Repeatability, Accuracy or Resolution

[BuckeyeDennis]

Here's another consideration, for what it's worth.

A Wixey measures the orientation of a surface relative to the gravitational vector. I have no reason to doubt that it does so within the claimed accuracy of +/- 0.1 degrees.

Now when you go and measure the orientation of a second surface, that orientation is also measured with respect to the gravitational vector. If what you really want to know is the orientation of Surface B relative to Surface A, all you have to do is subtract the two gravity-based measurements, just like in Dusty's example.

Except: both readings have a possible error of 0.1 degrees. Those errors may offset one another, or they may combine to double the relative-angle error. Bottom line, the guaranteed accuracy of the relative-angle measurement is now +/- 0.2 degrees.

[Gene Howe]

Tell me if this is the best way.
To get a bevel angle, I zero the Wixey on the blade then move it to the (already closely angled) table. Then adjust the table to the angle needed.
I wish magnets worked on aluminum.

[JPG]

Moving a measurement device about an irregular surface and getting differing readings has nothing to do with accuracy or repeatability. More than likely the irregularities will exceed the influences of technique.

When using the wixey(or similar), stiction has to be allowed for and is likely the determining factor of all it's accuracy deviations.

I think that is what I have been trying to say. Dusty seems to disagree for some reason.

I don't know why stiction is a factor. Stiction is the FORCE required to to overcome static cohesion and slide two stationary objects in contact with each other on a parallel surface. How does that translate to reading differences?

Stiction in the measuring device(the wixey 'bearings').

[BuckeyeDennis]

Moving a measurement device about an irregular surface and getting differing readings has nothing to do with accuracy or repeatability. More than likely the irregularities will exceed the influences of technique.

When using the wixey(or similar), stiction has to be allowed for and is likely the determining factor of all it's accuracy deviations.

I think that is what I have been trying to say. Dusty seems to disagree for some reason.

I don't know why stiction is a factor. Stiction is the FORCE required to to overcome static cohesion and slide two stationary objects in contact with each other on a parallel surface. How does that translate to reading differences?

Stiction in the measuring device(the wixey 'bearings').

I don't think a Wixey is a pendulum-type device. Probably based on a 2- or 3-axis MEMS accelerometer, with digital calibration curves. So no stiction.

[JPG]

. . .
I don't think a Wixey is a pendulum-type device. Probably based on a 2- or 3-axis MEMS accelerometer, with digital calibration curves. So no stiction.

Howcum it acts like it?

Well my igage one does.

[dusty]

Moving a measurement device about an irregular surface and getting differing readings has nothing to do with accuracy or repeatability. More than likely the irregularities will exceed the influences of technique.

When using the wixey(or similar), stiction has to be allowed for and is likely the determining factor of all it's accuracy deviations.

I think that is what I have been trying to say. Dusty seems to disagree for some reason.

I don't know why stiction is a factor. Stiction is the FORCE required to to overcome static cohesion and slide two stationary objects in contact with each other on a parallel surface. How does that translate to reading differences?

Stiction in the measuring device(the wixey 'bearings').

I don't think so. Read "How Does It Read". The specifications might be enlightening too.

http://www.wixey.com/anglegauge/faq/index.html#03

[BuckeyeDennis]

. . .
I don't think a Wixey is a pendulum-type device. Probably based on a 2- or 3-axis MEMS accelerometer, with digital calibration curves. So no stiction.

Howcum it acts like it?

Well my igage one does.

Because my hunch was wrong!

From Dusty's "How Does It Read" link, a Wixey does appears to be a pendulum-type device, with a non-contacting angle sensor. Except they call it a "rotating counterweight".

Back around 1980, I used a fairly high-end oil-damped pendulum sensor on a robotics project. Accuracy on that was only about half a degree, IIRC. But the sensor on that device was a potentiometer, which was probably the main source of stiction. I later replaced it with good accelerometers, which were quite expensive back then.

[dusty]

Tell me if this is the best way.
To get a bevel angle, I zero the Wixey on the blade then move it to the (already closely angled) table. Then adjust the table to the angle needed.
I wish magnets worked on aluminum.

What you are doing is correct and will give you the proper tilt angle but there is one thing to consider.

You must maintain an angular relationship between the position of the Wixey when attached to the blade and the Wixey when attached to the tilted table.

To do what you are saying, I use my Incra 120 Miter Gauge. The protractor is magnetic. You can set the Wixey on the protractor and it will adhere. With the miter gauge set at 90 degrees and the Wixey positioned against the vertical surface of the protractor, you have established that relationship.

Now when you tilt the table, the Wixey will give you an accurate tilt angle.

What happens if you do not maintain that angular relationship? Simply unlock the protractor and rotate the protractor to something other than 90 degrees. You will see the Wixey "tilt angle" change but the tilt angle of the main table has not. This is, in my opinion, a technique error.

[algale]

I have a Wixey-like Beall Tilt Box. I use it for angles other than 90/45.

For 90 degrees, I now prefer a small 4 inch engineer's square. For 45 degrees, I use a small 45 degree square like this one.

[Gene Howe]

Thanks Dusty. That makes sense. I never thought of using the miter gauge too.