Bridgeport Boss5 stepper to servo

Duc wrote: 1. Does the yaskawa need to be single mode encoder setup on the mesa card? Appears to be differential to me.

Yes, that is showing a differential receiver.

2. Where does the 7i77 get the 10v power for the analog drives?

The TB1 description on manual page 6 seems to suggest that it comes from the cable 5V or the TB1 5V connector depending on a jumper. (if it does, then presumably it is via a DC to DC converter)

andypugh wrote:

Duc wrote: 1. Does the yaskawa need to be single mode encoder setup on the mesa card? Appears to be differential to me.

Yes, that is showing a differential receiver.

2. Where does the 7i77 get the 10v power for the analog drives?

The TB1 description on manual page 6 seems to suggest that it comes from the cable 5V or the TB1 5V connector depending on a jumper. (if it does, then presumably it is via a DC to DC converter)

PCW confirmed that it comes from the 5V power



Cables
In the end, the china seller would not fix the encoder cables that were missing pin outs so I had to buy some yaskawa encoder cable extenders then swap the china end piece to the extender cable. Talk about tiny contacts.



Before tuning the drive make sure to have the limits switches wired in. This guy learned that lesson. :homer: Managed to slam the table against the Y-axis hard limit due to this. The table was jammed so hard that I need to hook a chain wrench to the unused part of the ball screw to force it off the hard stop. Area to left is where the chain wrench was used
image missing http://i1221.photobucket.com/albums/dd465/shefron/Machinery/Boss5%20servo/20151226_121216_zps4e8ucf4b.jpg


Picture of the hard stop that I hit. Its now off of it.
image missing http://i1221.photobucket.com/albums/dd465/shefron/Machinery/Boss5%20servo/5cce5748-36ca-43d2-b460-45b47179e2c0_zpsfisrpnbn.jpg

Velocity drives do not like a FF1 of 2. Caused the drive to move instantly once the power is turned on. Live and learn.


I was trying to fix this issue.
image missing http://i1221.photobucket.com/albums/dd465/shefron/Machinery/Boss5%20servo/Y-axis%20tuning_zpsdxwwemwg.png

Multiple things went into fixing the follow error shown in the picture above. Cant really pin it down to one thing since I was trying a lot of things at once. Here are a few things I did to fix the error.

Limit RPM of drive.
Max Velocity = RPM * Gear Reduction * Ballscrew
200 IPM = RPM * 1* (.2"/REV)
Max RPM I wanted the Y drive to go is 1,000 RPM. Servo system love speed but I will be researching Servo gearboxes later on since there isn't room for new pulleys.

Tuning setting in Linuxcnc
Located under machine --> Calibration - Can be adjusted on the fly and checked with the Halscope
10V = FF1 * Machine velocity (IPS)
IPS = IPM/ 60 -> converting Inches per minute to inches per second

FF1 should be around 3.03

P - Stiffness - I had to play with this value till the drives did not oscillate in place. Before I hooked the drives to the table the values were around 14,000 but in system were around 1,000.

Yaskawa drive adjustment.
Had to adjust the gain on the drive to limit the speed to 1,000 RPM when it saw 10V.

New Chart after some fiddling. A lot of the advice for fiddling with the Tuning came from PCW on Linuxcnc. Awesome support from the owner of Mesa.
image missing http://i1221.photobucket.com/albums/dd465/shefron/Machinery/Boss5%20servo/Y-axis%20159%20IPM_zpsnm1es1je.png

Had to finish up the pulleys for the Y-axis and Z-axis but really needed a break from tuning.

X-axis
Reason why I cant change the pulley ratio on the X-axis. No dam room but I can buy a servo gearbox later on.
image missing http://i1221.photobucket.com/albums/dd465/shefron/Machinery/Boss5%20servo/20151227_182836_zpso9y02xvf.jpg

Finished very similar to the Y-axis for the tig welding on the pulley adapter.

image missing http://i1221.photobucket.com/albums/dd465/shefron/Machinery/Boss5%20servo/20151227_213407_zpsrlldx6ch.jpg


Z-axis.
image missing http://i1221.photobucket.com/albums/dd465/shefron/Machinery/Boss5%20servo/20151228_130927_zpso3qp08js.jpg

Turned the boss off the Z-axis pulley.
image missing http://i1221.photobucket.com/albums/dd465/shefron/Machinery/Boss5%20servo/20151228_132925_zpszgtwt2ut.jpg

Fit like a champ
image missing http://i1221.photobucket.com/albums/dd465/shefron/Machinery/Boss5%20servo/20151228_132931_zps28dqmtgm.jpg

Forgot to finish the thread but the Yaskawa AC servos are working out excellent on the mill.


