PDA

View Full Version : No turning back, bought servo & control package



Mcgyver
08-10-2010, 11:22 PM
well I took the plunge, with a little encouragement from Macona I've bought a set of Yaskawa controllers and AC servos to start on the bport cnc series 1 retrofit. Part of me is really excited at the capabilties this will bring to the shop, the other part is almost groaning at taking on another major project - 2 lathes to scrape and rebuild, a t&Cg to finish scraping, a central coolant system 60% done etc etc...probably no different than the rest of you, irght?"

This is much worse than say some 16 year old who just got a license being handed the keys to a car....this is like handing the keys over to some kind yanked out the deepest outback who's never even been in a car.

Lots of dumb questions to follow Im sure :D

for those not familair with the series one cnc, they're a little bigger than a regular bridgeport. Mine's the rigid ram with a 30 taper. Here it is with the top removed (stands over 7' so won't get in the garage otherwise) when i got a couple years ago. Been moving it around the shop and looking at it since. They are imo a great base for a cnc retrofit, solid machines worth close to zero with a non working controller

http://i785.photobucket.com/albums/yy132/michael01000/bport/bport1.jpg

scmw
08-11-2010, 02:05 AM
Very Nice Mcgyver!!!

I purchased a pair of working Series II BP's in June. One's named Brutus and his sister is Big Bertha. Both currently have their original steppers as they started out as NC machines without any handles too. The seller who I bought them from had converted them to Mach 3. I'm currently learning the program on the job and loving it. Very easy to work with.

I'm keen on seeing what kind of performance you get with the servos. My mills will do about 30 IPS in their current config. Eventually I'll bite the bullet and replace the steppers with controllers and servos. I don't know how much that will cost but if it speeds things up without any sacrifice other than a few dead presidents, it should be worth it.

Does your Series I have an air assisted knee or is it only motorized? Mine use 40 PSI along with the stepper for the knee.
What size motor does it have on the spindle and is it 220 or 3 phase? Mine are 3 phase. I have a rotary converter to handle the 3 phase leg.

I agree that it's a great CNC retro candidate. The Series I and II mills are very rigid and have the strength required for serious cutting. You shouldn't see any flex at all with yours, especially with that ram. Nice Grab!

macona
08-11-2010, 02:39 AM
Very Nice Mcgyver!!!

I purchased a pair of working Series II BP's in June. One's named Brutus and his sister is Big Bertha. Both currently have their original steppers as they started out as NC machines without any handles too. The seller who I bought them from had converted them to Mach 3. I'm currently learning the program on the job and loving it. Very easy to work with.

I'm keen on seeing what kind of performance you get with the servos. My mills will do about 30 IPS in their current config. Eventually I'll bite the bullet and replace the steppers with controllers and servos. I don't know how much that will cost but if it speeds things up without any sacrifice other than a few dead presidents, it should be worth it.

Does your Series I have an air assisted knee or is it only motorized? Mine use 40 PSI along with the stepper for the knee.
What size motor does it have on the spindle and is it 220 or 3 phase? Mine are 3 phase. I have a rotary converter to handle the 3 phase leg.

I agree that it's a great CNC retro candidate. The Series I and II mills are very rigid and have the strength required for serious cutting. You shouldn't see any flex at all with yours, especially with that ram. Nice Grab!

Dang, 30 Inches Per Second??? ;)

The limiting factor in performance of the servos is the speed of the parallel port. I think I run my mill and lathe at 60kHz. Both around .0001 step resolution. With my old PM servos I can do 270ipm rapids with the parallel port. When I had a GRex installed I could do a full 400ipm (Rated 2000 RPM on the servos.

I have a set of Mitsubishi servos set aside for my mill. Who knows if or when I will get around to that though. Need to do the laser cutter next.

My supermax has the air balanced knee. I can get about 120ipm on the knee. Peck drills pretty good.

http://www.youtube.com/v/mtb7l6BS-Is&hl=en_US&fs=1?rel=0&hd=1

Mcgyver
08-11-2010, 08:27 AM
...duplicate

Mcgyver
08-11-2010, 08:27 AM
knee lift, umm, mines the armstong model :) Z motion is via the quill and the table has an old fashioned manual crank. As a newcomer, at for my first go around, getting something reliably working is the primary goal so i'm going to leave this configuration

It's 3 phase, almost all of my stuff is three phase run on a home made rotary converted. That was one of my questions.....I think i'm getting good balanced power based on measuring the voltage but I wonder wether it's going to be good enough for the controllers and and servos. Should i put inline with my generated 3 phase a small vfd to clean up the power, or is that a waste....or do i in fact need an even smaller vfd inline with each axis power supply? (ie maybe one VFD won't be able handle the difference of 1 or 2 or 3 motors running)

hitnmiss
08-11-2010, 02:08 PM
I'll be watching this thread with interest. I have a series 2 with a Heidenhien controller. Works well but some day it has to die.

My spindle motor is controlled by the controller, are you going to incorporate spindle speed into the controller?

Make sure you keep us up to date on your conversion!

Oh and SCMW, my mill is nicknamed Emma!

scmw
08-11-2010, 05:07 PM
Whoa, Jerry! Thanks for catching that. I meant IPM. 30 IPS would be a dream! However with the weight of the table being thrown around I'm sure that the box ways would wear down prematurely!

Armstrong model, I'll have to use that!

