Jump to content

Dolly Accident...Caught on Video!


Karl Lee
 Share

Recommended Posts

It looks like the system has a "link rod" between the dolly and the jib arm so that the centre hydraulic ram (pedestal) will give large vertical moves to the camera end. So the loaded arm can easily be out of balance without knowing it, hence the centre of mass for the total system gets displaced from the pedestal. As the arm rotates (in plan view) the system centre of mass may arc to a point outside the square footprint of the dolly wheels.

 

If the arm was first balanced, including checking any load shifts due to camera/camera operator, then it should be possible to keep the total system balanced. The operator's body movement could shift the centre of mass quite a lot. Can you disconect the "link rod" on these easily so you can balance the arm?

Edited by Gregg MacPherson
Link to comment
Share on other sites


If that's responding to what I just wrote, then a more realistic analogy would be balancing a dinner plate on top of a coffe cup. Without the arm being balanced the coffee cup is in the wrong place.

 

I get the concern about the height. Not to contradict that, but all the observations, calculations or intuitive adjustments to make the system balanced, and with a safety margin, happen in the plan view. Ideas about safety have to start there, regardless of height.

 

Once the system fails (tips) then a high CoM makes it a lot worse.

 

Coming back to the coffe cup. The total system that sits on the tracks has a centre of mass (CoM). The footprint of the dolly trucks is a rectangle, and this rectangle is the location where the "restoring" or "balancing" moments can be physically applied. If the system CoM goes to the edge of this rectangle it will tip. The CoM should stay at the centre, and the dolly system should be designed to keep it there.

Edited by Gregg MacPherson
Link to comment
Share on other sites

  • Premium Member

Again it is safer to overcompensate by keeping the center of mass lower than relying on a perfectly balanced set-up, particularly when a large part of your mass is coming from a moving human being. Of course you should balance everything as best as you can, but it is generally not a good idea to put too big of a crane on a small base (despite what Archimedes said...)

Link to comment
Share on other sites

I'm with both of you also :)

 

It is indeed enlightening to look at it in terms of restricted planes, but eventually it needs to be analysed in 3 Dimensions - many devices (like, um 'stuff') are in terms of their functionality planar (2D), but not in this case.

 

Gregg is quite correct about the plan view and it is the primary consideration, however adding height, especially to a dynamic object will increase the ease at which the CoM extends over the base constraints.

Link to comment
Share on other sites

I'm with both of you also :)

 

It is indeed enlightening to look at it in terms of restricted planes, but eventually it needs to be analysed in 3 Dimensions - many devices (like, um 'stuff') are in terms of their functionality planar (2D), but not in this case.

 

Gregg is quite correct about the plan view and it is the primary consideration, however adding height, especially to a dynamic object will increase the ease at which the CoM extends over the base constraints.

 

 

'Adding height' essentially increases the lever arm relative to the connecting point on the base, which in the 'balanced' condition the resulting torque is zero. However, any slight tilt, will allow that lever arm to have a non-zero torque, and if the platform does not have sufficient counter force then the force will result in the toppling.

 

From the look of the video, the tracks are not really 'wide' enough to support the man+camera when the arm has this mass at 90 deg from the track (and attendant wheels) and only because the arm is not fully extended, disaster didn't happen at this poin, and definitely does not have support to counter the tork in the forward direction along the track when the man+camera reach the forward position.

 

In addition to the just the 'static' situation, there is a dynamic situation where the momentum of the movement carries the load beyond where the operator 'wanted' to stop, also inducing a torque force...

 

I don't know about movie dollies and tracking, but in more industrial settings, a cart like this would have a set of weels under the track to assist in the countering of forward tilting, and a placement of the post with in a longer bed, either central, or one end or the other, with turns to the 'short end' elimiated with some sort of stop, for such conditions.

Edited by John E Clark
Link to comment
Share on other sites

 

It is indeed enlightening to look at it in terms of restricted planes, but eventually it needs to be analysed in 3 Dimensions -

 

 

I think I get why you are calling me out fot turning this into a 2D problem. But I'm not sure that is what I'm doing. Looking at the projections onto the groundplane of the moment arms, from the 3D physicality, is a good first step. It doesn't have to deny the rotational motion(s). Actuallly, I'm used to thinking first of statics, and of what may be called quasi statics, when we see an angular acceleration. For example when a glider wing begins angular acceleration into roll, the math is fairly easy. Normally the dynamics concepts are not invoked unless there is an oscillation which may increase or decay, or we have to include the structures flexibility, with the varying airloads and the possible harmonics ensuing.

