I couldn't resist a quick flight test with the TwinStar, and too much wind. The first attempts to turn on the stabilization were tragic, with the plane just spinning out of control. It seemed to be working fine on the ground, so I landed and checked it again. It still seemed fine, but the second flight acted just the same.
I'm sure Bill's laughing now, because he probably already realizes what I did. Yep, I had the aileron throw reversed. Doh! I had all this set up properly on the first board, but ended up ordering a second board as well. After the regulator changes, I installed the second board, which had the switches in their default position, NOT the position I put them in on the first board. I was just so convinced they were moving the right direction, I didn't question it until after the second flight.
Flipping the switch of course makes a world of difference. Stabilization works reasonably well for stock settings. The ailerons have too much gain, and the elevator not enough, but I'll start working on that.
Waypoint mode tried really hard to work, but the wind and poor flying qualities of the TwinStar kept it from being fully successful. I had a simple box set up, and it made the first 3 corners, but the last was upwind from far away. Ultimately, I became uncomfortable with my ability to see the plane from that distance, and brought it back manually. The main problem was the TwinStar wanting to balloon up, and more throttle makes that even worse. The AP really wasn't to blame for this.
I'm a little confused. I tried to better understand the difference between define or not CROSSTRACKING in the waypoints menu.
My english does not allow me to have a satisfactory comprehension of that.
Could you please explain how CROSSTRACKING work and the difference?
In wich cases is it better to use CROSSTRACKING or not?
Another couple of questions about GPS:
what exactly happen when GPS signal is lost in WP mode?
How many satellite the board needs to set home?
I ask this because I usually fly in an area with high mountains all around and then very bad GPS (4-7 satellites, often 4-5, depends on weather and moment).
I noticed that if I put the plane in wp mode (not crosstracking) it reaches the first wp at about 50m from the start position (home).
Since you mentioned working on new software, I would like to make a suggestion (request). In the quadrocopter world where I usually hang out, we also have gyros and ACC sensors of course. While the gyros should be calibrated every flight, the ACC really only needs to be done once, as long as you don't change the board mounting angle.
For our typical aircraft, that have no landing gear, I find it difficult to simultaneously plug in the battery, and hold the plane level, and still for 10 seconds. If the only requirement was to hold it still, you could just let it lay on the ground in whatever attitude gravity allows.
If this PIC has the ability to save the ACC cals in non-volatile memory, the suggestion would be to remove the ACC calibration from the initial power up routine. You'd have to create a trigger to do the cal only when needed, such as some combination of stick positions, or a jumper using one of the extra pins on the board.
Thanks for reading requests from the software challenged :-)
Rusty
I just made 4 success flights with the IMU on my old Robbe's Arcus.
Wow, I was very impressed by the efficiency of this device! Even with the base settings it works very well.
Stabilisation is awesome. From inverted flight it recover the model in a fraction of a second without overshoot or strange habits, and the same is from any flight attitude.
In wp mode it turn around very gently, holding the heading and the attitude very well. I only want it a bit responsive. I'm going to increase a little the P aileron's gain.
I'm not a programmer, I'm only a well experienced RC pilot. I can watch at the code and try to change something. Is for that that I appreciated a lot the clarity of your code, with all the explanations and, most important of all, the well explained values (what they means, meters, real degrees, binary degrees etc; what they do and wich are the maximums allowed).
Of coarse I have now to fine tune the Pid settings, and to comprehend the most I can of this device.
I tryed my plane without altitude hold (I disabled it), but I miss it. Now I have to find a solution, berhaps another esc.
Could you please tell me wich ones are already tryed and working?
I didn't get my 4.5V regulators until about 5:30 this afternoon, but it's installed, and so far I'd say it's a go.
With the 4.5V regulators, the VCC measures 4.451V. With a fully charged flight pack battery of 5.2V, there's no problem at all with the 3V pulse height from the Futaba FASST receiver.
Just for fun, I rigged up a variable regulator, so I could reduce the voltage of the receiver, while leaving the Red board at 5.2V. The reason for this is to turn down the voltage enough so that the receiver is forced to put out even lower amplitude pulses, just to see how low it can go. As the video shows, 2.65V pulse height is as low as I can get.
I'll certainly post any and all info about the regulator used, but I'd like to let Bill look over the data first, and see if there's anything else he would like me to test. There could easily be a better choice for the regulator as well, but I'm happy so far. I still have one stock red board, but barring any needed comparisons, I'll change out the regulator on that one tomorrow.
Bill, is there any way to see the GPS status (number of sats, signal levels, etc)? Since it's running right at the minimum voltage, it would be interesting to compare the full voltage version, to the reduced voltage. There's only a few tenths of a volt difference between running the stock board at 5.0V (VCC about 4.85V), and the modified board with a VCC of 4.5V, so I don't anticipate a noticeable difference in GPS performance.
Sadly, there won't be any flight tests this weekend for me. The cold front last night left us with "15-25 mph, gusting to 40 mph" wind forecasts. There's a little hope for late Sunday afternoon, but not too promising.
I just got my UavDevBoard. I programmed it with Aileron assist 1.8 and put in some waypoints.