Mcgyver
08-11-2010, 05:19 PM
A guy I know has a machine that goes up to 11800 ipm, thats about 200"/sec on the diagonal....its the rapids on a very expensive laser cutter. downright scary when its coming at you (behind glass, like thats going to stop it)

hitnmiss, the speed control is the manually adjusted cone shaped pulley style . Tool changer is also manual so hadn't really thought about it being an issue....change the tool, set the speed.

haven't nick named mine yet, we'll see how she behaves first

kf2qd
08-11-2010, 05:40 PM
You probably only want to go with a phase converter. A VFD will probably mess with the servos, or the servos with the VFD. (those servos are just a specialized VFD with more capabilities because of the motor.

Also check and see if you can hook those servos up single phase. Some of them will connect single phase with L1 to L1 and L2 to L2 & L3 on the drive. What wattage drives are you using? And did you buy the pre-wired connectors for the control connections? those connectors are real tight and the prewired ones will save a lot of talking to yourself....

MaxHeadRoom
08-11-2010, 05:50 PM
knee lift, umm, mines the armstong model :) Z motion is via the quill and the table has an old fashioned manual crank. As a newcomer, at for my first go around, getting something reliably working is the primary goal so i'm going to leave this configuration

It's 3 phase, almost all of my stuff is three phase run on a home made rotary converted. That was one of my questions....

In spite of having 3 phase in Have you checked to see if the controller and servo's are on single phase?
This is mostly the case, if so you just need a VFD for the spindle and everything else can be run off 240 1 phase.
This is how the majority of conversions are done.
Alternatively use a 1 phase in VFD and run the whole machine off 240 1 phase.
Max.

Mcgyver
08-11-2010, 06:51 PM
let me clarify/explain how a VFD came into my thinking....and why it should probbly go out again

I have 3phase via my rotary phase converter. The controllers I just bought were advertised, as 3 phase in. The manual shows 3P and 1P wiring, however i don't know if this means each servo can do either, or whether there are two different models offered; 1p/3P If they take 1p this whole thing is moot, but when i wrote the above i was thinking i need 3P into the controller - the sticker on the side of the servo says 3P in

I'd heard others complain that their home brew phase converters cause problems cnc electronics; don't know if its true or not but that that its not clean enough, even enough, right phase separation, whatever. So I was wondering if something, a VFD, should be used to deliver a higher quality 3 phase than my RPC is capable of (thinking the controllers needed 3p). I wasn't intending on using VFD to control anything, it would only be there to make sure the 3 phase was good

as i say I think its moot if the controllers will take 1P, I'll run 1p to them and RPC 3phase to the spindle motor....but that sticker is making me think its a 3P in model in which case will it tolerate perhaps less than perfect 3p from my RPC

the servos are Yaskawa 750watt Sigma II

http://www.jcdzwx.com/%E6%8A%80%E6%9C%AF%E8%B5%84%E6%96%99/%E4%BC%BA%E6%9C%8D%E5%99%A8%E8%AF%B4%E6%98%8E%E4%B 9%A6/%E5%AE%89%E5%B7%9D%E4%BC%BA%E6%9C%8D%E5%99%A8%E8%A F%B4%E6%98%8E%E4%B9%A6-SGDM.pdf

macona
08-11-2010, 10:50 PM
You probably only want to go with a phase converter. A VFD will probably mess with the servos, or the servos with the VFD. (those servos are just a specialized VFD with more capabilities because of the motor.

Also check and see if you can hook those servos up single phase. Some of them will connect single phase with L1 to L1 and L2 to L2 & L3 on the drive. What wattage drives are you using? And did you buy the pre-wired connectors for the control connections? those connectors are real tight and the prewired ones will save a lot of talking to yourself....


Hell with phase converters! VFD all the way! ;) A VFD will not interfere with a servo or vice versa. Virtualy all cnc machines have the spindle drive in the same cabinet as the servo drives.

The servos will run single phase. You usually have to cut back the acceleration a but to make sure you dont undervoltage the buss. I think I have my lathe's spindle servo set at 70% max torque and I have no problems running it off single phase.

The servo drives use single phase in to power the logic and 3 phase to power the buss. There is a contactor installed before the buss input. This contactor is either controlled by the drive itself or through the estop system of the controller. If something goes wrong buss power is dropped and there is no motion. Like I mentioned above, the drives will run on single phase you just need to derate a bit. The yaskawas may trip a phase loss error on single phase. I believe there is a setting in the parameters to ignore this. Another solution which is most commonly done is to run a jumper from one of the other phases to the other terminal. As long as it sees power it wont bring up a lost phase error.

And again, buy a VFD for the spindle. Dont even think about an RPC. I use a RPC on a three phase drill/tap press at my dads shop and it is no where near as smooth as real three phase.

Those little mini centronics are a pain. I scored some breakouts and cables for my next project.

clutch
08-12-2010, 07:08 PM
Anyone have a scan of the head drawing of this beast? I have bud that would love to have one.

Thanks,

Clutch

Mcgyver
08-12-2010, 07:26 PM
I have the electrical, hundreds of pages, but no mechanical :( sorry i can't help...I like to get one and scan and post for posterity

Hood
08-16-2010, 05:43 PM
Anyone have a scan of the head drawing of this beast? I have bud that would love to have one.