 

When the arm rotates about z (the vertical axis), it's not hard to consider the effect of angular acceleration (+ or -). Same for rotations about y or x (any axis in the horizontal plane). There are some concerns that intuitively come up for me about high acceleration rates about those axes, but the obvious and reccuring thought that I have is that an out of ballance jib arm is an invitation to an accident. It will produce side forces at the top of a high system as it rotates, hence moments (torques) that may not be able to be reacted by the forces at the relevent locations on that rectangle on the footprint of the trucks on the horizontal plane.

 

At the bottom line needs to be some rules or tables for the grip/crane op, so they can avoid accidents like the one in that video. A balanced jib arm is a very good start. If you can get the system CoM close to the centre of the rectangle described by the dolly trucks you have a good start with a safety margin. There may be cases where you want to consciously displace that CoM, to cover for what are anticipated as likely shifts in the CoM. Clever folk will do this and safely succeed.

 

If one does not know the actual location of the system CoM in relation to the magic rectangle, where the forces are applied that give the restoring moments then one is flying blind.

Link to comment
Share on other sites

I get the log in page.

 

I thought I had permanently deleted my FB account.

I have disabled my facebook account, and set up another under a non de plume (fictional name) so it's easier to occasionally look for friends etc. I was able to access that video OK. If someone has a video capture utility maybe they can post it on the forum or send it to you.

Link to comment
Share on other sites

I think I get why you are calling me out for turning this into a 2D problem.

I wasn't.

 

I agree with what you're saying. (apart from right now I guess :rolleyes: )

 

I also agree with David - therefore in lieu of the details I gave conditional agreement with your statement that:

 

 

(David) ... I don't think our ideas are really contradictory.

 

;)

Link to comment
Share on other sites

For those who have not seen this "crane", it's not like a traditional jib. The boom is operated by booming the dolly column, which is hard linked to the jib arm. And small balance shifts will be difficult to spot as the arm does not see saw freely.

 

The weight / height vs platform width is quite near its limit which makes it easy to fail. Kick out a couple dolly wedges and down you go.

 

I did use this thing on my last film for some shots looking down over a bed, but with a remote head and no operator, even then, we were quite cautious.

Link to comment
Share on other sites

I know this jib and worked successfully with it before. This accident was caused by a absolute lack of balance on the jib arm. No matter how high a jib arm is on that column, if it's in balance, things are fine, the jib is designed to carry a camera, DOP and AC. Thinking that the rod mounted between the jib arm and dolly replaces the purpose of the counterweights is a stupid thought and causes accidents like these.

 

For those who are interested, the jib used in the video was the Movietech DuoJib with the Movietech Magnum dolly.

 

Here a quote from the Duojib user manual: (Link, page 17)

 

Attention: Before set-up and at all times during operation, ensure that either Drive Ramp 3 or 4 is
selected. Do not operate the Duo Jib on Drive Ramp 1 or 2.
In the event of a power failure due to an electronical or battery cut out, it is recommended to have an
appropriate ladder to enable the camera crew to dismount.
Attention: Do not use pneumatic wheels when mounting or operating the Duo Jib!
1. Extend all 4 wheel arms of the Basic Dolly fully (pos. 1, page 8).
2. Mount the middle section (pos. 16, photo 3) of the Duo Jib onto the column by connecting it to the Euro Adapter
Mount.
3. Connect the Connector Rod (pos. 7, photo 3) to the Anchor Bracket on the column (pos. 9, photo 3). Insure that
the Automatic Locking Pins are locked securely.
4. Connect the Connector Rod (pos. 7, photo 3) to the Anchor Bracket of the Duo Jib of the Duo Jib
(pos.5, photo 3). Insure that the Automatic Locking Pins are locked securely.
5. Lock the middle section securely with the fixing lock (pos. 8 photo 3).
6. Mount the end section (pos. 4, photo 3) and tighten securely with the end section lock (pos. 18, photo 3).
7. Attach the Counterweight Triangle (pos. 3, photo 3) to end section.
8. Connect the parallelogram rod (pos. 19, photo 3) to middle section and counterweight triangle. Insure that the
connection pins (pos. 1 and 17, photo 3) connect correctly.
9. On the lower connection of the front section (pos. 14, photo 3), mount, High / Low Rig (pos. 13, photo 3).
Insure that locking lever ( pos. 12, photo 3) is locked securely.
10. Connect the Platform (pos. 10, photo 3) to High / Low Rig. Insure that locking lever (pos. 11, photo 3) is locked
securely. Maximum Payload capacity incl. 2 pers = 220kg / 485 lbs. For other versions see the following
pages.
11. Insure that all connections are locked and tightened correctly.
No more than two counterweights should be loaded whilst the platform is unloaded. As a general rule of thumb, for
one person on the platform attach 4 or 5 counterweights, for two persons on the platform attach 8 or
10 counterweights.
When operating the Duo Jib attention must be payed when mounting and dismounting the platform.
By two-man operation, remove the extra counterweights before one person dismounts i.e. no more than four or five
counterweights should be left connected. Before the second person dismounts, reduce the amount of counterweights to
three.
It is not recommended to drive the Duo Jib when the platform is unloaded and counterweights are
attached. At all times attention must be payed to balance.
Link to comment
Share on other sites

 

I know this jib and worked successfully with it before. This accident was caused by a absolute lack of balance on the jib arm. No matter how high a jib arm is on that column, if it's in balance, things are fine

 

 

Thanks for the link, it's much clearer now how it works - (the 'connector rod').

 

Sure if it's balanced, then it is balanced ;) ...very easy to convince yourself statically, and when it's low. But any accel/decel along the track especially with the arm aligned with the direction of track will create a dynamic consideration that is more complex. Irregularities in the track and/or dolly wheels and the height of the CoM all exacerbate the effects - for better or worse - the better case not usually taken notice of.

 

...then the camera operator puts his leg out.

Link to comment
Share on other sites

@Bjorn

Thanks for that. Mana from heaven (simple facts falling from the sky).

I counted about 3-1/3 counterweights on the jib arm, while some spare masses were sitting on the dolly platform.

Can one easily remove the link rod so that the jib arm is free and one can balance that, progressively, with the camera end fully loaded?

 

@Chris

Assuming a balanced jib arm and a centralized system Com, then there are safety margins. The designers will look at the moments created by the "dynamic" behaviour, and could easily give some recomendations on height, mass distributions, rotation rates and angular acceleration such that some safety margin is preserved. The angular velocity is not hard to measure. Once the grip has that down, then it can be measured how long he pushes for to get that, hence the angular acceleration. All put in simple tables, it could be easy

 

@Anybody

Did someone put their hand up earler quizing how jib arms can stay in balance with the camera at varying heights. If there was no parallelogram keeping the camera end level, then a teter toter like that can stay in balance easy. But when there is a parallelogram keeping the camera platform level, then the contribution from the platform towards the moment arm does not change with arm rotation (cam rising/falling). So there can be a contribution from the platform towards being out of balance. As the platform is always level, that little piece of the moment arm (lever arm giving the torque due to gravity), never changes, while the contribution to the moment arm from the gib arm as it rotates (cam up/down), is changing. As the camera end swings upwards past horizontal, this increment to the the moment arm on the camera end (increasing the moment arm), is a bad contribution. I haven't looked at how significant it is, but if the balance is out of whack to bein with, this may be the last straw.

Don't worry Chris, I'm sure everyone will get to the inertia with rotation issues eventually.

Link to comment
Share on other sites

I agree, it's easy - but it remains more complex than the static case.

 

Also once these limits on the acceleration (decel) are known as numbers, how does a grip then use them ? ...and then throw in a bad bit of track etc.

 

I think you're simply left with relying on the experience and general sense of the grip.

 

As interesting as these discussions are (for those inclined), they remain as academic as the reciprocal of our interest in becoming dolly grips :)

Link to comment
Share on other sites

 

Also once these limits on the acceleration (decel) are known as numbers, how does a grip then use them ? ...and then throw in a bad bit of track etc.

 

 

I don't think one can account for everything, but I think it's reasonable to expect the grip not to squander his safety margins.

If some maximum rotation rates or angular accelerations are given by the designers, then I think those same designers or the technical writers working for them could perhaps come up with some procedures that a grip could adopt. Measuring a rotation rate is not dificult. Measuring an angular acceleration is also possible with a simple procedure. How many seconds was the grip pushing to give that rotation rate. Yes Chris, some simplifying assumptions...the angular acceleration may not be constant, and so on....

 

If the grips understood the importance of centering the system CoM over the dolly truck footprint, balanced the jib arm, and had some guideline tables on height refered to mass, rotation rates and acceleration, some simplified procedures to check such, then a good safety margin could be preserved. Some of these things might have to be part of a small training seminar offered by the designers or rental companies. Maybe those kind of seminars already exist.

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

 Share

×
×
  • Create New...