Machsupport Forum
Mach Discussion => General Mach Discussion => Topic started by: evanse89 on October 07, 2013, 05:32:44 PM
-
I recently saved a pair of Pratt & Whitney Drill Mates from the scrap heap hoping that they would be good candidates for a retrofit project. The biggest obstacle I see with these machines is retaining use of the ATC, which is quite a unique design-very similar to a Fanuc Robodrill (http://www.youtube.com/watch?v=LkQZclaOSz8 first 0.07 shows tool change) or the Brother tapping machines, only completely mechanical-I am on here to ask if anyone has any advice on how to program this ATC.
The tools are held on a turret that rides on the front of the spindle, and rotate around the Y-Axis. The tool never leaves the turret on these machines, but is held loosely in its slot by a C-Clip and a bearing.
When it is time to perform a tool change the machine rises to its max working Z height, then it uses some sort of cam actuated by an additional Z-Axis movement upwards to swing the turret forward/away from the spindle, as it does this the cam releases the tool holder shank from the spindle, and the holder rests in the turret. With the turret positioned out away from the spindle it can rotate to the next tool. As Z lowers the turrent swings backwards/towards the spindle and the cam engages the holder, then it drops back down to the working Z height and off it goes.
I have never done an ATC in Mach3 so starting from the beginning my questions are:
Do I need to run the turret as my A-Axis?
Do I need to set soft limits to control where the Z-Axis goes to begin the tool change?
How the heck do I program all of that automatically into Mach3?!
I have seen some of the tutorial but I'm kind of intimidated...
Any help would be MUCH appreciated!
-
Unless the turret uses a stepper or servo motor then no you will not set it as an axis.
No need for soft limits for it, you will just command a position to move to, however softlimits are always a wise thing to have on any machine. If the axis has to move beyond soft limits to do its changing then your M6 macro can disable/re-enable softlimits as required.
All done in the M6start.m1s macro, how exactly will depend on the exact steps/sequence required, often a PLC is a good way to do this as it handles things externally and the macro only needs to signal to the PLC and wait for a signal back. All depends on exactly what is required and which is easier for you personally , VB or Ladder.
It will be intimidating but like everything else when you know the exact sequence required, know the I/O you have and what order etc then the battle is half won.
Hood
-
There is a small servo on the machine now that controls the turret location. Since all of the motors on these machines are analog servos I will probably be replacing the axis drives with some Nema 42 steppers and a 23 for the turret on one machine to test it.
I was wondering about limits just to prevent the machine from accidentally going to high and dropping a tool if it somehow skipped a step or through operator error-if I can set limits that will stop the machine but can turn off during the tool change macro that would be perfect, I just have no idea how to do that-my Mach3 experience is pretty limited so far, and I have exactly zero experience using a PLC...
-
Personally I would look at one of the analogue capable controllers that work with Mach and use servos with analogue Amps/Drives.
There are a few and I have used the CSMIO/IP-A and think its the dogs dangly bits.
The other two I know of are the DSPMC and the Kflop with Kanalog.
Hood
-
I didn't even think to see if I could find a way to run the older servos aside from fixing the existing Fanuc control equipment...
All of my experience so far has been with steppers and the servo tuning is a bit intimidating, also the cost of these kits seems a bit prohibitive; do you have an estimate for the cost to set up the CSMIO/IP-A of Kanalog systems (for 3 axis plus the ATC)? They both appear to be in the same price range.
Keeping the existing motors would be awesome if I could make it work!
-
I am not really sure of the total cost as with the CSMIO you would likely want the MPG module so that you could use for panel buttons and the MPG itsel. See screenshot of my panel.
The CSMIO is 24v for its I/O so is easy to integrate to an industrial machine.
Think the Kflop is 3.3v but the kanalog may be 24v, not sure on that, if not you will likely need extras to interface..
One word of warning, some Fanuc amps use funny signals for the control so make sure yours can accept +/- 10v input if you go down that road.
Hood
-
Think the Kflop is 3.3v but the kanalog may be 24v, not sure on that, if not you will likely need extras to interface..
KFlop is a mix of 3.3V and 5V tolerant (i.e. will only source 3.3V, but can withstand 5V input). The Kanalog can handle upto 24V through the optos, or 80V through the FET relay drivers, however the biggest drawback I find is the Kanalog only has 8 opto inputs. It does have other inputs available via a pin header, however they're low voltage. It also has eight 0-10Vanalog inputs, which you could probably use for digital inputs with a little bit code.