Got a urge to start setting up a 4th axis before I have a project need. No desire for a cheap unit that would only frustrate me or crap out later on. All Chinese and converted super spacers were off the table. I wanted at minimum a HAAS rotary but really wanted to justify a rotary in the Nikken, Tsudakoma, etc class of tables. We have a ton of Tsudakoma tables at work so I could use that as a way into tech support if needed. Ebay has alot of used rotary tables but difficult to weed thru the units unless you have a idea of what you want. Next was selecting a size that would fit on my bridgeport mill and not take up all the knee travel and weight limit. I was looking for a 160mm table but in the end I settled with a 200mm table.

Turns out a lot of people will take about 1/5 ebay price when you actually offer them money. Got this table for $1200 delivered to work with the tail support and hydraulic booster for the unit.
image missing http://i1221.photobucket.com/albums/dd465/shefron/Machinery/Boss5%20servo/ebay%20ad%201_zpska5etm6a.jpg

image missing http://i1221.photobucket.com/albums/dd465/shefron/Machinery/Boss5%20servo/s-l1603_zpsl1uh5adl.jpg


Once I got the unit home I was like a child at Christmas, Must open unit to see what I bought!!!!
Surprise surprise, the rotary had a Yaskawa SGMG-09A2ADS AC servo motor and not a DC servo motor like I was expecting.
image missing http://i1221.photobucket.com/albums/dd465/shefron/Machinery/Boss5%20servo/20160222_174220_zpso0md7p0m.jpg

Switches for the hydraulic brake indicator. Dual switch setup.
image missing http://i1221.photobucket.com/albums/dd465/shefron/Machinery/Boss5%20servo/20160225_164733_zpsliupdek4.jpg

Also opened up the covers to inspect the worm gear a bit and the motor gears. Dam near looked new. Beginning to think this had low hours and the job went under or the business did.
image missing http://i1221.photobucket.com/albums/dd465/shefron/Machinery/Boss5%20servo/20160306_175526_zpsfndgodj5.jpg

image missing http://i1221.photobucket.com/albums/dd465/shefron/Machinery/Boss5%20servo/20160222_175731_zpsbfgrju87.jpg

Now that I had a idea of what was in the rotary table, I could start researching interfacing the motor to my system.

Motor spec
Yaskawa SGMG-09A2ADS
850 Watt AC servo motor
8192 P/R incremental encoder
Sigma class of motors

At this point I would have loved to use a Sigma5 drive but all data points to that being a losing battle. According to Yaskawa website I need a SGDB-10ADG servo drive for this motor. I found a few on ebay but nobody was taking the low ball offers but manage to find a bid auction for 3 of them. I won one of them at $216 but others went for less than $200. One company was bidding against me for the units. Still cheap compared to normal prices.

Cables
I couldn't find shit for Tsudakoma cables on ebay and I didn't dare call the company for some. End connectors alone retail for $200 each. I did find the end connectors for $20 each on ebay and a 4 pack of 12ft supertrex cables (10 wire -16Ga) for $35.
MS3106F20-29S (encoder plug)
and
MS3102R28-11S, MR06-28-11P-A66-A106 (Motor power and I/O)

image missing http://i1221.photobucket.com/albums/dd465/shefron/Machinery/Boss5%20servo/supertrex_zpsczfzkyq2.jpg

Had to switch the rotary power plug to a female to match what I could find on ebay.
image missing http://i1221.photobucket.com/albums/dd465/shefron/Machinery/Boss5%20servo/connector%201_zpszpt0soac.jpg

Original plug
image missing http://i1221.photobucket.com/albums/dd465/shefron/Machinery/Boss5%20servo/s-l1602_zps8asgddhk.jpg

OK I will have to get back to the rotary updates later but here is the INI and Hal files that are giving me trouble for adding a linear encoder to the knee.

encoder 00 is for knee
Knee is W axis