Thanks,

Clutch


Yes :)
But no way to attach that I can see.
Hood

scmw
08-20-2010, 09:55 PM
Hey McG,

You must have those motors swapped out and running by now right? I'll bet there's enough swarf by your mill to reach your knees or is it just wishful thinking at this point? :p

Mcgyver
08-21-2010, 12:14 AM
Hey McG,

You must have those motors swapped out and running by now right? I'll bet there's enough swarf by your mill to reach your knees or is it just wishful thinking at this point? :p

oh man, do your overestimate me :)

starting this thread was a good thing though, my lethargy is counter acted by peer pressure.

A 57 lb box arrived from Korea less than 48 hours of the auction closing. Phenomenal service. My new best friend, My Korean buddy David is really a saint of a guy. I was prepared to pay 300-400 in tax and duty and when it arrived the fedex man wanted 13 in sales tax and 10 processing. I didn't understand?? Later, I discovered BFF David filled out the customs form a $!00 US ! :eek:

Ok, since i have the stuff, here's some pics. Its a very complete kit including connectors and breakout board. The four contactors i picked up locally for $100

http://i785.photobucket.com/albums/yy132/michael01000/bport/DSC_4591-large.jpg

Mcgyver
08-21-2010, 12:14 AM
http://i785.photobucket.com/albums/yy132/michael01000/bport/DSC_4592-large.jpg

http://i785.photobucket.com/albums/yy132/michael01000/bport/DSC_4594-large.jpg

Mcgyver
08-21-2010, 12:17 AM
Macona, these servos are very clear about three phase in - you still feel single can be input? I have 3phase, home brewed, but I think pretty good...voltages are balanced and its caused no trouble. Opinions on whether i should input the utilities single phase or my 3 phase?

http://i785.photobucket.com/albums/yy132/michael01000/bport/DSC_4593-large.jpg

Its going to be quite a job removing whats there. The second cabinet, the power cabinet, is jammed with transformers i don't think i need. If i gut everything and use just one cabinet i reclaim a few more sq ft

http://i785.photobucket.com/albums/yy132/michael01000/bport/DSC_4597-large.jpg

scmw
08-21-2010, 01:42 AM
I'll try to post some pics of my beasts when I get back on Monday. Mine started with the external cabs and the tape readers. Both are gone and the cab on the back of the mill houses everything now. I've got Gecko Vampires in each one as I'm still in stepperland. When you remove the tape reader, you might want to harvest it for spare parts and what are probably two decent motors.

I figure by the time on back on Monday you'll have it all wrapped up and milling right? :D

Seriously though, have a lot of fun with this and document the process so that you have a clue if and when something crashes. I'm not too far behind you in putting servos on mine.

djc
08-21-2010, 02:41 AM
...these servos are very clear about three phase in - you still feel single can be input?

The best advice I can give is try it and see. At worst, it simply won't work. It may throw up a warning or an error. It won't damage the servo.

I have a Mitsubishi MRJ2S 1Kw drive that says it's three phase only, but runs very nicely on single phase. I think the reason they ask for three phase input is that the current draw is then lower (e.g. on a 200W Mitsu. it's 1.5A on three phase and 2.2A on single phase).

Looking at the plate on the drive, I'd say that if you make sure you supply it from single phase that can easily supply 10A and through cable of suitable size for this, you should have no problems. Again, don't skimp on the cable from the drive to the motor.

macona
08-21-2010, 03:48 AM
Looks like a nice set of motors and drives! Everything there you need. I recommend getting some DIN rail to mount what you can on that, makes life easier! You got the contactors too. They make a special e-stop relay you can use to do you estop for the whole machine. It will also take in error signals from the drives. This can control these contactors. Dont leave them out!

Yes, they will work on single phase. Every three phase drive I have tried so far that specs 3 phase has worked on single, Mitsubishi MR-J, MR-H, MR-J2, Yaskawa, Allen Bradley, Glentek, Aerotech, etc. I cant remember if the Yaskawa has phase loss detection, if it does it should have a setting in the parameters to ignore it. If not just jumper the extra terminal to one of the others.

It not so much that the current draw is lower. It ripple. There is significantly more ripple from rectified single phase vs rectified three phase. To overcome this more capacitance is required in the main buss to filter this. More capacitance means more or bigger capacitors which cost more money and take more space so they can make as small of a drive. What can happen is the drive can trip an undervoltage error on a heavy load if the ripple bottoms too low. But since you are dealing with drives designed to run as low as 200v (Japan standard line voltage 100v, 200v, 400v) and figure 10% tolerance on that you get an undervoltage trip of probably around 180v. If you ever do run into an undervoltage trip just go into the settings and set the max torque lower and it will fix that problem.

You will want to make up a serial cable and download the yaskawa software. It beats the heck out of pushing the buttons on the front of the control!

And get yourself a fine tip soldering iron. The mini-centronics connectors are a nightmare! Some small heat shrink will help deal with stray wires in the connectors.

Edit, just noticed you bought the contactors. Doh!

Mcgyver
08-21-2010, 11:14 AM
I figure by the time on back on Monday you'll have it all wrapped up and milling right? :D
.

Only if you're on your way over to help, beer supplied.

