144 replies to this topic

[dankers]

Quote

The ESC that powers the receiver and now the camera servos gets quite  warm. I thought about using a UBEC but since I have used up all 6 servo  connections, I don't know how I'd do that

A servo Y lead, yes, it is that simple.

[sakumar]

dankers, on 07 September 2011 - 10:51 PM, said:

A servo Y lead, yes, it is that simple.

Should have thought of that. Thanks.

[SeismicCWave]

peabody124, on 07 September 2011 - 12:51 PM, said:

That would be f-ing awesome if that works.  Gonna go scrounge up some old cdroms.  So this is passive?
Cappello: just the same board.  It's a quad for four outputs and the last two are for the camera.

The first time I saw it was on a Bergen gasser. Larry Bergen called it the Canfield balancer. I used to have a wikipedia page about the setup but I can't find it right now. I still have it. It does not work all that well. At least not on the gasser.

http://www.helifreak...ad.php?t=180644

[jes1111]

Interesting idea! I've seen these balancers for truck tyres, etc. - but never knew they appeared in DVD drives as well.

Apparently the generic name for these devices is "Automatic Ball Balancer (ABB)".

http://yp.bmt.tue.nl/pdfs/4277.pdf

[peabody124]

BTW going back to the camera mount discussion.  Joe and I were talking trying to come up with an open design for a camera gimbal, finding a BOM for all the odd bearings, etc.  Is there much interest in this?  The goal would be something that can be made with the average CNC router.  I'm partial to the shape of the ones from Droidworx, Photoshipone and Helifreak are using.  Is there interest in this enough to create a thread about it?

[StarRider]

peabody124, on 08 September 2011 - 12:22 AM, said:

BTW going back to the camera mount discussion.  Joe and I were talking trying to come up with an open design for a camera gimbal, finding a BOM for all the odd bearings, etc.  Is there much interest in this?  The goal would be something that can be made with the average CNC router.  I'm partial to the shape of the ones from Droidworx, Photoshipone and Helifreak are using.  Is there interest in this enough to create a thread about it?

Yap, I'm VERY interested

Regards,
PA

[SeismicCWave]

peabody124, on 08 September 2011 - 12:22 AM, said:

BTW going back to the camera mount discussion.  Joe and I were talking trying to come up with an open design for a camera gimbal, finding a BOM for all the odd bearings, etc.  Is there much interest in this?  The goal would be something that can be made with the average CNC router.  I'm partial to the shape of the ones from Droidworx, Photoshipone and Helifreak are using.  Is there interest in this enough to create a thread about it?

Of course there is. I have a silly question. What is a BOM? Are you talking about my bowel movement?

DJ Vegh/Photoship One is very possessive about his design. Make sure we don't make it too much like his. Actually the ones by Photoship One requires a fairly precise fit on the roll mechanism. That makes cutting error very critical.

I actually like the one from A Vertical View. It is almost easier to fit different size camera mounting plate to the roll mechanism.

[AlPackin]

since its become a feature I am interested.  I just read the helisfreak thread on RCG today, looks promising but somebody showed up to pee on his parade at the end  

[Efliernz (Pete)]

sakumar, on 07 September 2011 - 10:10 PM, said:

The ESC that powers the receiver and now the camera servos gets quite warm. I thought about using a UBEC but since I have used up all 6 servo connections, I don't know how I'd do that.One nice to have feature would be to have an option to use a spare receiver channel to turn camera stabilization on or off.

To run one ubec to the rx / board and another to the servos connected to the same receiver...

1/ Common the negitive wires of the ubec's together - this is still the negitive of the servo.  
2/ Positive to the sevos is from the second UBEC.
3/ Signal to the servo comes from the receiver as usual.

If the servo jams, it will only effect the second ubec and not the receiver as the receiver is still fed by the first ubec + & -.
Sorry - no sketch or photo as I'm at work "working"

Pete

[peabody124]

SeismicCWave, on 08 September 2011 - 12:35 AM, said:

DJ Vegh/Photoship One is very possessive about his design. Make sure we don't make it too much like his. Actually the ones by Photoship One requires a fairly precise fit on the roll mechanism. That makes cutting error very critical.

