Calibration

[alanambrose]

Hi,

I always guessed that the cal camera was the same as the fid camera - mainly because it needs to be fairly small and presumably if you had written a bunch of camera code, you would tend to want to use some of the same code for the cal routines - which I'm suggesting implies the same camera. I can always be wrong. If I remember correctly the fid camera is a Watcom WAT-704R G3.8 - and you can still buy these new (for about £125 + VAT). It would be worth double checking I have remembered the right camera. The specs are here:

https://www.watec.co.jp/English/eng_spec/WAT-704R_G3.8_eng.pdf

Although the Wacom has AGC etc, I would have thought your eBay camera would work as it seems to have similar specs. I remember that there's some camera routines (maybe DOS based) that allow you to test the camera capture from the Imagenation board outside of the RV software. I can dig out my CAD for the holder for the Watcom camera if anyone wants to 3d print that.

Alan

[Mickyblueeyes]

Thanks Alan I will look see if I have the dos routines and go from there, the ebay cam has quite a good picture and can be focussed to quite a short distance maybe 10mm off the deck just a pity it isn't working. I didn't go for the Watcom because its quite long and needs to be a fair height above the grid to focus and have the dot at a usable size.

Micky

[alanambrose]

Hi Micky,

I think you boot into DOS and the utilities are in an 'Imagenation' directory - it was 3 years ago that I used them.

An image for the sleeve for the cal camera (assuming a WAT-704R) that I designed and had made is below - I've never had the time to test it though. From memory, it was meant to mount instead of the 'nozzle pick-up' - the thing with the two o-rings on - with some slits for the a-axis optical homing switch.

Alan

[Mickyblueeyes]

Thanks Alan, I have noticed a possible problem with the Ebay cam and the spec for the Watec mentioned in other posts, my Ebay Cam is PAL 50Hz and I have seen mention of the  EIA spec which is based on the NTSC standard, is the card expecting an NTSC style 60Hz signal?

Micky

[alanambrose]

Hi,

No I don't believe so - it's CCIR as far as I know. Unfortunately my RV is about 85 miles away atm otherwise I would check. One possible test is to check the Imagenation utility sees your FID & main cams OK. It might even give some camera data, sorry I can't remember whether it does. I have found though a copy of the PXC 200 docs and utilities:

http://anagram.net/nuts/Versatronics/PXC%20200%20card/

There are two versions of the utilities there - which version I used I'm not sure.

Alan

[Mickyblueeyes]

thanks for the files Alan, I will let you know how I get on.

Micky

[Jason]

Oooh, more PXC documentation. I'd started reverse engineering this so I could
replace the PXC board with a shim through to modern Windows capture subsystem
but it was slow going and I'd lost enthusiam a bit. This might give me the boost
to see it through, thank you!

[basemetal]

Its been a while since i posted, but i have had to stick my nose in our RV4's to sort out a little problem and so got caught up again in the forum

I did get my outstanding calibration bits from DC so i now have a complete original calibration kit - except the 4 spot lights that were placed in each corner of the machine to illuminate the grid. I have attached the pictures of the bits as calibration parts1 and calibration parts2. Somewhere on the forum is a pile of drawings in some obscure cad format that gives the dimensions, but for clarity i have attached them in pdf.

I was totally thrown by a drawing that showed orientation of the camera holder somewhere on the forum - but it appears wrong. I know that the fiducial camera was used when doing the calibration and know that the mirror needs to be held up with a cable tie or similar. I believe that the stepper motor is removed and the camera adaptor fitted ON TOP of the rotating bit (shown in Fitup of camera holder). The camera then looks down through the spindle, after removing the plastic plug. The camera is then held and adjusted with the hex head screws to get perfect concentricity.

Cheers

[basemetal]

Hmmm, for some reason it didn't post the pdf's. 2nd try.

Cheers

[basemetal]

This is a method that we have used when fixing the light ring on the fiducial camera to re-align it, or to get it to align with the arm picking tip when we have removed it.

Put a small blob of Prestik on a a PCB mounted in the machine. Poke the tool tip into the blob very gently using 10th increments. Switch to the fiducial camera and you will see just how much you are out by.

You can only adjust in x/y directions when the arm is in a specific position. The reason is you have to imagine three separate planes or grids (one for the fiducial camera, one for the fixed camera and one for the OTF camera) that all have to be related back to the master grid which is the arm placement grid. This step sets the fiducial camera to the placement tip.

Hope this helps someone.

[basemetal]

I found some of my notes that i made a number of years ago when DC did some adjustments on one of our RV4's and i have written them up. They really follow on from the camera calibration routine (need to be done immediately after doing that).

Cheers

[Jason]

They really follow on from the camera calibration routine (need to be done immediately after doing that).

Ah, this is brilliant, I had a long standing problem with rotated parts that I could never
cure using camera alignment alone. This is just the information that fixes it. Thank you!
When using the smallest parts, I had been laying boards out with as many components
at 0 degrees as possible    I won't need to do that any longer!

[trev]

I found some of my notes that i made a number of years ago when DC did some adjustments on one of our RV4's and i have written them up. They really follow on from the camera calibration routine (need to be done immediately after doing that).

Cheers

Thanks a lot for sharing these notes.