Re: rotations piezo system

On Nov 19, 4:49 am, corrie.fea...@xxxxxxxxxxxx wrote:
On Nov 18, 7:37 pm, Uncle Monster <uncle...@xxxxxxxxx> wrote:



On Nov 18, 6:05 am, Andrea Mazzolari <andrea.mazzol...@xxxxxxxxx>
wrote:

Have you looked into (absolute optical encoders) or
(ultra-high accuracy optical angle encoder systems)? I recall
reading about a new type some time ago that was extremely
accurate. A couple manufacturer's websites:

i have cheked those products, but it seams that they can not reach a
precision of 2 microradians.

I have contacted some manufacturers asking for a custom made solution.
In any case, in my opinion the best solution is based on piezo system.

I have found this:http://www.physikinstrumente.com/en/products/prdetail.php?sortnr=702500

Andrea

I wish I could find the information on the device
I read about. If I remember it used a cascade of
stages in one unit to achieve high precision. I
hope you will write about your application and
success with your project for us curious critters.

[8~{} Uncle Monster

I may be able to help you...

Renishaw make some very high precision angle encoders. In terms of
resolution, we can offer products with resolutions to 0.005 arc
seconds (that's 0.024 micro radians in real money). The question is...
do you need 2 microradians resolution, or 2 microradians accuracy?
Renishaw can help you with either case. If you need the former, then
we can achieve that quite easily with a variety of different ring
encoder products (take a look at renishaw.com and go to position
encoders, then select angle encoders). RESR and SiGNUM will both be
able to achieve that resolution - just select a ring size that fits
and go to the resolution table in the product data*** (also on-line)
for the correct interpolation rate. If you need 2 microradians
accuracy, then things get a little more tricky. There is a new product
called REXM that offers very high accuracy. The product is specified
to +/- 1 arc second total installed accuracy (+/- 4.8 micro radians),
but they are actually a lot better than that. If you choose one of the
larger ring sizes (300 mm +), the ring will achieve +/- 2 micro
radians total installed accuracy. To give you a guide, 183 mm rings
are specified at +/- 1 arc second total installed accuracy, but
actually typically achieve +/- 0.22 arc seconds.

The REXM is designed for use with two readheads. We make a box called
DSi that takes the signals from two heads and combines them so it
appears as one very high accuracy encoder.

If you need to work with a smaller ring, you could use correspondingly
more readheads and still achieve the required accuracy, but you will
have to average the counts inside your controller. If you want the
really highest accuracy, then you can use as many heads as you can fit
around the ring. When you use a lot of readheads, the limiting factor
on an encoder will be the sub-divisional error (sometimes called the
"interpolation error" or "cyclic error"). This is the error within one
scale period, so it will not be compensated by multiple readheads.
Renishaw's SiGNUM encoder has just 30 nm sub-divisional error,
compared to 200 nm for competitor products.

One final thing to mention... to get this level of accuracy, you
really need to use a modular encoder. Using an enclosed encoder that
has its own integral bearing will not work because the coupling errors
will take up your entire error budget.

If you have any other questions, please let me know.

Best regards,

Corrie.

Thanks, but it is Andrea who is looking for the encoder
and www.renishaw.com is one of the links I posted.

I saw that Renishaw manufactures an optical encoder
that utilizes more than one pickup on the same encoder
ring and that is the unit I thought would work for Andrea.

[8~{} Uncle Monster
.

Quantcast