CS8 to X8 conversion - Herkules ESC III V3

[BorisS]

Okay date is set. This friday we are going to do some testing in Regensburg with Andreas ! Andreas is setting up some equipment to take a closer look at what WKM is putting out over its ESCs ports.

Everything is ready to go hope the weather will be with us and nothing stupid happens until friday with the CS8. Just like two days again when my cleaning lady knocked my hexa from the table Luckily with not gear on only frame and motors View attachment 2577View attachment 2578

 

[redridinghood]

Boris, very nice of you to take the trip to Regensburg for helping Andreas.
One small discovery I made is that maybe the yaw gain is to blame for the erratic behavior Wookong is having with HerkII. When the yaw gain is too high the motors behave eraticaly turning on and off at a fast pace. We don't notice the same on generic esc's because of the slower ramping and responce rate.

 

[BorisS]

Red have you guys also experimented with vertical gains. Up to now i though it only has something do to with manual mode, but looking at the DJI posted parameter again that they posted months ago, it could also be related to this !

A. Pitch, roll:
For the first flight, pls go to an open area with no wind, test in Atti mode. Throttle slowly up to take the aircraft off the ground. See if It is oscillate in relevant direction. If everything seems fine, you can try to move the cyclic stick to feel how it responds to stick commands.
B. Push the stick to make the aircraft tilt, and then release immediately.Watch the response when it’s self leveling automatically. If it is too sluggish, increase the parameters by 10% -15% each time, until it wobbles when self leveling. Now the feeling will be more direct and straightforward.
C. But we do not want it wobble, so we need to slightly reduce the values.

B. Tail

The sensitivity of the tail is actually very easy to adjust, because there is only one basic sensitivity parameters, like general head-locking gyro. You can increase the gain to make it more sensitive and vice versa. As to multi-rotor platform, however, it is necessary to know that anti-torque for the blades controls the rotation of the tail. Due to the limited strength, it can’t shake like the tail of helicopters no matter how big the gain is, but will get motor to start and stop very agilely to affect the stability of the other direction.

If it is asymmetric about the speed to turn, please carefully check if the motors are installed upward straightly.

C. Height

There are two methods to verify if vertical gain is appropriate. Check if the height can be maintained when throttle stick stays central, or if there is a significant altitude change when flying cruise.

If it is the first time to fly, you can make it through the following way to reach a relatively appropriate gain. To increase vertical gain step up by 10% until vertical shock just happens or over-sensitive reaction from throttle stick. The phenomenon above mean too high gain, you can decrease the gain by 20% to get the relatively appropriate gain.

 

[RTRyder]

Boris, very nice of you to take the trip to Regensburg for helping Andreas.
One small discovery I made is that maybe the yaw gain is to blame for the erratic behavior Wookong is having with HerkII. When the yaw gain is too high the motors behave eraticaly turning on and off at a fast pace. We don't notice the same on generic esc's because of the slower ramping and responce rate.

DJI has a new manual out dated yesterday, I was reading through it this afternoon and started thinking about this. Some of the issues we're seeing with the large frames regardless of the ESC in use may well be due to the extra high settings on both yaw and vertical, problem is my CS is like a dog wagging its tail on yaw if the gain goes much below 200 so I don't see an easy solution. I did try angling the motors at one point to give it more yaw authority and lowered the yaw gain at the same time, all that did was make it wobble worse in FF and windy conditions.

Ken

 

[BorisS]

I dont know if this is appropriate, but after my testing with Andreas, I would be down with writing DJI a letter/email what ever with signatures of customers using bigger frames/setups asking them for some sort of white paper/case study. A clear step by step explanation on the gains setup components needed motors KV props etc to get the best out of their FC. Its has to be in their interest that we are all happy and i don't think it is impolite or to demanding to ask them to spend some time on this ?

Boris

 

[Siteline]

So for the last couple of sessions i've been focusing on that yaw movement that you are talking about RT and Red. We went out and got the craft dialed in forgetting to address the basic Yaw gains and when we got back home to have a look at the footage there was constant "tail wagging" through out. It wasn't really visible or obvious until looking at the footage. I've had the yaw gain maxed out at 300% in wind and feel it sort of shortens the wagging but it's still there and there seems to be a bigger bounce back once the actuated yaw stops at that setting. I too angled the motors only 1% in the corresponding direction hoping to see a difference and to be able to lower the yaw gain. It was quite windy today and I couldn't really find an optimal configuration??

 

[RTRyder]

I dont know if this is appropriate, but after my testing with Andreas, I would be down with writing DJI a letter/email what ever with signatures of customers using bigger frames/setups asking them for some sort of white paper/case study. A clear step by step explanation on the gains setup components needed motors KV props etc to get the best out of their FC. Its has to be in their interest that we are all happy and i don't think it is impolite or to demanding to ask them to spend some time on this ?

Boris

Worth a try but don't hold your breath waiting for them to do it. Thing that gets me is that if this were a government contract application or expensive industrial hardware then there would be a very explicit technical manual that explains all the different settings and how they work along with how to contact someone that would actually give a S**t and fix it if the user found something that didn't work right. Since the total sales of all the WKM to date probably equals about 1% of what one unit of similar military grade hardware would cost we get "disturbed" gain settings and poorly translated explanations that don't really explain anything.

Ken

 

[RTRyder]

So for the last couple of sessions i've been focusing on that yaw movement that you are talking about RT and Red. We went out and got the craft dialed in forgetting to address the basic Yaw gains and when we got back home to have a look at the footage there was constant "tail wagging" through out. It wasn't really visible or obvious until looking at the footage. I've had the yaw gain maxed out at 300% in wind and feel it sort of shortens the wagging but it's still there and there seems to be a bigger bounce back once the actuated yaw stops at that setting. I too angled the motors only 1% in the corresponding direction hoping to see a difference and to be able to lower the yaw gain. It was quite windy today and I couldn't really find an optimal configuration??

There's two issues in play with what you're seeing, one being the fact that the WKM needs to use significantly smaller props than the same multi would with other flight controller hardware, so the torque reaction to counter yaw is less than it would be with the bigger props. The other factor is that the higher the gain the quicker and harder it reacts on that axis which is exactly what you saw in use. The way I see it there's only three possible answers at present. One is don't fly in anything other than optimum conditions which just isn't realisitic. Next is find the best combination of settings and inflight adjustments to get reasonable results regardless of the conditons and that's what I've been doing since day 1 with large frame hex and it produces mediocre results. The last option is a bit extreme but it's a path I took recently until DJI can get a handle on what it takes to make a large multirotor fly well on their system, I swapped the WKM for the MK electronics stack that was on my AD6. I will say the CS6 flys significanly different WKM vs. MK but it also handles the wind 1000% better on MK than it ever did with the WKM. It's a heck of a lot more work to fly which makes me want to go back to the WKM but the video that comes out of the camera in windy conditons tells me the MK stack is the way to go for the time being. One of my WKM is now setup on a medium size quad for testing purposes and the unit from the CS is sitting on the bench waiting until there's a fix for what ails the firmware when used on large frames.

Ken

 

[Siteline]

So frustrating because on so many levels it's such a joy to fly. In the conditions we were flying today I wouldn't dare take my hands off the sticks to adjust the gains had i been flying the MK or at least the throttle. There are so many strengths with the movement of this FC that the MK just can't achieve because the very precise altitude control. I thought about returning to the MK but I can't bring myself to do it. I'm going to stick it out with WKM and hope they address our voices. They have the man power and the "pro" market is ripe for the picking.

 

[RTRyder]

So frustrating because on so many levels it's such a joy to fly. In the conditions we were flying today I wouldn't dare take my hands off the sticks to adjust the gains had i been flying the MK or at least the throttle. There are so many strengths with the movement of this FC that the MK just can't achieve because the very precise altitude control. I thought about returning to the MK but I can't bring myself to do it. I'm going to stick it out with WKM and hope they address our voices. They have the man power and the "pro" market is ripe for the picking.

I feel the same way, if it weren't for the fact I had the Mk hardware sitting there gathering dust I wouldn't have done it myself. On the other hand the issues with the WKM were just making it impossible to get the full benefit of what the CineStar is capable of. From the little time I've had the CS up and running on MK I see just how much more work it's going to be but the payoff is what I see on the memory card afterward, simply not possible to get the same result using WKM at the moment and I don't think their wonder gimbal is the right answer. Yes, I'm glad they developed it, but if the trade off is the frame bounces and wobbles all over the place while the gimbal stays still I'd rather fly a different system with my AV130 and spend less $ in the long run.

Ken

 

[BorisS]

Like said if someone helps me who's english is better than mine lets right them an email supported from several customers that they should address these issues. Or at least deliver something more convincing than this :

http://v.youku.com/v_show/id_XMzA2MTcyNzY0.html

Although looking at the video I more and more consider that they don't understand what most customers actually want. Having a camera on board on this octo wouldn't give the desired results we are looking for. And seeing how proudly they show that the radio is not being controlled, all they seem to have in mind is position hold or the attitude mode leveling at the cost of anything usable on the camera !

Boris

 

[RTRyder]

If I can just get to the point of getting some really good raw video with the current configuration of the CS, I have plenty of not so good raw video taken with the WKM in control that can be used for comparison. Still not sure they'd get it though, as you mention they seem to be hung up on how it can sit in the same spot until the battery goes dead and it comes down. Great, but what does that have to do with Aerial video work? I think Sidney gets it more than the others at DJI and I know that part of their plan has to be taking the stand that it all works right if you buy ALL of our gear, frame, motors, gimbal,etc. but at DJI prices that's about 3 times what it would cost otherwise.

And I don't think it's a good English translation you need, you want someone fluent in Mandarin so they'll understand what it is we're trying to tell them...

Ken

P.S. don't you ever sleep? It must be almost morning where you are...

 

[BorisS]

Haha i do but little only a few hours these days Trying to pack as much time in on multis as possible at the moment once one of my real estate projects start again i wont have much time. But i think i am going to leave the 8 hours car ride to nuernberg fair and back out for today. Although the gimbal and the s800 seem to have arrived at the fair !


Boris

 

[elossam]

I though before about the yaw control related with the other axus stability. The only way I found to get a reassonable yaw stability on low gains (about 140%) was using big surface airscrews like apc vs. Graupner and others. If you are right and wkm doesnt like big blades then we are arriving to the same result from different directions. At the moment I'm waiting for a good day to test with some wind to test it with a high disturb gain value ans see what this param is used for. If nothing changes then I will exchange the wkm by a Hoverfly and keep the wkm in the toy quad units.

 

[elossam]

Just comming back from the garden after flying two bats using the max disturb gain value allowed, 200%. I' m not sure but seems to be better, not a stable flight for filming unedited but way better handling high wind gusts. To be true it was a difficult test for the wkm. My garden has some hight trees very close, the main building ans several zones that makes the wind come from different directions and speeds with descending rotors etc... I was able to heard the engines fighting alot but the unit remains quite controlable. A little fast forward input shows very high gain settins because doing it against wind makes the ufo oscilate very fast but doing it in wind direction is smooth. I cant reduce the gain because that affects the hovering stability in wing conditions, the only way to do that is a mix between elev and/or ail and the x2 port to reduce gain in atti mode according to the stick imput required to start the FF motion. For me this was the last test. This evening I will mount the Hoverfly again to see how it performs since the GPS implementation and give this way some time to DJI to think about the question of Wkm and "big" multirrotors.

 

[BorisS]

Scheiting in my pants, but not having any other option. I tested the bird inside today. Due to the weather in austria at the moment other way !. Cant go to Andreas for some testing without having had the bird in the air. Nervous like it was my first take off ever !

 

[redridinghood]

Looks good Boris. Why was your Cinestar grounded? Did you crash?

 

[BorisS]

Nope luckily no crash. Had an issue with one of the Herks. Sent it to Andreas for a checkup. Im the meantime the Hers V4 came out. Agreed to test them and go to Regenburg for further tests. Than the Herks V4 took longer than planned for the first units to be completed. All of a sudden about 2 months have passed ! In the meantime I already had the new herks, but my Heatsink CNC guy managed to mess up the new Heatsink although he did it before. Couple things were working against me !

Boris

 

[elossam]

What trouble do you had with the herkules. Have had one with puffy lines close to some fets apparently for no reasson.

 

[BorisS]

Once more i managed to pull the screws to tight on my Heat sink and give the board some flex resulting in one of the components to loosen at the soldering joints.

Elossam what do yo mean with puffy lines ?

Boris