This is an area where there are many misconceptions about what is required to keep a camera steady in the air. The level at which people will go to achieving perfection is a personal one and for most of us the commitment has to be balanced with that which is affordable and practical. It is also horses for courses, many will be happy just producing their own vimeo or utube streams. Others may need broadcast TV or even Cinematic quality rushes. The fact is that most modern IS lens, when mounted vibration free, can give a quality that will suffice for a lot of general uses without any moving camera mount at all.
As a starter, here are some observations. Most people will have invested heavily in an under-slung type of camera mount. It is a very practical solution but not altogether an ideal one. In an effort to gain the best possible results lets consider some facts. If you hung a servo motor in space and then attached a weight of equal mass to the output arm you would get an opposite and equal reaction. i.e. the both components would move in opposition to each other. So, based on that fact. If you place a heavy camera on a platform and try to move it you will get a reaction in the opposite direction. So to stop that occurring we need a fairly large mass in the form of a camera mount. That means additional weight which we don't want. So one thing that can be done to reduce some of that unwanted reaction is to make sure that the axis of movements passes thro. the cameras center of mass.
Checking out many videos and posts it seems that the most common problem is that of vibration. So we attach our mount to the model with silicon or rubber soft mounts. whereas that cures one problem it now introduces another. As the camera moves about it causes our friend to reappear. (the opposite reaction force). So to reduce that effect we need to place our soft mounts as far apart as possible. The worlds best electronic camera mounts use five axis of stabilization but that is not really what we need here because we already have what is termed an outer axis - the model itself. It is very good in the Z axis (yaw) and as such we can use the model to steer the camera in that axis. So for pitch (nick) and roll we tend to use the two outputs from the MK FC. But these outputs use gyros that are 300 degrees/sec (Full scale range) which have an output of only 2 mv/deg./sec. which will oscillate long before we get to sensible stabilization for an inner axis. Alternatively a gyro that is only 20 deg/sec. (full scale range) has an output of 50 mv/deg/sec and is not expensive as it is the type used inside many IS camera lenses. So that's it for a starter, hopefully you guys can add your own theories and share your experiences.
As a starter, here are some observations. Most people will have invested heavily in an under-slung type of camera mount. It is a very practical solution but not altogether an ideal one. In an effort to gain the best possible results lets consider some facts. If you hung a servo motor in space and then attached a weight of equal mass to the output arm you would get an opposite and equal reaction. i.e. the both components would move in opposition to each other. So, based on that fact. If you place a heavy camera on a platform and try to move it you will get a reaction in the opposite direction. So to stop that occurring we need a fairly large mass in the form of a camera mount. That means additional weight which we don't want. So one thing that can be done to reduce some of that unwanted reaction is to make sure that the axis of movements passes thro. the cameras center of mass.
Checking out many videos and posts it seems that the most common problem is that of vibration. So we attach our mount to the model with silicon or rubber soft mounts. whereas that cures one problem it now introduces another. As the camera moves about it causes our friend to reappear. (the opposite reaction force). So to reduce that effect we need to place our soft mounts as far apart as possible. The worlds best electronic camera mounts use five axis of stabilization but that is not really what we need here because we already have what is termed an outer axis - the model itself. It is very good in the Z axis (yaw) and as such we can use the model to steer the camera in that axis. So for pitch (nick) and roll we tend to use the two outputs from the MK FC. But these outputs use gyros that are 300 degrees/sec (Full scale range) which have an output of only 2 mv/deg./sec. which will oscillate long before we get to sensible stabilization for an inner axis. Alternatively a gyro that is only 20 deg/sec. (full scale range) has an output of 50 mv/deg/sec and is not expensive as it is the type used inside many IS camera lenses. So that's it for a starter, hopefully you guys can add your own theories and share your experiences.
Last edited by a moderator:
