Universal table woes (long) - Page 3
Close
Login to Your Account
Page 3 of 9 FirstFirst 12345 ... LastLast
Results 41 to 60 of 175
  1. #41
    Join Date
    Mar 2005
    Location
    diamond ohio
    Posts
    287
    Post Thanks / Like
    Likes (Given)
    0
    Likes (Received)
    64

    Post

    nice pictures ross,
    i understand now. sean, for your type of work a three jaw chuck and a plate with a register should be good enough (.005 tir) most of my stuff needs to be more accurate. i do what ross does get the part running true to the rotary table and tram in the x,y after. another note my back plate that i use locates in the pilot in my table (gets me real close) and with a buck adjustable chuck you can really get it right on. i think bison makes an adjustable 6 inch chuck for around $250.00 less back plate. keep in mind if you use the table bore for location you are cutting down the amount of thru capacity for long shafts like ross has.
    dave

  2. #42
    Join Date
    Dec 2000
    Location
    Coos Bay, OR
    Posts
    3,853
    Post Thanks / Like
    Likes (Given)
    42
    Likes (Received)
    56

    Post

    Yeah, I don't want to give the impression that I'm just hacking away at everything. I do sometimes need accuracy. It's the times when I really don't need accuracy that I dislike going through the setup.

    The Deckel R-tables we are speaking of have a steel ring at the entrance of the thru hole of the table. This is normally protected by a plastic plug. Surely this is removed plug is to allow drills to pass but I think it might also be protecting some accuracy that ring has in relation to the rotation of the table and the ring could be used to indicated the spindle to the center *directly to the table*. I'm not 100 percent sure on that and I really should indicated it for TIR. If the ring *is* true to the center of the table, then I will seriously consider adapting a collet closer to it where the drawbar handle resides under the table.

    I think it would be super to have a collet closer where the collet is aligned to the table center and the collet face is nearly flush with the top of the table.

    My biggest reservation is "Why didn't Deckel do it?". They have a history of not missing stuff like that if it is practical.

    Sean

  3. #43
    Join Date
    Dec 2002
    Location
    Benicia California USA
    Posts
    8,025
    Post Thanks / Like
    Likes (Given)
    1793
    Likes (Received)
    2377

    Post

    Of course the ovious setup is to have the chuck with a register to the center bore on the table. Then geference the machine to home and move to the center hole and dial it up. Record the "X" and "Y" location of the bore (provided the table is flat and plump). Enter the center location in a small programn that you keep in memory using a high program number so it does not get in the way of your working programs.
    Then when you need to get to the table center, simply "Home" the machine, make your little move program active, and press cycle start....Machine goes directly to being centered over the hole....of course if the head is not in exact tram or the table is not exactlu true in both planes of tilt all this is for nothing...but it is quick!
    Cheers Ross

  4. #44
    Join Date
    Dec 2000
    Location
    Coos Bay, OR
    Posts
    3,853
    Post Thanks / Like
    Likes (Given)
    42
    Likes (Received)
    56

    Post

    Yep, that's the quick way if you're runnng a NC. My R-table is mounted on my manual FP-3.
    Can the "Activ" dro do this?

    Sean

  5. #45
    Join Date
    Dec 2002
    Location
    Benicia California USA
    Posts
    8,025
    Post Thanks / Like
    Likes (Given)
    1793
    Likes (Received)
    2377

    Post

    Sean:
    Can't speak for all the "Aktiv" mnachines, but on my FP2 fitted with the TNC 113 the short answer is yes.

    There are 4 assignable reference datums on the control. One can reference the machine while any Datum is actieve. To reference, press the "Ref" key on the control. This "Locks" the display of all 3 axis. You then move each axis to the minus end of travel. Once the slide passes the reference mark within the DRO scale the reference is then set for that axis and the display begins reading. Once all axis are referenced, move the machine to center on the table through hole and set the "X" and "Y" to zero while using the same Datum that the machine was rferenced using...
    Now when the power is off and then restored by referencing the machine while in the correct datum and then making a programed move to "X 0.0" and "Y 0.0" the machine will return to the center position. You may use the rapid feed override to speed the positioning to the set point.
    As far as i know, all the Heidenhain DRO's have this reference setting capibility and at least two selectable datums...so even the standard DRO fitted machines can hold the table center position if desired.

    Cheers Ross

  6. #46
    Join Date
    Dec 2000
    Location
    Coos Bay, OR
    Posts
    3,853
    Post Thanks / Like
    Likes (Given)
    42
    Likes (Received)
    56

    Post

    So *thaaaat's* what the "REF" button does

    I'll have to go and try that.

    Sean

  7. #47
    Join Date
    Dec 2000
    Location
    Coastal Dogpatch, SC, USA
    Posts
    51,292
    Post Thanks / Like
    Likes (Given)
    2788
    Likes (Received)
    5609

    Default

    Quote Originally Posted by AlfaGTA View Post
    Well i just could not stand it. Last night even though it was getting late i went out to the shop and pulled the table top off my FP2. This is the smaller version of table that is on my FP4NC.


    I thought we later determined your FP2 universal table was actually the same physical size as the FP4NC universal table ?

    Reason I'm re reading this thread is the universal table on this Maho I just bought is giving no readout on the screen, so now I'm worried it has encoder troubles.

  8. #48
    Join Date
    Jul 2005
    Location
    San Francisco, CA
    Posts
    807
    Post Thanks / Like
    Likes (Given)
    11
    Likes (Received)
    81

    Default

    Well what I find interesting is that in the picture from the first post the Angle encoder has 2 labels the first E7 and the second ??K3. This is the coded manufacuter year and month which would mean they are 6 years apart...Why??

  9. #49
    Join Date
    Dec 2002
    Location
    Benicia California USA
    Posts
    8,025
    Post Thanks / Like
    Likes (Given)
    1793
    Likes (Received)
    2377

    Default

    As i remember the "guts" under the table top between the FP4NC and the FP2 are not the same. The encoder on the FP2 was a bit smaller as i recall and the setup of the board and dip switches was not even the same. (don't remember the FP2 having the dip switches at all)
    Stupid of me not to have taken detail photos of the two and record all the part numbers....Did not have my camera home when i took the FP2 apart, poor excuse , but there it is.
    As to the encoder....talked with some service engineers at Heidenhain and the short version is they don't know squat about the encoder used on my FP4NC table. If it has any trouble it was off to Germany and a 3 month lead time for any service, plus $$$$$$$$$$$$$! Can't remember now but the sum of $6K sticks in my head. The HH people suggested that i contact a field calibrating service to check and adjust the reader output. They use (the HH folks) dedicated test equipment that as i was told gives only the signal data in a sort of go/no go format. No body at Heidenhain could tell me how to field adjust the encoder using a meter or a scope. And since there are two reading heads you have to get both syncronized i think within specific range or the accuracy will go to hell. (note on the Deckel unit the scale is connected to the rotary table so it is part of the table and if removed to adjust on the bench will not have any rotary bearing, so some type of fixture would be required for any work done off the table.)
    I will be interested to see how you make out on this if indeed you have encoder problems. Perhaps you will find the guy that knows how to set up one of these in the field and can give you the easy step by step.
    By the way, i took the table off since i started this posting and looked again at the encoder and reader heads. I found a piece of dirt lodged on one of the lenses of one reader head. I cleaned everything and now the table is back to running with both lamps on and it reads correctly...but i do keep an eye on it. i verify the true "0" position with the dial indicator any time i run a job using the rotary .



