Jump to content
Daniel Wee

My little NAV/OSD project - under construction

Recommended Posts

Another Question:

You got the strength of the RSSI Signal in your OSD. How did you

do that? Have you got a receiver how is able to send the RSSI signal

to a Servo port?

Or have you done it different?

Geko

Share this post


Link to post
Share on other sites

I don't have RSSI in the OSD. I think you saw the GPS signal strength which is different from the RSSI. I can have RSSI but didn't want to risk injecting RFI into the receiver IF stages.

Daniel

Share this post


Link to post
Share on other sites

oops sorry so it was a different OSD...

Hmm but it would be the best thing for

flying. You could fly quite far away because

you always know how strong the singal is...

So you haven't had problemes with distance flying?

Geko

Share this post


Link to post
Share on other sites

So far I've only gone as far as about 1.2km but without any loss of control. In my case I am not nearly as concerned with loss of control now because I have the auto-pilot that will automatically kick-in in case of signal loss. The auto-pilot will navigate the plane home to me even if I turn the transmitter off.

I originally envisioned an RSSI display on the OSD but that would involve:-

1. Dismantling the receiver

2. Drawing out and buffering the RSSI line from the receiver

3. In this process, risk injecting RFI back into the receiver

It seems that the potential loss of performance outweighs any potential gain. I have read of tests where the 72MHz TX/RX performed well out to about 3 miles or so using the default configuration. Further improvements can be gained through using better TX antennas on the ground. As such, it didn't seem to me a priority to implement the RSSI display after considering all the cons. Don't forget also that the RSSI will be different for every different make of receiver so that adds another level of complication.

Daniel

Share this post


Link to post
Share on other sites

Yes that's true.

I forgot about the auto pilot. With this function

you don't need the RSSI signal, of course.

Geko

Share this post


Link to post
Share on other sites

Daniel,

Could you post the latest code hex file (2.6) and a current sensor pic/schematic (if you've got it). I decided to etch this on my own using a toner transfer method. Thanks!

Share this post


Link to post
Share on other sites

Here you go:-

NAV/OSD software 2.6

The latest version includes support for Imperial units and some other minor improvements. I'll try to get the current sensor schematics done and post that. In the mean time, I am continuing to fine tune the parameters. In particular, I think I will include the option of tweaking the accuracy of the various sensors - in particular the current sensor, in my next revision. The problem is that I don't know the repeatability of the sensor as yet. I have just built a second sensor and will soon be able to tell if it is repeatable (precision-wise that is).

I am still waiting for my second BOB-4SG to arrive and I will then test the new layout. If you're not in a hurry, give me a few days to test this new board layout and I'll be able to tell you if it works.

Daniel

Share this post


Link to post
Share on other sites

Why do I need to use FMA? My auto-pilot already works. And I totally can't see the relevance of your link at all - it sounds like a troll.

Daniel

Edited by Daniel Wee

Share this post


Link to post
Share on other sites

Ack! I just realized a blunder I made with regard to component selection. I had been using the Allegro ACS750 sensor for my current sensing, thinking that this was the most suitable part. Upon re-reading their web-page, I realize that the ACS755 is the correct part to use. Previously, the part measured 240A over the 5V full-scale, resulting in a best case of 234 mA/bit, which is dismal resolution to say the least. In practice, this works out to be 247 mA/bit or so.

Now, the ACS755SCB-050, which is what I am going to use - once I get the samples (will take about 10 days more to get here), we're talking now about (in practice - with a theoretical full-scale of 83.3A) a resolution of 81 mA/bit. Even if I were to use a the 100A part (ACS755xCB-100) we're still talking about roughly 162 mA/bit - far better than the 247 mA/bit I'm getting now.

Note: Design change

So, in short - design change to use ACS755SCB-050 (or ACS755LCB-050). To you guys who are building this thing - please note this change. I will be changing the code to reflect this change once I get hold of the said part.

Daniel

Edited by Daniel Wee

Share this post


Link to post
Share on other sites

Daniel,

How important is your ground-plane? Can we use a single sided copper PCB board to avoid clearing the drill holes from the ground plane? Thanks!

Share this post


Link to post
Share on other sites

I believe the board will work even without the ground plane since it doesn't carry any important signal paths that are not also carried on the underside. I did the ground plane in the interests of keeping RFI down as much as possible but the board should work without it.

Daniel

Share this post


Link to post
Share on other sites

sorry, i posted an answer in the wrong discusiongroup i gues

since fma gives you a horizon indication the tip was to link it to a attitude indicator

i flew a foamy movistar with a felsweb transmitter and was interested in a head-up display. i have a multyplex magister and a mikado logo 20 .my aim was taking fotographs

Share this post


Link to post
Share on other sites

Another change to the PCB layout - non-critical. I'm just changing it so that it's consistent with my other prototypes. This has to do with the connector pinout for the current sensor. The ground and the sensor output will be swapped - pins 2 and 3.

New PCB files:-

Minor change to PCB

Daniel

Share this post


Link to post
Share on other sites
Why do I need to use FMA? My auto-pilot already works. And I totally can't see the relevance of your link at all - it sounds like a troll.

Daniel

sorry, i posted an answer in the wrong discusiongroup i gues

since fma gives you a horizon indication the tip was to link it to a attitude indicator

i flew a foamy movistar with a felsweb transmitter and was interested in a head-up display. i have a multyplex magister and a mikado logo 20 .my aim was taking fotographs /and as being a sound recordist tell how a troll sounds ?

Share this post


Link to post
Share on other sites

Ok. Board is etched and most of the components are soldered. Just waiting on the resistors to arrive. Those 1/8w resistors are hard to find!

