AlexMos and other brushless gimbal controllers....what is the truth?

[Bartman]

Hi everyone,

As some of you know I call around from time to time to ask what some of you are up to and I'm trying to figure out what the honest-to-goodness truth is about brushless gimbal controllers.

Is there a majority of users that have achieved the very high quality results we have all been hoping for? What is the state of the market......clones? legit manufacturers and suppliers?

Seeing wat was broadcast last night on Discovery Channel using DJI/Zenmuse equipment I'm wondering if the generic brushless gimbal market is keeping up.

Discuss??

Thanks and looking forward to your replies!

Bart

 

[BrianMac]

I recently bought the Infinity M Gimbal with an AlexMos 2 axis controller. I am amazed at how good this thing is! I have a Cine6 that is pretty stock with MK electronics. I was flying it with an AV130 mount slinging a Sony NX30 (like the 760) around. I had even tried the SkyLine for a bit but gave up trying to get it configured correctly. The brushless system is much more responsive and smooth. A night and day difference from the servo gimbals I've used. I haven't done anything extensive with it so far, and haven't had it out in any meaningful wind. But I can tell it's definitely a much more smooth option for camera stabilization. And the price? Can't beat it! The InfinityM is probably in the mid range of price, and it's not a lightweight. It's aluminum! So you probably want to use something with a little umph. My BLs do get a bit warm with this config so I'll need to look into getting some heat sinks on them soon.

The downside? Getting the bugger set up. It's not an easy task to get it first mechanically balanced with the camera on the gimbal and then set up the electronics. But if you are patient and understand the physics of the gimbal it's not impossible.

Overall I'm very impressed.

 

[Tomstoy2]

Myself, I picked up a cm2000 gimbal when they first became available last year.
Good little gimbal. Had a motor I replaced and I ended up replacine the clone board for an AM board, but for 5he price was still happy.

Now with the next generation gimbals to soon come out I'm pretty excited.
to have a glitch free gimbal with no jitters will really be sweet!
Like so many others, I can only get so much out of my set up, but not hit the magic setting!

 

[Bartman]

thanks for the feedback Brian. which version of the 2-axis controller are you using? i have a Ver1 on the bench awaiting my stab at a gimbal but it's bogged down along with everything else at the moment!

 

[BrianMac]

It's the Simple BGC. I think I'm running v 2.3 on the firmware.

 

[PeteDee]

I also god one of the ARRIS CM-2000 gimbals when they first came out but decided that I wanted something more compact for transport so got a TBS Discovery Pro, getting very good results.

Pete

[video]http://www.youtube.com/channel/UCwZSpD9HB_QV6LgwOHdCv5Q[/video]

 

[BrianMac]

Dang, now I want a Disco Pro!

 

[jfro]

I had good luck with 2 mid size gimbals and alexmos v1 boards running the first couple versions of his firmware. On my HF Pro, I had to dial back the gains to get smooth & steady video...

However, made a change to my 12lb MR and ever since have been fighting some oscillations. Getting closer to fixing, but still not as smooth on 2 axis as a few months ago. Cabling and sensor placement are critical. Very sensitive...

I now have 3 gimbals, 1 of which is a Tarot GoPro3 which came ready to go. It flies pretty good. It uses some form of clone card, not alexmos.

It's very confusing trying to figure out what is legitimate Alexmos or not as I think the clone peeps have figured out how to install the legit Alexmos firmware.

I go here to look to see if the company is listed as a legitimate site.

http://www.basecamelectronics.com/wheretobuy/

Edit.... Just flew couple test flights on Alexmos and wrapped aluminum around the gimbal wires. Much improved. Not 100%, but couple more flights and might have it. So strange, cause I didn't have any issues until I swapped 30 for 40 amp esc's.

 

[SMP]

^ Yup.

 

[jfro]

Put ferrite rings on both ends of IMU, and gimbal motors at Alexmos side. Put 8 ferrite rings on the my 40amps ESC's.

Still get some slight jittering. I have 3 gimbals I've messed with using 2 alexmos controllers. Also have 1 Tarot gorpo gimbal. Tarot pretty much plug and play and works 98% of the time.

My alexmos setups took a while to get down, but then flew very nicely. Then made change from 30-40amps escs and a some issues with my HFP board. Ever since, I've been fighting some slight vibrations. Changed flight controllers, changed from xy8 to x8. Put new bearings in 6 or 8 motors.

Just pulling my hair out. 1 minute I think I almost have it solved and then boom, back come the vibrations. Mostly when movement is starting or reversing.

 

[Bartman]

jfro, it's been my opinion for a while that reactive gimbal controllers would never really achieve "perfect" because they're always going to be a step behind the flight controller. i figured that when the intentions of the flight controller could be factored into the gimbal control scheme then something closer to perfect would be possible. i don't know what DJI does to mate the Zenmuse up to the WKM or A2 but maybe they're doing predictive control of the camera mount?

i really don't know what the truth is behind all of this stuff. i wonder what the acceptable reaction-lag is before the human eye can't see it anymore. my stuff is all servo driven still so i'm probably the wrong guy to talk to about brushless but i'm curious as to how it's coming along.

 

[jes1111]

You're certainly not wrong about being "behind the game", bart - but the main problem with these brushless gimbals is lack of torque. The measured torque of these motors is piddly compared to even the tiniest RC servo - something less than 0.25 kg.cm. (less than 3.5 oz.in.) for a "Nex5" size of motor. It'll work fine if the balance is perfect (which is probably why DJI limit the camera/lens) but as soon as you have a small amount of imbalance then the motor is struggling to keep up with the demand for torque and that's when the jitters start. I feel it's a kind of "out of the frying pan and into the fire" situation - we cursed servos and now we are cursing brushless motors. The answer has to be more torque - but the trade-off is brutal: more weight, more money and the backlash problem if you introduce gearing to increase the torque.

Predictive input may help some but, for example, a Cineflex doesn't have any link to the heli pilot's controls and it has to cope with fearsome vibrations - yet it does an impeccable job - you just have to throw engineering grunt at it (and therefore money).

I don't understand the new Alexmos 3-axis product - who can carry the weight of the motors that that board is capable of driving? Nuts!

Brushless motors are a better starting point than RC servos, but we've some way to go yet on this one.

 

[Bartman]

.................Brushless motors are a better starting point than RC servos, but we've some way to go yet on this one.

i can't necessarily agree 100% with you on this. regarding my gimbals, i'm doing more photography than anything so it suits me to have 900 oz-in of torque on each axis at the camera tray and the ability to change lenses without moving the camera body if i don't want to.

i'm getting pretty nice video results as well but it takes some time to get everything tuned.

 

[jes1111]

I hear you. In your case, torque is more than sufficient - but backlash is your enemy. Obviously belt drive masks the backlash somewhat so you may get little apparent free movement of the camera (which means its going to be fine for stills), but nothing can make up for the delay (and the jolt) within the multi-stage gearbox inside in that servo every time it reverses direction (which is the big problem for video). And there's the dilemma - servos have the torque but also backlash, brushless has no backlash but insufficient torque.

 

[Bartman]

well, i'm using pretty high quality servos and i don't think there's much backlash, if any. maybe after a lot of use. i'm using gears, fwiw, there's no issue with imbalance like there would be with belts, the gears mesh so there's virtually no play. with brass and aluminum gears as they wear in you just tighten them up. it all works very well actually.

 

[SleepyC]

Brushless is it’s own beast. it takes MUCh patience and time to get it right. When right it’s awesome. But right can be like 80% of the time.
Wind, temp, pressure, diff cam all play into it. It’s a never ending quest.

 

[jfro]

What puzzles me is I got video I was happy with earlier in the summer. But I can't get back to where I was before making some changes.

I'm building a 3rd mr now. When done, I'm going to put my HFP FC back on the x8 and give it a try again. On the HFP, I had the gains dialed down so to make it feel relatively soft, or not snappy. It helped on keep the little judders out of the video.

If the gopro3 had manual exposure, I'd just put brushless gimbal aside for a while and let technology catch up and just fly the gopro for a that just sort of works. .

I need to have a win soon with my larger brushless gimbals, cause I'm getting frustrated with the amount of time spent on them without any clear cut lessons learned on what to do and not. Maybe a fesh MR build will shed some light on whatever I'm doing wrong or missing.

 

[Droider]

The new daddy on the block my well be the G-lock system. either servo or brushless its looking good to me.

I am still on servos and flying the camera NOT the multi. I just can't justify the time or the money on a technology that is still so unreliable. Think about it and name a three axis or single axis controller that really works… the best on the block is the Radian for servos.. Now that controller has just been taken a massive step further in the G-Lock system.. and it will work.

The designer is a brit with a passion and wants it right, so a few people including my self have been testing it on different platforms.

I just know its going to be a ground breaker.

Dave

 

[Kari]

The new daddy on the block my well be the G-lock system. either servo or brushless its looking good to me.

I am still on servos and flying the camera NOT the multi. I just can't justify the time or the money on a technology that is still so unreliable. Think about it and name a three axis or single axis controller that really works… the best on the block is the Radian for servos.. Now that controller has just been taken a massive step further in the G-Lock system.. and it will work.

The designer is a brit with a passion and wants it right, so a few people including my self have been testing it on different platforms.

I just know its going to be a ground breaker.

Dave

Can't wait to get g-lock brushless up and running. I have three Radians waiting on my desk for upgrading all my gimbals from alexmos. Not that i'm not happy to Alexmos boards, i'm getting fantastic results at the moment with my all alexmos boards but it will just be so simple to tune, cleaner setups etc.

 

[jfro]

The new daddy on the block my well be the G-lock system. either servo or brushless its looking good to me.

I am still on servos and flying the camera NOT the multi. I just can't justify the time or the money on a technology that is still so unreliable. Think about it and name a three axis or single axis controller that really works… the best on the block is the Radian for servos.. Now that controller has just been taken a massive step further in the G-Lock system.. and it will work.

The designer is a brit with a passion and wants it right, so a few people including my self have been testing it on different platforms.

I just know its going to be a ground breaker.

Dave

I just heard about the g-lock last week. Did a quick check but didn't get a feel for the costs. Any estimates on what the electronics are going to run and what's needed?

I do see lots of decent video on Vimeo and Youtube from the Alexmos based gimbals. It would be nice to hear from some of the peeps that are having success with it about how they are getting it done.