CarbonCore Cortex....Owners' Thread

When you say wobbles, what exactly do you mean, and what sort of weight are you carrying, do the wobbles come when your are flying or stationary, or descending or what?
If you wanted to move your IMU there is the possibility to put it on top.
Another possibility is to actually put it underneath, if you added spacers and a little platform.
Can you post any videos of these wobbles?

Hi guys,

Thanks for all the input! :eek:) This is the only video I have of it flying, but it's not easy to see the little wobble that it does after applying stick input.

https://www.youtube.com/watch?v=0fVhK6_sdA8

Like i do on all the rigs, i use a DJI bluetooth interface and do the gains from my iPad during flight. I've always found this to be the easiest way to fine tune. I normally stand in a field with a helper and just tell the helper what to change and to what and i can see the results immediately. I've done a lot of gains tuning (around 20+ rigs so far), but I must admit on all Naza setups the Naza has been very close to the CG. I've had discussions with CC about this and it does seem that the Naza might not be the best FC as it doesn't have the ability to offset the FC. So, the only choice may be to move the FC to the center of the top plate! I'll keep you posted!
 

Quinton

Active Member
OK I had a bit of playing around with my retracts today since I finally got thrm back together.

At the moment it looks a bit like the following when you have intelligent gear off...

F1 Goes all the way up, then back down (problem is when you disconnect power the servos lose power) and it will hit ground. Seems to be a 2 way switch only.
F2 Is opposite of F3 and F4 and is only a 2 way switch.
F3 and F4 is not a 2 way switch, you can control it from a pot or slider, however when you power up you do need to ensure you have it off the ground, as it will twitch landing gear up then back down again. (The servos will stay locked when power is removed however could this depend on where the endpoints are set, as it may simply be a case I had it further set than F1/F2

Seems like the A2 wont play ball at all with landing gear that does not have intelligent checks built in like the DJI ones, probably the Droidworx ones will also work.

I updated all my firmware again today, and noticed that sometimes now that there is an alarm when the retractables move and power is not cut to them, must be some kind of FC safety thing built into latest firmware.
So what choices do we have now using the A2 (and please don't say use a Wookong)

Anyone know anywhere in the UK where we can get the DJI intelligent electronics on their own, or is there any other solution.
Maybe Cameron can come up with some kind of solution, after all I cant imagine the Wookong will start to be faded out and we may not be able to get hold of them, also the Naza sis ruled out, so not much of a choice really.


F2
 

Mrtarango32

Member
View attachment 17015
Figuring out how I'm going to run the escs.
I know CC won't like what I've done.
But it is my build.

What are your guy's thoughts on bullet connectors to the esc/power board. Run em or just solder directly to the board?
 

Attachments

  • ImageUploadedByTapatalk1395961548.521491.jpg
    ImageUploadedByTapatalk1395961548.521491.jpg
    104.8 KB · Views: 470


sk8brd

Member
i like to soder directly to the board if possible. it may save you a tiny bit of space. if your using bullets on the motor wires i would shrink over the bullets incase they vibrate loose they won't separate.
 

Mrtarango32

Member
i like to soder directly to the board if possible. it may save you a tiny bit of space. if your using bullets on the motor wires i would shrink over the bullets incase they vibrate loose they won't separate.

Ya I think that's where I'm leading too also.
 


sk8brd

Member
there you go loss some little weight and made some more space.

side note-i looked at the cc builds on this forum and i'm imho not huge on the idea of having the fc/imu so close to the esc's and led on that bottom plate. generally people avoid doing that in fear of rf interference. i would try to get those electronics on the top plate especially since there's also bec's down there for the retracts-another source of noise. the battery's on the bottom of the frame would of been or the pdb/ esc's/bec's on the bottom as Quinton mentioned a good rf free setup. i know it works the way it is everything on middle plate but just my personal views. also the a2 has a built in rx thats not good if there are a lot of noise nearby. you can check other builds if you want but people place their rx far away from everything electronic usually like the gps puck,led, etc

since carbon is a good rf blocker the top would be ideal for imu and flight controller and or middle plate if nothing else is on there.
 
Last edited by a moderator:


The retract solution!

Revive this thread?

Sounds good to me :)

Here is some more info on the A2/retract saga. Here is what I've tried:

1.) DJI A2 - F1 output to factory installed Cortex retract servos. No good. Like a few of us mentioned, it's always backwards (ie. A2 assistant always shows the gear up when it's down and vice versa). Also, If you ignore the assistant position and power it (with the gear down), it will retract it, then lower it (also referred to a "twitch" by others). Which isn't cool.

