Machsupport Forum
General CNC Chat => Show"N"Tell ( Your Machines) => Topic started by: Dan13 on March 07, 2012, 09:23:57 AM
-
Hi,
I have this little lathe which hasn't really been used in the last couple of years, so I decided I'd better convert it to CNC which I could really use.
(http://i887.photobucket.com/albums/ac80/yosefi83/9x20%20Conversion%20Photos/CJ0623.jpg)
It is a Chinese conventional 9x20 lathe. They make them on several factories in China with different quality. This particular one was quite poorly made. The good thing in these lathes is usually their heavy cast bed and headstock and that's why it qualified as a good candidate for a conversion :)
Going to strip off everything and fit ball screws to both axes, with stepper motors to drive them and make up a new tray with appropriate guards. The plan is also to replace the AC induction motor driving the spindle with an AC servo motor. Quite some rework is needed to make all I want.
First, stripped off everything:
(http://i887.photobucket.com/albums/ac80/yosefi83/9x20%20Conversion%20Photos/BareBed.jpg)
Headstock got new bearings installed and after mounting it back I checked for alignment to the Z axis:
(http://i887.photobucket.com/albums/ac80/yosefi83/9x20%20Conversion%20Photos/HeadstockAlignmentTest-1.jpg)
It was found to be around 0.1mm out over 150mm. Simply filing the V-groove on the headstock in the right spots brought it to within 0.02mm over 150mm, which is a satisfactory result. It was a long process: filing a bit, mounting it back and testing and then back dismantling and filing more based on the test result. Took quite a few iterations, but it paid off :)
(http://i887.photobucket.com/albums/ac80/yosefi83/9x20%20Conversion%20Photos/AlignHeadstock.jpg)
-
There was no enough room to put a substantial ball screw in place of the poor leadscrew, so I decided I to put the cross slide on linear guides to have a bit more clearance for the ball nut. On the other hand though, I didn't want to significantly reduce the clearance over the cross slide. I had very low profile Bosch Rexroth preloaded linear guides (10mm high) and I decided to you them, but also had to mill a bit into the cross slide as well as the carriage to achieve the required clearance.
Here is the carriage after milling the recess and drilling the holes for the linear blocks:
(http://i887.photobucket.com/albums/ac80/yosefi83/9x20%20Conversion%20Photos/CarriageTop.jpg)
Going to mount the linear slides upside down as I figured it would be more convenient, so the carriage will have the linear blocks and guides will be mounted on the cross slide. The mounting holes for the linear blocks got counter-bored from the back:
(http://i887.photobucket.com/albums/ac80/yosefi83/9x20%20Conversion%20Photos/CarriageBottom.jpg)
Also, in the above photo, a lub channel has been milled into the flat sliding surface. The V-groove won't need that as it has a slightly smaller angle than 90° and there is always some clearance between the top of it and the V-guide.
On the left side, milled flat the surface for the Z ball nut assembly to be mounted.
This is the cross slide after milling the recess and drilling and tapping M3 holes for the guides:
(http://i887.photobucket.com/albums/ac80/yosefi83/9x20%20Conversion%20Photos/CrossSlideBottom.jpg)
Here's a top view, mounting holes for the X ball nut assembly have been drilled and counter-bored:
(http://i887.photobucket.com/albums/ac80/yosefi83/9x20%20Conversion%20Photos/CrossSlideTop.jpg)
Will fill the unused holes with epoxy later.
Dan
-
The original carriage gibs weren't satisfactory for me. I wanted something more substantial and something that can be easily adjusted and locked. The original ones had the gibs mounted on screws which could be tightened for to adjust, but nothing to lock those screws themselves from unlocking.
I designed new gibs to replace the original ones. Here is my rear gib:
(http://i887.photobucket.com/albums/ac80/yosefi83/9x20%20Conversion%20Photos/RearGibAssy.jpg)
Couldn't find cast iron stock large enough to make it so made it two parts. The base is mild steel and the slide is cast iron. The slide bar sits on a step and is bolted to the base. It has the lub channel milled.
The 3 holes on the base match with the original bolt locations and will be used to tighten it against the slide for adjustment. The slots on the back flange will have bolts to secure it in place once adjustment has been completed.
Dan
-
Excellent write up. I'm looking forward to watching your progress. I have a 11x24 Rockwell that I want to convert. You're making me really want to start now.
Mike
-
The front gib was more tricky in terms of mounting and size to prevent interference with the Z ball nut underneath. This is what I ended up with:
(http://i887.photobucket.com/albums/ac80/yosefi83/9x20%20Conversion%20Photos/FrontGibAssy.jpg)
Again it's two pieces for the same reason, base being steel and the sliding bar cast iron. The two M6 holes at either end match up with the original holes on the carriage two bolts there will be used for adjustment. The 3 larger holes match with the 3 holes I bored in the carriage (bottom view photo in the previous post). These will have dowel pins.
It's hard to see on the picture, but there are M4 holes on the sides perpendicular to the M6 ones. They are for M4 locking set screws.
Dan
-
Looking real good so far Dan, nice work !
Have you decided on which ballscrews yet ? Buying new made to order ...eeeBay ? Chineee ?
Just curious as I am planning to do similar, again. Good screws don't come cheap, usually.
Thanks for the post,
Russ
-
Thanks Russ.
For the Z axis I have a 14mm diameter, ground NSK ball screw I got from eBay. It's used but it looks in a very good shape. For the X, I got a Hiwin 12mm diameter C5 rolled ball screw from 1aballscrew.co.uk (http://1aballscrew.co.uk) who tend to offer good pricing and also do the end machining.
Dan
-
Are you planning on keeping it flat or using it as a slant-bed?
I turned mine into a slant and it helps keep the chips out of the z- screw. Mine is far from done but it works and has made tons of parts.
-
Hi,
I am going to keep it flat. I have sheet metal guards designed to cover the screws so should be no problem there.
