Building of a Robust Spline Profile Filter using Labview

[Simone 27]

Hi Andrew,

Sorry for the delay again, i'm still banging my head off a wall with the physical build! I'm having alot of trouble trying to manipulate the part into moving and turning the laser on. It's looking like I might have to take the components off and try to power them in a seperate rig. (Don't really have time for this at this stage of the year though) I see what you are saying about the opening of two files. Thank you for the adjusted code. The filter is working perfectly giving me the required graph @ data 2. I'm still not sure how weighting equation would be implemented on the Spline Filter though. And if you have any thoughts on generating a control signal to power a laser without any access to information on the laser I'm all ears 🙂

I suppose it wouldn't be any fun if it was easy 🙂

[Andy.G]

Hey Simone,

As far as I’m aware, lasers in CD and DVD players work by shining onto a reflective surface of a CD/DVD and the reflection or lack thereof, will provide a series of binary data strings which then your PC will convert into a audio or visual track.

Therefore in order to have the laser determine the topography of a surface you would have to use a sensor that is capable of detecting the time taken for the light to reflect back from the surface. As I’m sure you are aware this will happen almost instantaneously, therefore the device you are using must be capable of measuring the differences in time taken for the light to reflect. I would suggest that this data is stored into memory, as there is very little chance that you will be able to do any further post processing on this data during the run time of the laser as you would lose the speed and determinism needed for accurate measurement.

I would strong suggest the use of a compact RIO for the use of this project, as the on board FPGA should provide the necessary acquisition speed to control the camera. As for the laser, I would keep it on until the objects surface is covered, this way you would only need to control the camera. You must however also consider the control of the device that will be moving the either the laser and camera or the substance you are measuring.

What’s the setup so far? How have you approached this?

[Simone 27]

Once the laser is turned on it will try to read a pit or no pit situation on the disk surface to export binary information to the player. As the disk has a wobble when rotating the laser must move in relation to this. It achieves this by use of a focusing technique where a voice coil motor will move the laser unit in and out towards the disk based on the error signal generated by a four quadrant photo detector which has the beam reflected back onto it.

It is this error signal that i am trying to tap into as it will be directly proportional to the micro peaks and valleys of the surface profile. My issue with the laser in the external bluray drive is how to actually turn it on and leave it on, by figuring out the signal that tells the unit to turn on its laser and reproducing it.

I have found a write dig port example in labview that i'm hoping i'll be able to use to move the stepper motor by soldering onto it directly and am trying to look at that today. I've had a fun hour trying some of the tightest soldering imaginable but that's done now.

I realize i will still have an issue of relating X coordinates to the Y coordinates i will gather from the laser movement, one possible alternative would be to use a machine table for the movement rather than the motor but this is not ideal.

I'm loving that every answered question brings another ten questions to the table 🙂

[Andy.G]

Hey Simone,

How is the error signal determined within the output of the photo-detector? What is the data output of the device, assuming that the error signal is bundled with the data taken from the disk. The fact that we don’t know much about the Blue Ray player makes life very difficult (why can’t Sony just release their intellectual property!).

What I would suggest is a scope, rip the device open, boot it and monitor the input of the Blue Ray system. If you can replicate the signal you should be able to at least activate the device. As for leaving it on, I would assume it remained on? It would a really expensive process to have an AICS built just to pulse the laser at high frequency.

I take it the stepper motor will just be rotating the disc you wish to measure? This should be fairly easy to control using the NI Motion SW. There is a really good online tutorial here. This will allow you to allocate the movement of the stepper motor via coordinates and therefore should make life a little easier.

In the mean time if you’re stuck on any LV programming, post it and I will see what I can do.

P.S. We would be very interested to see the results of the project with some pictures, videos and maybe a case study? I will speak to the marketing chaps here at NI and we can see if we can make it worth your while (get your work published).  

[Simone 27]

The stepper motor is used to drive the laser across the length of the disk(or in my case the polished surface) Your right about using a scope to monitor the signal that turns on the laser. Tricky work but i hope to get into it next week. I'm sure there is a LV feature that can monitor/record and reproduce a signal.

In a perfect world with more than one motion controller in our institute that motion sw would be great 🙂 The way i think it could be done without a driver would be to use the Write Dig Port example in LV and modify it to send the following sequence to the four cables of the motor(bi-polar stepper motor)

Phase1 00110011 

Phase2 11001100 

Phase3 01100110 

Phase4 10011001 

I'm not sure how i can do this though and ensure that the 4 pulses are sent at the same time.

As for the P.S. bit, thank you very much! It's nice to be reassured that what i started months ago is actually a good idea, that can get lost in the frustration sometimes 🙂 I wish i didn't have 6 other subjects distracting me, i've really enjoyed the research into this project.

[Andy.G]

Hey Simone,

Yes its quite easy to monitor and record the signal, I’m pretty sure it will be a digital pulse so this should not be a problem.

IF you wish to send the pulses at the same time you can set the write function to single channel multiple samples, this will allow you to send a binary pulse train to the output of your DAQ device.

What DAQ device do you have including modules?

[Simone 27]

I'm using a NI PXI-1033 that has two modules, NI PXI-6221 M Series and 6533 Digital I/O

[Simone 27]

Hi Andrew,

I'm just trying to tidy up the last bits of code today. If you were available for a few questions it would be greatly appreciated I'm trying to run the filtered signal through a high pass filter to effectively straighten out the spline filter mean line so i can start my numerical calculations but am having trouble converting the data type to something the new filter will read. Also do you have any pointers on the implementation of the filter equations mentioned earlier in this thread to the spline filter.

Le meas.

[Simone 27]

Sorry to be a little clearer...

I now have 480 xy data points taken from a white light interferometer to an excel file, taken from the excel file to labview, opened in labview and combined through a spline filter which has given me a mean line., this is then graphed, I now wish to pass the new graph of the original xy data and the mean line created by the spline filter through a high pass filter to remove the mean line (straighten it out) i then want to graph this. Next to remove the edge effect i wish to take data points 10-470 and create a new graph also i will use this last set of data to calculate my Ra values.

Le meas.

[Simone 27]

Something like this for the selecting of points and filtering...