Trying all the settigs before the big day (tomorrow if I can) I wasn't able to have my esc working.
The esc is connected to the board to have altitude hold.
It seems that the esc (a Scorpion) try to arm before to have any good signal from the board (10").
So it doesn't arm at all.
Forgive me if this was already asked but I wasn't able to find an answer.
For some reason, your last couple posts don't give me the "reply to this" option, so I can't reply directly below them. Have I ever mentioned how much I hate this ning format...
FWIW, I'm not any sort of expert in hardware design. I gave some thought to the level shifters, and Sparkfun has a neat little board for less than $2 that will shift a couple signals. Unfortunately, it would require a 3V supply, which isn't readily available on the receiver. By the time you shifted 4 channels, it would get really messy with cables, even if I did a plug in adapter board.
I had also considered the diode solution, but would rather change the regulator. I'm pretty confident that a 4.5V regulator will take care of everything nicely. If I'm wrong, it certainly won't be the first time. If I'm REALLY wrong, I have another CPU on order as a spare :-)
Hi Bill,
Thanks for your suggestions.
As soon as I get home, I'll surely try on low voltage power supply.
The strange thing that I've haven't already figured out, is why input 1 and 2 had no problem at all!
I know that the first 2 inputs are on one port on the PIC and input 3 and 4 are on the D port. Maybe this port has a little different trigger? Is there a possibility that we need a different configuration?
I'll post the results as soon i get home to test.
See you later,
Andrea.
FYI - I don't use channel 3 but have had no problems with the channel 4 input for selecting the mode. I use a Futaba R607FS (2.4GHz) and a Dimensions Engineering SportBEC to power everything. I've flown with this configuration and haven't noticed any problems so far.
Hi Bill,
I really need some assistance.
I've tested so far 4(!!) different red boards, but no one seemed to correctly parse the signal on input 3 and 4.
I tested them using stock sparkfun self-test firmware and also with self-compiled test firmware downloaded from here.
I've also made tests with AileronAssist firmware (without much success due to bad input channel 4 recongition).
I don't really know what I'm doing wrong, and it seems quite strange that all the 4 boards where bad.
I'm testing with futaba radio equipment, powered with 5/6 Volt stabilized power supply.
Input 1 and 2 are perfect, but input 3 and 4 aren't working at all, or seems to sometimes read one pulse sporadically.
Do you have any suggestion?
You can also contact me by email, so we can go to further investigation.
Thanks.
Andrea.
Replies
I'm sure Bill's laughing now, because he probably already realizes what I did. Yep, I had the aileron throw reversed. Doh! I had all this set up properly on the first board, but ended up ordering a second board as well. After the regulator changes, I installed the second board, which had the switches in their default position, NOT the position I put them in on the first board. I was just so convinced they were moving the right direction, I didn't question it until after the second flight.
Flipping the switch of course makes a world of difference. Stabilization works reasonably well for stock settings. The ailerons have too much gain, and the elevator not enough, but I'll start working on that.
Waypoint mode tried really hard to work, but the wind and poor flying qualities of the TwinStar kept it from being fully successful. I had a simple box set up, and it made the first 3 corners, but the last was upwind from far away. Ultimately, I became uncomfortable with my ability to see the plane from that distance, and brought it back manually. The main problem was the TwinStar wanting to balloon up, and more throttle makes that even worse. The AP really wasn't to blame for this.
All in all, a very promising first attempt :-)
Thanks,
Rusty
I'm a little confused. I tried to better understand the difference between define or not CROSSTRACKING in the waypoints menu.
My english does not allow me to have a satisfactory comprehension of that.
Could you please explain how CROSSTRACKING work and the difference?
In wich cases is it better to use CROSSTRACKING or not?
Another couple of questions about GPS:
what exactly happen when GPS signal is lost in WP mode?
How many satellite the board needs to set home?
I ask this because I usually fly in an area with high mountains all around and then very bad GPS (4-7 satellites, often 4-5, depends on weather and moment).
I noticed that if I put the plane in wp mode (not crosstracking) it reaches the first wp at about 50m from the start position (home).
Thanks!
Ric
Since you mentioned working on new software, I would like to make a suggestion (request). In the quadrocopter world where I usually hang out, we also have gyros and ACC sensors of course. While the gyros should be calibrated every flight, the ACC really only needs to be done once, as long as you don't change the board mounting angle.
For our typical aircraft, that have no landing gear, I find it difficult to simultaneously plug in the battery, and hold the plane level, and still for 10 seconds. If the only requirement was to hold it still, you could just let it lay on the ground in whatever attitude gravity allows.
If this PIC has the ability to save the ACC cals in non-volatile memory, the suggestion would be to remove the ACC calibration from the initial power up routine. You'd have to create a trigger to do the cal only when needed, such as some combination of stick positions, or a jumper using one of the extra pins on the board.
Thanks for reading requests from the software challenged :-)
Rusty
I just made 4 success flights with the IMU on my old Robbe's Arcus.
Wow, I was very impressed by the efficiency of this device! Even with the base settings it works very well.
