Repairing Heindenhain VRZ 753B + LS803 scales - Page 5
Close
Login to Your Account
Page 5 of 5 FirstFirst ... 345
Results 81 to 98 of 98
  1. #81
    Join Date
    Sep 2011
    Country
    GERMANY
    Posts
    2,970
    Post Thanks / Like
    Likes (Given)
    1902
    Likes (Received)
    1089

    Default

    Hi Bob,

    Quote Originally Posted by HBOB View Post
    Of interest to me is the reference to using a IR type LED as a replacement for the scales light bulb. Wondering if you tried the so called near or the far IR type?
    These Heidenhain scales use silicon solar cells. Those have a spectral response which extends up to wavelengths a bit longer than one micron (1000 nm). Spectral Response | PVEducation . The near IR LEDs emit at wavelengths up to around 950nm, so those are a good match. The far IR LEDs emit at much longer wavelengths, where the silicon solar cell won't respond. So use near IR LEDs.

    And what were/are your thoughts on LEDs usage as a bulb replacement.
    I've replaced one burned out bulb with an IR LED and am happy with the result. So if I have any more bulbs burn out I will do the same.

    Would you consider sharing some more information .about the 11 micro amp to ? volt converter you made? Know that I would be interested in a schematic an how you calibrated it. Also think other readers in the present or the future would find that information very useful.
    OK, I will do this at some point in the future. Don't have the time to do it now.
    Last edited by ballen; 01-16-2019 at 02:11 AM.

  2. #82
    Join Date
    Dec 2002
    Location
    Benicia California USA
    Posts
    8,290
    Post Thanks / Like
    Likes (Given)
    1928
    Likes (Received)
    2619

    Default

    Bruce has had better luck than i have with the LED conversion....
    Spent some time and energy making carriers and mounting LED's on the reader heads for the rotary encoder on the universal table...

    Had some success getting the scale to read and output a reasonable shaped signal, but could never get enough output to be reliable.....
    Eventually i went back to the incandescent lamp to get the encoder to provide a reliable output.....

    Made new carrier to hold the LED....Original and new .................



    Output side of LED




    New carrier..............



    I have drawings for the carrier if you would like.....



    Cheers Ross

  3. Likes ballen, Colt45, DennisCA liked this post
  4. #83
    Join Date
    Mar 2017
    Country
    LATVIA
    Posts
    510
    Post Thanks / Like
    Likes (Given)
    131
    Likes (Received)
    256

    Default

    Quote Originally Posted by jz79 View Post
    the old glue (most likely it is silicone) is still firmly stuck to the glass, my plan was to leave it there so it continues to work as the buffer for vibration and thermal expansion differences between the glass and the aluminum case

    anyway, it does seem to be some sort of a silicone, because the urethane dabs I placed on it cured and didn't adhere at all, could be easily picked off

    I ordered Dow DOWSIL 3145 link to product info pdf), from the description it looks to be the perfect stuff for this, pretty pricey, ~60eur for a small tube of 90ml, maybe some cheap sanitary silicone would also work, but I'm afraid of any unwanted additives those may contain, solvents or worse - acids etc, that might damage the deposited measuring stuff on the glass

    and as a backup I also ordered Loctite 770 primer, supposedly it lets CA glues to stick to silicone, but that will be the last resort, I'm afraid of the CA fogging up the glass scale, I could perhaps warm the glass up little bit and cool the aluminum case somewhat for any possible fumes to settle on the case, but that will be the last option
    small update, the Dowsil 3145 silicone glue worked really well, the bond to the old silicone glue on the glass scale was as strong as the parent material, and it adhered to the anodized aluminum case very well as well, so in case anyone has to glue the glass back into the case - this is the right glue to use

  5. Likes ballen liked this post
  6. #84
    Join Date
    Mar 2017
    Country
    LATVIA
    Posts
    510
    Post Thanks / Like
    Likes (Given)
    131
    Likes (Received)
    256

    Default

    not wanting to make a new thread for this (used search, but didn't find anything on the keywords I tried), a quick question regarding sine wave Heidenain scales:
    is there a module for sine wave output scales that can combine outputs of 2 scales and then send the signal to DRO or TNC?