-
Could you help me piece together a simple kit to run this so I can get a better estime of the cost for a 3 axis (with potential a 4th) plus ATC? The servo setup sounds ideal but it also sounds a lot more technical than the stepper systems I am used to so I dont know where I need servo specific parts like drivers or extra wires/connectors for encoders, etc...
What servo drives can I use with these systems-can I run anything or does the analog encoder require something trick? I think I should be able to hack my way through the programming, the electrical stuff is what really gets me overwhelmed at the moment.
I will post the Fanuc motor specs as soon as I am able to.
-
If going the CSMIO route you will have the CSMIO/IP-A itself and I would recommend also the MPG module, not the pendant, just the module, unless you wish to have the pendant of course ;). You may require an extra I/O module as well depending on the amount you will require.
The only real advice I can give is study what you have, try and work out what connections you will need and that will help you decide.
The analogue output controllers for Mach can use any analogue controlled amplifiers, in other words if your amplifier/drive gets its command from a +/- 10v signal they will work with these controllers.
All of these controllers will also require that you can feed the encoders back to them so either the encoders on the motor need to be 5v ttl signals or that any simulated/pass through signals from your drive are that.
To explain the last poart I will give an example. Some of my machines use Sine/Cosine encoders on the motors, these are not suitable for the controller. These signals however are read by the servo drive for its own use and then the servo drive will condition these signals and pass them out as square wave signals which are suitable for the controller to use.
Sorry its not a huge amount of help but without knowing the specifics involved its impossible to say what is required.
Hood
-
Spindle Motor:
Fanuc AC Spindle Motor Model 3 Type A068-1003-B 100
Torque cont 3.7 kw
Amp cont 17 A
200 V
1500/6000 rpm
Its square and really heavy so I assume it really is an AC motor and thats not just some random part of the name
X and Y Motors:
DC Servo Motor Model 5M Type A066-0642-B012
Torque cont 5.9 Nm
Amp cont 9 A
151 V
2000 rpm
Z Motor just looks like a scaled version of X and Y I will get the specs on that ASAP, I just pulled those from pictures on my phone...
I can look through the piles of books for wiring diagrams and potentially find info on the encoders. Would it be possible to put new encoders on?
What kind of drivers can I run these servos off? Like i said I'm 100% new to servos, and to anything of this scale....
My ultimate goal is just to get these machines running well enough that I can cover my expenses, take my girlfriend out to dinner for putting up with this project, and buy a case of beer for the guy who has been storing these for me. If I can change tools, do some light cuts on steel plate, and cut aluminum at 40ipm I would be super excited-I'd honestly be really excited to see it work at all.
-
Do you have the amplifiers/drives that were there originally? If not then the spindle may be a problem depending on the whether its encoder or resolver and if encoder what type.
The axis motors being DC should be easier, there are plenty of DC drives available from low end right up to industrial quality. My friend actually has some mid range type drives for sale but sadly he is in the UK so probably not an option.
If you can find info on the spindle encoder then I may be able to suggest a drive.
Hood
-
I have the original axis and spindle drives for the both machines, as well as a couple of spares but from what I remember each machine had one bad axis drive-I just don't know which one is which. My initial plan was to find out which (if any) spare drives worked and fix whatever other issues arose but the machine is a tape reader and runs on an older control which I have no experience with my plan switched to retrofit, which is where this thread started.
If I can use the stock servo motors that would be great, it (sort of) saves money, and increases the machines potential. I'm just skeptical about using the old boards because I don't really know which (if any) boards are actually functional anymore.
-
But both spindle drives should be functional-there was never any mention of them being broken.
Now that I think about it there may have been a 3rd machine because I do have an entire extra set of motors as well as the spare boards...
-
That is good that you have the spindle drives as they would have been the hardest to match up. AC servo motors are often best matched with the drives they were meant to work with, some drives can be set up for different motors but a lot of them you can only use specific motors as there are not any options to set custom motor parameters in them.
DC motors however are easy as almost any DC drive will work as long as it is capable of delivering the current and voltages that the motor wants.
Encoders may have to be swapped but usually DC motors just have 5v square wave encoders so hopefully that will be what yours are.
Hood
-
Is there an easy way to check they type of encoder? I have a lot of documentation for these machines that I could sort through.