    Cheers Ross

  10. #50
    Join Date
    Dec 2002
    Location
    Benicia California USA
    Posts
    8,025
    Post Thanks / Like
    Likes (Given)
    1793
    Likes (Received)
    2377

    Default

    Quote Originally Posted by Hansdie View Post
    Well what I find interesting is that in the picture from the first post the Angle encoder has 2 labels the first E7 and the second ??K3. This is the coded manufacuter year and month which would mean they are 6 years apart...Why??
    Well the two parts in question are the reading head assembly and the glass scale i believe. Guess the easy answer is that one of the items must have been replaced at some time, as to not being original with the table. They are both the same series number so i would guess this does not cause any real trouble as long as the scale and reading heads were properly aligned when/if it was replaced. Table originally was purchased (by me) from FPS in Germany and was guaranteed as being rebuilt and calibrated....Were they qualified to do this work? I have no idea. Have i ever tested the rotation as to accuracy? No, only so far as it agrees with some Brown & Sharpe angle blocks i have, and as i said it returns to zero now relative to the mechanical zero set point after a proper cleaning.

    Cheers Ross

  11. #51
    Join Date
    Dec 2000
    Location
    Coastal Dogpatch, SC, USA
    Posts
    51,292
    Post Thanks / Like
    Likes (Given)
    2788
    Likes (Received)
    5609

    Default

    Quote Originally Posted by AlfaGTA View Post
    As to the encoder....talked with some service engineers at Heidenhain and the short version is they don't know squat about the encoder used on my FP4NC table. If it has any trouble it was off to Germany and a 3 month lead time for any service, plus $$$$$$$$$$$$$! Can't remember now but the sum of $6K sticks in my head.
    I had the figure of $1,800 in my head for a Maho table encoder but it turns out that was for a used one that MahoDoc had, but no longer has. A new one from DMG is $7,000 !
    I will be interested to see how you make out on this if indeed you have encoder problems.
    Well I dodged a bullet on this one.... just for the heck of it I took the cover off the junction box below the table where the encoder cable connects to the orange cable that goes to the measurment module in the CPU box and the damn thing wasn't even connected !! Encoder cable end wrapped up on a plastic bag and the other connector with a screw on plug in place. So, I wiped some oil off (hydraulic line for rotary table brake system runs thru same box..must have a tiny leak), connected it up and now it works perfectly !

    The curiosity of course is why on earth would previous owners have gone to the trouble of disconnecting the thing ?....the mysteries of previous owners tinkerings never cease to amaze...

  12. #52
    Join Date
    Dec 2000
    Location
    Coastal Dogpatch, SC, USA
    Posts
    51,292
    Post Thanks / Like
    Likes (Given)
    2788
    Likes (Received)
    5609

    Default

    Dredging up this old thread as yesterday I thought I might have encoder problems on my 2038 universal table (originally on manual 1980 vintage FP2) Previous owner cut the original 9 pin round connectors from all scales, including the rotary one so all I have is wires hanging out (which is especially bizarre with the table encoder considering it was probably never connected to anything)

    So, as there is only about 8 inches of rotary encoder (ERO 720, btw) wire still left, I figure out a way to spice an L-Com 15pin D Sub cable to those tiny wires, shrink tube over each wire, and then shrink tube over the entire splice....and it works.... at least on my Heidenhein ND780.

    So yesterday I connect it to a used Positip 880 for the first time and it works there as well...but as soon as I cut power on to the Deckel electrical box the previously black numbers for the C axis turn red and there is an error that says "Encoder X4- Defective Frequency Error. Press CE" WTF ?

    If I press CE the error goes away and it still works and seems to be just as accurate...but the numbers for X4 (C axis) stay red, which is slightly annoying. All I can figure is the Positip 880 is either more sensitive to errors than the ND780.... or, more likely, shows that sort of error if it detects an RF situation...i.e. perhaps I either need to shield tha area where I spliced the two cables or I need a better shielded cable than the L-Com standard model ? (L-Com makes a better shielded cable but it is only available with female D Sub 15 connector on one end... I went with the normal one as you could get female at each end.... that way I have essential two connectors for the price of one cable)

    Having said all that, I couldn't resist having a look at the actual encoder, so pulled the Ross trick by raising the table nearly all the way via Z axis, connecting a strap across the Y ram, unbolting the table top, and lowering the Z. Bottom line is it looks like brand new in there, including the red grease on the ways....encoder is pristine, needs no cleaning.

