Machsupport Forum
Mach Discussion => General Mach Discussion => Topic started by: Davek0974 on February 21, 2018, 06:31:46 AM
-
I'm slowly making my way to an ATC project only BP conversion, got tool length on the knee, will be fitting a BT30 spindle this weekend, got new tooling etc,
Have read many great threads on here about ATC's but can find nothing on spindle orientation - seems with the BT30 having dogs, without this in place the ATC is a non-starter??
Now, I have an encoder on my spindle as it does rigid tapping, my first thought was can Mach read the Index pulse on this? Maybe that would be a way forwards, i just have no idea, the spindle drive is a standard 3Hp VFD driven AC motor the VFD has an "Inching" input option IIRC.
Whats the way forwards here?
-
The way my BTC-1 does it is to inch the spindle and drop a shot pin into a ramped slot and when the shot pin reaches the bottom of the ramp/can't rotate anymore it sends a signal to stop inching and the spindle is orientated properly.
Mike
-
Hello Dave,
i am also stuck on exact the same point on by TonGIL TNV40 retrofit.
i also have CSLab,Encoder on spindle to a ENC module and VFD 0-10V for 0-6000 RPM as spindle drive.
i worked on the spindle orientation for hour's without any succes.
on one of my other machines i have a step/dir driven spindle, on problem there.
so let's see somebody has a good idea, but i think there is no way to get a 0-10V spindle to do exact positioning.
on a other machine i have a lenze VFD witch allows me to do the positiong on the driver.
Thomas
-
The way my BTC-1 does it is to inch the spindle and drop a shot pin into a ramped slot and when the shot pin reaches the bottom of the ramp/can't rotate anymore it sends a signal to stop inching and the spindle is orientated properly.
Mike
that is also a way i was thinking about. any photos ?
-
Yep, this point could be a major go/no-go point as changing controllers etc is out for the present due to funds and the fact that its all new anyway :(
I have see mechanical methods but i dislike those so far - cams, shoot-bolts etc.
There are also many mentions that on a BT30 size spindle the dogs are not needed anyway as the drive is taken by the taper, no idea if that is so or a good idea but it would certainly fix the issue cheaply !
-
Hi Dave,
this is from smurph:
http://www.machsupport.com/forum/index.php/topic,35694.msg248927.html#msg248927 (http://www.machsupport.com/forum/index.php/topic,35694.msg248927.html#msg248927)
The spindle dogs have to engage, a BT spindle taper is to steep to take the cutting torque like a Morse taper. Additionally if the dogs did not
line up the taper would not meet anyway.
Craig
-
Not sure how you will line it up accurately enough without mechanical or electrical control capable of the required accuracy.
I have a Hurco with a BT 30 I was hoping to get going one day. I was going to drive the spindle with a servo and use the homing in the drive to orient the spindle for tool changes. Originally it used a shot pin.
-
Hmm, ok a shot-pin or bolt seems it may be the only way then with my motor/drive setup.
I'll need to get the jog function programmed and working off a push-button in the drive first to do some tests it seems.
I knew i should have gone for a servo drive when i rebuilt ;)
-
Ok, before i go too deep on mechanical indexing etc - all of this is a way off yet, lets have a re-visit to my setup..
I have a Bridgeport Knee mill with CNC everything :) It has recently been tagged as a KMVMC (knee mill vmc) :)
It has a 3Hp AC motor running sensorless vector drive with VFD at a 1:1 ratio, max speed is wound up to 125Hz to give me 4000rpm.
Is there a way to calculate a size of AC servo motor that would slot in and give me the same (or better) performance??
It seems most common ones only go to 2500rpm so would need to run at 1:1.6 ratio.
Just as a cost effectiveness exercise.
-
I'm pretty sure the Hurco did a slow run of the spindle then killed the spindle (will free wheel for a little bit). After killing the spindle and some set time it would trigger the shot pin. When the spindle coasted into position so that the shot pin and hole lined up it would bring the spindle to a stop at the correct position. After tool was changed, shot pin would retract. Be sure to think about fail safes if you go this route. Shot pin would ideally be gravity return. I would also want to be able to check that the shot pin is at home position before continuing.
-
The way my BTC-1 does it is to inch the spindle and drop a shot pin into a ramped slot and when the shot pin reaches the bottom of the ramp/can't rotate anymore it sends a signal to stop inching and the spindle is orientated properly.
Mike
that is also a way i was thinking about. any photos ?
Sorry I do not have any photos.
If yiou really need them I could take some this weekend.
Mike
-
I don’t think this problem is a big as you think with sensorless vector VFD, I once built a pump with that kind of VFD and it ran reliably down to 18 rpm with a load, or 1/100 of base speed. My machines also had to stop and line up with rails that plugged in. Taper your drive dogs as much as you can too, so you don’t need to be perfect. Run the motor at the lowest possible speed and tell it to stop on the sensor. Worse that will happen is that it will overshoot a bit. As long as the overshoot is reliably consistent you can adjust sensor position to accomodate.
-
Another thought about this issue. Things like this work much better when they are designed to be extremely tolerant of misalignment rather than being very accurately made. So I htink a better approach here is to have the tool gripper mounted to be spring loaded upward. You press it against the spindle and if the dogs are not lined up the compressing of the spring triggers a limit switch that causes the spindle to rotate until the dogs drop into the slot and the limit switch then indicates that the alignment has been achieved, fasten the tool and retract the gripper. No spindle switch needed.
-
The way my BTC-1 does it is to inch the spindle and drop a shot pin into a ramped slot and when the shot pin reaches the bottom of the ramp/can't rotate anymore it sends a signal to stop inching and the spindle is orientated properly.
Mike
that is also a way i was thinking about. any photos ?
Sorry I do not have any photos.
If yiou really need them I could take some this weekend.
Mike
Mike i would realy appreciate, to get maybe new ideas.
Thank You Thomas
-
All sounding good, lots of options - pictures are always good :)
I gather that as nobody is tackling the AC synchronous -> AC servo swap question, it really is as difficult as i thought ??
:)
-
Hi Dave,
I recently bought a second hand 1.8kW Allen Bradley AC servo and drive for a spindle. I needed a spindle motor with some decent torque at slower speeds for steel.
This servo has turned out great. The main reason I went with a servo is that they are the most powerful/torque-y motor for the size.
Its the first time I've really ever played with a modern AC servo and I am impressed, very, very impressed! They are extremely capable and while it was not even
on the radar, stopping the spindle to line up the drive dogs is trivial with a decent servo. My 'zero window' is 4 counts of an 8000 count encoder ie 10.8 arc min!
Easily good enough to line up surely!
I had always been put off by the cost of servos but after playing with one I wont be looking for any other solution anytime.
Craig
-
Hi Dave,
for the Moment i am 'flirting' with two Solutions:
1st chinese spindle Controller (Manual attached)
i have servo Motors and Controller from this compnay on all my machines,
they are working without any Problems. still waiting for a offer for this Controller.
pro
-gives me all posibilities
-modern,standard equipment
contra
-maybe cost
-Need's to be rewired
-will take some time to get
2nd i call it dogcatcher
-will be a sloted plate (5mm makronlon, because i am getting scared to drive metal against metal)
-driven by a 20mm pneumatic cylinder
-8mm proximity will give the Trigger for drive in
-8mm proximity detect dog
pro
-have all the material laying around
-cheap
contra
-part's to make will also take time
-not sure i will work
i am just brainstormin, all comments wellcome.
Thomas
I dont think this problem is a big as you think with sensorless vector VFD, I once built a pump with that kind of VFD and it ran reliably down to 18 rpm with a load, or 1/100 of base speed. My machines also had to stop and line up with rails that plugged in. Taper your drive dogs as much as you can too, so you dont need to be perfect. Run the motor at the lowest possible speed and tell it to stop on the sensor. Worse that will happen is that it will overshoot a bit. As long as the overshoot is reliably consistent you can adjust sensor position to accomodate.
i have tryed this solution, but was not able to get the stopping Point good enough.
Another thought about this issue. Things like this work much better when they are designed to be extremely tolerant of misalignment rather than being very accurately made. So I htink a better approach here is to have the tool gripper mounted to be spring loaded upward. You press it against the spindle and if the dogs are not lined up the compressing of the spring triggers a limit switch that causes the spindle to rotate until the dogs drop into the slot and the limit switch then indicates that the alignment has been achieved, fasten the tool and retract the gripper. No spindle switch needed.
i have a 'umbrella' toolchanger witch drives in/out, and uses zhe Z-axis for tool Change, so no way.
-
Hi Dave,
I recently bought a second hand 1.8kW Allen Bradley AC servo and drive for a spindle. I needed a spindle motor with some decent torque at slower speeds for steel.
This servo has turned out great. The main reason I went with a servo is that they are the most powerful/torque-y motor for the size.
Its the first time I've really ever played with a modern AC servo and I am impressed, very, very impressed! They are extremely capable and while it was not even
on the radar, stopping the spindle to line up the drive dogs is trivial with a decent servo. My 'zero window' is 4 counts of an 8000 count encoder ie 10.8 arc min!
Easily good enough to line up surely!
I had always been put off by the cost of servos but after playing with one I wont be looking for any other solution anytime.
Craig
How big was the machine, what sort of cuts have you taken with it?
I was thinking to just throw on a 2.6kw servo and see what happens, my experience of servos on the axis drives has shown that they are extremely powerful for the size. I can get a 2.6kw 10Nm cont motor and drive delivered for £500
I would need to gear it up to get 4000 rpm by 1:1.6 as they only do 2500rpm but at top speed it would not be using max power anyway.
-
Hi Dave,
its 6Nm and 3500 rpm in a baby (X,Y,Z = 200,200,200) bed mill. I have taken 6mm deep cuts with a 16mm endmill in steel with cooling but my machine is flexing
and dancing but throwing chips like its going out of fashion, really impressive.
Prior to getting this Allen Bradley unit I bought a 2.6kW 12Nm 3000rpm Vickers unit, again second hand. When it turned up I would say it was new or maybe
re-manufactured. The date on it suggests 1994. Only downside is it has a 4 pole resolver which severely limits my choices of drive. My interest is electronics so I
decided to make a Field Oriented Control drive for it. Still working on it, been a long learning curve!
As it turns out this is a big servo, too big for my machine, I think it would topple over and my machine can't really handle it all anyway. My plan is to use it as a lathe
spindle motor.
If a hobby is measured by what you learn in the pursuit of it then this servo and drive is a very VERY good hobby, I've had to learn all sorts to stay on track!
Craig
-
Hi Dave,
could you not try it direct rather than all the expense and shagging around to gear it?
Of course there is nothing so permanent as a temporary thing.....
May also be possible to push the 2500 rpm. You may have read about 'field weakening', an extremely clever means of manipulating the back EMF constant to
achieve higher rotational speeds for a given input voltage.
I have not seen it mentioned in any servo drive manuals that I've read but you can be sure that my Vickers servo (3000rpm rated, 4000rpm max) drive will have it!
Craig
-
That 16mm tool at 6mm deep in steel - was that profiling or slotting, only asking as there is a big difference in power used.
I have read something on field weakening but nothing further, i would have doubts about pushing these cheap import servos too far.
The setup is HTD belt drive now at 1:1 so going 1:1.6 would just be a new pulley/belt.
-
Hi Dave,
that was profiling, probably 25% step over. My machine starts flexing before the power limit of the servo is reached.
The essential concept of field oriented control is that the applied field is calculated and phased such that it is 90 degrees to the magnetic field of the rotor
and therefore produces the highest torque for a given current. If instead the applied field is at a greater or lesser angle then a fraction of the applied field
will either reinforce or oppose the rotor field. Should the rotor field be weakened in this manner then the motor will produce a smaller back EMF at rated speed
or alternately a greater speed for the same back EMF.
A servo operated in this fashion might produce 10Nm at rated speed of 2500rpm but with field weakening produce 7.5Nm at 3500rpm, ie you can trade torque
for speed, very clever. You may have noted that some industrial machine specifications might list a maximum traverse speed of 36m/min but a max cutting speed
of 20m/min. What they are referring to is that with field weakening its possible to get higher speeds but not retain the torque and thrust to cut.
I'm not sure you aren't doing the cheap imported servos a disservice by suggesting they are not capable. Commonly you'll see a servo with a maximum speed limit
well in excess of its rated speed. Such a servo has the capacity to use field weakening. I suspect that the quality and reliability of the drive will come under pressure
well before the servo wilts.
Craig
-
Interesting stuff.
Yes quite possible the import can take it but I just don't like pushing the envelope on that stuff. :)
Somedays your'e just happy it turns on - Huanyang VFD's for example ;)