Just want to be clear, if these are the square wave encoders I should be able to use the spindle drive, all stock motors, and pretty much any 9A/150+/-V DC servo driver and 3/4/5/6(?) axis breakout board, and either the DSPMC, KAnalog/KFlop or CSMIO/IP-A board and run this off of Mach3?
Any recommendations on DC Servo drivers/breakout boards?
-
The encoder or resolver on the spindle motor wont matter as you will be pairing it with the drive. The only issue you would have is if you wanted spindle speed or rigid tapping, if the encoder was not suitable for the controller (csmio/dspmc etc) then you would have to fit another encoder somewhere els on the spindle.
There are more Step/Dir drives for DC available in the low to mid range market than there are analogue but I think AMC make some decent ones.
Hood
-
Z Axis:
DC Servo Motor Model 10M Type A06B-0651-B212
Torque cont 12 Nm
Amp cont 12 A
165 V
1500 rpm
Spindle encoder:
Optical Shaft Encoder
DC 5V
1024 counts/turn
-
Sounds like the spindle encoder will be ttl but there is a possibility it is not. If you could find a part number on it you may be able to find a data sheet.
Regarding the axis motors, as said basically any DC drive should do as long as its capable of the voltage/current the motors require. There are two options for you however, do you go the cheaper way and get the more hobby orientated drives which will be Step/Dir input or do you go a bit more industrial style and get analogue capable, personally I would look for the second option purely because I like the fact that the analogue controllers always know where they are due to encoder feedback to them.
Suppose there is a third option and that is to look out for second hand DC drives, they often appear on eBay and you can get some very good deals as most people are wanting AC drives nowadays.
Hood
-
Saving money by not having to buy new motors and knowing that I can retain the positioning feedback make me more willing to put up some extra cash towards better drives (A-M-C?).
This is all a bit more technical than I am used to so excuse me if this question sounds silly but what is the difference between the KFlop/KAnalog system and the CSMIO/IP-A system?
I have seen some info on threads about KFlop and they seem to have excellent technical support, but haven't seen much about CSMIO. They are both in the same price range as far as I can tell so I'm not really leaning in either direction yet.
-
http://www.ebay.com/itm/MIYANO-BNC-34-CNC-TAMAGAWA-OPTICAL-SHAFT-ENCODER-TS1508-N91-TSI508N91-/161009529900?pt=LH_DefaultDomain_0&hash=item257cea742c (http://www.ebay.com/itm/MIYANO-BNC-34-CNC-TAMAGAWA-OPTICAL-SHAFT-ENCODER-TS1508-N91-TSI508N91-/161009529900?pt=LH_DefaultDomain_0&hash=item257cea742c)
This is the type of encoder I am dealing with. I can't find any data sheets online so I will have to sort through the heap of documentation that came with the machines.
-
As far as I am concerned the biggest drawback to the Kflop is that to do much more than the standard setup you really need to know C or maybe C++ or some other similar programming language.
The CSMIO has most options already done and is well thought out and you can use VB script to access I/O that is non-standard to Mach.
Hood
-
I found a thread on CNCZone from earlier this year where someone (local) is using the CSMIO system, so I should be able to use that as a reference.
I also discovered after further investigation that the ATC turret is not controlled by a 4th motor, but there is actually a gear tucked away on the spindle that engages with one on in the turret, and spindle orientation will rotate the turret to the correct position. That reduces the need for a 4th motor and drive but now it is absolutely necessary to use the existing spindle drive, or find a replacement should they be broken. A PLC should be able to control that and the Z position correct?
-
Is there an easy way to verify that these motors and/or drivers work before I go buying drives? Like I said I have nearly enough to build a 3rd machine out of this I'm just not sure what (if any) works and what doesn't.
-
There are quite a lot of people using the various CSMIO's just not many people posting. I have done a couple of build threads of my machines that I use the CSMIO's on. There is a CS-Lab section on this forum, just in case you didnt see it.
I am not sure if you will even need a PLC, the CSMIO has a lot of I/O and with a wee bit of work you may be able to do it all within the M6 macro.
If you can hook up power to the drives and make all connections needed (enables etc) then you can test things with a battery box. A battery box is basically a 9v battery with a pot so you can alter the voltage. I have made a couple up before and I was slightly more elaborate in that I had 2 9v batteries and thus I could get -9v through to +9v when turning the pot.
What you use the battery box for is to simulate the +/- 10v input to the drive, this will make the motor turn and the amount of voltas and polarity of the voltage will determine how fast and which direction.
Hood
-
KFlop is the main controller board with only low voltage inputs/outputs, and the Kanalog is essentially an add-on board that provides analog inputs/outputs, aswell as relay drivers (8 x 80Vmax outputs) and 8 each of 24V tolerant inputs/outputs.
The main difference between the CSMIO's and KFlop, is the CSMIO is essentially setup through Mach, whereas the KFlop has it's own configuration (and CNC) software.
For basic operation KFlop setup involves configuring axis via it's own dedicated software, then copying the generated C code into a init.c file which is then linked to Mach. Anything that relies on critical timing (i.e. such as homing where motion has to be controlled quickly on a switch change) or custom functions, have to be coded in C. There are lots of examples included with the software (you can download it from the dynomotion website and have a look at it), and Tom Kerekes provides brilliant support for any queries.
However, if you're not comfortable dealing with C code, it's maybe not the bext option for you.
As for testing drives, a simple method Tom Kerekes recommends, is a 1.5V battery combined with a 1k resistor. Connect them in series to the drive analog input, and see if the motor spins. The low voltage means things shouldn't suddenly take of, and the combination with the 1k resistor should minimise the risk of damaging anything by connecting things wrong.
-
CSMIO/IP-A has a threading module, I assume that is for rigid tapping, but that wouldn't be necessary to control spindle positioning for the ATC would it? The spindle is an AC servo so the drive should be able to take care of that.
Would it be necessary to buy any additional I/O modules for this set up? It can handle the 6 axes, and the only extra stuff I would be working with would be some additional limit switches for the ATC.
I'm also just wondering what kind of additional components this would require, steppers are easy and you just need a bunch of wire from the motor to the driver and then to the breakout board, there aren't really specialized plugs, PSU, etc...
-
What I would like to do would be to go bare bones on this one. Get the minimum number of components necessary, maybe some of the lower end drivers, and see how this retrofit compares to the other conversions. If it works great then Ill order the full package for the second machine, and whatever pieces I felt were missing for the first one.
I should be able to get away with just the CSMIO/IP-A controller (and Threading module if necessary for the ATC) and some of the less expensive AMC drives, I have a PC and Mach3 already, so then I just need motor/encoder cables (comes with the drive maybe?), wires for limit switches, ethernet, etc... what kind of Power Supply do the servo drives run off of? I have 2x151V, 1x165V and 1x200V, plus whatever CSMIO needs...
-
Yes you do need the encoder module for rigid tapping, in reality with the IP-A it should not be required but I think the reason I was given at the time was they had standardised the plugins and thus it was needed. Maybe in the future (Mach4) it will not be required but ......
Talking of rigid tapping, it is very nice indeed with the CSMIO, previously it was good but you had a few things you needed in addition to the M84 line, now all you need is the start position and the M84 line. Here is a video I did just the last day showing rigid tapping.
http://www.youtube.com/watch?v=WUO3pSb0JwQ
Regarding spindle orientation, that can be handled in the plugin, there is a section there that allows you to stop the spindle at any position, not sure how well it works as I have not tried as my toolchange requires that I am spinning at about 200rpm rather than having to stop.
You wont get encoder cables etc with the drives, you will have to make them up yourself, I would imagine you will have them already if you have the rest of the stuff fro the machines. You may have to cut off an end and wire to whatever plug the drive requires.
Power supply will depend on the drive, for example all my servo drives take mains electricity (240v single or 415v three phase, depending on the drive) Lower end drives tend to require you have a power supply and the power supply will usually have to be DC of approx the voltage you wish to output to the motors.
CSMIO needs a small 24v DC supply.
Hood
-
I feel really confident that the CSMIO is the way to go after I saw a machine with it installed running over the weekend. I will hopefully be able to try and test some motor drives today to see what is and isn't functional-at which point I will need to determine which drives I need to order.
If the drive is not connected to a motor is there an easy way to see if it is functional? Same goes for the motor not being connected to the drive? You'll have to forgive me fo asking these questions as I am by no means an electrician, its all still black magic to me at this point...
-
Was that Mikes Shinx router?
As for testing drives without being connected to motors, then not really and even if you can test some functions it will depend on the type/make of drive as to how far you can test.
