Brown & Sharpe MicroVal Axis Error Fixed - Page 2
Close
Login to Your Account
Page 2 of 2 FirstFirst 12
Results 21 to 38 of 38
  1. #21
    Join Date
    Jul 2018
    Country
    UNITED STATES
    State/Province
    New Hampshire
    Posts
    171
    Post Thanks / Like
    Likes (Given)
    1
    Likes (Received)
    35

    Default

    I want to suggest something to consider. With most optical read heads the
    A and B phase have mostly separate circuits. The optical sensor and circuitry
    will be entirely separate. The light source may or may not be separate.
    Since both your A and B phase are bad it is likely that it is not the sensor
    part of the read head that is bad. More likely is is some of the common
    components.

  2. #22
    Join Date
    Oct 2021
    Country
    SWEDEN
    Posts
    18
    Post Thanks / Like
    Likes (Given)
    5
    Likes (Received)
    0

    Default

    Im not sure how you mean now... do you mean a head from a working axis versus a head from the non working axis?

    Your previous pictures shows the sinus you shall get here...
    No square waves... the modulation created by the filter fixed on the read head and the moving scale glass piece are photo etched so that they produce a sinusoidial signal.


    So im missing something here...

    ( And please i would really apreciate a picture of the component values of the cable monted pcb.)

  3. #23
    Join Date
    Oct 2021
    Country
    SWEDEN
    Posts
    18
    Post Thanks / Like
    Likes (Given)
    5
    Likes (Received)
    0

    Default

    Exactly, Im on your track Mr_CNC_guy..
    With the reservation that instead of two, there are four separate electro/oprical circuits in these read heads, but the signals are combined two and two to produce the A/B or sin/cos signals.
    I think that the electrical stuff is ok… but there is something strange with the optical stuff….

    To further explain whats going on in these read heads i searched for som examples and found some….

    The read heads use the principle of the moiré effect, ie the interaction between two separate patterns to produce light and dark areas when the patterns interact.
    Some pictures…

    952fe8e6-10ce-4789-8997-e649f721c3e7.jpg
    d2c88b71-5432-4fd7-b683-09bbc5546b90.jpeg
    fb28f55a-026e-4042-b445-d2a2cb49bb8f.jpg

    The first shows the principle setupof the sensor system, allthough there are four like these next to each other. The filter closest to the Led and photoelement is fixed to the read head. The scale below moves in relation to the read head, and thereby causing the moaire effect , which causes light to reflect more or less into the sensor, depending on the filters relative positions.
    The second picture shows an attempt to illustrate the four areas that the four sensors collects light in. And the third image shows the moiré effect when the filers move relative to eachother. (Animated gif does not seam to work here… but see moiré effect on wikipedia for a working example)

    The four sensor areas will se differnt amount of light due to the moiré effect. The red and blue areas will be opposits of each other, ie max light in red are, while the blue gets the least. And same goes for the green and yellow.

    The output signals from the sensors is corresponding to the amount of light hitting the sensor.
    The red and blue sensors are used as a differential pair,and the green and yellow is the other pair. Representing the SIN/COS signals. The differential pair is summed in a diff amp to create a single ended signal, where the opposits add upp, and their identical offset is cancelled. (Diff amp does; Red minus Blue signal. And when red is at its max, the blue is at its lowest, the signal becomes the sum of these two signals) Another diff amp summs up the green and yellow signals.

    So after the diff amps we have two signals of identical sinusodial signals with the same amplitude (when the read head is moved along the scale) but with different phase. These can be ploted on an x-y graph, and will then ilustrate a vector plotted in a unit circle defined in trigonometry. By using trigonometry math we can calculate an angle of the vector, i.e 0-360 degrees witch represents a movement between the two ‘filters’ by one line. Which is in the ballpark of 10-40 micro meters on these type of scales. (Im not shure of which resolution the B&Scales actually have.)

    If one of the filters is missing… there will be the same amount of light hitting all the sensors all the time due to the missing moiré patterns…

    That is what I think is going on in the faulty scales…. Or maybe. Not going on….