I just can't see anything that i need to keep in the old unit. Looks like there'll be enough transforms reclaimed that i can take another shot at my spot welder. yet another unfinished project

On the three phase, thanks for clarifying...since I have my rotary 3 phase and its being run to the machine anyway for the spindle, is there any reason why you would use it for input on the contactors?

thanks for the help gents....as i say this new to me so i appreciated the hand holding.

macona
08-22-2010, 02:09 AM
Nope, dont do it. Just run them off single phase. Also now is the time to add a VFD to the spindle. They are cheap and you get more and smoother power from the motor. RPC 3 phase is not true 3 phase in the least. One other very, very nice thing, especially for cnc, is the VFD can brake the motor to a stop very fast. Some older CNCs used a pneumatic brake to stop the spindle. Problem with this was the brake wears out and then it is a real pain to get to it.

Mcgyver
08-22-2010, 04:39 PM
Nope, dont do it. Just run them off single phase. Also now is the time to add a VFD to the spindle. They are cheap and you get more and smoother power from the motor. RPC 3 phase is not true 3 phase in the least. One other very, very nice thing, especially for cnc, is the VFD can brake the motor to a stop very fast. Some older CNCs used a pneumatic brake to stop the spindle. Problem with this was the brake wears out and then it is a real pain to get to it.


So with that, I don't really need anything in back box either (loaded with transformers)

Got the side box off today, no great mystery but close to a days labour of dirty work (on a humid day). I'm getting excited though about reclaiming that couple of extra square feet......with 2.5 extra square feet its gonna feel like a dance floor in there. The shop is so bloody crowded i had to keep moving the machine to get at stuff to take the box off. Ugh

Standard electrical i'll keep, but any value in any of these boards?

Oh yeah, what did you mean about making up a serial cable and downloading software.....is it something to do with fine tuning the drives via computer? when this is all done and said, they don't have to be regularly accessed do they?

macona
08-23-2010, 02:04 AM
On the front of the drive there is a small connector below the display. Older drives had a DB9 and newer ones used a small mini-centronics. This is a serial port. There is software you can download from Yaskawa that will allow you to configure and tune your drive. It also allows you to test the drives and run the motors with the software. It will also show you a graph of all sorts of variables like follow error, motor current, and other stuff.

The manual shows how to make the cable. I just chopped up a regular serial to make the one I did.

The drives themselves have an auto tune routine built in. Run it once and you will never have to again unless you change something mechanically.

lazlo
08-23-2010, 02:47 PM
Macona, these servos are very clear about three phase in - you still feel single can be input?

Yes, you can feed them single-phase, but the problem is you have to derate the power by 40%. The -8 suffix means they're 750 Watt servo drives, so you should be driving a 450W servo with them.

Do these take step/direction input? The SGDM was only sold in the Asian market, so I'm not familiar with it, but several of the Yaskawa servo drives I've purchased are analog input only.

djc
08-23-2010, 03:11 PM
S...what did you mean about making up a serial cable and downloading software.....is it something to do with fine tuning the drives via computer? when this is all done and said, they don't have to be regularly accessed do they?

I can only echo what Macona said above. It's not so much fine tuning as simply accessing, reading and writing down (on a piece of paper) the myriad of parameters these drives use. It's so much easier on an computer screen than by doing the four button shuffle on the drive itself.

The Mitsu. comms. program keeps a record of alarm history, a record of what parameters were changed and when and can reset to default if you screw up. It shows the state of all the inputs, and can display things like RPM, number of pulses received, pulse frequency and number of revolutions turned.

Depending on how you put the drives into the machine and where the machine is positioned, it may be hard in the future to get to the drives to make adjustments or to troubleshoot something that causes the servo motor to fault. If there is a cable hanging inside the cabinet door that you can hook out and plug into a laptop, it is a lot less hassle.

One thing you may have to do manually on the drive is set the mode of communication. Read the manual, but they probably talk both RS422 and RS232. The latter is what you want for one drive at a time, directly into the PC. When you have the system mocked up on the bench, with all axes working, it may be worth investigating the other mode as this allows you to talk to each drive in turn through one connection to the PC (it does require a small bit of extra hardware though).

Mcgyver
08-24-2010, 04:02 PM
Do these take step/direction input? .

I hope so. My interpretation of the manual is that they do (of course its tedious and 600 pages and I'm a newb so anything is possible)} and they were bundled with a break out board and wiring instructions for a mach 3 set up

So on this phase question again......I get that my generated 3P isn't perfect, but if they can take single P and work enough magic on it to make it perfect 3phase ....wouldn't taking my maybe rough 3P and cleaning it up be a walk in the park for the drive?

macona
08-25-2010, 04:26 AM
No, dont mess with it.

lazlo
08-27-2010, 07:15 PM
I agree, don't mess with it, but you're going to have 60% of the drive power running on single phase that you do on 3 phase. That's enough to fire-up the servos, but it will be interesting to see what happens when you get in the cut.

macona
08-27-2010, 11:22 PM
He should get more than 60% out of these. The drives are designed to run at 200v so even with single phase ripple at 240v he shouldnt have much loss. I have a yaskawa motor like his and it is expecting 200v to the motor. The big 3.5kw mitsubishi I have has a motor voltage of 145v.

My friend with the Milltronics VMC has 900w Yaskawas on it and he is running single phase and has never had an issue. He is running full rapids too.

KiloBravo
10-05-2010, 02:17 PM
I am in the same boat. I just bought a Boss 4 and want to retrofit the controls.
An auction just closed by FA-PARTS that looks like the exact package you purchased. What was the price ? The one I was looking at was $1500. Would I have to pay a (duty tax) in the US ?

Macona I sent you a PM before I found this post. I forgot they added a Digital Machinst to the forum.

Regards,
Kevin

Mcgyver
10-05-2010, 04:43 PM
same guy, FA Parts. mine was 1800 but i got a 4th axis, either a spare or one day i'll build a 4th....the price was pre negotiated and he put an auction up for me with the 4 and i hit the ask 2 minutes later. He offered to transact directly but not knowing and being on the other side of the world the 100 for fleabay/paypal was worth it imo

FA might just be the greatest vendor out there, without any prompting he filled the customs form out at $100 so there was only a $10 processing charge from fedex (as you probably noticed, shipping is included). There is no duty on these into Canada, but at 1800 I'd expected the sales tax extortionist to take note. I love you FA Parts.

Little progress to report, the side box is off...that's a fairly big job by time scroungable parts are saved....and wiring and conduit has been installed.

KiloBravo
10-05-2010, 08:42 PM
Mcgyver can you PM me his contact info ? I would appreciate it.

Side note someone sold Boss 8 servos and controllers on PM today for $500 for three axis !!!!! Sorry I missed that one.

Regards,
Kevin

jacampb2
10-06-2010, 06:39 PM
Sorry, wrong thread.

KiloBravo
10-07-2010, 12:41 PM
Hey what do you guys think of this auction for a Bridgeport rebuild ?
They have a package deal on some 750W Panasonic motors.

However, the drivers are 220V 3 phase in and 116V 3 phase out. They are matched to the Panasonic motors, but if I have to derate to run on single phase will that affect the torque too much ??? I don't have three phase at my house.

http://cgi.ebay.com/PANASONIC-750W-AC-SERVO-DIRVER-MOTOR-3AXIS-CNC-ROUTER_W0QQitemZ260672362572QQcategoryZ78193QQcmdZ ViewItemQQ_trksidZp3286.m7QQ_trkparmsZalgo%3DLVI%2 6itu%3DUCI%26otn%3D2%26po%3DLVI%26ps%3D63%26clkid% 3D5154146875163713440

macona
10-07-2010, 07:23 PM
I have not used panasonic servos so I really cant comment much on them. I know Yaskawa and Mitsubishi both have great support.

Otherwise they will work fine for a bridgeport. Assuming you are moving the quill and no the knee I would look for something around a 250 or 400 watt motor to move that. 750 is massive overkill for a quill. You could send them a message about either getting an additional 400w or get two 750s and a 400w.

You wont need to derate. These drives are designed to run down below 200v in japanese systems. With standard power at about 240v the voltage drop due to ripple will still be within parameters. Elmo Motion has a nice graph showing losses on a 3 phase drive ran on single phase versus current draw.

KiloBravo
10-07-2010, 08:59 PM
The motors I mentioned are 116 Volt 3 phase I believe the other ones he sold were 220Volt 3 phase.

:confused: So would the 116 Volt motors be enough for the Bridgeport ?

Mcgyver
10-14-2010, 07:35 PM
I'll try to make this ia bit of a blog, posting lots of pics to compensate for the lack of progress. I came into this with almost no cnc experience so even the basics the vets take for granted can be a mystery...so maybe the next newb will get something out it

Getting the side box of was the better part of half a day. Saved what i could, most of the caps, rails etc. Its removal freed up enough space several machines got re arranged to take advantage.

http://i785.photobucket.com/albums/yy132/michael01000/bport/DSC_4599mod-large.jpg

http://i785.photobucket.com/albums/yy132/michael01000/bport/DSC_4600mod-large.jpg

Mcgyver
10-14-2010, 07:41 PM
Next up is sorting out the wiring for the package I bought.

Here's a shot of the controller.

http://i785.photobucket.com/albums/yy132/michael01000/bport/DSC_5327-large.jpg

UVW are the hook ups for the servo. At the other end, the servo motor has a short unshielded connector, similar to the plastic ones found a pc power supply (whats the proper name for this?)


http://i785.photobucket.com/albums/yy132/michael01000/bport/DSC_5325-large.jpg

FA Parts also included nice connectors so i can have the motors plug into the side of the box.

http://i785.photobucket.com/albums/yy132/michael01000/bport/DSC_5316-large.jpg

Mcgyver
10-14-2010, 07:43 PM
also supplied were these connectors so i can make a cable from the servo's encoder to the controller. These are shielded and go from CNC2 on the controlle to the encoder on the servo

http://i785.photobucket.com/albums/yy132/michael01000/bport/DSC_5324-large.jpg

Finally, there is the centronics connector the goes from CNC1 to the break out board


http://i785.photobucket.com/albums/yy132/michael01000/bport/DSC_5322-large.jpg

Mcgyver
10-14-2010, 07:49 PM
Questions.

encoder -> CNC2, shielding required, correct? what are these connectors called? any views on what wire will work best?

servo - > UVW. Four wire required, servo lead is not shielded, should they be? gauge or other specs i should be thinking about?

break out board to -> CNC1, any recomendations?

I need some help on ideas for conduit from box to servos and some hint the right way to ground things.

Lastly, CNC3 is for a computer connection for diagnostics etc - macona suggests making up a serial cable; what type connector is it on the controller?

thanks!

macona
10-14-2010, 11:55 PM
The white connector are the AMP mate-n-lok connectors. Very common. Digi-key has them,.