Motors are dependant on type, AC servos require a drive but DC Brushed are easy to test, just connect to a car battery and they should rotate, reverse the polarity and they will go the other way. How fast will depend on the voltage constant but that is likely on the label, it will say so many volts per 1000RPM.
Hood
-
Yes I was really impressed with how well it functioned, it looked every bit like it was designed to use all of those components, very much UNLIKE my gantry project that used stepper kits. Feedback, acceleration without skipping, and running at reasonable speeds for a machine of that size, it was definitely not like my machine at all...
When I am selecting motor drives what do I need to pay attention to? The A-M-C drives are definitely a bit more spendy than I was hoping for on my test machine but if there is nothing comparable I might just have to bit the bullet.
-
If you are wanting to use the IP-A then you will need a drive that accepts analogue signals. I dont think there are any of the lower end drives normally associated with Mach that can accept that, they are all Step/Dir as far as I am aware.
If you keep a watch out on eBay you will likely find some drives suitable, what you need to look for are ones that have the voltage and current capacity of the motors you want to use.
I am not sure what kind of cost the AMC drives are as I have never bought any but if they are relatively expensive it may be better looking at complete AC Servo motors and drives. They are not that cheap either but you can often pick up nice ones for a reasonable price on eBay and often there are Samsung Fara drives and motors in the 750W to 1Kw range on eBay from Korea.
Hood
-
I will check eBay, I am also talking to someone at Zapp Automation.
I contacted MachMotion just to see what a complete kit would run and they said that the Fanuc motors might use some sort of proprietary encoder that might not interface with aftermarket drivers, is that true? And if it was couldn't I just replace the encoders-the ones I have are old school anyways and with the updated processing power in the CSMIO I should be able to run higher resolution encoders anyways.
-
Are you in the UK?
Some Fanucs do indeed have propitiatory encoders but I think that is more modern AC servos but couldnt be 100% certain.
The motors I have on the Chiron are Sine Cosine encoders and I have the output from the drive interpolated to, if I recall, 35,000 pulses per rev and that is just a walk in the park for the CSMIO.
Hood
-
I have new concerns after reviewing the wiring diagrams. The motor encoders go straight from the motor to the "master PCB," I was under the impression that the drives usually handled the encoder feedback, but if these motors need to communicate with the PCB I have pretty limited options:
Repair the stock control PCB and drives, downside is I'm unsure what does and doesn't work and this looks like it would be a nightmare.
Replace the motor drives (with A-M-C) at $400+/- per axis, the easiest method if I can just rip out the stock garbage and plop new drives in, downside to this is that once the motors croak I'm stuck with drives that run motors I haven't had much luck finding replacements for.
Replace the motors and drives (Zapp Automation 1.3kW) at $1000+/- per axis, most expensive upfront and is pretty much my original plan.
-
Can I just put new encoders on the existing motors, and the replacement drives will handle whatever feedback that provides?
The different AMC drives support different types of feedback depending on the model, Tachometer, Incremental, Halls or +/-10VDC Position.
-
That is normal regarding encoders going to the control on DC brushed setups.
They will likely have a tach for speed reference and that is fed back to the amplifier (drive) . The encoders are used by the control (Fanuc in your case ) so that they know the position and can send +/-10v signals to the amps to move the motors. So if changing to CSMIO/IP-A you would have a similar setup but the encoders would hook to the CSMIO.
There are some Samsung 1.5Kw motors and drives on eBay at the moment, http://www.ebay.co.uk/itm/ALLEN-BRADLEY-1-5KW-SERVO-PACK-CSD3-DRIVER-SAMSUNG-MOTOR-3-AXIS-CNC-ROUTER-/281178207322?pt=LH_DefaultDomain_0&hash=item417786f45a
They are 3 phase input however so maybe you only have single phase?
Hood
Hood
-
I am about to move and the new space will likely be getting a phase converter.
So I should still be able to get away with running the stock drives and the stock servos if need be, as the encoder feedback will go through CSMIO, and the drives pretty much just tell the motor how fast to spin and which direction?
But if I went with AMC drives the encoder would feed into the drive? The cheap one (still $500) says it can run using a tachometer.
-
Yes if you used the stock amps you would just use the encoders for the CSMIO, the Tach on the motors would go to the amps as they do now.
If you went AMC drives then you could choose whether just to get ones with tach feedback or you could choose encoder feedback and then you would take the encoder output from the drives to the CSMIO.