Thanks for the picture. I like your collet addition. Might consider it on my Emco C5 as I usually use bar stock there. Where did you get yours?
Dan
-
The collet closer is a Royal that was originally on a southbend 9" It uses 3c collets. I bought it on e-bay.
I added turcite to the bottom of my z axis carriage and had to rework my gibs as well. I have been thinking of adding linear rails as you are doing to yours.
I like yours so far, keep up the good work!
-
Here is the bearing housing for the X axis:
(http://i887.photobucket.com/albums/ac80/yosefi83/9x20%20Conversion%20Photos/X_BearingHousing.jpg)
I went for a needle thrust bearings arrangement here. It is two needle thrust bearings closed from either side of a step inside the bore. They are combined with a deep groove ball bearing at the end to provide axial support for the belt tension. Here is the section of the arrangement:
(http://i887.photobucket.com/albums/ac80/yosefi83/9x20%20Conversion%20Photos/X_BearingSupport.jpg)
The reason for this over angular contact ball bearings is that angular contacts at that size are pretty limited in the axial loads they can take. The needle thrust bearings on the other hand can handle axial loads several orders of magnitude higher than the angular contacts. Preloading is fairly easy too. And as a bonus they are only a fraction of the price of angular contacts :)
The bearing housing will be mounted on the back of the carriage (as the motor is going to sit on the back). There are M6 mounting holes on the back of the carriage seen on the carriage picture.
The two bosses on the bearing housing are for mounting the motor mounting plate.
By the way, it's not shiny in the picture as it got treated with a light solution of phosphatic acid to prevent corrosion. Works well.
Dan
-
Looking good Dan :-)
Hood
-
Thanks Hood!
Sorry for the slow update. Have been too busy lately.
This is the ball screw I am going to use for the Z axis:
(http://i887.photobucket.com/albums/ac80/yosefi83/9x20%20Conversion%20Photos/Z_BallScrew.jpg)
It will be mounted on the front, where the original lead screw used to be. I made a mounting block for the ball nut out of aluminium:
(http://i887.photobucket.com/albums/ac80/yosefi83/9x20%20Conversion%20Photos/Z_NutMount.jpg)
The mounting block will be bolted to the bottom of the saddle. And here is the mounting block I made for the bearings housing:
(http://i887.photobucket.com/albums/ac80/yosefi83/9x20%20Conversion%20Photos/Z_DriveMount.jpg)
The bearings housing will go into the round bore and the stepper motor will be mounted on the opposite side. I will be using timing belt drive between them. The elongated bore for the motor will allow for belt tensionning.
Dan
-
Made a ramp for triggering Z home switch:
(http://i887.photobucket.com/albums/ac80/yosefi83/9x20%20Conversion%20Photos/Z_HomeTrigger.jpg)
It's made from steel and mounted on the bottom of the ball screw mounting block:
(http://i887.photobucket.com/albums/ac80/yosefi83/9x20%20Conversion%20Photos/Z_HomeTrigger2.jpg)
Also made a bearing support for the Z ball screw floating end:
(http://i887.photobucket.com/albums/ac80/yosefi83/9x20%20Conversion%20Photos/Z_BearingSupport.jpg)
Dan
-
Here is the X ball screw with a locking nut I made:
(http://i887.photobucket.com/albums/ac80/yosefi83/9x20%20Conversion%20Photos/X_BallScrew.jpg)
It's 12mm diameter 5mm pitch Hiwin ball screw, preloaded to zero backlash. And forgot to mention the Z ball ball screw specs in the previous post: it is 14mm diameter, 5mm pitch.
The ball nut will be mounted to the cross slide with this mounting block:
(http://i887.photobucket.com/albums/ac80/yosefi83/9x20%20Conversion%20Photos/X_NutMount.jpg)
Dan
-
Bored all the timing pulleys. These is the X axis set:
(http://i887.photobucket.com/albums/ac80/yosefi83/9x20%20Conversion%20Photos/X_Pulleys.jpg)
It's a 2:1 reduction as I wanted to increase resolution on the X. The Z is 1:1 since I don't need that high of a resolution but better to have the speed there. This is the Z set with a key slot cut in the ball screw pulley (didn't have enough thickness after boring the 10mm diameter hole to allow for set screws as in the others):
(http://i887.photobucket.com/albums/ac80/yosefi83/9x20%20Conversion%20Photos/Z_Pulleys.jpg)
And the spindle set:
(http://i887.photobucket.com/albums/ac80/yosefi83/9x20%20Conversion%20Photos/SpindlePulleys.jpg)
Dan
-
Coming along nicely, should be up and running by Monday :)
Hood
-
YES , monday, we will have a video of the lathe running and turning parts, can't wait to see ;D ;D
Jeff
-
LOL ;D
Would love to see it running by Monday myself, but there's a lot of work to be done yet. Next I am going to paint it. Then I am expecting the sheet metal parts to arrive next week and I will have to do some welding and then paint the enclosure and cabinet. Will then assemble it all and start working on the electronics. If all goes well and I don't get distracted by work I expect it running in about 3 weeks.
Dan
-
Got it painted:
(http://i887.photobucket.com/albums/ac80/yosefi83/9x20%20Conversion%20Photos/Painted.jpg)
I figured I wouldn't paint the aluminium parts. Wasn't sure how this paint would hold up on aluminium so figured I would rather have them anodized. So another little delay :(
Dan
-
If you use etch primer on Alu the paint lasts well. The other thing we do on alu for boats is let it weather for a few months to get the oxide coating then paint and it lasts well. The latter method however would not be suitable for you unless you took the bits to sea for a while and let the salt air do its work ;D
Anodising will work well though and is probably the best for something like a lathe.
Hood
-
Thanks for sharing the methods, Hood. While I could think there would be one like the first, the second one is interesting. Not as practical being a few months, but interesting nevertheless. Anyway, parts are sent out and it should take a few days.
Dan
-
Assembled the cross slide and linear guides and the X axis ball screw. Here are a couple of pictures:
(http://i887.photobucket.com/albums/ac80/yosefi83/9x20%20Conversion%20Photos/LinearSlides.jpg)
(http://i887.photobucket.com/albums/ac80/yosefi83/9x20%20Conversion%20Photos/X_Drive.jpg)
It's just a preliminary assembly test so not all screws are there and it's not tightened yet. All fits nicely and look like it's going to work :)
Also made a spacer ring to mount the pulley on the spindle. It was needed because the original pulley was wider being 3 stage and plus there was a gear on the spindle. Just turned down the the original pulley and cut off the required length. Here is the spindle pulley mounted:
(http://i887.photobucket.com/albums/ac80/yosefi83/9x20%20Conversion%20Photos/SpindlePulley.jpg)
Dan
-
After the headstock had been aligned parallel to the ways by filing its V-guide, it became a bit lower and the tailstock needed to be aligned accordingly. I took apart the top part of the tailstock and milled 0.3mm off from the tailstock base:
(http://i887.photobucket.com/albums/ac80/yosefi83/9x20%20Conversion%20Photos/TailstockBaseMilled.jpg)
Got it aligned to within 0.03mm vertically. It's quite a satisfactory for me.
(http://i887.photobucket.com/albums/ac80/yosefi83/9x20%20Conversion%20Photos/TailstockAlignment.jpg)
Dan
-
Looking good Dan, curious though why you have the trucks on the cross slide?
Hood
-
Thanks Hood. The linear blocks are attached to the carriage for a few reasons. First flipping them around would have meant I had to drill 24 through holes in the cross slide right where the T-slots are, which preferred not to. Second, I did the design before I thought of putting a sheet metal guard on top to keep chips away and when the rails are on the top part they will always stay clean. With a guard it won't mean much, but it was a consideration at the beginning. And lastly I wanted to gain more X travel. The carriage is shorter then the cross slide and I couldn't mount the same length rails on the carriage.
I know that it may be problematic and I also had Graham tell me that I may have overhang problems this way. When the blocks are on the moving part the load on them is always the same regardless of their position. In my configuration the blocks are stationary, meaning the load on them will vary with the cross slide position. It's same as on a mill, where gaining X travel is desired and then you have a long table riding on a short bed. But the little difference is that on a mill the load is always at the bed centre ;).
Anyway, I can always limit the stroke if I see chatter problems at the extents. And if it's really bad I will consider flipping them over.
Dan
-
To be honest Dan it will likely be fine seeing as its a relatively light duty machine and you wont be taking massive cuts, I was just curious as I think I would have chopped the cross slide down by half and had all the trucks on it, in fact I probably would have just made up a sub plate as a cross slide and that would have given me the option to mount different configs of top slide on such as gang tool setups.
I have been toying with the idea of making a wee lathe style machine as I have some beautiful heavy duty roller slides that came from a scrapped Bullard lathe and I need to use them for something ;D
Hood
-
To be honest Dan it will likely be fine seeing as its a relatively light duty machine and you wont be taking massive cuts, I was just curious as I think I would have chopped the cross slide down by half and had all the trucks on it, in fact I probably would have just made up a sub plate as a cross slide and that would have given me the option to mount different configs of top slide on such as gang tool setups.
I see what you mean. Gang tooling was also in mind, so thought I could utilise the T-slots.
I have been toying with the idea of making a wee lathe style machine as I have some beautiful heavy duty roller slides that came from a scrapped Bullard lathe and I need to use them for something ;D
Hood
;D Having them lying around is good enough a reason.
Dan
-
Been quite a while since the last update. Was a huge delay with the sheet metal parts. Eventually I got them and made some progress though :)
This is the cabinet assembled after welding and sanding:
(http://i887.photobucket.com/albums/ac80/yosefi83/9x20%20Conversion%20Photos/Cabinet1.jpg)
And showing how the door opens:
(http://i887.photobucket.com/albums/ac80/yosefi83/9x20%20Conversion%20Photos/Cabinet2.jpg)
I made the door open this way for troubleshooting convenience, so I can see the monitor at all times.
Working on the wiring now. And added some of the sheet metal parts to the lathe also. Will take some photos later.
Dan
-
Like the idea you had for the door, space saver as well I would imagine :)
Hood
-
Thanks :)
Tried mounting the monitor and found that the door carrying arms are a bit weak. Their section is 12x3mm and there are four of them. I think I will replace them. Will get them cut from a 5mm instead and may also increase the width to around 15mm.
By the way, the picture above doesn't show the door fully opened and makes it look like part of the cabinet is covered. When fully opened it is almost in the same line with the cabinet edge and you get full access to the cabinet interior.
Dan
-
Maybe just geting the edge bent to a right angle would allow you to use the thinner material.
Hood
-
The shape is too odd to try and bend the edge. And even if I wanted to I had to order new ones cut anyway (significantly wider to allow for the bend) so getting them cut from a thicker material in the first place is cheaper, faster and easier ;)
Dan
-
Hood,
It was in your wee lathe topic, I think, that I asked you about how you implement Estop in your control. You said that Estop was cutting power to your drives. This is how I want to do this on this one, but I am a bit confused. If the servo drive outputs a fault signal I want it to trigger the Estop sequence which in turn will shut the drive off and depending on the wiring (fault line active low or high) this will either mean that I won't be able to get out of this state or it will get powered up right after it powers off. Question is how do you do this?
Dan
-
Have a manual override button on your panel that bypasses the fault signals.
Hood
-
So after hitting the Estop button you hold that override button while releasing the Estop?
Dan
-
The way I do it is my E-stop cuts power to the contactors and also sends an E-Stop signal to Mach. If I bring it back on it powers the contactors coils again and I can reset in Mach.
My fault signals from the drives, limits and E-Stop are also used to take the enables away from the drives and it is the limits I have the override button for. I do all this in a PLC as its the easiest way but could be done the same way with relays I suppose.
So if a drive faults I can reset the drive then reset Mach but if a limit is hit I need to press that button until I reset Mach then jog off, I can then release it.
If you were wanting to have the fault signals in the actual E-Stop string then you would have to press the button until the drives enabled and sent the no-fault signal.
One thing I was going to incorporate into my limit/fault/E-Stop was the overtravel inputs to my drives as that should stop them immediately but as I found they do stop instantly then I never needed to. Possibly on a lathe like yours with minimal friction it may be something to look at.
Hood
-
Ah... so a fault signal from a drive won't cut off power to the drives on your machines. That explains it. Hmm... The little lathe only has a servo on the spindle. The axes are steppers. So I think I will do the same then. A fault signal from the servo will cut power to the steppers and send a signal to Mach3. Somehow I overlooked this and wanted the servo cut power to itself on a fault ;D
Dan
-
That should work fine especially if you can also have the fault signal throwing an E-Stop to Mach.
Hood
-
A small update. Ball screw guards mounted:
(http://i887.photobucket.com/albums/ac80/yosefi83/9x20%20Conversion%20Photos/BallScrewGuards.jpg)
Dan
-
Looks great ! Cant wait to see it moving.
JH
-
Thanks JH. I am getting close.
Dan
-
Made a PCB last week for all the buttons on the front panel:
(http://i887.photobucket.com/albums/ac80/yosefi83/9x20%20Conversion%20Photos/ButtonBoard.jpg)
Soldering was a pain with all the jumpers that were needed. It took a few hours spread across a couple of days, and this is after all the components have been soldered to it:
(http://i887.photobucket.com/albums/ac80/yosefi83/9x20%20Conversion%20Photos/ButtonsBoard2.jpg)
The buttons are tactile buttons and they will have caps with different colours. For the overrides I used mechanical encoders.
Dan
-
Have been working on the wiring for the past few days. Interesting how many wires there are for machine which merely has 3 motors. Here is the component plate, a moment before I mounted it inside the cabinet, with all the components wired:
(http://i887.photobucket.com/albums/ac80/yosefi83/9x20%20Conversion%20Photos/Wiring1.jpg)
And all this to eventually control 3 motors ;)
I decided that cutting power to the spindle drive on an Estop event was not what I wanted, since I didn't want the spindle to freewheel to a stop, but rather stop instantly, so I am using a contactor (bottom right) to cut the motor phases from the drive and switch in resistors between them instead, to dynamically brake the motor. I feel more comfortable with this.
Dan
-
Vice work Dan, very neat, tidy and compact.
Will be anxious to see this one finished !
Again, NICE !
Carry on,
Russ
-
Looking good Dan. Does the Ultra 5000 have limit inputs? I was thinking of using them to halt motion but found I didnt need it on the axis and on the spindles I have brakes.
Hood
-
Thanks guys.
Hood, the Ultra 5000 has an embedded controller and is intended for stand alone applications. It doesn't have any inputs/outputs tied to any hardware feature and it has a microcontroller which you can program in C to do whatever you want, so I believe you could program it for limit inputs. Are you suggesting this in place of disconnecting the motor from the drive on an Estop? Doesn't look safe to me...?
Dan
-
Was just thinking out loud and curious of what the 5000 actually were like in case I came across one at the right price ;)
My thoughts on previous builds was to cut the power to the drive(like I do) but if the friction of the slides wasnt enough to brake things instantly (or very close) I was going to use the spare contact on the contactor to apply the overtravel inputs thus allowing the drive to instantly halt the motor.
Hood
-
Be aware that the Ultra 5000 don't accept any external signals (step/direction or other). They were not meant to be used with an external controller. They do have an option for electronic gear follower, however, and this is the mode I am planning to use mine in.
cut the power to the drive(like I do) but if the friction of the slides wasnt enough to brake things instantly (or very close) I was going to use the spare contact on the contactor to apply the overtravel inputs thus allowing the drive to instantly halt the motor.
And that is while the drive is not powered already? Not sure I would trust this. I have followed the recommendation in the Ultra manual for switching in braking resistors and it's been working great and I think I will stick with this approach.
Dan
-
How are you doing the follower? Using a SS and quad output?
Not reaally sure what you were meaning by "And that is while the drive is not powered already?"
What I was meaning was I was thinking of cutting the AC to the drive and at the same time having the spare contact of the contactor act as the overtravels into the drive. Hit the E-Stop, the drives AC gets cut, this will take a second or so to use up the power in the caps but with the overtravel getting seen by the drive it should stop instantly.
The resistors are a good way to go for sure. I dont see the wiring diag in the manual I have but there are quite a few manuals on the 2098 drives so likely its in one of the others.
Hood
-
Yes, ESS and quadrature output. Spindle doesn't have the quad output yet, but Greg is working on this.
That's right. This was what I was meaning regarding cutting the AC to the drive - you are relying on the bus voltage that's left in the drive just for a few seconds to brake the motor and this was what I was not comfortable with.
You can find the Emergency Stop description and diagram in this manual for instance: 2098-in001_-en-p on page 132.
Dan
-
Might give that a try when its done as it may work better than Step/Dir spindle in the SS/ESS
The axis motor would stop instantly with an ovetravel so even less than a second to dissipate the power wouldnt be an issue. However the big drives I use have shunt circuitry and the wee drives I use active shunts, what it would be like without the shunts I dont know.
I just downloaded that manual and see the circuit you are talking about, strange thing is I cant seem to find anything similar in the 3000's manual I have. Have done a search and neither "Emergency Stop" or "Contactor Wiring" or" Emergency Stop Contactor Wiring" gives a hit, wonder why they would have it in the 5000 manual but not the 3000 ones?
Hood
-
Strange. Don't know. It's also in the Ultra 100 manual - same diagram.
Why do you think quadrature will work better than differential step and direction (you have line drivers I am sure)?
Dan
-
Its not the step/dir being the problem as such but rather the implementation of Step/Dir spindle in the SS and ESS not being perfect. If you try and use Spindle Override it will seem fine then all of a sudden the pulse stream will stop for a fraction of a second then it will be ok again and thus I have SRO disabled on my machines. The same thing actually also happens during normal spindle operation but it is a lot less common. I likely notice it more because I have a heavy spindle ( Large spindle with gears, 250mm dia chuck etc) with a lot of momentum and it gives a right clunk when it happens.
Hopefully Greg will have a look at things when he is doing the quad spindle for you, I gave up hoping he would sort it.
Hood
-
Ah... that. I thought I was the only one reporting this to Greg. I don't tend to use SRO, but I see what you describe when I have CSS enabled - it would operate fine for minutes and then would miss some pulses and the drive would fault. Hope to be able to report how it goes with the quad signals.
Dan
-
Afraid not Dan, I dirst mentioned it to Greg about 3 or 4 years ago when I fitted the servo to the lathes spindle. Then I did the Beaver Mill and it to had the same issues. Next was the Bridgeport with the Scorpion controller and it to has the issue. The Scorpion in fact shares all the niggles the USB SS has for me where the ESS got rid of them all with the exception of the spindle issue and it also reintroduced an old USB SS issue where the pulses would be cut abruptly to the spindle if you pressed stop after a feedhold.
Maybe with you getting the Quad spindle looked at it will make Greg look at his coding but afraid I got fed up asking and waiting.
Hood
-
Well... hope the quad signal will fix this or as you suggest will get Greg to look into that part of the code.
Hope to have an answer for the quad thing soon.
Dan
-
Fingers crossed it works.
Hood
-
Got the motors mounted.
X axis motor mounted:
(http://i887.photobucket.com/albums/ac80/yosefi83/9x20%20Conversion%20Photos/X_MotorMounted.jpg)
Z axis motor mounted:
(http://i887.photobucket.com/albums/ac80/yosefi83/9x20%20Conversion%20Photos/Z_MotorMounted.jpg)
I think I mentioned this previously, but just in case: the motors are 400 steps/rev NEMA23 steppers from Keling.
Spindle motor mounted:
(http://i887.photobucket.com/albums/ac80/yosefi83/9x20%20Conversion%20Photos/SpindleMotorMounted.jpg)
It's an 800 watt Parker AC servo motor which is to be controlled by the Ultra 5000 drive just mentioned. The motor is rated at 7500rpm so I geared it down to get more torque. Won't need more than 3000rpm on this machine anyway.
(http://i887.photobucket.com/albums/ac80/yosefi83/9x20%20Conversion%20Photos/SpindleDrive.jpg)
And with the cover in place:
(http://i887.photobucket.com/albums/ac80/yosefi83/9x20%20Conversion%20Photos/SpindleDriveCover.jpg)
Dan
-
This is the cabinet with all the components mounted inside and wired:
(http://i887.photobucket.com/albums/ac80/yosefi83/9x20%20Conversion%20Photos/Wiring2.jpg)
Buttons caps are not fitted yet. There is one thing that I have already found out about the tactile buttons - they are absolutely not suitable as jogging buttons. After releasing a button it doesn't spring back instantaneously and the axis continues moving for a fraction of a second after the button has been released. Wish I knew this when I was designing the control :(
There are quiet as many components on the back of the door as well (PC motherboard and hard drive) ;)
(http://i887.photobucket.com/albums/ac80/yosefi83/9x20%20Conversion%20Photos/Wiring3.jpg)
Dan
-
Here is a short video of the first test running the axes (sorry for the poor quality as I tried to keep the file size reasonable):
http://www.youtube-nocookie.com/v/oYIXuNLr62c?version=3&hl=en_US&rel=0
Rapids are 8,000mm/min for the Z and half that for the X.
Dan
-
Looking real pro there Dan !
Very nice indeed.
About the tactile switches, are you sure what you are seeing is not the decel after the button is released ?
These snap action switches are normally fast and "clean".
Can't see the vid, says it private, get an error and it won't load here.
Nice work,
Russ
-
Hi Russ,
Thanks for the compliments.
I am sure about the buttons as when I jog from the keyboard it's perfectly responsive. With the keyboard I am able to do a very short touch on the key and the axis would hardly accelerate, moving just a bit. With the tactile buttons, on the other hand, no matter how short of a touch you try to do it would always move that minimal amount, managing to accelerate to full speed (or close anyway).
Sorry for the video. Please check now.
Dan
-
Dan,
I wonder if it could be due to being an external button, not just the button type.
Could verify by swaping with a good limit switch or other known good button ?
Might be the electrical "route" of the external circuit causing the delay ?
I'm just chatting, I'm certain you have a keen grip on it.
Anxious to see what you find out.
Thanks again,
Russ
-
A good idea, Russ. Will have to test with another switch and same wire length to be sure. Could be the wire inductance playing a role (thought currents being sunk by the ESS are extremely low)... hmm... but then as far as I can tell it does start immediately, but it's when I release that it lags. I will do the test to find out.
Thanks.
Dan
-
Looking good Dan, regarding the external buttons, it may just be the way it is via the ESS. I am not sure if the ESS handles the jogging internally or has to pass it to Mach, if the latter then that could be where the delay is, where a keyboard is connected right to Mach and not through a plugin.
Hood
-
Thanks Hood.
You and Russ were right about the external buttons. I just played with a simple push button in place of the tactile ones and it demonstrated same behaviour. Thought may be the controller frequency would have an effect on this, but it absolutely didn't. Too bad it's the way it is as the jogging buttons are almost useless :(
Dan
P.S.
LOL... just reread my post from yesterday and it's funny how I assumed it was wire inductance.... probably was late enough to make me think not clearly ;D
-
How are you doing the Jog buttons? you have them set directly in Ports and Pins or are you using an OEM trigger?
Not sure if it would make any difference either way but may be worth trying the opposite of what you have.
