[N4NV]

It seems like all I can do is wait. I am waiting for my timing pulleys and belts, and my metering valves for the lube system. The big wait is to solve my motor drive problem. I want to make sure the motors and encoders I purchased are going to work before I go through all the trouble of making mounts for them. I have spent 15 hours over the last week trying to get my Gecko 320 to run my motors, plus almost 3 hours on the phone with Mariss over the course of 4 phone calls.

At this point I have ferrites on the AC power, DC power, and +5V power for the breakout board. I have the 470uF cap on the motor input, shielded 22 gage cable running to the encoder (I tried a cat 5e cable as well). All three motors behave the same, no matter how I set the pots on the drive. They seem to be randomly hunting for position and when I command a rotation, they sound like they have rocks in the bearings. The motors have run for over 24 hours since installing new brushes, they look fully seated. I have a clean signal from the break out board, 0V to 5V. I have a perfect signal from the encoder on both channel A and B (I ran the motor from a separate power supply while monitoring the channels). The motor is grounded; there is a shielded power cable to the motor. The encoder cable is grounded on only one end, the drive end. I changed the printer cable. The printer cable, motor power cable and encoder cable are separated by at least 1' until they get to the drive. I have tried several motor tuning combinations in Mach 3. I tried feeding the drive with a function generator. The only thing I have not changed is the drive. I ordered a couple more. I will try them when they get here.

I sent the drive I have back to Gecko for evaluation. I hope it is the drive because I am at a loss as to what else could be the problem.

Vince

[N4NV]

My timing pulleys came in yesterday. The small ones have 5/16" bore and I need 1/2" for my motor shafts. I just happen to read in another thread about a way to bore the hole larger and keep it in the center of the pulley. The flanges of the pulley are pressed on and not turned with the pulley. If you try to hold the pulley by the flanges, the bore will not be true. Besides which, the flanges did not look that strong to clamp in a chuck. WhiteTiger suggested turning a piece of pipe so the inside diameter was the same as the outside of the pulley. Then turn the outside so it is true with the inside. Cut the pipe into three pieces and place one under each jaw of the chuck.

I found a brass fitting on my wife’s plumbing truck (she does high purity process plumbing and pipe fitting) and turned like the suggestion. Placing it between the jaws and the pulley kept the jaws from clamping on the flanges and held the pulley true. From here it was just a matter of boring the hole larger. Both pulleys came out great.

Vince

[RotarySMP]

Thanks for saring those photo Vince. I have to the same thing on some flanged pulleys I have.

[N4NV]

After over 20 hours of trouble shooting and changing every component in drive system, I finally got my Gecko 320 to turn my motor properly. Everything change I made helped a little. The last thing I did that fixed the problem comptetely was to add ferrite beads to the motor wires.

Vince

[N4NV]

One worked the other does not. I sent back one drive on Sunday and ordered two others. I got one of the new ones to work perfectly (by adding a ferrite to the motor leads). I then moved all the wires from the motor that was working great to the second drive in the exact same order they were on the first drive. It does not work, it sounds just like the one I sent back and is constantly hunting for position. I then moved all the wires back to the first drive, one at a time. The motor works great. Both drives have the Gain and Damping set the same. No adjustments on the second drive have any effect. I had brown hair when I started this project. Now I am mostly grey. Could I have a bad drive? I moved the wires back and forth twice with the same results.

Vince

[N4NV]

Both drives are now working! I talked with Mariss again this morning. The problem was with the Damping and Gain. I thought I had tried just about every combination possible including setting both drives the same. For whatever reason, on the second drive there is a lot less tolerance in the settings than the first drive. By making smaller and smaller adjustments, I was able to find the sweet spot. A difference in 5% of the rotation of the adjustments would be the difference between working and not working.

Now I can start on my motor mounts and my lube system. While I was working on the motor problem, I received my 8 Bijur B1114 metering valves.

Vince

[CNCfun&games]

HI Vince,
could you post the ordering info. on the Ferrite Beads?
Thanks......W.Smith

[N4NV]

HI Vince,
could you post the ordering info. on the Ferrite Beads?
Thanks......W.Smith

I did not order them. I found them at a local surplus store called Halted Specialties. They did not even know they had them. They were $0.60 each. In the picture you can see what they look like. I don't know much about them. They looked big enough to run my wire through so I gave them a try.

Vince

[N4NV]

I think I should stress how sensitive the Gecko drive is to noise and the settings of the dampening and gain. With the same motor, and cables, the two drives required different settings to work properly. BTW, I tried both the shielded 22 gage wire and Cat5e cable for the encoder and the 5e looks a little better on the scope, even though it is unshielded. Mariss said it was a toss up. With all the steps he had me go through to trouble shoot, I found that there are about 20 ways you can screw up. I only managed 3 of them and it still took 20+ hours to straighten it out.

OK, back to work. Since I received my Bijur B1114 metering valves I decided to finish the one shot way lube system. The B1114 is a size 2 and what Jason at D&L recommended. My mill came with zerk fittings (which as you can guess, someone in the past pumped it full of grease). I purchased a cheap one shot pump and an 8 port manifold from Enco. I think I mentioned that my wife was a high purity process plumber. She let me borrow her 1/4” tubing bender and gave me a bunch of scrap tubing and fittings to play with. I just had to make sure it was “clean” scrap and not contaminated with “one step” gas. I wanted to go with stainless tubing as much as possible. The plastic tubing systems just look like a weak spot in the system. I had to use ¼” reinforced plastic tubing in three places to accommodate the movement of the saddle. I “pre-filled” the longer tubes with way oil otherwise the short ones would be pumping out oil for days before the longer ones filled. As you can see, it was a really tight fit. I had to grind down a wrench to fit the nuts on the manifold. It works. I have way oil oozing out of the ways.

At work we share a building with a sheet metal shop. I keep his machines running electrically, and he lets me dumpster dive in his scrap pile and use his equipment. I made some way covers out of .080 aluminum with some felt glued to the bottom. I will also make my motor mounts out of scrap .080 aluminum to check my measurements before I start cutting the ½” aluminum plate.

Vince

[N4NV]

I re-arranged some things in my life and now I am able to get the ball screw conversion kit from Hiwin. They are on order and should get here in about 7 working days. I started cutting on the X drive end plate today and made some good progress. For the last several days I have been measuring and re-measuring which has been a pain because each time I have to take off the power drive. I drew up the end plate in AutoCAD, then plotted it out to scale. It did not look quite right so I did some more measurements. Each time I was getting slightly different results. After I thought I was good, I drilled out a scrap piece of sheet metal. This fit OK so I drilled the end plate.

In the pictures you can see my paper template, the piece of sheet metal and the final end plate. Because my mill has the factory power drive, the X lead screw ends before the end plate. The power drive has a stub shaft that engages the X lead screw. I still don’t know what I am going to do about a bearing support for this end of the new ball screw. If I use a bearing block, it will move my timing pulley out too far. I may have to bore a hole in the end plate and press in the bearing. Until I get the ball screw kit, I don’t know quite what I will have to do. For now I just drilled a 3/8” hole to check for alignment.

Vince

[CNCfun&games]

Vince,
The two Y axis plates in note 11 are leftovers, if you want one to play with let me
know - (they are .75 thk AL).......W. Smith

[N4NV]

Vince,
The two Y axis plates in note 11 are leftovers, if you want one to play with let me know - (they are .75 thk AL).......W. Smith

I don't know how to find note 11. Thank you for the offer. I already have the material for the Y axis. I would be interested in a picture that shows how you worked out the bearing support.

Vince

[CNCfun&games]

Vince,
Sorry for the confusion...guess this is note or post # 33(upper rt. hand corner in blue
area.....W. Smith

[N4NV]

OK, I thought that was it, but I got confused by the "Posts:19" number in the upper right. I have been copying your installation somewhat. I am interested in a little more detail of your Z axis, such as how you set up the bearings, what type, what holds them in place etc.

Vince

[N4NV]

I found a used heavy duty oscilloscope cart for $40 that I am going to modify for use as my computer and monitor cart next to the mill. I also mounted the sides to my X axis box, lots of drilling and taping. I used a thread forming tap for the first time. I was hoping to end up with a little stronger threads. I can’t finish it though because I don’t know how deep to make it until I get my ball screw kit.