    An interesting aside is the ERO 720 rotary has two reader heads but only one of them is wired up...and you can tell from the pristine solder connectors it was never wired up....like a "spare" reader head.

    So, back to the DRO... I discovered you can "turn off" error warnings for any particular encoder on the Positip. If I do that, even with the machine running full blast (spindle on, power feed running) the rotary encoder numbers stay black, no error, and seems accurate. But still, I wonder.....should I try and shield the cable better, like wrapping tin foil* around where the cables are spliced....or might this indicate some deterioation in the encoder electronics ?

    ----------

    *If not tin foil (really, aluminum foil) ...what ? Does there exist "split" braded wire for this purpose ?

  13. #53
    Join Date
    Mar 2010
    Country
    UNITED STATES
    State/Province
    South Carolina
    Posts
    17
    Post Thanks / Like
    Likes (Given)
    2
    Likes (Received)
    4

    Default

    The braided wire around the cable acts as a faraday cage and gets grounded on one end to stop RF from entering the signal wires.

  14. #54
    Join Date
    Dec 2000
    Location
    Coastal Dogpatch, SC, USA
    Posts
    51,292
    Post Thanks / Like
    Likes (Given)
    2788
    Likes (Received)
    5609

    Default

    Quote Originally Posted by Kunz-mann View Post
    The braided wire around the cable acts as a faraday cage and gets grounded on one end to stop RF from entering the signal wires.
    Yes indeed...so what is the solution....find some braided wire to wrap around the splice or might aluminum foil work just as well ? (either way making sure the addition touches the braided wire of both cables)

    I guess my inclination would be to remove the current shrink tubing, extract some braided wire from an old junk cable, cut it down the middle with scissors and hope it doesn't fall apart, set it in place, wrap electrical tape around it, and see if that helps.

    If not then I probably should buy the better L-Com cable that has better shield. Kind of annoying the rotary encoder seems to be "hard wired".....This would be sooooo much easier if the damn thing just had a connector on it already to I could connect the L-Com cable directly to that connector rather than splicing wire to wire..... maybe it does, hidden behind the round ring underneath where the wire exits ??

    -------------

    As an aside, it is interesting that Heidenhain went hog wild with shielding on the LS503 linear scales.... at first there appears to be only 5 wires exiting the cable, when 9 are needed. But closer inspection reveals that 4 of them are actually 2 wires inside of yet another shielded cable ! So, shielding inside of more shielding ! Wonder why they didn't do that with the ERO 720 rotary encoder as well ?

  15. #55
    Join Date
    Oct 2012
    Country
    UNITED STATES
    State/Province
    Virginia
    Posts
    10,264
    Post Thanks / Like
    Likes (Given)
    5843
    Likes (Received)
    4628

    Default

    Quote Originally Posted by Milacron View Post
    Yes indeed...so what is the solution....find some braided wire to wrap around the splice or might aluminum foil work just as well ? (either way making sure the addition touches the braided wire of both cables)

    I guess my inclination would be to remove the current shrink tubing, extract some braided wire from an old junk cable, cut it down the middle with scissors and hope it doesn't fall apart, set it in place, wrap electrical tape around it, and see if that helps.

    If not then I probably should buy the better L-Com cable that has better shield. Kind of annoying the rotary encoder seems to be "hard wired".....this would be sooooo much easier if the damn thing just had a connector on it already.... maybe it does, hidden behind the round ring underneath where the wire exits ??
    "Tin" vs Aluminium foil - if you had a source - would be easier. Al is too challenging to make reliable contact to, its oxide being such a good insulator.

    Tinned copper braid is a PITA w/r not puncturing fingers, but more reliably terminated.

    Two handiest sources include sacrificing a foot or three of TV/Media coax, salvaging what braid yah need. RBK to doo that, but one usually dasn't have to go far to find the material.

    Same again with robbing Stainless braid off plumbing hookups if yah need larger diameters.