Hood
-
Hood,
I had them set directly in Port and Pins. Just tried using an OEM trigger instead and it behaved absolutely strange - a single short button press and the axis doesn't stop till it hits soft limits.
Dan
-
It may not work well via OEMs but just on the offchance, did you have the active state set wrong for the OEM Trigger input?
Two presses would likely start/stop if you do.
Hood
-
Nice Job Dan!
-
Thanks JH. Working on the spindle servo now. Programming the drive to do all the stuff I want. A few little things still left then to get it fully running: fitting the home switches, the cable carrier for the carriage, fitting a guard on the chuck with a motor disable switch and fixing the tray and guards. Got them welded already, but the tray came out horrible. It's stainless and was TIG welded, but still got seriously (and I mean SERIOUSLY) distorted.
Dan
-
I am having a problem trying to tune the Ultra 5000 servo drive. Autotuning doesn't appear to work as the values it calculates don't give good performance and are the same values regardless of the mass connected to the motor.
I can't get it to hold position. It keeps producing this annoying buzzing noise. And it eats up all the current set in the current limit. If the current limit is small the buzzing sound stops and the current drawn in a standing state approaches zero as expected. But at this currents the system cannot operate as the motor doesn't have enough power.
Tried to play with all the parameters, but still can't get it stand still without producing the noise and drawing so much current.
I was able to tune the servo to fair results with the chuck removed, but have no luck with it mounted.
Does anybody have any thoughts? Must be something I am overlooking or doing wrong. JH? Hood? You set up far more servo systems than I did.
Dan
-
Hi Dan,
Till the pros get back ....
I've only used the DDM and DSD drives but had this happen when I had the wrong motor set up for the drive.
Might double check ? ? ?
Russ
-
Dan, just read your PM and was about to reply but saw you had posted here so might as well also reply here.
Lathes spindles can be a real PITA to tune and I am really needing to redo mine since I added the chucking cyl;inder.
Anyway, AutoTune has never really worked that great for me, I have found that I need to set 50% or more for the current in auto tuning to get it to even work and setting nearer 100% gave best results.
Now to your buzzing, mess with your filter, it defaults to 150 but I have found I usually need to alter that, on some spindles I need more and some a lot less, I have only 1 on the lathes spindle and I think on the Beavers spindle I have 200 so mess about and see but I reckon you may have to go lower, try 20 for a start.
Hood
-
I just remembered its not an AB motor you have so what Russ says may be worth checking, if you do not have the custom motor set correctly you may get issues.
Also worth noting is having a low count encoder makes tuning a lot more difficult as an example here are a few screenshots of exactly the same PID settings and same manual tuning settings. The only difference is I have interpolated the encoder (smart encoder with 1024lines per rev) by x4, x256 and x1024. As you can see there is a huge difference between the x4 and the other two.
Hood
-
Thanks guys. I checked the custom motor parameters and found a small discrepancy: torque constant was set to 0.389 instead of 0.336. I fixed it, but this is a small error and I don't think it would be that much of a problem. But I will play with tuning tomorrow to check. Oh... and there is a parameter in the Ultraware motor editor called "Rated voltage" - the one that I filled in there was "Maximum bus voltage" from the motor specs. Assume this one is correct.
Hood, I am not sure which parameter you mean when you say you had to increase the current to 50% or more.
I tried the filter, but at low values the system is very loose and increasing the filter value brings back the buzzing.
Your point regarding the low current encoder might be the case here as the encoder on this one is only 1000 lines per revolution. The images you attached are very interesting. Didn't realise it would play such a significant role on a spindle.
Dan
-
Oh... and one another question. In the custom motor setting window there is a parameter called "Inertia", may be if I enter there a value accounting for the chuck inertia it will help? Or shouldn't I do this?
Dan
-
Its the step current in the auto tuning window, may be totally different in the 5000 drives though.
Regarding the inertia I dont know if I would do that or not as it states its the rotor inertia and not the system inertia.
I think the encoder actually affects the spindle more as it is a constantly changing mass and also you are varying the loads quite considerably all the time. The low count encoder doesnt allow the drive to respond nearly so quickly to changes. I can set the encoder to 4x and twist the motor shaft and I can feel it moving and fighting me but with exactly the same tuning but the encoder set to 256x I cant even feel the shaft move as the drive corrects it much more quickly. On an axis the loads are not changing so dramatically so its not quite as important.
Going back to the step current for auto tuning, how I discovered that was like you mentioned auto tuning always seemed to throw the same numbers which when I came out of auto tune and enabled the drive the motor shook and jumped and then shortly after faulted. That is when I went and tried manual tuning. I set the scope to channel A as current command and set the scale to 50% of the drives peak which is about 47 amps (94 peak on this drive) but when I started the manual tuning the current was way off screen. So I figured the drive was not able to use the set parameters as it was needing to use more current but that had been capped by the 50% setting. I increased the scale to 80% of the drives Max current and it was on screen so I determined due to the mass I was needing a much higher current in auto tuning. I tried that and got much better results but they were still not great so I went back to manual tuning. As I have a back gear in my lathes headstock I have backlash between the spindle and driven shaft so I also only tune the spindle in one direction, this likely wont be a problem for you however as you are direct drive to the spindle.
The mills were much easier to tune as they had very little mass compared to the lathe and they didnt vary nearly so much.
Hood
-
Ah... no, it's the same with the 5000. Also have the step current there, but again no matter what (changing the step current as well) the autotuning gives same parameters.
I tried the inertia thing and it didn't make any difference at all.
I went back and removed the belt and so tried to tune the motor alone, but interestingly I was not able. Exact same values that worked for me before, didn't this time and I couldn't tune it any better than with the chuck together. Really have no clue what's happened.
Dan
-
Just have to try messing with different numbers and see how you get on. Set the I low, maybe 5 or so and do manual tuning and set the P in increments of 100 and see what you get. If you get it sounding kind of half decent then increase the I a bit at a time and see how you go.