    The two filers must have identical line spacings, otherwise there would probably be no noticeable moiré effect visible…

    Edit… hmm by thinking a little bit of it… to call this moiré effect is technically wrong… but the effects are very similar. As the line spacing is identical on the two filters, it is probably more correct to call this an interference effect…
    The difference between the four areas is produced by offsetting their pattern a quarter of their linespacing for each area…
    Attached Thumbnails Attached Thumbnails 95e2a79d-b626-4dbb-aad4-1125f163d4b3.jpg  
    Last edited by malman; 12-04-2021 at 11:15 PM. Reason: Clarification

  4. #24
    Join Date
    Jul 2018
    Country
    UNITED STATES
    State/Province
    New Hampshire
    Posts
    171
    Post Thanks / Like
    Likes (Given)
    1
    Likes (Received)
    35

    Default

    From the picture of the two read heads, one good and one bad, it looks like
    the filter is missing. That would be the common fault to effect both
    the A and B phases. You will not get the necessary interference pattern.

  5. #25
    Join Date
    Oct 2021
    Country
    SWEDEN
    Posts
    18
    Post Thanks / Like
    Likes (Given)
    5
    Likes (Received)
    0

    Default

    Quote Originally Posted by Mr_CNC_guy View Post
    From the picture of the two read heads, one good and one bad, it looks like
    the filter is missing. That would be the common fault to effect both
    the A and B phases. You will not get the necessary interference pattern.
    Exactly....

    agftemp, I think you should remove a read head from one of the working axis to compare... and see if there is a filter there, the one we think is missing on the troublesome axis…

  6. #26
    Join Date
    Aug 2017
    Country
    UNITED STATES
    State/Province
    Michigan
    Posts
    31
    Post Thanks / Like
    Likes (Given)
    0
    Likes (Received)
    2

    Default

    The pictures below show the comparison between a good reading and a non reading head. The glass element is not present on the non-reading head. I know this looks obvious, but I didn't have any knowledge of how these operated and I wasn't sure if the glass was for protection or if it served a function.

    Also I didn't have any history of the machine and it appeared that the glass was removed. I'm assuming that was because the head stopped reading and someone tried to diagnose it. In addition I bought another head from ebay and it did not have the glass.

    I thought it might be possible to determine if one of the LEDs or sensors had failed. This is why I used a camera to verify the LEDs worked and I thought it would be possible to check the function of the sensors by measuring voltage at the connector for each sensor. Without moving the axis, each sensor output ~1.5v so they were all producing a signal. Next I measured voltage of each sensor while moving the axis. The result looked like noise. I repeated the procedure for the good reading and non-reading heads and got the same results.

    With this information it appears that you guys are correct and that the glass element is needed. Tomorrow I will look at the glass under a microscope to see if there are any visible lines.

    Now, the next question is where can I get a glass element or a replacement reading head? Does anyone have any recommendations?

    Thanks,
    Alan
    Attached Thumbnails Attached Thumbnails reading-head.jpg   non-reading-head.jpg  

  7. #27
    Join Date
    Aug 2017
    Country
    UNITED STATES
    State/Province
    Michigan
    Posts
    31
    Post Thanks / Like
    Likes (Given)
    0
    Likes (Received)
    2

    Default

    Malman I want to return the favor and provide pictures of the cable mounted PCB, but can you explain more what you are looking for.

    Thanks,
    Alan

  8. #28
    Join Date
    Oct 2021
    Country
    SWEDEN
    Posts
    18
    Post Thanks / Like
    Likes (Given)
    5
    Likes (Received)
    0

    Default

    Great that you found the issue…
    I now hope you can find a complete readhead for your unit.

    The info I’m looking for is clarification of the component values of the resistor net and the potentiometers on the pcb depictured in your post… thread post #12.
    ( the right picture )
    So either some pictures of the component texts on these components, or just write them down and forward to me.