Encoder: There is cabling designed especially for encoders. There are several companies that sell it. It is shielded and twisted pair. IGUS is good. Get a motion rated cable. http://cableorganizer.com/igus/cf211-data-cables/

Servo: You can get away with not using a shield. Some times it is a good idea but I dont think I would worry about it. Same as above, get a motion rated cable. Spend the money and do it right the first time. Again, Igus and Kable Schlepp are good. When you feel a good cable its like a overcooked noodle.

For generic signal lines and the like you can chop up old serial, parallel, or SCSI cables for the wiring. There is no real current handling required.

CN3 is a mini-centronics. 3M, among others make these. Looks like a 20 pin.

For soldering these connectors you are going to want a good fine point soldering iron. They are a PITA. Small heat shrink too.

Erik Brewster
10-15-2010, 10:52 AM
Questions.
what are these connectors called?


Firewire / IEEE1394 / i.Link. Don't know why, but I've never found the connector kits very cheaply.

Erik Brewster
10-15-2010, 11:00 AM
get a motion rated cable

I've had good luck with this cable. It's very reasonably priced and is available to ship in any length you want. The sales people are also willing to deal with small customers.

http://www.northwiredirect.com/roboticscableforcontinuousflexing.aspx
http://www.northwiredirect.com/enduroflexerp-2.aspx

They even sell firewire specific cable
http://www.northwiredirect.com/firewire.aspx

I'm not convinced that you need it -- Yaskawa encoder signals are just a serial burst, then incremental, I think (been a long time). In any case, it's no 800 MB/s stuff like the latest Firewire. Any twisted pair will do. Shield wouldn't hurt, thought I bet you'ld be fine without it.

Mcgyver
11-23-2010, 10:31 AM
typical male, I hate shopping...causes me more procrastination than the work...

quick Q, the 4 wire flex (20g) from Northwire is a fraction of the cost of that from CableOrganizer. The only difference i can see is that Cableorganizer is twisted pair - opinions on whether that will matter for the servo cable??.....same on the encoder cable; so long as its shielded, do i have to have the twisted pair?

thanks

Erik Brewster
11-23-2010, 09:13 PM
You might get away with it, as it isn't the twist so much as the proximity of the wire pairs to each other. I don't think I would risk it, though. Twisted pair really works and is proved in countless items you use everyday. It would really be a shame if you went and made the cables and they were flaky.

Maybe you could use the DeviceNet cables that Northwire sells? Those have two twisted pairs per cable. Not sure why they have two different twisted pair gauges, but it should work.
http://www.northwiredirect.com/devicenetthin.aspx

Or some of the vision cable...
http://www.northwiredirect.com/gigevision.aspx

In any case, you should give them a call. They were helpful to me.

macona
11-23-2010, 09:49 PM
In general you want twisted pair with encoders. Especially if you are running single ended. You might be able to get away with it if you were doing differential encoders.

Mcgyver
11-24-2010, 07:32 AM
thanks again both of you for the continued help.....

Mcgyver
05-15-2011, 06:37 PM
He should get more than 60% out of these. The drives are designed to run at 200v so even with single phase ripple at 240v he shouldnt have much loss. I have a yaskawa motor like his and it is expecting 200v to the motor. The big 3.5kw mitsubishi I have has a motor voltage of 145v.

time flies, numerous other projects interrupt but back at this.....

macona, in another thread you mention not to belt down the servos....since you have the same motors, can you confirm this advice would hold for my servos and mill? the original steppers were all 1:1 and you're saying I think I should i do the same with the servos. It would be easy to get either way as even with 1:1 I'm using a jack shaft so that i get the motor mounting i want and avoid over levering the motor shaft.

would this machine be able to take advantage of the un-belted down motors? or would things be too fast then?

thanks.

djc
05-16-2011, 03:34 PM
...would this machine be able to take advantage of the un-belted down motors or would things be too fast then?

Only you can tell us....

What is the max. (or rated) RPM of the motor? What is the pitch of your ballscrew? Hence, if the motor is going full tilt, what is the speed of your axis? Does this number frighten you? How does it compare with other machines with broadly the same slideway arrangement?

MaxHeadRoom
05-16-2011, 04:42 PM
Your advantages in reduction, providing you still get the required rapids, is that torque is increased by the ratio of the reduction and the motor/load inertia ratio is decreased by the square of the reduction, the latter means that higher accel/decel rates are possible.
Max.

macona
05-16-2011, 10:19 PM
There is really no point in belting them down. They have constant torque throughout their RPM range and even 1:1 you are going to have so much power its stupid. You would be able to snap off 1/2" end mills with motors half the size. I have.

Mcgyver
05-16-2011, 11:33 PM
thanks Jerry.

Guys you're right, i pooched out on being a good wanna be engineer on this one, I'm barely in the consciously incompetent quadrant insofar as safe speeds go and couldn't tell you the max speed without spending half a day in the 5,000 page manual :D. But the motors and mills are well known and my tired brain in my tired head hurts already with all i have to figure out so needed the informed view of someone who knows them.

I did figure out a simple design to for the motor conversion. Uses a jack shaft as i didn't want to cantilever the short stubby shaft on the motor. Its a bolt on and is identical for each axis using the original pulleys/belts. pics to follow, I want to finish it and confirm its going to be ok.

macona
05-17-2011, 01:55 AM
I think the top sustained speed of your motors is 3000 RPM. With a 5 TPI ball screw that means a top rapid speed of 600ipm @ 1:1. If you set up the drives for .0001" per pulse from the controller and you use the parallel port with Mach3 you will top out in the 300 - 350 ipm range before you start having issues. Mach does support pulsing up to 100khz or so but I have never had luck past about 65khz on multiple machines.