I don't think the general shape with the belt is something he can claim ownership (or patent?).  It's the same as this one from droidworx http://www.kousupers...era-gimbal.html .  Here is another similar design http://www.copterwor...60/Default.aspx .  It just strikes me that going between those runners will probably really keep the roll smooth and rigid.  I agree about the manufacturing tolerances so let's take this over to a new thread http://forums.openpi...-gimbal-design/

[dankers]

peabody124, on 08 September 2011 - 01:39 AM, said:

I don't think the general shape with the belt is something he can claim ownership (or patent?).  It's the same as this one from droidworx http://www.kousupers...era-gimbal.html .  Here is another similar design http://www.copterwor...60/Default.aspx .  It just strikes me that going between those runners will probably really keep the roll smooth and rigid.  I agree about the manufacturing tolerances so let's take this over to a new thread http://forums.openpi...-gimbal-design/

New thread, awesome and very excited to see what people here can do.

There is no fun copying anyway, not our style. So not the plan and lets do something ourselves, sure influence from other designs and as those links show, there are many similar but as a community we have so many skilled people here and some very smart engineers, I think we can easily do as well as the very expensive ones.

[aerialcameraman]

I think we are all in agreement that vibration is the number one destroyer of good footage. Yet when I look around me I see the same old motor mount designs that have set a president since Holger cobbled up his first  attempt.  We often build up a mental shape association with that which we feel looks right. The concept of putting the prop. on the wrong end of the motor beggars belief. Why place the imbalance and aerodynamic loads as far away from the mounting plate as you can get it?  Once you do it this way you don't go back.

I would draw a similar analogy to having a large, heavy, yet very flimsy camera mount hanging like a sack of bent carrots beneath the model. The further the center of mass is away from the plain of lift the more energy it takes to keep it level after a lateral acceleration.

Just out of interest, a slight reverse pendulum is the most efficient config.

Attached Files

•  Arm.jpg   234.26K   311 downloads
•  Front.jpg   193.88K   326 downloads

Edited by aerialcameraman, 08 September 2011 - 10:16 AM.

[Fuwu]

Very nice!

The stabilisation looks really accurate, can't wait to test it!

Thanks!!!

[Herculito]

Fantastic. I am waiting for my cc to try.

Thanks, great job.

[Dane]

I need to do further tuning but not bad for 2am's work....



Great job CC team!

[Dane]

aerialcameraman, on 08 September 2011 - 09:59 AM, said:

I think we are all in agreement that vibration is the number one destroyer of good footage. Yet when I look around me I see the same old motor mount designs that have set a president since Holger cobbled up his first  attempt.  We often build up a mental shape association with that which we feel looks right. The concept of putting the prop. on the wrong end of the motor beggars belief. Why place the imbalance and aerodynamic loads as far away from the mounting plate as you can get it?  Once you do it this way you don't go back.

I would draw a similar analogy to having a large, heavy, yet very flimsy camera mount hanging like a sack of bent carrots beneath the model. The further the center of mass is away from the plain of lift the more energy it takes to keep it level after a lateral acceleration.

Just out of interest, a slight reverse pendulum is the most efficient config.

Wow thats a very serious rig.... NICE!

[jet760]

HI can anyone tell me what I need to set the camera setting on for the GCS
thanks

[Dane]

jet760, on 09 September 2011 - 01:25 AM, said:

HI can anyone tell me what I need to set the camera setting on for the GCS
thanks

varies per gimbal/camera mount

[peabody124]

http://wiki.openpilo...a Stabilization

If something is unclear please tell us and we can improve the documentation.

[aerialcameraman]

Dane, on 08 September 2011 - 11:25 PM, said:

Wow thats a very serious rig.... NICE!

Having seen so many posts about MK's etc that came down in the sea. I decided that it was time to kill two birds with one stone. Make it float and gain a bunch of other features at the same time.