    I saved the CMM i got from going into the dumpster, but all electronics/computer system was all ready gone. So I have to reproduce the interface electronics myself. I am an electrical engineer by trade, so i see no major issues to build a rudimentary measuring platform with what i got and a raspberry pi or something similar.
    I have figured out how to connect and interface the read head, I’m just not sure about the appropriate currents that shall be applied to the read heads. So if i get hold of the component values I’m home with this part.
    I have an idea about current levels and component values, but if i get it wrong i might fry the read head, and i prefer not to try that if i can help it…

    After verifying the read heads I have to tackle the hose/tube swap as they fall into pieces by just looking at them..

    I guess that a picture of the back side of the same PCB would not hurt either, just for completeness…

    Edit: Btw.. the read heads seams to be available from the current owner of the brands, Hexagon. I asked for some info from then, but they had no info on these machines, but they replied that they got 4 read heads in stock here in Sweden..
    Im not sure i want to know the prices of those, but at least they are not unobtanioum… they were a so called slow article, i.e. not many sold over the years….

  9. #29
    Join Date
    Aug 2017
    Country
    UNITED STATES
    State/Province
    Michigan
    Posts
    31
    Post Thanks / Like
    Likes (Given)
    0
    Likes (Received)
    2

    Default

    I am bummed that I can't fix the read head myself. I don't need a CMM, just wanted to get it working to learn how they operate. Aside from the low initial price, I've had to spend money on the hose, and power supply. I'm scared to call Hexagon to find out the price. That is why I was so excited to find one on ebay, and it was only 30 miles away from my home. Unfortunately it also didn't have the glass on it. I've only encountered three read heads, and two of them didn't have the glass on them, seems unusual and suspicious. I took more pictures of the potentiometers and posted them below. Hopefully this is what you were looking for. If not, let me know.
    Attached Thumbnails Attached Thumbnails pot-pic-3.jpg   pot-pic-2.jpg   pot-pic-1.jpg  

  10. #30
    Join Date
    Jul 2018
    Country
    UNITED STATES
    State/Province
    New Hampshire
    Posts
    171
    Post Thanks / Like
    Likes (Given)
    1
    Likes (Received)
    35

    Default

    Could you replace the entire scale with a cheap Chinese scale?

  11. #31
    Join Date
    Oct 2021
    Country
    SWEDEN
    Posts
    18
    Post Thanks / Like
    Likes (Given)
    5
    Likes (Received)
    0

    Default

    Thanks a lot Alan.

    So close, but still not there…..

    The numbers i am after did not show up in the pictures…

    First, the potentiometers… im not shure about the three others but the VR1 marking says that it is a Bourns 3328H potentiometer.. but the last numbers was hard to read but i believe they are 10x (x is hard to figure out in the second image.. 2,3,6,8,9. The other pots seams to be of another brand or type… but if we can get the number of vr1 the others don't matter… they will be the same resistance.

    And then to the most interesting part… the yellow resistor pack RP1.
    As of Murphys laws, of course it had it markings to the ‘hard to see’ side…

    I think that the first line of markings ends 102, which is the resistor configuration number, if it is a bourns part this means that it has 4 individual resistors, connected pin 1-2, 3-4, 5-6, 7-8. Which makes sense in this app. There are also other configurations.
    The second line shall have the actual resistance values… can this be read or are they totally blocked by vr1? If so, can you measure the resistance of these resistors (in circuit) with a multimeter?

    And just for reference, snap a picture of the bottom side of the pcb, showing the copper traces.

  12. #32
    Join Date
    Aug 2017
    Country
    UNITED STATES
    State/Province
    Michigan
    Posts
    31
    Post Thanks / Like
    Likes (Given)
    0
    Likes (Received)
    2

    Default

    I took more pictures. Hopefully you will be able to get the information you need. It is only possible to add 5 pictures per post so I will do this in a few listings.

