Printed Circuit Board Assembly & PCB Design Forum

SMT electronics assembly manufacturing forum.

SMTnet
»
Electronics Forum
»
Help CSM84 mechanical alignment

Help CSM84 mechanical alignment

Subscribe to Thread

I am still learning the CSM84, but have gotten one board pro... - May 03, 2007 by jmelson

Your Head 1&2 have Chucks, is Head 3 a Chuckless Head? The ... - May 04, 2007 by DSD

Thanks! Well, the procedure in the book says to first c... - May 04, 2007 by jmelson

Try this. Remove the chucking assembly from Head 3. Insert ... - May 04, 2007 by Darby

Yes, I took the chuck off head 3, and made my own nozzle, wh... - May 07, 2007 by jmelson

jmelson

#50097

Help CSM84 mechanical alignment | 3 May, 2007

I am still learning the CSM84, but have gotten one board programmed and working. Now, I'm doing the next board, and need to use the mechanical alignment unit for a 22 x 22 mm TQFP. This chip is too large to clear the jaws on any head. Can the CSM hold the jaws open while it picks up with the nozzle, or do I have to take the jaws off one nozzle for this?

But, the real problem is when I program that component for mech. alignment (1 or 2, makes no diff) I get a "PCB DATA ERROR" when I try to run. The program stops initializing with the part number of the part that calls for the mech. alignment unit. If I change that component to not call for the MA then it loads up and runs to that part, which it is unable to pick up from the feeder due to the jaws. So, I have something wrong in the setup, but I can't figure out what is wrong.

Another quirk is that the machine setup data shows the mech alignment unit at -22.22 47.56, while the unit is at the right rear of the machine, around 571.0 562.0 in machine coordinates. Unless these coordinates are referenced off some other position, they don't make sense. The example screen in the manual also shows coordinates similar to my 571,562 numbers. But, changing these numbers doesn't seem to make any difference in the initializing problem above. Also, the beam sensor cannot be brought to the center of the mech alignment unit, it runs into the travel limits!

One other problem is that head 3 does not dip down to pick up parts under program control. I can manually command it and it works fine. Is there some setting that has told it to not lower head 3?

This is the most basic CSM, no V, no Z, and no III, Type PA130620.

Thanks,

Jon

reply »

DSD

#50098

Help CSM84 mechanical alignment | 4 May, 2007

Your Head 1&2 have Chucks, is Head 3 a Chuckless Head? The Chuckless Head or Precision Head is the intended Head for use with the Centering Station only. Using Head 1 & 2 is not good as there is too much nozzle play. In your Machine Configuration is HEAD3 set to AANC? In your Machine Configuration is M.CENTERING 1 set to YES In your Machine Configuration is M.CENTERING 2 set to NO. Does a reasonable XY position exist for Head 3 in the HEAD-OFFSET menu? Does a reasonable XY position exist for the Centering Station in the POSITION menu? It sounds like your coordinate for the Centering unit is incorrect!!! Drive HEAD 3 accross to point the nozzle in the center of the Centering Unit while in the POSITION menu. Teach with Head 3 Nozzle over the center of the Centering Unit. This is rough only. Once you have the machine working, you should Calibrate the machine to CAD standard using the hidden "CALIB" utility.

The error PCB DATA error can mean many things, but generally it is telling you that the machine physically cannot go to the position requested for several possible reasons or utilisation of an option is inhibiting it.

reply »

jmelson

#50100

Help CSM84 mechanical alignment | 4 May, 2007

Thanks!

Well, the procedure in the book says to first calibrate the mech aligner with the beam sensor, but that exceeds the X travel. I guess you can just eyeball it off the nozzles and then use trial and error. While poking around on the machine, I found out the whole rear feeder plate was loose! I'm guessing they had this off to swap something before selling me the machine. But, there is no offset for the rear feeders, the mech aligner offset was out of range, too.

The first reason I couldn't initialize the program was that you need to set the head for mechanical aligning to IC, not standard. (Maybe others work, too.) Isn't AANC an auto nozzle changer? I don't have that, or vision, or Z. I have jaws on all 3 heads. The head offsets are all consistent to some reference, but the beam sensor probably was moved at some time, so they no longer align to that. I'm fixing that offset in small steps.

I don't think you can teach it with the nozzle directly, it wants the beam sensor to be used for all coordinate entry except feeder teaching, which works off the camera.

Anyway, I set head 2 for IC, and removed the jaw chuck. (Right now, head 3 has a tiny nozzle in it and won't pick up anything bigger than a flea. Since I won't be placing fleas, I'm going to make a better nozzle for big chips, and remove the jaws but keep the chuck body on there, I'm sure it helps keep the nozzle holder centered.) I moved the aligning station to a place where the beam sensor could get to the center, and the whole thing worked in general. Not perfect alignment, but it made it through all the steps without croaking. I think I'll put the aligner station back and figure the offsets to it from nozzle 3.

I'll have to learn how to use the CALIB utility.

There's so much the manuals don't cover!

Thanks,

Jon

reply »

Darby

#50111

Help CSM84 mechanical alignment | 4 May, 2007

Try this. Remove the chucking assembly from Head 3. Insert a D nozzle. this is the largest nozzle that a Standard R-DC Synchro head will take.

Data In > System > Machine Config > Mechanical OP > Head 3 Spec = Vision.

Data In > System > Machine Config > Mechanical OP > M Centering 1 = Yes

If your Centering unit is a two stage unit then also set M Centering 2 = Yes You should then be able to roughly teach the MCU position using Head 3. The MCU considers itself to be at NEGATIVE 90 degrees as the long jaws are at 90 deg to the rails. Rough XY will be around 508 and 634.(Mine are anyway) Using CAD data place the component on double sided tape at 0, 180,90,-90. Note the offsets and fiddle with the co-ordinates of the MCU until you get it right. This is great fun. Remember it thinks it is at -90. Send me your machine, component and board files and I'll take a peek.

reply »

jmelson

#50136

Help CSM84 mechanical alignment | 7 May, 2007

Yes, I took the chuck off head 3, and made my own nozzle, which is working quite well. I can't set head 3 to vision, as this machine doesn't have the vision system. I set it to "IC" and that seems to work fine. I set mech alignment 1 and 2 to yes, as I need the high position for TQFPs. I eyeballed it with the nozzle and corrected for the offset between nozzle and beam sensor, and it worked well enough to get a part onto the board. Once I figured out that the 90 degree position of the alignment station meant you corrected the X offset to move the chip in the Y direction, I was able to dial in the offset until the chip placement was dead on. Thanks, everyone, for the help. I still need to make a new nozzle for head two, as that was made by the previous owner of the machine, and may be worn out. I get poor vacuum and discarded chips. But, this nozzle I just made works so well, I'm sure I can make a small chip version, too.

Thanks again,

Jon

reply »