Machsupport Forum
Mach Discussion => General Mach Discussion => Topic started by: fast89 on July 18, 2016, 08:06:04 PM
-
I'm doing my research for a 3-3.5kw servo drive and motor that i can use as a lathe spindle (c axis). It has to be single phase input. I would also like 3000-4000rpm out of it. I'm having trouble finding anything that can be run on single phase 200-240v. Anyone have experience in this area?
-
I think you will struggle finding that, biggest I can recall seeing are about 2Kw.
Hood
-
I found 3.5kW in my recent searching, but you won't get the speed, the bigger motors (>1kw) seem limited to 1500-2000rpm max.
-
looks like parker has some single phase drives that take step/dir signals. Now if i could find a motor i might be in business.
-
Hi,
I bought a servo off ebay a while back 2.8kw at 3000 rpm. I cant find a drive to run it at least that I can afford.
This servo is quite old, mid 90's, but looks brand new. It is fitted with a resolver and restricts the choices for a drive.
I tried sensorless vector drive but am unimpressed.
I made the decision to build one, very much easier said than done. This hobby, or should I call it an obsession, requires
you learn all sorts of new stuff. If success is determined by what you learn along the way the this project is a boomer!
Looking around both new and secondhand there are plenty of servos to choose from and cheaply if secondhand. Drives
are a different kettle of fish altogether.
Amongst the bigger industrial drives it is quite common to separate the power supply from the inverter. It allows flexibility
to spec the supply to your existing regulations and whether you have your own onsite transformer.
There has been a big upsurge in VFD use in rural NZ for irrigation pumps. Now the power supply authorities are having
to demand farmers fit these units because of the degradation of the supply. They can be nearly as much as a VFD.
There is nothing to prevent you from running single phase but you will need a dedicated 32A or maybe even a 50A supply.
The power company will probably come down on you to fit a conditioning unit otherwise you will degrade your neighbours
supply. Additionally most servos above 2.2kw are rated for 400v line or 560v DC link. If you run such a servo from 230v line
it will not achieve rated speed. A PFC boost in a conditioning unit will overcome the problem but it will SUCK major current
and you going to have some BIG link capacitors.
Search for single to 3 phase converters but be warned they are expensive and you DON'T want junk because they blow
up. I've had several thru work, Chinese stuff and the proud owners are livid!
Craig
-
Ok just looked at the specs of the motors I have on my Computurn lathe
Z is H-4075 Allen Bradley and 3000rpm and 9.9Nm cont stall torque so KW is 3.14Kw
X is 4000rpm H-4050 Allen Bradley and 6.8Nm cont stall torque so KW is 2.87Kw
I also have F-4075 and F4050 motors which are similar specs, 2.95Kw and 2.92Kw
I run them from DSD-030 drives which are single phase.
Hood
-
Also worth mentioning, you have to make sure you do not get the HV drives as they are 3 phase, so DSD-HV030 is no use for you, it has to be DSD-030.
There are also DDM drives which were a previous model and you can get a DDM-030 drive. There is also the DDM-075 drive which can be run single or 3 phase 240v but you will have to make sure your supply is good enough.
There is also the DSD-075 drive but I think that is strictly 3phase 230v unlike its older version (the DDM)
Hood
-
Hood,
Thanks for the info. I found some good prices on a DSD-030 drive and a MPL_A4530K motor. That would get me 2.5kw and 4000rpm. I think this will be enough as Its a small Hardinge HSL lathe. But i hate to under power things. I find myself always pushing the envelope.
-
I would worry a bit about the MPL-A4530K motor, it is 19amps cont and 62 peak, the 030 drive is just 15/30.
I would look for an H or F 4050 I would think, the F being probably better for a lathe spindle as it has more inertia.
Hood
-
I got that motor from allen bradleys specs. It is limited by the 030 drive. If i used the bigger 075 drive it would put out more but those are 3 phase. On another note, do you have a link for the manual for the 030 drive. My quick searching didn't find it yet. I really didn't think i'd have this much trouble finding a single phase drive to go up to 3.5kw.
-
The reduction in performance is what I would be wary of, you are going to get it tripping if you start pushing it.
You should find the manual no probs from Rockwells literature Library.
One of the ones I have is
Publication number 2098-IN003E-EN-P
Hood
-
ok, so i have a drive and motor mounted and wired. Time to set some parameters in the Ultra3000 drive. But allen bradley won't let me download the software because i'm not the registered owner of the product and/or the serial doesn't match their records. Does anyone have the Ultraware software they can share?
-
ok nevermind i think i got it.
-
Hood,
Do use step/dir or step up/step down mode? I have it rotating at an rpm thru Ultraware but not so successful in mach3. On a position command, it oscillates wildly and faults out
-
Hood,
On a position command, it oscillates wildly and faults out
That sounds like drive/motor tuning needs to be done.
-
Ok, It needed to be set on step up/ step down mode. Now i can manually jog the spindle but i still have problems. I keep getting an e21 error which is auxiliary encoder error. It seems i should be able to disable the auxiliary encoder but i cannot find the setting anywhere in ultraware. I'm using version 1.83.
-
and the motor will only go in one direction. manually jogging the spindle thru mach3 seems to be fine, just one direction only. If i command a G0 C180. then back to 0 then 180 again, it will get an e21 error.
-
Not sure what controller you are using but if it is from Mach then Step/Dir is what you want to be using.
You should also really be using differential Step/Dir signals rather just single ended, so maybe you will need to make up a line driver.
E21 error is to do with the step/dir signals, they are not good enough and hence the drive is faulting with that error.
Hood
-
hood,
the drive is an ultra3000 2098-dsd-030x I'm using a breakout board that has single ended or differential. Currently i'm using differential. I'm a little confused on the inputs. I thought i would have to specify somewhere what the actual input is.
-
You have to tell the drive that you are using Step/Dir inputs, see first screenshot.
Next you have to tell it if you want any gearing between the step inputs and what the drive sees (second pic). For example if you are just using the parallel port then you may not get the spindle turning very fast due to the pulse limitations of the parallel port and the kernel you have set. So what you would have to do is set some gearing in the first box. If you had it 1:4 then what the drive would do is multiply every pulse from Mach by 4.
As far as the inputs for the Step/Dir they are on specific pins, 4,5,6,and 7 if I recall. You may also need pin 2 connected to the same Gnd that your differential signals are connected to.
-
I have those settings correct. Not sure on the gearing yet, it needs to work a little better and more stable before i'll worry about that. The ground on pin 2 could be the issue. I'll give that a shot tomorrow. Right now its time for a pint.
-
Grounded pin 2 with no change. Even tried step and direction single ended because it is a different set of pins on the B.O.B. Motor will still only rotate in one direction and errors out very easily. Has to be something with the breakout board or mach3. From Ultraware, it will run any rpm just fine.
-
Ok I presume you have done a tune in the Ultraware and it is looking good?
What motor do you have?
What steps per unit and Velocity do you have set?
Is it always E21 that is the error?
What breakout board are you using?
Hood
-
I've done only minimal tuning. I haven't gotten the oscilloscope function to work in Ultraware. I'm using MPL-A4530K-HK22AA for the motor. It has a 2000line encoder. The steps are about 22.4 (should be 22.2222). For now it is close enough on the steps and getting 360* revolution. Any velocity over 4000deg/min gets more and more unstable. It's always an e21 error unless the proportion gain is too high and it starts to oscillate and will get an e23 alarm. I'm using a machmotion apollo I breakout board. (its in all my machines). I'm wondering if i need to combine the 5v from the drive with the 5v on the board?
-
No, don't connect the 5v unless your breakout requires 5v power.
Can you export the drive setup and attach the uxf so I can have a look?
You will probably have to zip it so that the forum will accept.
-
here is my turn xml and the exported file from the allen bradley
-
sorry, here is the mach xml file
-
I don't see anything obviously wrong in either of the files.
I got the E21 error on my Beaver Mill one time when I used a certain plugin for the SmoothStepper, Greg had altered something for me but he must have also done something to the Step/Dir pulses as it would constantly give me problems, he redid the plugin and all was fine again so there was an issue somewhere with the timing.
It sounds like you may also be getting the same issue but whether it is the board outputting erroneous pulse signals or wiring I am not sure. With you only getting one direction that could be the way you have the axis set up, rotary axes can be set like that if I recall and if you wish them to go the opposite way then you need to command a negative move.
-
I was really hoping you would find something simple. I have messed with this thing for days. My Teco ac servos are working fine on this machine. I don't have a smooth stepper. I don't think i even have any plugins enables that could affect this. Grrrr. Might be time for a plan b.
-
Have you tried setting it up as a spindle or as a linear C axis?
