Originally Posted by eyeinsky1

I know this is a great advantage to have an aircraft that as you add and remove equipment such a cameras, AV downlinks and batteries for different needs, the weight and balance of the airplane is not so critical. In fact you can use this to change the characteristic of your flight. For example: On a windy day you will have lots of lift, your aircraft will be working hard adjusting altitude a fwd C of G would be better in these conditions. In low wind condition ultimate video opportunities you can run an aft C of G to get a nose up flight attitude the aircraft is much floatier and will maintain the slow fights better. Another thing we get is a virtual stall avoidance system in your aircraft. Not to say it will prevent the crash but it should minimise the wing drop spiral dive situation that we all have found our self in before, with our heavily modified over weight aircraft operations.

Here’s a video of my gyro stabilised Twinstar that had pitch gyro come loose in flight. I will be adding the KK board to this ship to soon as they arrive.

http://vimeo.com/26526847

Yep, Wings in particular are a pain regarding cg, as batteries ect are usually embedded so if you add something that adds weight to the nose or rear, it can be a pain to get the cg right again.

I had another flight with it even further back. The only thing I really notice with it further back is the roll response is less and less.

I'm out to the field shortly to test fly the standard aeroplane OpenAero without accelerometer on the HK EPP FPV. It looks good on the ground, the servo are much quieter with the OpenAero sw.

Wish me luck.
Jerry

Best of luck. I'm sure it will go well

This is all very exciting. It helps so much to have people try it out. I wish I had as many test pilots on the NeXtcopterPlus thread
Jerry, I am pleased to hear that you think the servos are quieter. I had a bit of a brainwave regarding synchronising the servo updates with the RC input interrupts. The RC interrupts is what causes the servos to jitter. The remaining noise comes from hardware limitations, but I'm pretty happy with it.

Originally Posted by NDw

Yep all working now.
I did two more flight tests, much calmer today. The first one went perfect so for the second one I said I'd try moving the cg back to make it very tail heavy to see what'd happen. I moved it about 1.5 inches back, and my wing was very close to being tail heavy as it was. There is no way it'd fly 1.5 inches further back without the gyros.
With it it still flew perfectly.

That looks amazingly stable! Since the day was calmer, can you tell me how the control response felt? Did you feel that the system was trying to dampen *you* as well as the external instabilities?

Standard Aeroplane OpenAero KK control HK EPP FPV test flight was successful. Wind was low sky’s were foggy, the OpenAero sw performed excellent, we perform two test fwd and aft C of G operation.

HK EPP FPV KK OpenAero Test Flight.mov (8 min 25 sec)

Originally Posted by eyeinsky1

Standard Aeroplane OpenAero KK control HK EPP FPV test flight was successful. Wind was low sky’s were foggy, the OpenAero sw performed excellent, we perform two test fwd and aft C of G operation.

That looks great Jerry. I hope you and NDw don't mind me linking to your videos in the front page
Did you think the lack of elevator response was from the CG being still too far forward, or it was being dampened by the software? If it was the software, it would still have the same throw, just be more sluggish getting there. If you just didn't have enough elevator regardless of how long you held up or down elevator, then maybe it was the CG?
BTW, I love your other surf videos. Surfing in Nova Scotia - cool.

Originally Posted by HappySundays

That looks great Jerry. I hope you and NDw don't mind me linking to your videos in the front page
Did you think the lack of elevator response was from the CG being still too far forward, or it was being dampened by the software? If it was the software, it would still have the same throw, just be more sluggish getting there. If you just didn't have enough elevator regardless of how long you held up or down elevator, then maybe it was the CG?
BTW, I love your other surf videos. Surfing in Nova Scotia - cool.

Thanks Surfing is big here

The elevator has always been an issue the servo is little to weak cold weather makes everything little stiffer too, the GG does effect elevator too but as far as I'm concerned the OpenAero worked great.

Cheers Jerry

Great vid Jerry.

Originally Posted by HappySundays

That looks amazingly stable! Since the day was calmer, can you tell me how the control response felt? Did you feel that the system was trying to dampen *you* as well as the external instabilities?

Yep it def has an effect on the responsivness. Like the rates were lowered. I'm sure if you lower the gain it would be more responsive but then less stable.

I like it more stable. You just predict your turns more rather than trying to turn at the last second.

Originally Posted by NDw

Great vid Jerry.
Yep it def has an effect on the responsivness. Like the rates were lowered. I'm sure if you lower the gain it would be more responsive but then less stable.
I like it more stable. You just predict your turns more rather than trying to turn at the last second.

Thanks for the feedback. In the original code, the gyro feedback is calculated from the difference between your input and the external rotations of the model. Now if you do it that way, as you drop the gyro gain, the control also drops and at zero gain you have no control at all.
So, for OpenAero, I changed it so that it only responds to model rotation. You can set the gain to zero and still have full control. However as the gyro feedback is now only based on model rotation, is doesn't take into effect the rotations that you ask of it via your input. There was a chance that it would have tried to null you out completely, but I don't think it has the range to completely overpower the RC input.
Anyway, if all you have to do is increase your rate a little, then I see no pressing need to improve it yet. You probably need more input the higher the gyro gain. I could probably automatically increase the RC response as the gain is increased but I'll leave that for a rainy day

Slightly off topic, but, I have a KK Multicopter, v5.5d+ board that refuses to 'talk' over the serial lines (I have checked all connections/solder joints/components and the only thing left is the chip).

I have now removed and replaced the Atmega 168PA chip with a brand new virgin chip (I have a hot air solder reflow station so replacement is a doddle).

I can't get to burn a bootloader to the virgin chip - I've tried with avrdude and Arduino - response is the same, "target doesn't answer. 1"

Is there anthing 'special' I need to do to initialise/boot this board with a virgin Atmega 168PA chip??

Nigel.

Originally Posted by Devonboy

I can't get to burn a bootloader to the virgin chip - I've tried with avrdude and Arduino - response is the same, "target doesn't answer. 1"

Is there anthing 'special' I need to do to initialise/boot this board with a virgin Atmega 168PA chip??

Nigel.

Hi Nigel,
Yes there is! I had exactly the same trouble as you when I swapped a 48P for a 168PA on a HK board. Brand new chips are for some reason unable to be written to at full speed. On my USBASP programmer there is a second jumper which is used to select the slower speed. Once programmed with LazyZero's tool I found it could then use normal full speed.
To be honest I never got the serial loader for KK+ to work for me, I just burn it using the USBASP.

Thanks for the response HappySundays.

I am also using a USBasp programmer and also tried with Lazyzero's software.
I used avrdude in command line mode and told the programmer to burn at <200kHz (-B5 'switch' and get 187500 Hz) as this is what a virgin chip apparently needs.
Still no joy, although avrdude complained that it is "not able to set SCK period - please check for usbasp firmware update" (I am using the latest fischl driver http://www.fischl.de/usbasp/ ) - although I often see that message and it still programs OK, albeit, maybe at a faster rate on the old chip which already had the pre-installed bootloader.
I have this kind of USBasp (in fact, the very same one as the guy hacked)
http://jethomson.wordpress.com/2011/...sp-programmer/
The guy refers to the J2 connectors as
"First insert a jumper wire into the two holes circled in yellow in the picture above. Doing this connects RST on the 10 pin header to PC6 (RESET). "
Other than that, I don't seem to be able to put that board into 'slow mode' ??
I'm very new to KK boards, but I seem to be learning fast and the hard way!

I only tried using a Sparkfun FTDI based serial board to program firmware via a COM port and to read your Aero GUI (with correct cable(s)).
I never get to program via serial or read the GUI, so assumed my original chip was 'deaf' as it never replied to an input of any kind (the FTDI is flashing madly on the Tx LED, but nothing on the Rx LED - I know the FTDI is OK as it works fine in Arduino and MultiWii !).

Darn, this is so frustrating, never had any trouble with MultiWii

Nigel.

Originally Posted by HappySundays

Thanks for the feedback. In the original code, the gyro feedback is calculated from the difference between your input and the external rotations of the model. Now if you do it that way, as you drop the gyro gain, the control also drops and at zero gain you have no control at all.
So, for OpenAero, I changed it so that it only responds to model rotation. You can set the gain to zero and still have full control. However as the gyro feedback is now only based on model rotation, is doesn't take into effect the rotations that you ask of it via your input. There was a chance that it would have tried to null you out completely, but I don't think it has the range to completely overpower the RC input.
Anyway, if all you have to do is increase your rate a little, then I see no pressing need to improve it yet. You probably need more input the higher the gyro gain. I could probably automatically increase the RC response as the gain is increased but I'll leave that for a rainy day

Yep imo it is excellent as it is.

So I see in the first post you mention that in CPPM mode you can do some extra things like turn off the gyros if you want to via channel 7, and that the frsky recivers can do this.

I'm using the 8 channel non telemetry frsky reciver atm with my small wing. How do you use CPPM?

Thanks

Originally Posted by Devonboy

Thanks for the response HappySundays.
I am also using a USBasp programmer and also tried with Lazyzero's software.
I used avrdude in command line mode and told the programmer to burn at <200kHz (-B5 'switch' and get 187500 Hz) as this is what a virgin chip apparently needs.
Still no joy, although avrdude complained that it is "not able to set SCK period - please check for usbasp firmware update" (I am using the latest fischl driver http://www.fischl.de/usbasp/ ) - although I often see that message and it still programs OK, albeit, maybe at a faster rate on the old chip which already had the pre-installed bootloader.

Yep, I see this error all the time too. The firmware they are referring to is the firmware *inside* your USBASP, not the windows driver. I don't have the required bits to reflash my USBASP so I just figure I can keep using it as I have been.

I have this kind of USBasp (in fact, the very same one as the guy hacked)
http://jethomson.wordpress.com/2011/...sp-programmer/
The guy refers to the J2 connectors as
"First insert a jumper wire into the two holes circled in yellow in the picture above. Doing this connects RST on the 10 pin header to PC6 (RESET). "
Other than that, I don't seem to be able to put that board into 'slow mode' ??
I'm very new to KK boards, but I seem to be learning fast and the hard way!

That USBASP doesn't seem to have the "slow" jumper. I figure that Lazy's latest software tries to change the clock speed as required but you need to have the right firmware in your USBASP to do so. I agree that it's all a bit much lol. I just had a quick look online and I see that some USBASP boards don't have the Slow Mode jumper. So, unless they have the right firmware, I don't know how that is supposed to work...

I only tried using a Sparkfun FTDI based serial board to program firmware via a COM port and to read your Aero GUI (with correct cable(s)).
I never get to program via serial or read the GUI, so assumed my original chip was 'deaf' as it never replied to an input of any kind (the FTDI is flashing madly on the Tx LED, but nothing on the Rx LED - I know the FTDI is OK as it works fine in Arduino and MultiWii !).

Well, unless your board is a KK+ *and* still has the bootloader inside the Atmel chip (very unlikely if you have replaced the chip) then the serial loader will never work. I could never get it to work anyway! lol. Ok, next you have to know that the GUI uses a quite different pinout of that port than the serial loader. You have to make up a special cable for the serial port to work with the GUI. Details on the first post.
The problem with KK is that it was very badly designed in regards to pin usage. It has a perfectly good UART inside the chip, but the original designer connected the TX pin to the Aileron input.... so, doing the GUI while still using the aileron input was a challenge