    I know there is such a thing for TTL scales, but hadn't had the luck to google up one for Heidenhains old sine wave scales, perhaps I'm using wrong keywords

    reason for asking is the quill and column relation of the Picomax 51, quill has a rotary encoder that is coupled to the Z axis of the TNC, but the column has no scale (but has limit switches coupled to M codes that you can use in your programs to make tool changes quicker for instance), the problem is, whenever the column goes up/down, the Z value goes out the window, I thought about installing a scale for the column and combine the output of the quill and column to be sent to the Z input on the TNC, I don't expect micron precision or repeatability, I understand some errors may occur combining those signals, but still it would be a big improvement on the current setup

  7. #85
    Join Date
    Mar 2017
    Country
    LATVIA
    Posts
    510
    Post Thanks / Like
    Likes (Given)
    131
    Likes (Received)
    256

    Default

    I thought I'd post this short vid showing how the filter on the glass scale works, this time the scale is round though, notice the order of rotation of the 4 squares on the glass disc, I'm barely applying any pressure on the handwheel for the rotation of the squares to start, and then backwards


    visual representation why they are drawing circles on a scope

    couple pics of the insides of the HR 152/5000 rotary encoder, it came with the Mikron WF41 TNC355, didn't work, plugged in ROD 450/5000 encoder from Picomax, and it worked, so these 2 seem to be interchangeable

    HR 152/5000, will require some cleaning...
    20200318_194231.jpg 20200318_201131.jpg

  8. #86
    Join Date
    Oct 2017
    Country
    FINLAND
    Posts
    734
    Post Thanks / Like
    Likes (Given)
    121
    Likes (Received)
    226

    Default

    If I may use this thread to ask a Heidenhain question. I've noticed a weird behaviour in my VRZ 713B, it never goes back to zero. Never ever I cannot get it to zero except by zeroing the counter. It'll go to 0.010 of 0 but never to actually 0. instead it'll jump to -0.010 or -.015, same if I got from the other side. It skips the 0. I do not think this is proper behaviour but I do not know what the culprit could be, something in the electronics perhaps. It did have a fried diode bridge and old capacitors, but maybe some other components too where less than stellar shape.

  9. #87
    Join Date
    Sep 2011
    Country
    GERMANY
    Posts
    2,970
    Post Thanks / Like
    Likes (Given)
    1902
    Likes (Received)
    1089

    Default

    Hi Dennis,

    my Heidenhain does not do what you are describing, so I think something is wrong.

    The first thing is to understand if the problem is with the display or with the scales. Does this happen on all the axes? Can you go to other whole numbers, for example to 10.00 or 10.000? Does your display have a switch on the back to select XXX.XX versus XXX.XXX display format? Does the switch setting have any effect on the problem you have described?

    Cheers,
    Bruce

  10. #88
    Join Date
    Oct 2017
    Country
    FINLAND
    Posts
    734
    Post Thanks / Like
    Likes (Given)
    121
    Likes (Received)
    226

    Default

    I only have the one axis set up yet so I have only tested the Y axis. I have no problems hitting exactly 10.000 or 9.000, only 0.000 is a problem and it doesn't matter what resolution I use. I think it is an error in the DRO, because zeroing the scale anywhere does not make a difference.




  11. #89
    Join Date
    Sep 2011
    Country
    GERMANY
    Posts
    2,970
    Post Thanks / Like
    Likes (Given)
    1902
    Likes (Received)
    1089

    Default

    There is circuitry to handle the change in sign (positive versus negative) that occurs at zero. That might not be working correctly. Does the +/- key work correctly to enter values? If you connect the Y-axis scale to the X-axis input of the DRO, do you have the same issue with the X-axis channel?

  12. #90
    Join Date
    Oct 2017
    Country
    FINLAND
    Posts
    734
    Post Thanks / Like
    Likes (Given)
    121
    Likes (Received)
    226

    Default

    I spotted this some time ago and I believe I swapped the connections and the behaviour was identical. I will need to recheck to be certain but I believe so. I will try to enter some manual values too.

  13. Likes ballen liked this post
  14. #91
    Join Date
    Sep 2002
    Location
    People's Republic
    Posts
    4,383
    Post Thanks / Like
    Likes (Given)
    419
    Likes (Received)
    2670

    Default

    i believe the 'counting' side of the box is different from the 'display' side of the box. What I am saying is the scale reading side has no idea what is going with the display, so it has to be on the output side.