Stabilisation is awesome. From inverted flight it recover the model in a fraction of a second without overshoot or strange habits, and the same is from any flight attitude.
In wp mode it turn around very gently, holding the heading and the attitude very well. I only want it a bit responsive. I'm going to increase a little the P aileron's gain.
I'm not a programmer, I'm only a well experienced RC pilot. I can watch at the code and try to change something. Is for that that I appreciated a lot the clarity of your code, with all the explanations and, most important of all, the well explained values (what they means, meters, real degrees, binary degrees etc; what they do and wich are the maximums allowed).
Of coarse I have now to fine tune the Pid settings, and to comprehend the most I can of this device.
I tryed my plane without altitude hold (I disabled it), but I miss it. Now I have to find a solution, berhaps another esc.
Could you please tell me wich ones are already tryed and working?
Thanks for such a great device and program.
Best regards.
Ric
I didn't get my 4.5V regulators until about 5:30 this afternoon, but it's installed, and so far I'd say it's a go.
With the 4.5V regulators, the VCC measures 4.451V. With a fully charged flight pack battery of 5.2V, there's no problem at all with the 3V pulse height from the Futaba FASST receiver.
Just for fun, I rigged up a variable regulator, so I could reduce the voltage of the receiver, while leaving the Red board at 5.2V. The reason for this is to turn down the voltage enough so that the receiver is forced to put out even lower amplitude pulses, just to see how low it can go. As the video shows, 2.65V pulse height is as low as I can get.
BTW, the video was done in a single take, so lets keep the video critique to a minimum :-)
http://www.radrotary.com/RedBoard_4.5VCC.wmv
I'll certainly post any and all info about the regulator used, but I'd like to let Bill look over the data first, and see if there's anything else he would like me to test. There could easily be a better choice for the regulator as well, but I'm happy so far. I still have one stock red board, but barring any needed comparisons, I'll change out the regulator on that one tomorrow.
Bill, is there any way to see the GPS status (number of sats, signal levels, etc)? Since it's running right at the minimum voltage, it would be interesting to compare the full voltage version, to the reduced voltage. There's only a few tenths of a volt difference between running the stock board at 5.0V (VCC about 4.85V), and the modified board with a VCC of 4.5V, so I don't anticipate a noticeable difference in GPS performance.
Sadly, there won't be any flight tests this weekend for me. The cold front last night left us with "15-25 mph, gusting to 40 mph" wind forecasts. There's a little hope for late Sunday afternoon, but not too promising.
Cheers,
Rusty (well equipped Den)
I just got my UavDevBoard. I programmed it with Aileron assist 1.8 and put in some waypoints.
Trying all the settigs before the big day (tomorrow if I can) I wasn't able to have my esc working.
The esc is connected to the board to have altitude hold.
It seems that the esc (a Scorpion) try to arm before to have any good signal from the board (10").
So it doesn't arm at all.
Forgive me if this was already asked but I wasn't able to find an answer.
Any help?
Thanks and best regards.
Ric
For some reason, your last couple posts don't give me the "reply to this" option, so I can't reply directly below them. Have I ever mentioned how much I hate this ning format...
FWIW, I'm not any sort of expert in hardware design. I gave some thought to the level shifters, and Sparkfun has a neat little board for less than $2 that will shift a couple signals. Unfortunately, it would require a 3V supply, which isn't readily available on the receiver. By the time you shifted 4 channels, it would get really messy with cables, even if I did a plug in adapter board.
I had also considered the diode solution, but would rather change the regulator. I'm pretty confident that a 4.5V regulator will take care of everything nicely. If I'm wrong, it certainly won't be the first time. If I'm REALLY wrong, I have another CPU on order as a spare :-)
Rusty
Thanks for your suggestions.
As soon as I get home, I'll surely try on low voltage power supply.
The strange thing that I've haven't already figured out, is why input 1 and 2 had no problem at all!
I know that the first 2 inputs are on one port on the PIC and input 3 and 4 are on the D port. Maybe this port has a little different trigger? Is there a possibility that we need a different configuration?
I'll post the results as soon i get home to test.
See you later,
Andrea.
FYI - I don't use channel 3 but have had no problems with the channel 4 input for selecting the mode. I use a Futaba R607FS (2.4GHz) and a Dimensions Engineering SportBEC to power everything. I've flown with this configuration and haven't noticed any problems so far.
I really need some assistance.
I've tested so far 4(!!) different red boards, but no one seemed to correctly parse the signal on input 3 and 4.
I tested them using stock sparkfun self-test firmware and also with self-compiled test firmware downloaded from here.
I've also made tests with AileronAssist firmware (without much success due to bad input channel 4 recongition).
I don't really know what I'm doing wrong, and it seems quite strange that all the 4 boards where bad.
I'm testing with futaba radio equipment, powered with 5/6 Volt stabilized power supply.
Input 1 and 2 are perfect, but input 3 and 4 aren't working at all, or seems to sometimes read one pulse sporadically.
Do you have any suggestion?
You can also contact me by email, so we can go to further investigation.
Thanks.
Andrea.
-
4
-
5
-
6
-
7
-
8
of 11 Next