    Bill

  16. #56
    Join Date
    Dec 2007
    Location
    Central Ohio USA
    Posts
    3,807
    Post Thanks / Like
    Likes (Given)
    73
    Likes (Received)
    1709

    Default

    Milicron wrote: "I guess my inclination would be to remove the current shrink tubing, extract some braided wire from an old junk cable, cut it down the middle with scissors and hope it doesn't fall apart, set it in place, wrap electrical tape around it, and see if that helps"

    If you obtain some braided wire for this, better to push it together like a "Chinese finger puzzle", make it fat, then draw the other cable thru it, than to cut it and tape back together.

    Besides video cable, microphone cable will have the braided shield. Also, auto parts stores (which may be open today) will have braided ground straps, the ends of which can be cut off and then you'll have your tube of braid... at least a short one.

    Actual foil, wrapped in a layer or two around the cable and positively connected (electrically) will suffice for testing purposes. The stuff in computer cables is mylar foil, similar to kids' balloons.

    But yes, getting to the real connector and replacing the cable assembly is ultimately the proper path.

    Chip

  17. Likes Milacron liked this post
  18. #57
    Join Date
    Dec 2000
    Location
    Coastal Dogpatch, SC, USA
    Posts
    51,292
    Post Thanks / Like
    Likes (Given)
    2788
    Likes (Received)
    5609

    Default

    Since Ross's old photos have disappeared, some might find it of vague interest to see what one of the universal table encoders looks like with the table top off and the protective sheet metal cover removed. Thus the below...

    img_3257.jpg img_3258.jpg img_3259.jpg

    The 5 volt lamp is on the wired reader head, in the middle photo.... the unwired "spare" reader head at 6 o'clock.

    Note the 6 bolts one has to remove to remove the table top (besides the obvious 4 bolts one loosens to rotate the table....nuts/washers are completely removed for this) need to be pretty close in alignment to their original positions when reassembling the table top, so best to insert a couple of those bolts back in their holes (inside of T slots) when raising Z again to line them up. There is also the alignment a large O ring to consider.

  19. #58
    Join Date
    Dec 2002
    Location
    Benicia California USA
    Posts
    8,025
    Post Thanks / Like
    Likes (Given)
    1793
    Likes (Received)
    2377

    Default

    Looking at your posted photo....looks like you removed all the Allen screws from the encoder mounting ring....
    In fact there are two different screws there... There are three small Allen's that hold the cover in place i believe...Then there are an additional 4 Allen's (larger) with spring washers that set in cutouts in
    the cover...Those hold the encoder mounting ring to the table mounting flange.....(part that rotates)

    The larger larger Allen's having been removed, you now need to check the alignment of the encoder so that it runs or rotates dead true with the table rotation Even if you did not disturb the ring, likely it will shift when the screws are reinstalled and tightened...The encoder is made to have some play on the screws to allow alignment.....Best done using a microscope and reading the actual lines of the encoder.....At the very least a tenths
    indicator (or better) reading the ID of the mounting disc.....
    Note: Not necessary to remove the larger Allen bolts that fit within the cutouts of the cover shield.....No reason to remove these unless you are planning on removing the encoder, which i would not reccomend
    as getting the reading head precise alignment to the photo cells when re-installing will require some tooling an "O" scope and some electronics...(test rig)


    Additional note...The sloping notch machined into the steel mounting ring is there to allow access to the adjusting Allen screws on the lamp assembly.(still difficult)..these screws mask the light output through the condensing lens and is how the lamp is adjusted to give a good 180* phase shift (required) on the two sets of photo cells....

    Can you make that DRO display the angle in Thousands of a degree instead of minutes and seconds?

    Cheers Ros

  20. #59
    Join Date
    Dec 2010
    Location
    Sydney Au
    Posts
    429
    Post Thanks / Like
    Likes (Given)
    47
    Likes (Received)
    46

    Default

    Standard flex has 3 conductors with larger gauge than the others which makes matching a pain...remnant below.

    Couldn't find anything exact locally, but if there's a Helukabel supplier over there, that's the replacement supplier I used, for something similar. I think they're a German or Austrian company??

    010.jpg

  21. #60
    Join Date
    Dec 2000
    Location
    Coastal Dogpatch, SC, USA
    Posts
    51,292
    Post Thanks / Like
    Likes (Given)
    2788
    Likes (Received)
    5609

    Default

    Quote Originally Posted by AlfaGTA View Post
    Looking at your posted photo....looks like you removed all the Allen screws from the encoder mounting ring....
    In fact there are two different screws there... There are three small Allen's that hold the cover in place i believe...Then there are an additional 4 Allen's (larger) with spring washers that set in cutouts in
    the cover...Those hold the encoder mounting ring to the table mounting flange.....(part that rotates)

    The larger larger Allen's having been removed, you now need to check the alignment of the encoder so that it runs or rotates dead true with the table rotation Even if you did not disturb the ring, likely it will shift when the screws are reinstalled and tightened...The encoder is made to have some play on the screws to allow alignment.....Best done using a microscope and reading the actual lines of the encoder.....At the very least a tenths
    indicator (or better) reading the ID of the mounting disc.....
    Note: Not necessary to remove the larger Allen bolts that fit within the cutouts of the cover shield.....No reason to remove these unless you are planning on removing the encoder, which i would not reccomend
    as getting the reading head precise alignment to the photo cells when re-installing will require some tooling an "O" scope and some electronics...(test rig)


    Additional note...The sloping notch machined into the steel mounting ring is there to allow access to the adjusting Allen screws on the lamp assembly.(still difficult)..these screws mask the light output through the condensing lens and is how the lamp is adjusted to give a good 180* phase shift (required) on the two sets of photo cells....

    Can you make that DRO display the angle in Thousands of a degree instead of minutes and seconds?
    Yeah, I noticed after after I removed the light green sheet metal cover there was no need to remove the hex screws that sat in the notches of said cover, and then got to thinking there must be a reason for that, so promptly put them back in after I took the photo. They dropped right in what seemed to be dead center of their tapped holes so I don't think there was any movement while they were out.

    If there was some shift in the firmament, I'm not seeing it in the end results...the readings seem exactly the same as they were, comparing the analog scale on the table to the readings on the DRO....but then maybe they are off when measured with more precise instruments...will have to check that.

    Yes, the DRO will display in thousandths of a degree...I purposely set it coarser thinking that might help with the "frequency error" but it made no difference so I might as well set it back to a finer reading.

    ============

    Re the six larger hex screws that sit inside the T slots... are you saying that not only is there no need to remove those....one should not remove those due to potential alignment with encoder issues ? In other words that big plate is supposed to come off with the table top ?? If so seems like that would create more alignment issues than leaving it in place by removing the screws...not sure I follow you there. I removed them as the table top would not come off otherwise.... at least it seemed that way at the time.

    But bottom line is I can still zero the table with the dial indicator with witness mark dead center zero, set DRO at zero, rotate table 180 degrees on the DRO and the witness mark is exactly where it was before....which was not dead on 180 degrees on the analog band.

    Did you see my previous post about that last week ? I just assumed the encoder was more accurate than the analog scale.....but maybe not...maybe my encoder was already out of adjustment ?? I have yet to check what matters more, which would be to make sure it trams exactly the same indicator reading along the full X travel at 0 and 180.

    ==============================

    (on edit)

    Here's what I posted last week, that no one responded about..

    ...changed it to 9,000 and now it's right. At least I hope so...DRO reading is 0.176 degrees off from perceived dead center of 180 degree witness mark.....(about .010" difference distance from 180 metal band mark to 180.0000 degrees on DRO) Presume it is difficult to get the metal degree measurement band that surrounds the table as accurate as the encoder but you tell me.

    So....how is this on your universal tables ? If you have the analog band scale dead on zero and rotate it to 180 via digital means, is the witness mark on the band scale dead on 180 or off just a little, like mine ?


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
  •