I spent the last several evenings drawing up the Y axis mounting plate. I then drilled it out of a scrap piece of sheet metal to check the measurements. It came out pretty close, but I think I will rotate it a few degrees to give the motor a little more clearance from the knee. It turns out that with the timing belt around the dial of the Y axis, it has virtually the same diameter (within .005”) of the belt around my timing pulley. I used the center distance designer tool at the SDP-SI website:
http://www.sdp-si.com/Cd/default.htm
which worked out perfect for the final dimensions. Now I can start drilling and cutting on the ½” Al plate.

Every evening I have been running each of the servo motors from my Gecko 320 while to motor is mounted in a vise. This is to seat the new brushes I installed. The last motor should be seated tonight. It really does make a difference in the very small motions of the motor but anything above a few FPM and you can’t tell the difference.

Vince

[N4NV]

Tonight I got the Y axis motor mounting plate drilled. The plate is .500 aluminum, 6”x16”. I still have to trim it and figure out how I am going to mount sides to it to hold the cover in place.

My biggest challenge is mounting the timing pulley on the Y lead screw (ball screw when it gets here). In the picture the timing belt is on the indicator dial. It just so happens that with the belt on the indicator dial, the OD of the belt is within .005” of the OD when the belt is on the pulley. This lets me take measurements with just placing the belt as seen in the picture. If I replace the indicator dial with the pulley, the belt is at least .250” out too far from the mounting plate. The shaft on the motor is not long enough to get the small pulley in line with the larger one. If you look close in the picture you can see that the small pulley is well past the end of the shaft. I think the solution will be to machine the bearing plate down at least .250”.

Building a cover for the whole thing will also be a challenge. I could do it out of sheet metal, but I prefer to use .375 aluminum bar stock to give some extra rigidity to the motor mounting plate.

[mxtras]

Here are a few pics that may interest you - the old screw ends compared to the replacement Hiwins, the old brackets compared to the new and the intereference issue and it's resolution (a chamfer).

Scott

[mxtras]

And here's another potentially helpful pic - I did not remember that I even took this pic. The photo of the fix came out blurry - a common problem for me when I take photos too close....long story but it has to do with a fall that the camera was subjected to.

This was the alignment of the replacement dogbone when using the factory Bridgeport locating features for the Y axis screw mount. In other words - there is a fairly substantial dimensional difference between the factory bracket and it's replacement. The solution I chose (after many hours of screwing with it) was to elongate the mounting holes. This was done after verifying that the bearing bracket was in fact seated squarely and with the factory dowels in place. I chose this route as a way to establish parallelism with the ways - it was right with the acme screws so it will be correct with the ballscrews - that was the thought process and as it turned out, I think it was the right decision.

I have not posted this stuff in my thread yet. Sorry for the severe delays in updating my own thread.

Scott

[N4NV]

Great pictures, they answer a lot of questions. That dog bone is out of wack a lot more than I had pictured in my mind. I got to thinking, if your motor mounting plate is between the table end bracket and the knee and if that motor plate is not perfectly parallel, could that not affect the alignment?

I think I have figured out both the X and Y axis. On the X, I am going to turn the ball screw to I have a bearing surface under the table next to the end with the drive. This will allow me to add a bearing support block on the backside of the motor mounting plate under the table. I may loose at most .750” of table travel on that end.

For the Y I am going to machine about .250" off the back of the cross feed bearing bracket. Then I will turn down the dial holder (part #5 in the Bridgeport manual). Then I can bore the timing pulley to mount on to the dial holder. This will keep steel to steel contact on the radial bearings.

Vince

[N4NV]

Tonight I removed .250” from the cross feed bearing bracket. What I didn’t realize was how much the socket head screw holes were counter sunk. After removing the material (I set it up in a 4-jaw chuck in my lathe), I saw that there was not enough meat under the heads of the screw to securely hold the bearing bracket in place. I will have to use hex head bolts and washers instead. This will end up inside the drive box so it will not show. I just don’t know how else I could make it work. I also took .100” off the motor support plate under the bearing bracket to further move my timing pulley towards the knee.

In the picture, you can now see the motor shaft just extending past the pulley when the belt is parallel.

Vince