Hood
-
Found the problem finally :)!!! It appears that Autotuning is seriously broken in the Ultra 5000. Once you have used this feature the drive goes crazy and is virtually impossible to tune. After using the Autotuning you have to reset the drive in order to get things back to normal.
No this crazy buzzing has stopped for me and I can at least try and tune the system :)
Dan
-
Thats good to hear :)
Hood
-
I have used a vacumn cleaner hose from a dyson to cover the lead screw on the Z Axis on a 1440 lathe
-
You mean it fully stretches and contracts with the nut move? That's nice.
Dan
-
Yes I cut it long enough to stretch and compress on both ends to keep the lead screw covered. I also put plenty of grease in there. I built this 3 years ago and can drill a 1.718" dia thru solid stock at .005" per rev.
-
check out my videos on you tube under pqmcorp 4 videos of machines I have built from scratch The lathe is now completed
-
Just watched your videos. You have one huge workshop there... and it looks so clean in the same time... I liked it. The lathe looks good too. In my opinion, with the Chinese stepper drivers, the motors or louder as compared to Gecko. And speaking of this, I would love to try out Granite Devices drives with stepper motors. They claim that they drive the stepper with a sine wave and I think it would make the motor run quieter and smoother. However, their drives are a bit expensive.
Dan
-
I know my shop / equipment is larger than most hobby shops. I am a 5 S (Lean) expert for my day job why it is clean and organized... I used the chinese motors and drives from Keling Technologies because of price. They work fine and when you are machining they cannot be heard.
Thanks for the comment and viewing my videos
Joe
-
Thank you for sharing, Joe.
Dan
-
Dan, did you ever get the spindle tuned? Did you ever get the plugin sorted so you can run it from the SS with the quadrature signal?
Hood
-
Hi Hood,
I was not able to tune the spindle :( Just looks like the Ultra 5000 has the wrong parameters for tuning a servo. Can't get it hold position and be stiff at the same time. Either can get it position tuned or be stiff. Put in an AMT programmable encoder instead of the original 1000 ppr one the other day. Set it to 2048 ppr and tried to tune it. Had only less than an hour to play with it, but it didn't feel any better. In fact may be even worse, but I am not sure. The belt, although pretty tight, starts behaving as a musical instrument string when I try to increase the position gain. And in the same time I get overshoot on position. The point when I am happy with the position gain and there is no overshoot, the spindle loses stiffness completely.
Regarding the plugin, yes, Greg sent me a version with quadrature signal for the spindle, but he said he didn't think it would solve the problem we were experiencing, but it would be nice if it did in some way. If you like to play with I can send it to you. It works very well here, and in spite of what Greg said I haven't been able yet to get any symptoms of the problem we were seeing with the spindle. SRO seems to be working fine as well as CSS. But I only had a few minutes to play with it.
Dan
Dan
-
Hopefully you will get the better of it. Can you set the resoltion of these encoders higher? You may have more chance getting it tuned if you can.
I have a word doc that I got from someone who in turn got it from Allen Bradley, it is for tuning the Ultra 3000's but to be honest its not that detailed but may help, Have attached it here.
Regarding the ESS plugin, yes if you could send I will give it a go and see.
Hood
-
Hood,
Thanks for the doc. The Ultra 5000 has a somewhat different tuning screen, but I think I can try implementing some of the recommendations, like the scope setup. Wonder if they could provide a similar doc for the 5000 if I asked.
I emailed you the plugin. Please let know how it works for you.
Dan
-
Thanks Dan will give it a go later today if I get a chance, I have to swap a servo and drive on my manual lathes spindle first though.
AB may have info but hope you have more luck that I have ever had, all they ask is if you have a service contract and if you dont they are not willing to talk. It may of course be different in other parts of the world.
Hood
-
It may of course be different in other parts of the world.
:D :D Not with AB.
-
It may of course be different in other parts of the world.
:D :D Not with AB.
I dont know either, I think once Brett managed to get some info on the Sercos drives for me, then again it may have been because he put on his best, posh, redneck accent and said if he didnt give the info him and brother Bubba would come and visit ;D
Hood
-
Tried the plugin and it seemed fine for a while but then the pulses stopped for a split second so afraid its not fixed for me.
Hood
-
We'll have to keep waiting for Greg to fix this then :(
Dan
-
Maybe one day he will but well........
I have just purchased the CSMIO IP-S, should be getting it in a week or so. It has a few drawbacks at the moment but also a lot of pluses. I plan to put it on the wee lathe first then if it comes good I will be swapping it over onto the big lathe and the ESS will go on the wee one.
Time will tell if its the right thing to do I suppose.
Hood
-
That's an interesting controller. I was not aware of it. Will be interesting to see how it works for you. Found the introductory thread for this controller and your last question is back from May and was not addressed, which makes me wonder if they still work on it and all issues have been sorted.
Dan
-
We have talked via email since then and I am fairly confident the threading can be sorted. It works reasonably well as is, just the pullout delay means you will end up with a groove at the end of the thread which is not the best. The encoder modules are due to be on sale this week so I decided to get one. They also have a more basic controller but sadly it seems there will not be any threading in it which is a shame.
Heres a few more vids from the guy in Poland that seems to be their tester for the lathe.
http://www.youtube.com/user/dziabi3?feature=watch
Hood
-
The lathe has been running for quite some time now. Sorry for not updating with a proper video, been short on time as always :(
I have been experiencing troubles with the ESS though. It keeps running out of data and no matter what I do it keeps doing this. Reduced the controller frequency to as low as 250Hz and it's at least usable, but absolutely frustrating with the delay on Feedhold, FRO and SRO. Anyway, seeking for a solution and for a correct way to do the wiring, decided to use a separate 5V supply for the ESS, to get proper isolation from the PC. What I had is a single PC PSU for all the logic - both 5V and 12V. However, unfortunately, I have discovered something weird with the Ultra 3000 - it appears that its Common (ECOM) for Step/Dir input is grounded to the body and hence to the ground of my machine.
Is it really so (Hood?) ? How do I separate then the 5V logic from the rest?
Dan
-
Dont think you can Dan, have you tried an older plugin?
I had issues with the one you sent for quad spindle and newer ones were even worse so I have not updated it for a long time.
I am still using 10a plugin and dont have running out of data issues but as said did get them with newer ones.
Hood
-
Hi Hood,
That's a shame with the grounding then :(
Currently using the latest one which supposedly fixes this issue. Haven't tried older plugins. Could you please email me the one that works for you? What controller frequency do you have with it?
Dan
-
I have the frequency at 2KHz. Te attached plugin does not have quad spindle so you may not be able to use it.
Hood
-
Thanks Hood. Swapped the Ultra 5000 for a 3000, so using Step/Dir now ;)
Dan
-
Let us know if it helps, certainly never had an issue with the step/dir reference being grounded.
What version of Mach are you using? If 066/067 I would junk it right away as it has issues, mostly in Turn but have also found a few in Mill, maybe I should re-mention it to Brian now there are issues with mill as well as he may do something about it ;)
Hood
-
I have heard that the latest couple of versions had bugs. Don't remember though what version I have on this lathe. Was just at the machine, but forgot to check. will check tomorrow.
I soldered in an optocoupler on the Step/Dir line between the ESS and the Ultra 3000 and another one for the Index going back to ESS. These are the only two lines sharing same ground with the ESS. Thus I was able to use a separate power supply for the ESS. The immediate result I noticed is that moving the Servo power connector had absolutely no effect. Before that, when I moved the connector up/down the GND pin would disengage and the axes (steppers) start moving at a steady feed. I think it comes to show that now I have the wiring done properly with good isolation and it will save me more possible headaches in the future. Besides, I did suspect it was losing a step every now and then, but couldn't confirm this. Hopefully this got solved with this move as well.
haven't yet fully tested and run a program. Replaced two double channel optos, but on both the second channel didn't work so the direction line was not working. have some more spares, so hopefully will find a owrking one tomorrow. Also, was not able to get the Index line through the opto. Will use a scope tomorrow to check this too.
Dan
-
Got the direction line working today. Turns out I was connecting the second channel of the opto with wrong polarity ;D Now it works fine. Don't know if it's my imagination but I think the steppers got a smoother mid-band sound with this isolation.
Also partially got the index signal sorted. I am getting it through the opto and it works fine. But I am using the unbuffered output from the Ultra 3000. Now I wanted to use the buffered one so that I can divide the counts inside the drive (hopefully it divides the index too), since I have a 2:1 reduction. However there is some strange problem with the buffered Index line - it outputs exactly 8Khz signal at all times - even when the motor is still. A and B buffered outputs are fine, it's just the Index that does this weirdness. Looks like a bad drive, though it is a new one. Will probably have to use an external opto-sensor for the Index :(
Dan
-
Thats a strange one, why not just go back to unbuffered and use the ratio in Pulley setup?
Hood
-
This is what I have been doing - either under pulley setup or in the ESS config. Problem is that I found that I couldn't re-ran the same thread. It doesn't always pick up the index in the same position of the chuck. Thought if it would be done in the drive it might be different.
Dan
-
May be worth resetting to factory defaults to see if the drive works. Also might be worth checking the line driver, maybe a solder short on it.
Hood
-
OK. Didn't think about resetting to factory defaults. May be worth a try indeed. No line driver - using a scope right on the drive's connector.
Dan
-
I was meaning the line driver on the logic board.
Hood
-
Ah... it does look like a hardware issue. Not sure if I would be willing to take it apart though. Haven't yet tried resetting factory defaults. Will try that first. Mean time it's divided in the ESS config as it used to be.
Ran a a couple of short programs a few tens of times each. Had the ESS running out of data once at 4Mhz controller frequency. Reduced it to 2Khz and so far it looks good. Using the current plugin (ESS_v10h2d1a) and Mach version .062. It's hard to tell though if the problem has gone. Only time will tell. Haven't tried your plugin since I want to know what exactly solves this. If I see the problem again I will try your plugin.
Mach3 .062 has a strange issue however. If I feedhold and then start again it would make unpredictable moves at rapid feed for a few seconds (just skipping all over the place with the toolpath showing exactly what it does) and then follow remaining toolpath correctly.
Dan
-
If your eyes are good and soldering skills reasonable its fairly easy to replace the receivers or drivers, I had to do it on one drive that I made a balls of soldering a cable up ;D
Hood
-
Thanks for the encouragement, Hood. Still more like a shot in the dark, but I will give it a go when time permits and if factory defaults fails.
Dan
-
Was still experiencing those annoying runs out of data (controller now at 500Hz). So tried the plugin you posted. This introduced a new problem: every time I used a run from here - as soon as the OK on the prep move window hit - a SmoothStepper Watch Dog window popped up! It suggests increasing the trigger time, but that is already set to the highest 3.1 sec. Wasted a few hours and haven't made a single part yet! Can't get from start to end without a ESS error!
And an update about threading: ran several tens of thread cut tests today with the index being halved in the Smooth Stepper config and good news is that as long as I don't disable the spindle servo and rotate the chuck manually, it seems to pick up correctly each time, running same thread over and over again.
Anyhow it's sad as the lathe is useless since can't make a part without getting an error from the ESS.
Dan
-
Not so good about the time out issues, I have only suffered from that once and it was when I tried a newer plugin, possibly the one you passed on. Regarding the trigger time, not sure what mine is set at but will go and look, I have cetainly never changed it so should be whatever default is.
Hood
-
I am thinking it's PC dependent. Wonder though if there're any network settings I could change in Windows to sort this.
Dan
-
Just looked and mine is set at 2.0 which as said must be default.
Not sure about the network, mine is just windows XP default network set up. ESS is connected direct to the picos on-board port and I have a USB network adapter for connecting the lathe to the workshops network.
Hood