    I wonder if there is a tech or service manual rattling round the internet for these. I think there are a ton of internal settings that I hope i never need to know....but mine is a bit newer

  15. #92
    Join Date
    Oct 2017
    Country
    FINLAND
    Posts
    734
    Post Thanks / Like
    Likes (Given)
    121
    Likes (Received)
    226

    Default

    I swapped the leads and no difference. No problem entering a number and flipping it from positive to negative either.

  16. Likes ballen liked this post
  17. #93
    Join Date
    Mar 2017
    Country
    LATVIA
    Posts
    510
    Post Thanks / Like
    Likes (Given)
    131
    Likes (Received)
    256

    Default

    put a dial indicator in parallel with that axis and zero out the DRO in mid travel of that indicator (adjust the indicator to zero as well), then traverse over the zero mark 10-20 times from lets say -1,5mm to +1,5mm and then travel back to zero on the dial indicator 0 and check what the DRO reads - this way you will see if the error is just in inability of displaying a zero on the DRO or there is something wrong with the counter circuit itself

    you can also, after those 10-20 times of back and forth, go to -1,00mm and +1,00mm on the dial indicator and compare it to the DRO

    I find it difficult to imagine what could be damaged in the counter circuit for it not being able to display the just the zero, if there was some bit stuck in "1" or "0" in some IC, then there surely would be other counting errors present, that is why you should do the dial indicator test to verify that 1,00mm of actual travel verified on the DI is indeed 1,00mm on the DRO

  18. #94
    Join Date
    Sep 2011
    Country
    GERMANY
    Posts
    2,970
    Post Thanks / Like
    Likes (Given)
    1902
    Likes (Received)
    1089

    Default

    Dennis, this does sound odd, and I can not imagine what might be wrong. However I suggest that you first check the power supply lines (both pre- and post-regulator) for correct voltage and noise, and replace any electrolytic or tantalum capacitors or other components needed to get that clean. I have fixed 3 broken HH DROs. In all of them, the problem was in the power supply. These problems include burned out diodes in the full-wave rectifier, high ESR capacitors, and burned out voltage regulators. Cheers, Bruce

  19. #95
    Join Date
    May 2018
    Country
    CANADA
    State/Province
    Quebec
    Posts
    141
    Post Thanks / Like
    Likes (Given)
    110
    Likes (Received)
    51

    Default

    I'm thinking about buying a DRO Pros 3-axis magnetic scale kit.
    I have their 2-axis lathe-specific DRO on my Chinesium lathe. It was easy to install, and has been trouble-free since I installed it. I also like the simple LED display - it's large and readable and uncluttered. These magnetic scales are fairly compact at 1.1" or 2.8cm wide, plus the thickness of a protective cover to make sure they don't get banged or crushed.

    Quote Originally Posted by ballen View Post
    The X-axis. The hammer-tone steel cover is flush with the vertical table, so does not interfere. The sealing lips are on the bottom side. However I might at some point re-machine the mounting blocks to move the body of the scale about 7mm to the right, because when the table is shifted as far away from the operator as possible, the read head comes right to the end of its travel in the scale (within a fraction of a mm) which makes me nervous.
    Did you ever go to the trouble of moving the scales?

    Quote Originally Posted by ballen View Post
    The gray signal-connection box was originally located in the middle of the area where I have added the one-shot oiler, so I have moved the box a bit towards the front of the machine.

    Attachment 106830
    I'm guessing the gap between the scales and the table is to give you access to the feed trip dogs bolts? Do you ever use those?

  20. #96
    Join Date
    Sep 2011
    Country
    GERMANY
    Posts
    2,970
    Post Thanks / Like
    Likes (Given)
    1902
    Likes (Received)
    1089

    Default

    Hi Siggi,

    Sorry that I never replied, this slipped by!

    I am skeptical that these magnetic scales are accurate or will hold their accuracy. The ones that DRO Pros sells are manufactured by Electronica in India, and a good friend of mine checked their calibration over a period of several years against a reference. There were two things that were disturbing. The first is that the overall length calibration (initially) was incorrect by an amount that was ten times larger than the claimed accuracy of the scale. The second was that the length calibration drifted significantly over the first year or two after he put this into service. If you look at how these scales are constructed (a magnetised substrate in a plastic carrier, glued to a thin stainless steel strip) there are plenty of possible sources for such drift. For example residual stress release in the stainless steel strip. There are other problems with the design as well, to do with how the magnetic domains are oriented and how they will decay over time.

