BobL
22nd Feb 2016, 09:29 PM
Finally got a chance t take some pics of this today.
The is a bit of an exploratory project where I'm hoping to learn from mistakes. Hence its been made from scrap parts and in a way that can be easily altered.
The balancer is loosely based on this https://youtu.be/xP_Fc8actok (http://metalworkforums.com/redirect-to/?redirect=https%3A%2F%2Fyoutu.be%2FxP_Fc8actok)
Please note that as it is this rig is potentially dangerous and should not be copied without significant modification.
Here are a couple of pics of the whole rig.
361202
Here you can see how Unistrut Track has been used to position the main components so the components can be easily moved around or broken down for storage or reuse.
361203
A is a transfer pulley to arrange for the belt tension to be perpendicular to the direction that the floating carriage (contained inside C) can move.
This direction is parallel to the bench top allowing the vibration to be examined in just one dimension which makes things easier to manage.
B is a fixed shaft support and also contains the rotary position PD sensor (the black box and the SS wheel next to it on the shaft)
D is a SS test/balance wheel to which 3mm screws containing test/balance weights can be added anywhere on the wheel.
The object to be balanced is placed onto the shaft next to the test wheel.
Cat 5 cable is used to wire the two sensors to E which is a junction box that allows easier electrical access to the Accelerometer and rotary position outputs.
The test shaft sits on two pairs of bearings (one pair of bearings on the fixed carriage and the other pair on the floating carriage)and is currently only held down by the belt tension.
This allows the shaft size to be more easily changed although this will then require dedicated rotary encoder wheel, shaft pulley and test whee for each shaft size.
To tension the belt both the fixed and floating carriage sit on height adjustment platforms
It's far too dangerous to operate at speed and I need to add some bearings to the top that would prevent the shaft flying off.
I feel comfortable running the test wheel up to 30 Hz but won't try anything out of balance until added security is implemented.
Ultimately a steel mesh cage will be made to go over the whole shebang
361204
Here is a close up of the floating carriage (F) assembly
The carriage is suspended from 16 small bearings that allows the carriage to rock in the direction given by the red arrow.
E is the rotary position marker and A is the box that holds the Photodiode and a Photo detector/sensor.
B is the test wheel
361205
Here you can see the back of the rotary position wheel (E) half of which is painted matt black and the other half is left polished. As the B&W divide passes across the PD this drives the Black box output high and VV.
361206
Here is the PD box - its the dark blue dot near the top of the box.
361207
This junction box allows the output of the X/Y/Z components of the Accelerometer to be accessed.
These are the 3 BNC connectors in line near the top of the box.
Only one of these makes any physical sense and that the direction of the floating carriage.
The solo BNC connector on the lower half is the Rotary position state.
The black plug on the LHS of the box is 5V DC in from an old mobile phone charger.
I have hooked these up to a CRO and could see something meaningful but not very clearly.
Sorry no photos of the CRO trace as the CRO is currently out of action.
A first order balancing can be obtained by looking at the pair of CRO signals on the screen at the same time to determine the magnitude of the vibration and approx position of where weight has to be added to reduce the vibration.
Weights can be added to the test wheel to reduce the magnitude of the measured Vibe.
This is repeated and weights added and subtracted accordingly.
Sounds easy but is actually VERY confusing
There are more refined/precise/elegant ways of doing this that I am currently exploring.
Most of the components for this project came from stuff left over from other projects or scrap metal.
Even the nice SS wheels were picked up as scrap from a metal merchant
The 3P 2HP 2850 RPM motor was purchased several years ago for $30
The 4HP VFD is the VFD I use for general motor testing
Purchased specifically for this project were
- 16 8x22x7 mm bearings $13
- 4 , 17 x 40 x 12 mm bearings $11
- 2 belts $28
- the accelerometer cost US$1.83
- the rotary position PD cost US$1.30
- various plastic boxes, connectors, glands, nuts and bolts etc cost about another $30
All up about $100.
As I said in another post this is just the easy or beginning bit , now I need to make sense of the outputs and use those to make the balancing corrections.
The is a bit of an exploratory project where I'm hoping to learn from mistakes. Hence its been made from scrap parts and in a way that can be easily altered.
The balancer is loosely based on this https://youtu.be/xP_Fc8actok (http://metalworkforums.com/redirect-to/?redirect=https%3A%2F%2Fyoutu.be%2FxP_Fc8actok)
Please note that as it is this rig is potentially dangerous and should not be copied without significant modification.
Here are a couple of pics of the whole rig.
361202
Here you can see how Unistrut Track has been used to position the main components so the components can be easily moved around or broken down for storage or reuse.
361203
A is a transfer pulley to arrange for the belt tension to be perpendicular to the direction that the floating carriage (contained inside C) can move.
This direction is parallel to the bench top allowing the vibration to be examined in just one dimension which makes things easier to manage.
B is a fixed shaft support and also contains the rotary position PD sensor (the black box and the SS wheel next to it on the shaft)
D is a SS test/balance wheel to which 3mm screws containing test/balance weights can be added anywhere on the wheel.
The object to be balanced is placed onto the shaft next to the test wheel.
Cat 5 cable is used to wire the two sensors to E which is a junction box that allows easier electrical access to the Accelerometer and rotary position outputs.
The test shaft sits on two pairs of bearings (one pair of bearings on the fixed carriage and the other pair on the floating carriage)and is currently only held down by the belt tension.
This allows the shaft size to be more easily changed although this will then require dedicated rotary encoder wheel, shaft pulley and test whee for each shaft size.
To tension the belt both the fixed and floating carriage sit on height adjustment platforms
It's far too dangerous to operate at speed and I need to add some bearings to the top that would prevent the shaft flying off.
I feel comfortable running the test wheel up to 30 Hz but won't try anything out of balance until added security is implemented.
Ultimately a steel mesh cage will be made to go over the whole shebang
361204
Here is a close up of the floating carriage (F) assembly
The carriage is suspended from 16 small bearings that allows the carriage to rock in the direction given by the red arrow.
E is the rotary position marker and A is the box that holds the Photodiode and a Photo detector/sensor.
B is the test wheel
361205
Here you can see the back of the rotary position wheel (E) half of which is painted matt black and the other half is left polished. As the B&W divide passes across the PD this drives the Black box output high and VV.
361206
Here is the PD box - its the dark blue dot near the top of the box.
361207
This junction box allows the output of the X/Y/Z components of the Accelerometer to be accessed.
These are the 3 BNC connectors in line near the top of the box.
Only one of these makes any physical sense and that the direction of the floating carriage.
The solo BNC connector on the lower half is the Rotary position state.
The black plug on the LHS of the box is 5V DC in from an old mobile phone charger.
I have hooked these up to a CRO and could see something meaningful but not very clearly.
Sorry no photos of the CRO trace as the CRO is currently out of action.
A first order balancing can be obtained by looking at the pair of CRO signals on the screen at the same time to determine the magnitude of the vibration and approx position of where weight has to be added to reduce the vibration.
Weights can be added to the test wheel to reduce the magnitude of the measured Vibe.
This is repeated and weights added and subtracted accordingly.
Sounds easy but is actually VERY confusing
There are more refined/precise/elegant ways of doing this that I am currently exploring.
Most of the components for this project came from stuff left over from other projects or scrap metal.
Even the nice SS wheels were picked up as scrap from a metal merchant
The 3P 2HP 2850 RPM motor was purchased several years ago for $30
The 4HP VFD is the VFD I use for general motor testing
Purchased specifically for this project were
- 16 8x22x7 mm bearings $13
- 4 , 17 x 40 x 12 mm bearings $11
- 2 belts $28
- the accelerometer cost US$1.83
- the rotary position PD cost US$1.30
- various plastic boxes, connectors, glands, nuts and bolts etc cost about another $30
All up about $100.
As I said in another post this is just the easy or beginning bit , now I need to make sense of the outputs and use those to make the balancing corrections.