2.) DJI A2 - F1 output to DJI s800 retract control module. Not the perfect solution. I purchased HV (7.4v capable) servos to do this because DJI listed it as having a 7.4v output, which I just found out is a lie and its 6v to the servos. I also had to move one of the servos to the other side of the retract mount, as this is how the s800 is laid out. Anyway, the problem is that the s800 retract control module cuts the power to the servos 3 seconds after it extends or retracts. So, when I retract the gear in flight, it will droop back down. Annoying and useless. So I've tried some springs and couldn't get it how I'd be satisfied.

3.) DJI A2 - F1 output to Cortex retract servos in opposite configuration, meaning I literally unmounted them from the retract mounts and put them on the opposite side of the mount, so they function in reverse... and guess what?! It works perfect! A2 assistant is right and it doesn't retract them when you power it up!! You can't adjust the endpoint position, which would be nice if that was an option in the A2 assistant, but it works as it is. The only thing is that you might need slightly longer screws as the other side of the retract mount isn't countersunk. I could have saved some time and money (not needing the DJI s800 control module or HV servos), if I would have thought of this sooner. Oh, well.

So, it would be wise if Carboncore used this information to their advantage and countersunk the mounts on the opposite side and assembled them with the servos opposite as they are now. Walla, finished, end of story. If you weren't using an A2 and wanted to plug the retracts into your servo, you'd just need to reverse it in your tx.

When, hope that makes sense.

Paul
 


Quinton

Active Member
Well done you, and thanks for sharing.
Did it work with the Intelligent landing gear turned on at all, or did you not get that far?
Sometimes the simplest answers stare you in the face.
 

kloner

Aerial DP
i swear my dw retracts plug and played. had the intelligent thing on, what a joke, i'm always down low.....
 

Quinton

Active Member
i swear my dw retracts plug and played. had the intelligent thing on, what a joke, i'm always down low.....

I have the DW retracts on a Wookong, and the difference with them is that they will never retract until you flick the switch "twice" which is the smart thing to do.
They also remember their last position.
Would be nice to find out where you could purchase something like that.
 

I'm trying to share (ah, well, remember) all the details about the retract thing. Here are a couple more notes.

Since you can't adjust the endpoints, it just so happens to be about right on the travel. Though I'd like them to retract up a bit more. As it is, they are pretty much straight out to the sides. So, here is my potential solution. I ordered two of these the other day:
http://www.servocity.com/html/180o_servo_stretcher.html#.Uz9ic203C5I

This allows you to change your endpoints (cw and ccw independently), as well as get more than 90 degrees out of the servo. This way I should be able to push it right to the mechanical limits of the slot the bearing rides in, without overloading/binding the servos.

I was happy that the endpoint right now in the gear down position is far enough that you can't fold the leg up (it's past the "cam over" point) with the power off.

At this point, I don't think the servo-stretcher thing is needed. But I had already ordered a couple so I'll try it.

The biggest downfall that I can see right now is that they are really fast. Like they've had too many Redbulls fast. I could slow it down on the DX8, but that wouldn't do any good using the F1 output.

So, I just remembered that I've got a couple of these servo slower downer thingies laying around from another project:
http://www.rcdude.com/servlet/the-25/Dionysus-Dual-adjustable-servo/Detail

But wait, there's more. This unit might be the trick. It allows you to use one of its outputs for a reversed signal (so you theoretically wouldn't have to physically swap the servos around) and also allows you to change the speed (no endpoint adjustment thought):
http://www.rcgroups.com/forums/showthread.php?t=1189626

As far as the "Intelligent Gear" option, I haven't actually enabled it, but there is no reason it won't work. The reason it wouldn't work before is obviously that they were backwards. Now that they are going the correct direction, it'll work.

I turned on the warning reminder on the DX8 for the switch I am using for the gear. This way the tx beeps like crazy if I turn it on with the switch in the wrong position. Nothing too exciting there.

Thanks,
Paul
 
Last edited by a moderator:

CarbonCore

Member
<!--[if gte mso 9]><xml> <w:WordDocument> <w:View>Normal</w:View> <w:Zoom>0</w:Zoom> <w:Compatibility> <w:BreakWrappedTables/> <w:SnapToGridInCell/> <w:WrapTextWithPunct/> <w:UseAsianBreakRules/> </w:Compatibility> <w:BrowserLevel>MicrosoftInternetExplorer4</w:BrowserLevel> </w:WordDocument> </xml><![endif]--> Hello all and thanks for sharing your experience with the Cortex and all your suggestions!
Good Idea Quinton about starting a Cortex Flyer FB page – I setup the CarbonCore Multicopter page for exactly that, now with about 4200 likes.

BunnyGirl – I may have a solution for you and am sending it with your next Cortex ;)

Quinton – I promise I will look into the A2 with the servo retracts and if I come up with a solution I will share it!
Also, MrTarango32 is using u5 motors with the intention of lifting 8-12Kgs, if I’ve quoted that correctly.

Paul – Got your A2 tip. I will be very surprised if in-Tx programming cannot be done to reverse one of the servos. It seems the combined Rx-FC causes a lot of confusion.

But if one retract works from the intelligent gear option, then why not just connect the retract servos on a Y-lead? That’s why they’re assembled with rotational symmetry, and not a mirror-image of each other.

CopterSquad have just called me today, to tell me that a CarbonCore Hexacopter was not flying properly on an A2, with the recommended gain settings, after updating the firmware from 1.1 to 1.2. “It felt woolly”. They have said they noticed that the maximum gains on the A2 1.2 firmware are now 500! So I’m wondering if that means you must now double the gain settings to make them equivalent to the old gain settings. I won’t know until I’ve flown with an A2. I have a couple here to try.

I have flown several Cortex with the WooKong flight controller. Granted DJI may discontinue the WKM at some point, and may have already stopped ‘support’.
I know the WKM works in the Cortex and that’s what I now recommend.

I also flew the Cortex for 6 months prior to its release with different combinations of motors and props. As far as I can tell, with a co-axial setup, flying with the same props top and bottom, is like flying with different props.

It has also come to light that the NazaV2 may have a slower processor than the WKM and A2, and does not handle the yaw on an X8 as well.

The NazaV2 does not have flight controller offsets and so is not suitable in a Cortex unless it is fitted centrally, above the power board or on the top plate.
The gimbal carbon plate (for Alexmos controller) has the same hole pattern as the power board and Cortex lower plate.

I, CarbonCore, am not a DJI Reseller or rep and am not obliged to provide support on DJI products.

I cannot monitor all the forums to answer questions posted on forums, sorry.

If you have a genuine question about a CarbonCore product, please email me on the CarbonCore Multicopter Facebook page.

Current CarbonCore customers have my contact details.
<!--[if gte mso 10]> <style> /* Style Definitions */ table.MsoNormalTable {mso-style-name:"Table Normal"; mso-tstyle-rowband-size:0; mso-tstyle-colband-size:0; mso-style-noshow:yes; mso-style-parent:""; mso-padding-alt:0cm 5.4pt 0cm 5.4pt; mso-para-margin:0cm; mso-para-margin-bottom:.0001pt; mso-pagination:widow-orphan; font-size:10.0pt; font-family:"Times New Roman";} </style> <![endif]-->
 

Paul – Got your A2 tip. I will be very surprised if in-Tx programming cannot be done to reverse one of the servos. It seems the combined Rx-FC causes a lot of confusion.

I'm not the only one that has mentioned this and I'm nearly 100% certain you can not. I think it's practically a bug on DJI's part. They would probably say it isn't, because I believe that they have their s800/s1000 in mind. It is really strange. If you reverse the channel on your TX, it will make the gear function opposite. BUT, in the assistant software, it also switches it. It's hard to describe. Maybe I should get a video. Let me try to explain better. This is all from memory, as the Cortex is in my office and I'm home for the weekend. I believe that there is a dir selection in the assistant for this but I'm not 100% sure on that but I'll add it here anyway.

Option A: TX switch position 0, channel normal (in tx settings), normal dir in assistant, gear physically down, assistant says gear is up.
Option B: TX switch position 0, channel reversed (in tx settings), normal dir in assistant, gear physically up, assistant says gear is down.
Option C: TX switch position 0, channel normal (in tx settings), rev dir in assistant, gear physically down, assistant says gear is up.
Option D: TX switch position 0, channel reversed (in tx settings), rev dir in assistant, gear physically up, assistant says gear is down.

Here are the two problems that arise from this:
1.) you can't use the intelligent gear option. This is because if you did and you had a RTH landing, it would be so generous as to retract your landing gear for you before it lands, instead of putting it down. That'd be a bad thing.
2.) The "twitch". Or what several of us have experinced and that is that when you power up the Cortex with the gear down, it will quickly retract the gear and then put it back down. Not so lovely as it'll smash your gimbal into the ground. Luckily, I found this out the easy way (suspending it to test), but others haven't been so lucky. I am 100% certain that this is because DJI made it this way so that when it is connected to a s800/s1000, it gives it a "gear down" command instantly on start up (which, since it is backwards remember), it retracts the gear, then it sees that we are requesting a "gear up" (as it thinks anyway, which is not true since again, it is backwards) so it puts them back down. Lovely.

But if one retract works from the intelligent gear option, then why not just connect the retract servos on a Y-lead? That’s why they’re assembled with rotational symmetry, and not a mirror-image of each other.

But neither of them work. They are both backwards. I am connected through the Y-cable that came with the Cortex. They are assembled in symmetry, just opposite of how they were. A simple $3 reversing unit would fix it. Or if you assembled them with the servos physically reversed, which wouldn't negatively effect people on other FC's, as they could just reverse it in their tx.

CopterSquad have just called me today, to tell me that a CarbonCore Hexacopter was not flying properly on an A2, with the recommended gain settings, after updating the firmware from 1.1 to 1.2. “It felt woolly”. They have said they noticed that the maximum gains on the A2 1.2 firmware are now 500! So I’m wondering if that means you must now double the gain settings to make them equivalent to the old gain settings. I won’t know until I’ve flown with an A2. I have a couple here to try.

I'm not sure what firmware my A2 is on (off the top of my head), but I believe it is current. Something seems very strange that the gains would need to be that high. I'm running around 175-185 on the basic gains right now (with Zenmuse attached).

I also flew the Cortex for 6 months prior to its release with different combinations of motors and props. As far as I can tell, with a co-axial setup, flying with the same props top and bottom, is like flying with different props.

It has also come to light that the NazaV2 may have a slower processor than the WKM and A2, and does not handle the yaw on an X8 as well.

The NazaV2 does not have flight controller offsets and so is not suitable in a Cortex unless it is fitted centrally, above the power board or on the top plate.
The gimbal carbon plate (for Alexmos controller) has the same hole pattern as the power board and Cortex lower plate.

I, CarbonCore, am not a DJI Reseller or rep and am not obliged to provide support on DJI products.

I cannot monitor all the forums to answer questions posted on forums, sorry.

If you have a genuine question about a CarbonCore product, please email me on the CarbonCore Multicopter Facebook page.

Current CarbonCore customers have my contact details.
<!--[if gte mso 10]> <style> /* Style Definitions */ table.MsoNormalTable {mso-style-name:"Table Normal"; mso-tstyle-rowband-size:0; mso-tstyle-colband-size:0; mso-style-noshow:yes; mso-style-parent:""; mso-padding-alt:0cm 5.4pt 0cm 5.4pt; mso-para-margin:0cm; mso-para-margin-bottom:.0001pt; mso-pagination:widow-orphan; font-size:10.0pt; font-family:"Times New Roman";} </style> <![endif]-->

Thanks for chiming in. All the info we can gather together the better. That's all I'm here for is to share my findings and learn from others as well. I think mine is flying pretty good. I think I'm getting to the point of feeling more comfortable with it than the Trex700e, but not as comfortable as I am with the Phantom... but that's just because the cost. haha
 


CarbonCore

Member
Hi Matt, Tiger Order is on the way in, so more spare props sooner than later hopefully.

Very good post Swarfworks - Thanks again!
 

Quinton

Active Member
View attachment 21033

This is how I'll have the x8 motor/prop rotation.

Sorry I have not been posting in a while but I have been away dealing with personal things and will now be ready to test again this week.
Does anyone know any information regarding what exactly these numbers equate to regarding the motors, as I can not find anything from DJI.
What exactly is the relationship in the numbers, do they control speed or something else.

Kopterworx have mentioned that it is not good to mix different props when using DJI Flight Controller, and I have no reason not to doubt them, as I am sure they have spent a lot of time/effort and money finding out exactly what is the best setup, but I would like to know why, and I doubt they would be willing to share such information.

I have been reading a lot of stuff, and a lot of the things I have read mention that using a smaller prop on the bottom would be beneficial, which is opposite to the way it is currently being used.

As we have no firm data on why they are being used with a larger prop on the bottom, has anyone done any real world testing, and I am not just talking about looking at something and saying, yes thats better.
I have tried 16 on top and bottom, and it is quite clear that it is not as good as 15 top and 16 bottom, however I have also tried 15 on top and bottom, and it seemed to be just as good as 15/16 together.

I know CarbonCore say they have done a lot of testing and its what they found best, however with not actually having their test data, I would like to try and find out for myself exactly "why" this would be the best setup.
Has anyone played with the % in this image, and know exactly what they do, do they change the speed of the motors or something else?
Its really hard to tell how it affects things, the only ones that can really answer that are probably DJI, as surely the Flight Controller will always try and even things out to each motor no matter how the props are set up.

Anyone any thoughts on this, doing bench tests would not really come to any conclusion, as we have no idea how the flight controller is counteracting what we would like to achieve.
Its hard to believe there is no real firm data out in the open to prove what works better, and quite frankly a LOT of stuff I have heard is usually mentioned because they heard it from another source and it is simply not true in a lot of cases.
 

Top