    Based on that, and on conversations with a friend who does DROs and metrology as his job, I have only used glass-based scales on my own machines, with a single exception (a Renishaw optical scale on a metal substrate, on the long axis of my cylindrical grinder).


    Did you ever go to the trouble of moving the scales?
    No, I came to my senses. One of my late father's favorite sayings was, "if it ain't broke, don't fix it". It ain't broke.

    I'm guessing the gap between the scales and the table is to give you access to the feed trip dogs bolts? Do you ever use those?
    Yes, I do, but it's very rare. I did use the Y feed stops a month ago, because I was doing a bunch of max-feed-speed (400mm/min) passes, in situation where I would crash the Y-ram against the vise if I went too far. So I carefully set up the stops, just as a precaution. That's typically the only kind of situation where I use the stops, but it sure is nice to have them for that peace of mind.

    Cheers,
    Bruce

  21. #97
    Join Date
    May 2018
    Country
    CANADA
    State/Province
    Quebec
    Posts
    141
    Post Thanks / Like
    Likes (Given)
    110
    Likes (Received)
    51

    Default

    Quote Originally Posted by ballen View Post
    I am skeptical that these magnetic scales are accurate or will hold their accuracy. The ones that DRO Pros sells are manufactured by Electronica in India, and a good friend of mine checked their calibration over a period of several years against a reference. There were two things that were disturbing. The first is that the overall length calibration (initially) was incorrect by an amount that was ten times larger than the claimed accuracy of the scale. The second was that the length calibration drifted significantly over the first year or two after he put this into service. If you look at how these scales are constructed (a magnetised substrate in a plastic carrier, glued to a thin stainless steel strip) there are plenty of possible sources for such drift. For example residual stress release in the stainless steel strip. There are other problems with the design as well, to do with how the magnetic domains are oriented and how they will decay over time.
    Interesting - I don't suppose you happen to remember the magnitudes?

    Even knowing this, I think I may go with those scales anyway. Between the price and the size->ease of mounting, I think it's a good compromise for my "needs", such as they are.

  22. #98
    Join Date
    Sep 2011
    Country
    GERMANY
    Posts
    2,970
    Post Thanks / Like
    Likes (Given)
    1902
    Likes (Received)
    1089

    Default

    Quote Originally Posted by sigurasg View Post
    Interesting - I don't suppose you happen to remember the magnitudes?
    See here: the "as delivered" error over 24 inches was around 0.014". Best Mill DRO for the money Most of that error could be removed via "two point calibration". But "as delivered" from the factory the scale was not accurate.

    These days, on standard glass scales, the grating pitch is 20 microns. Provided that the system is not dirty and the head is not missing counts, this means that to get 5 micron resolution requires no interpolation, because the quadrature pattern rotates through 360 degrees of phase every 20 microns. Each quadrature has two edge transitions per 20 microns. So between the two quadratures, there is one edge transition every 5 microns, so no interpolation is needed. Thus, for example, to get 1 micron resolution requires a factor of five interpolation.

    The magnetic scales typically have a 1mm or 2mm pitch. Thus to get 5 micron resolution requires a factor of 50 or 100 interpolation respectively. To get 1 micron resolution requires x250 or x500 interpolation. That has a lot of negative implications for accuracy.

    Between the price and the size->ease of mounting, I think it's a good compromise for my "needs", such as they are.
    Don't do it! When you have a DRO, you will rely on it. If it's not correct all of the time, bad things will happen. This is a situation where spending enough money to get an accurate and good-quality system is smart.

    Regarding the ease of mounting, it is straightforward to mount glass scales on Deckel FP2s, and there are tons of examples out there. Yes, it may take some hours spread over a few days, but at the end of it you will have a system that will be totally reliable all of the time, and will require zero maintenance or attention. The mill itself can be used to make whatever custom mounting hardware is needed. It's worth investing that time.
    Last edited by ballen; 05-12-2020 at 04:36 AM.

  23. Likes AlfaGTA liked this post

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
  •