    I tried to measure the resistance on the long yellow component. The values did not hold and would drift from a low value down to zero. Really strange and not sure why this was happening.
    Attached Thumbnails Attached Thumbnails 20211206_200734.jpg   20211206_200749-1-.jpg   20211206_200758.jpg   20211206_200820.jpg   20211206_200839.jpg  


  13. #33
    Join Date
    Aug 2017
    Country
    UNITED STATES
    State/Province
    Michigan
    Posts
    31
    Post Thanks / Like
    Likes (Given)
    0
    Likes (Received)
    2

    Default

    More pics. They show some different views.
    Attached Thumbnails Attached Thumbnails 20211206_200859.jpg   20211206_200905.jpg   20211206_200914-1-.jpg   20211206_201822-1-.jpg   20211206_201853.jpg  


  14. #34
    Join Date
    Aug 2017
    Country
    UNITED STATES
    State/Province
    Michigan
    Posts
    31
    Post Thanks / Like
    Likes (Given)
    0
    Likes (Received)
    2

    Default

    Last of the pictures I took. Let me know if this is able to help you.
    Attached Thumbnails Attached Thumbnails 20211206_202538.jpg   20211206_202432.jpg   20211206_202353.jpg  

  15. #35
    Join Date
    Jul 2018
    Country
    UNITED STATES
    State/Province
    New Hampshire
    Posts
    171
    Post Thanks / Like
    Likes (Given)
    1
    Likes (Received)
    35

    Default

    If you can read the numbers on the long brown part I could probably
    identify it for you. It is almost certainly a resistor package.
    They made them in a few formats. Sometimes they even printed the
    internal schematic on the side. Unfortunately the numbers are on
    the side against the other components.

  16. #36
    Join Date
    Oct 2021
    Country
    SWEDEN
    Posts
    18
    Post Thanks / Like
    Likes (Given)
    5
    Likes (Received)
    0

    Default

    Ahh.. yet another Christmas present...
    Big thanks for the pictures...
    unfortunately it seams to be hard to nail the right info..

    I hope you are willing to try another variant.

    i have drawn some guides in one of your pictures.

    pot-pic-2-edit.jpg

    This pot is marked with a Bourns logo and the text 3329H- and then i think it is 10? as shown in the picture.
    can you read this text? (The other three pots seams to be of another brand so they are not to be read)

    And for the resistors.
    As the yellow resistor pack is a single in line component it can be gently bent away from the pots so that the text can be read.

    (Regarding the measure values, i might have tricked you.. what i meant with my text 'can be measured in circuit' ment that is probably can be measured onboard the PCB, (but i should also have written that the PCB shall not be connected to the following interface electronics ))

    Best regards /Klas

  17. #37
    Join Date
    Oct 2021
    Country
    SWEDEN
    Posts
    18
    Post Thanks / Like
    Likes (Given)
    5
    Likes (Received)
    0

    Default

    Btw.. i realized i have made an mistake about the previous schematic... it was drawn with info from the Deva023 manual, witch i now realized is not valid for the original interfaces. (They seam to have their own pinout into their own interpolator)

    I have drawn a new schematic using pictures in this thread about these B&S Microval interfaces, which has fixed cables. ( There seams to be multiple versions of interfaces )

    This is how i think these interfaces is connected.

    .img_1135.jpg

    I could not make out some colors of the cables from the PCB to the Read Head, and the connections from the cable to the read head is a educated guess.

    Edit: the way the resistors are connected might explain why the measure a bit strange… if the pot is turn to one of the endstops it practically shorts the resistors..
    Last edited by malman; 12-29-2021 at 05:10 PM. Reason: better picture

  18. #38
    Join Date
    Aug 2017
    Country
    UNITED STATES
    State/Province
    Michigan
    Posts
    31
    Post Thanks / Like
    Likes (Given)
    0
    Likes (Received)
    2

    Default

    The journey continues. I was able to buy another CMM for less than what a new one cost. Turns out the new CMM was refurbished with all new read heads. They put two of the heads back in the plastic bag so I have them as spares. The third head was included and guess what, the glass had come off it. I suspect the glass came off causing the need for the repair, and while they were making the fix, they replaced all the heads. Unfortunately they changed the size of the read head so I can't use them directly on the old CMM. In addition, the glass is shorter so I need to "extend" it by adding another piece to keep the read head from tilting. I am going to see if I can rig together a solution by using the glass along with something to effectively extend the glass. Got my fingers crossed that this might work. The pic below shows the new smaller read head with the glass.
    Attached Thumbnails Attached Thumbnails 20220101_113213.jpg  


Tags for this Thread

Bookmarks

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •