[Deet]

measure the hover current, set that number into OSD, done

[Plüschi]

My OSD picture is quite bad, characters difficult to read, lines irregular. This is due to a small monitor with lousy resolution. I think it has 240 lines only. Is there a way to adapt MWOSD to this bad monitors? Maybe double icon size?

[shikra]

@subaru - excellent. I had a hunch that would be it Can be supported but would be a recode of the pulse measuring routine. I'd prefer to avoid as I think most rx will support standard PWM. Excelelnt - and thanks for testing / follow up.

@Pluschi - yes - agree its too fine for the low res monitors. Its great on my fatshark aviators Next release will have current font + a second thick font option

@ABL - nice one! and glad someone takes the humour As with Pluschi - next release will have the option of thicker font.

@LVneptune - exactly as Deet say - just adjust so that OSD displays same value as your ampmeter at somewhere around hover throttle...
The equation is an approx - based upon testing of a number of motors / props + some data I downloaded from the web. If calibrate around the hover point, as this is is an avergae, over time the accuracy is pretty good. Use it as a guide until gain confidence. Change of prop / weight will require different calibration...

--

[shikra]

@Muddiman - current from multiwii board is not in the current code - it looks easy to add - if I put something together next week could you test for me?

Edit - just to add - note if you connect via MultiWii FC, it only sends the maH, so not possible to read directly the current draw... better to connect to the OSD if you can

[HFMan]

Shikra,

Got to fly the latest firmware today- looking great! Both my main 6S voltage and 12V video voltage are measuring perfect, and the Flytron current sensor is just about dead on now. I was having so much fun I wasn't watching the RSSI much, but was never more than half mile out so I'm sure it was fine. The only thing I did notice is the arrow that is supposed to point towards home doesn't seem accurate much at all- like there is a large amount of lag in the GPS data. I presume that is more something in the Naze32 and not the OSD, right?

We had my Cinetank and a high powered Flip FPV up today chasing each other doing laps around a 1/4 mile middle-school track- what a HOOT that was. The OSD was nice to watch as we were ripping through the turns.

I'll keep testing/reporting, but great job so far!