300ipm sounds like it's a lot but it is not. I have actually cut aluminum at somewhere around 100ipm in my machine, but had to slow down due to lack of spindle power.

Also having the possibilities of high travel speeds means you can play with high speed machining where you take a lot of light passes really fast.

Mcgyver
05-17-2011, 09:42 PM
another Q, the motors' shaft has nothing, no flat, no key way, nadda. For this application do you think a taper lock has sufficient holding power on a plain shaft without a key?

macona
05-18-2011, 12:41 AM
Taper lock, split clamp, or something like a trantorque fitting.

http://www.fennerdrives.com/keyless_bushings/trantorque_home.aspx

Mcgyver
09-30-2011, 04:06 PM
progress slowly being made.....here's a blow by blow

new servo vs old stepper. Some sort of jack shaft adapter is needed - servo motor shaft is not long enough, especialy if an adapter plate is used. These motors are expensive so i didn't want to risk cantelevering the shaft out more with an extension.

http://i785.photobucket.com/albums/yy132/michael01000/bport/DSC_6105-large.jpg

shot of where the stepper came off of. the three steppers are identical so whatever solution i come up with gets done x 3

http://i785.photobucket.com/albums/yy132/michael01000/bport/DSC_6106-large-1.jpg

raw materials - i regret using 4140 for what will be the bearing housing, too tough to get a decent finish on it

http://i785.photobucket.com/albums/yy132/michael01000/bport/DSC_6560-large.jpg

tubing is 6x3x.25"

Mcgyver
09-30-2011, 04:14 PM
first thing is the plate the servos will sit on. Servos get bolted to this, which is slotted, to permit belt tensioning between the servo and jack shaft.

cleaning it up clamped to an angle plate

http://i785.photobucket.com/albums/yy132/michael01000/bport/DSC_6567-large.jpg

mass production. All done in the mill with the plates clamped together, I dropped some pins in the bolt holes after reaming them as a I couldn't mill the slot in all three plates at once. You can see I'm using three T slot stops so i can swap the plates in and out

http://i785.photobucket.com/albums/yy132/michael01000/bport/DSC_6883-large.jpg

http://i785.photobucket.com/albums/yy132/michael01000/bport/DSC_7170-large.jpg

Mcgyver
09-30-2011, 04:17 PM
after bandsawing the rectangular tubing, I square up the ends with a flycutter also using the angle plate

http://i785.photobucket.com/albums/yy132/michael01000/bport/DSC_6581-large.jpg

here's a better shot of the T slot stops - very handy for repeatability

http://i785.photobucket.com/albums/yy132/michael01000/bport/DSC_6723-large.jpg

I drilled and bored through both sides as a single setup. I also tried to clamp right over using thre of each as the tube surface isn't flat and i wanted to minimize distortion

http://i785.photobucket.com/albums/yy132/michael01000/bport/DSC_6728-large.jpg

http://i785.photobucket.com/albums/yy132/michael01000/bport/DSC_6732-large.jpg

Mcgyver
09-30-2011, 04:19 PM
boring through the 6x3" was a challenge - needed a long boring bar. Unless its dead sharp HSS it would chatter

http://i785.photobucket.com/albums/yy132/michael01000/bport/DSC_6836-large.jpg

http://i785.photobucket.com/albums/yy132/michael01000/bport/DSC_6838-large.jpg

mock up assembly

http://i785.photobucket.com/albums/yy132/michael01000/bport/DSC_7173-large.jpg

Mcgyver
09-30-2011, 04:29 PM
the jack shaft end needs two bearing housings. these were rough drilled and turned

http://i785.photobucket.com/albums/yy132/michael01000/bport/DSC_7498-large.jpg

then using a metric bore gauge and gauge blocks, I got the perfect bore for the bearings. My plan breaks the rules a bit in that the only what to assemble is one loose on the bearing ID, one loose on the OD....normally you're supposed to have loose/loose on both ID's OR both OD's....we'll see how it works

http://i785.photobucket.com/albums/yy132/michael01000/bport/DSC_9516-large.jpg

here's one way to turn to depth, after taking a facing cut with the crossfeed locked, set the carriage stop using an adjustable parallel. made it easier to repeat the process

http://i785.photobucket.com/albums/yy132/michael01000/bport/DSC_9520-large.jpg

finish turning done in the 4 jaw, using my favorite indicator a, a Verdict 10ths one. I set the compound over so that each .001" graduation was .0025 mm - this made it comparatively easy using the bore gauge to hit my target by better that .01mm. Boring or turning to fit roller element bearings is probably the most accurate work an average sort if called up on to do

http://i785.photobucket.com/albums/yy132/michael01000/bport/DSC_9519-large.jpg

Mcgyver
09-30-2011, 04:31 PM
gratuitous turning shot

http://i785.photobucket.com/albums/yy132/michael01000/bport/DSC_9526-large.jpg

arrangement of servo and mounting plate

http://i785.photobucket.com/albums/yy132/michael01000/bport/DSC_9529-large.jpg

http://i785.photobucket.com/albums/yy132/michael01000/bport/DSC_9531-large.jpg

DFMiller
09-30-2011, 05:31 PM
Looking good.
Is this your first foray into CNC?
Dave

Mcgyver
09-30-2011, 09:50 PM
Thanks Dave, yup, first time. lots to learn!