-
I have not tried setting as a spindle yet. i think i can get that working. I did try as a linear axis and that made no difference. I reset the eprom to factory settings and started over. That didn't make any difference either. Did a little more tuning and it is more responsive now but still only goes in one direction from mach3 even though i can see the dir. pin change from 5v to low. From Ultraware, it will do everything it is supposed to, for, rev, and rpm. This just keeps coming back to poor pulse stream from mach3. I finally got the oscilloscope function to work. Yeah the pulses don't resemble a square wave at all. Maybe you have good luck with these because of the smooth stepper outputting cleaner pulses?
-
I swapped with the y axis and it does the same thing. Only goes in 1 direction. Looking at the oscilloscope, going forward has nice distinct pulses although not close to a square wave. Going in reverse is a mess of jumbled pulses. Its no wonder it doesn't move, it doesn't know which way to go. The pulses are erratic and unevenly timed/spaced. Maybe it is just damaged inputs on the drive. My other drives work fine with the pulses that mach3 and the B.O.B are putting out.
-
It does sound like there may be an issue with the drive :(
One thing you could try if you wanted to test it out is take the encoder output from one of your other drives and input them to that one. You could then set it up as a slave drive and encoder follower and see how it performs. If it works ok then it can't be the inputs that are screwed as they are the same inputs used for Step/Dir.
-
Yes, it was the drive. Bought another one for $300 on ebay, threw it in there today and the spindle rotates forward and reverse as a-axis just fine. Now to figure out the spindle setup. I'm using parallel ports so i am greatly limited in pulse frequency. I'm thinking maybe some electronic gearing to lower the required steps for a given rpm. Or maybe using the command +/- inputs. I wish there was the option of naming an input pin for forward and reverse and 0-10v for speed.
-
You could do a workaround for the analogue voltage, I have done it in the past.What I had was a double pole double throw relay and used that to invert the 0-10v so that you were feeding the drive with -10 to 0 or 0 to + 10 depending on the way the spindle was to rotate. I used the M3 to enable another relay to put the voltage to the drive and the M4 operated it as well but also swapped the original relay.
Having said that Step/Dir would be a better option in my opinion, it would give you much better control and would even allow you to rigid tap if you wanted by using the SwapAxis() in Mach and changing it to a rotational axis.
Your motor is 4000rpm and if I recall your kernel in Mach is 60KHz, that would mean the max RPM without electronic gearing would be 450rpm so to get full rpm you would have to set a 1:9 for the electronic gearing.
Hood
-
I don't see gearing as an option on the digital inputs though.
-
If you look at post 19 I attached some pics, the bottom one is where you would enter the gearing against Preset 0.
-
Yes, but how do i switch between presets. Its not an option in the digital input drop down menus.
-
There are Presets in the Inputs, see pic, however you don't need them if all you are wanting to do is set one electronic gearing only.
Preset 0 means you do not have to have any inputs configured as that will be the default scaling, ie 1:9, 1 master pulse = 9 drive pulses.
-
Ahh. got it. With 360 steps in mach3 and the drive set to 1:87, i get the rpm i want and well within mach3 pulse limits. Next problem: At 4000rpm i have aver 20,000 counts of error. Which is fine but occasionally it gets an e19 alarm because of it. I can't find where to increase the amount of allowable error.
-
I wouldn't increase the error myself, be better tuning it better.
If however you want to change the allowable error then it is from the fault page if I recall, think there is an option top right of that page. Will check to make sure though as the memory s not as good as it once was :D
Hood
-
Was almost right, it is on the Fault page itself rather than a sub page.
Hood
-
One more question. My other drives enable with a ground signal but the ultra3000 wants 24v. I don't see any way to make it enable active low. I can't separate this enable from the rest of the drives, they are all enables from the same signal . If i can't change it, i'll just do a relay i guess.
-
Nope, afraid not, sounds like your other drives must be Chinese or at least Asian. They seem to prefer the Inputs that way for some reason.
Hood
-
Teco, so yes they are chinese. I've had great luck with them though. They are in all my machines. So why does M5 disable my output 1, (spindle drive enable). very strange.
-
Teco, so yes they are chinese. I've had great luck with them though. They are in all my machines.
Not saying they are bad drives in fact I have used Samsung Fara drives and they are excellent. Just saying it seems to be the preferred method of the Asians for their inputs where as in the West they tend to have sinking inputs.
So why does M5 disable my output 1, (spindle drive enable). very strange.
Ah now that is a nice wee quirk.
You have M3 spindle relay set to OutPut 1. Now I hear you say, yes but I have the spindle relays disabled. That is the quirk, for some reason even though you have it disabled it will still switch it off with M5 if it is on. Easy way to solve the problem, enable the relays, change the output to one that is not used then disable them again :)
Hood