Position Resolution and Accuracy of a PSD

[simasa]

Hello,
I want to use a PSD (Position Sensitive Device) and a laser beam for the position measurement. Laser beam will be stationary and PSD will be moving (50 mm/s).

PSD’s are available in different sizes (4 x 4 to 20 x 20) mm. Bigger the size of the PSD, less the resolution and accuracy. It is mainly due to more noise in the signal for bigger PSD surface. Position resolution may also depends on some other factors like stability of laser beam.

From one corner to other of the PSD; a voltage signal of 1-10V. To get the position resolution of 1µm, about 1 mV signal must be readable for 10 x 10 mm2 PSD. Its mean noise should be less than 1 mV.

Now there is no theoretical way to calculate reliable resolution and accuracy of such an application. One should try.
Before I try, perhaps some one could benefit me from his experience.

Which maximum resolution and accuracy can be achieved from this setup?

Regards
simasa

 

[IRstuff]

You need to provide more information.

> How is the signal to be read?
> What is the readout noise?
> How is the signal configured? A CW signal would have different noise than a pulsed signal.
> What is the size of the laser spot relative to your desired measurement resolution?
> What is the spatial profile of the laser spot?

TTFN

FAQ731-376

 

[simasa]

Thanks for the reply.
I am currently thinking if I should try to develop thins application or not. The hardware and signal configuration is not fixed at this point.

The things which I can fix now is:
about 9x9 size PSD

This is correct that that spot size, relative size of spot to PSD, signal configuration, readout noise all play a role.

Which company PSD to buy, laser of which properties to buy, what should be the optimal PSD to spot size ratio, what is optimal spatial profile (perhaps Gaussian), what is the best way for signal configuration, PSD safety against surrounding light, how to calibrate a PSD for optimize linearly, etc. would be actually second step questions which I may also like to ask in this forum.

At this moment, let’s assume that all these things are optimal to achieve maximum position resolution and position accuracy.

I need position resolution of 1µm and better and accuracy also within 5µm.
Can I practically achieve thes targets with such setup?

Some one having experience with such a use of PSD may be able to answer.

If possible, I would be working on finding optimal hardware and optimal signal configuration.

I would be thankful for any further assistance.

Regards
simasa

 

[IRstuff]

Still unclear whether you've given enough information.

These guys claim 1 ppm spatial resolution is achievable:

http://www.on-trak.com/theory.html

TTFN

FAQ731-376

 

[simasa]

Thanks for the reply,
Yes some manufacturers claim 1µm resolution. But it may be the ideal case with PSD in stationary condition.
If some body may have experience with practically realizing some position measurement application using a PSD, that would be great information for me.

 

[IRstuff]

Well, looking back at your OP, I would question whether the overall concept is practicable. 1 um resolution at 50 mm/s is about 5 microseconds measurement time to get any real level of precision.

Do you have a stable track and stable drive? Is the system on an air table?

TTFN

FAQ731-376

 

[IRstuff]

Most other systems that I can recall use either laser interferometers or encoders for anything with large travel.

TTFN

FAQ731-376