DFMiller
10-01-2011, 03:48 PM
Mike,
I look forward to your progress.
Is that a R2E3?
I briefly owned a share in one. I wish it would have fit in my shop. It was sort of working but need a brain transplant. The other shareholders did not want to do it. It had great possibilities.

Welcome to the dark side. You are a talented machinist so you only have to work the rest of the stuff.

Dave

Mcgyver
10-02-2011, 09:27 AM
Is that a R2E3?


got me there, but I don't think so....googling R2E3 they seem to have a lot of cowling and electronics around the head that mine didn't have.....but with all that stripped away as shown in the following link, the underlying mechanical looks the same

http://chriscoleman.com/stuff/

DFMiller
10-02-2011, 12:30 PM
Mike,
I cant remember if you gave any details earlier. Is this a conversion from a manual to CNC or just an upgrade? From most of the pictures its a upgrade to modern controls. That is a much better way to go.

What are the X Y and Z movements and how large is the table? Also what is the spindle. If its QC 30 I think I have some tooling around that I will have to send your way. I got it and never used it. Much easier to send East then South. ;-)

Dave

Mcgyver
10-02-2011, 01:08 PM
Hi Dave, its a conversion to servos of what I think is series 1 cnc - was never too particular about was i getting....saw it, liked it and after an exchange tooling i wasn't using it followed me home. There's a pic of it on the first page. I got it not working and a lot of what i read said the old electronics were temperamental at best so i didn't attempt to get it working. I got a good ebay deal on used high end servos with some guidance from Jerry and that's where I'm at.

Just came in from turning the jack shafts; turning loose or interference fits for roller bearings is, for my shop anyway, about the most exacting lathe work there is. I was doing some experimenting and I think I've got consistently turning to tenths down, more on that in a bit

its 30 taper, will take regular 30 taper or those quick lock ones. If you've got some extras that would be great - we'll figure out a fair exchange, I always have extra stuff kicking about as well

thanks!

DFMiller
10-02-2011, 07:14 PM
Mike,
Looks like a nice piece of iron. It also looks like the small table which limits the work envelope but also means it will not be sagging that much. ;-)
Have you decided what controls you are going to use? Going to follow the flock with Mach or going to try out something else?

My best success with interference fits seem to be using the loctite method.

I will keep my eye out for my QC30 stuff. I know I have some around here.
I found my Quikk Switch 200 collet holders and my QC40 DA 180 holder but not my QC30 stuff.
And a couple of Rutex 3 axis drive board sets and 6 of so RS485 Stepper drives.
I have too much stuff. I forgot I had most of that.

Dave

Mcgyver
12-28-2011, 08:05 PM
ok, some more build photos......there's a lot of holes to poke in the bearing housing. I found a beater bearing, bolted it to a hunk of steel and indicated it to centre. position the spindle for a hole - ie offset one axis by 1/2 the PDC and drill and ream a 1/4" hole.

http://i785.photobucket.com/albums/yy132/michael01000/DSC_9540-large.jpg

Then using x/y coordinates move over to a 90 degree indexed position - 1/2 the PDC of the second axis. The table doesn't move again. Load a blank, drill/ream then index it using a pin to the previously drilled hole. It makes quick work of them

http://i785.photobucket.com/albums/yy132/michael01000/DSC_9543-large.jpg

The shafts are straight forward turning but a lot of it. Bearing fits call for accurate work, one's a press fit, the other sliding. I know that breaks with convention...but if i burn out a $6 bearing every decade or two I won't care

http://i785.photobucket.com/albums/yy132/michael01000/DSC_9645-large.jpg

The shafts are cut for a 3/16 key in the horizontal mill and the bearing is pressed on one end.

http://i785.photobucket.com/albums/yy132/michael01000/DSC_9756-large.jpg

Mcgyver
12-28-2011, 08:09 PM
Similar to the shaft, one housing is a press the other sliding..

http://i785.photobucket.com/albums/yy132/michael01000/DSC_9753-large.jpg

This bottom press fit one gets bolted to the bottom of the rectungular tubing

http://i785.photobucket.com/albums/yy132/michael01000/DSC_9755-large.jpg

The timing pully inside on the jackshaft is drilled and reamed to 3/4" then broached

http://i785.photobucket.com/albums/yy132/michael01000/DSC_9741mod-large.jpg

next it's drilled and tapped for locking set screws

http://i785.photobucket.com/albums/yy132/michael01000/DSC_9748-large.jpg

Mcgyver
12-28-2011, 08:17 PM
I used two set screws, one over the key to secure it

http://i785.photobucket.com/albums/yy132/michael01000/DSC_9751-large.jpg

Here's a trial assembly, everything turns smoothly! I'd originally thought I'd but jbweld or something between the housings and tubing, assemble and lightly tighten, as way to overcome the expected non-flat surface of the tubing. so far hasn't seemed necessary

http://i785.photobucket.com/albums/yy132/michael01000/DSC_9759-large.jpg

Here's the unit install with the AC brushless servo, so far so good

http://i785.photobucket.com/albums/yy132/michael01000/DSC_9760-large.jpg

With guards in place. Since the bolt-on replicates the old stepper's bolt pattern an identical bolt-on assembly installs on the other two axis.

http://i785.photobucket.com/albums/yy132/michael01000/DSC_9761-large.jpg

Now its onto wiring