Daniel - I think I've got the pin assignments mapped correctly. Would you please list them for the Bob-4, GPS, and ICSP program header. I just want to double-check before I fry something! :D

Also, did you just solder wires between the two boards to connect them?

Too bad on the current sensor. I just got my samples of the ACS750. No problem, though. Allegro is very open with their samples.

Thanks!

Edited by docphi

Share this post


Link to post
Share on other sites
Another change to the PCB layout - non-critical. I'm just changing it so that it's consistent with my other prototypes. This has to do with the connector pinout for the current sensor. The ground and the sensor output will be swapped - pins 2 and 3.

New PCB files:-

Minor change to PCB

Daniel

Hi Daniel,

I think your compressed file is corrupt. I test it out and it (WIN RAR) keeps telling me the file is corrupted.

John

Share this post


Link to post
Share on other sites

Okay, I've re-uploaded the PCB files. Should be okay now - had some trouble uploading them yesterday.

As for the interconnects - yes, just a few wires between the controller board and the BOB-4SG board. As for the exact connections, they are as follows:-

On the photo (component side view) - the connections, from left to right of the 5-holes to the BOB-4:-

1. goes to SS0\ pin-4 on the BOB-4SG

2. goes to 3.3V pin-1 on BOB-4SG

3. goes to MISO pin-21 on BOB-4SG

4. goes to MOSI pin-22 on BOB-4SG

5. goes to SPCK pin-12 on BOB-4SG

GPS connector, L-R:-

1. to GPS RX1 pin-4 (green)

2. to GPS TX1 pin-3 (white)

3. to GPS BAT pin-8 (orange)

4. not connected (or to 1PPS pin-7 if you like)

5. to GPS GND pin-2 (black)

6. to GPS Vin pin-1 (red)

I think that's the correct pinout (I've not visually verified this though)

I should also mention at this point that on the BOB-4SG you need to make a number of connections to ground to configure the board correctly. I will post this in a few days (once I get time to pull it out to see what I configured).

Daniel

Edited by Daniel Wee

Share this post


Link to post
Share on other sites

I'm going to hold off on the current sensor for the moment, unless there is a demand for it. The reason being that I am going to re-work it to use the ACS755 which will actually work a little differently from the ACS750. But if you really want the ACS750 design, just let me know.

Daniel

Share this post


Link to post
Share on other sites

Sounds good. We'll wait for the new current sensor design. Could you list pin assignments for the ICSP header? It looks like it's:

L-R:

1-Vpp (pic pin 1)

2-Vdd (pic pin 20)

3-Gnd (pic pin 19)

4-PGD (pic pin 17)

5-PGC (pic pin 18)

Thanks!

Edited by docphi

Share this post


Link to post
Share on other sites

A bit more testing today - I've de-commissioned my Cessna and now everything is fitted into my Phoenix Trainer EP. It's a very clean setup - about 1.35kg AUW, inclusive of a Prolite TP2100 3S battery. I intended to test out the autopilot in this setup but conditions were too gusty for me to get a proper trim on the plane. I did discover that the altitude control is working very nicely though.

So - with this setup - I flew for about 20minutes - drawing only 1511 mAh out of the pack! Amazing - by this estimate, I am projecting around 27 minutes of flight from a full pack. This is a very nice surprise - very very efficient flight.

On another note - I started losing the r/c control at around 0.8km ground distance (400m up) which actually put the plane into auto-pilot. Now I know the failsafe part works. This plane is using a slightly inferior receiver than my previous Cessna (which got me out to 1km+ without loss of control. That was using the JR RS10DS - a full 10-channel PCM dual conversion receiver. With my current testing, I'm using a R770S 7-channel PCM single conversion receiver. I don't know if the variation in distance is due to the different receiver or if it is due to some other factors. Nevertheless, 0.8km is not enough for me - even from the R770S so I will be looking into the re-location of the receiver for more separation from the other electronics.

On yet another note - I am using a new 500mW module from Lawmate, and discovered that, as with my older module, it was prone to physical vibrations. Vibrations in the module resulted in horizontal lines across the screen. I strapped it down with some cable ties which improved things considerably. I'm not sure which is the offending part though.

Daniel

Share this post


Link to post
Share on other sites

Parts are all here. Board is etched. PIC is programmed. Time to get soldering...

Daniel, what is the value of the resistor next to the transistor? One is a 4K7 resistor, the other one is not on your schematic.

Thanks!

Edited by docphi

Share this post


Link to post
Share on other sites

Should be a 10k. The one right next to the transistor is a 4k7 and the one closer to the diode is a 10k I believe. I'll post the latest firmware in abit - made some minor changes and improvements.

My ACS755-050 came today so I will be working on the new current sensor. Also, my second BOB-4SG also arrived so I will be testing now on a secondary prototype for code changes.

On another note - I managed to get out to fly some today - it stopped raining and there was lovely weather, no wind at all. I earlier had some problems with the receiver going into failsafe at about 0.8km so I re-arranged some cables and put on a ferrite choke on the group of cables. Today I managed to reach 1.4km (ground distance, LOS is 1.62km) and an altitude of 821m (just under 2500ft) - both are new records for me - without any lockouts or loss of control.

I am also using the new Lawmate 500mW. I've noted before that the module needed to be strapped down. I added two more cable ties and today the picture was perfect - no horizontal lines at all.

I didn't really get to set up the autopilot today as I was too busy testing the distance (and trying to set a new record). I'll try to set it up properly next flight. I found that I don't really like altitude control - maybe I should just go for altitude hold. For example, when I am out there at 800m, I don't really want to descend - I just want the plane to come back. I may put in a feature that allows altitude hold as an option. I'll be thinking about this in the next week or so.

Way to go docphi! Keep us posted on your progress - I want to see you succeed!

Daniel

Share this post


Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.


×
×
  • Create New...