Hood
-
Your motors do have a Tach I assume?
Hood
-
Yes, the documentation I have also mentioned motors with and without resolvers(?) so I don't know what I have in addition to the tachometer.
If the spindle drive is shot would it be possible to replace that set up with a regular 3 phase motor, encoder and VFD? I believe spindle positioning only needs to be accurate within a couple degrees and as the gears mesh it will line up the spindle and ATC turret if that matters.
-
It should be possible to use an Induction motor and VFD with encoder as the CSMIO can orient the spindle and I think it will work with a VFD/encoder setup
Hood
-
I think I might have already asked this but can I just put a new encoder on a motor? I don't know what I have on the Fanuc motors now or if they work, and I'm also wondering for the spindle, if I would need to find a motor with an encoder or if I can just buy a ___HP 3 phase motor and stick an encoder on it. It seems like an induction motor, encoder and VFD would be a lot less expensive than just a replacement AC Servo.
-
Heres what I have to work with at the moment, just FYI. The second machine is blue, both were "rescued" from Boeing.
-
I don't know if it is possible to tell what kind of encoder I have from how many wires there are-also hopefully if they need wires replaced that doesn't mean they're likely toast, because I have 3 spare motors with wires cut and Y on one machine (shown in picture) has some broken connections.
-
It is hard to say for definite as to whether you could replace the encoders with others if these prove unsuitable. On normal DC motors, with analogue amplifiers being used, the encoders are only there to provide feedback to the control, the speed reference is provided to the amplifier from a tacho.
If however you were to replace the amps with other drives then yes you could swap encoders as long as the drives can accept the feedback you will put on. In other words a lot of modern drives just need encoder signals so that they can determine the speed.
So basically it all boils down to what you have and what you keep.
The info above is not really enough to say what you have and I suspect Fanuc code numbers on the encoders will not shine much light on the situation, it is however worth a try, so if you can have a look under the yellow caps and maybe even take a pic and post here.
Not sure what your second last pic is showing as its a bit small.
If swapping out the spindle motor to an induction motor and VFD then yes fitting an encoder is all that is needed. You wouldnt even need the encoder on the motor, you could have it belt driven from the spindle or whatever if that was an easier option. I would however say personally I would try to use the Fanuc motors if at all possible.
Hood
-
Reinspecting the spindle label I found that it says it is an "AC Spindle Motor Model 3," I think that when I was writing down all of the info I might have just assumed it said Servo and not Spindle, whoops...That makes me think that a 220V VFD should be able to replace the spindle drive if need be, is that correct?
Talking with Mmoe about his Shinx and CSMIO he said I could potentially rewire the motor drives (fanuc velocity control unit a06b-6047-h002 and h00(2/3)) to run on single phase power, any insight into how to do that?
-
I ask because I will have 220 in the shop and will be getting a phase converter but it is not the top priority.
Would I need a tach AND some other encoder for the spindle to get positioning and spindle speed control? Also is a Z Axis brake critical, since I'm going to gut the machine I might be putting some sort of counterweight/strut into the column.
-
The spindle motor and drive you have will likely be similar to what you would have with a normal 3 phase induction motor and a VFD. However the motor will likely be dual winding and the torque curves likely much better than that of the normal VFD/Motor.
It may well be possible to connect things up on single phase, your axis motors are DC and good chance your drives themselves actually require DC input and thus your power supply is a separate unit consisting of a transformer/rectifier and some capacitors. You may be able to connect your transformer up differently. One other thing is in the USA you have two live wires for your 220v where in the UK we have 240v single phase which consists of a live and neutral. So if you connected your two lives and then just jumpered one of them over to the third connection it may work fine. Then again I am no electronics or electrical expert so take from this what you will.
Regarding spindle positioning, it is likely the spindle drive you have at the moment has the power to do it internally, if you went standard motor/VFD then you would probably have to use the CSMIO to orientate the spindle, for that you would need an encoder and also the encoder module from CS-Lab. That is a wise thing to get anyway as it will allow you to do rigid tapping.
I think a brake on the Z is likely the best thing, even if you fit a counterbalance it may still be wise to have one. It doesnt have to be on the motor, you could fit it anywhere you can, for example on my Chiron I didnt have a motor with a brake so I retrofitted a brake to the end of the ball screw. I had an old motor I had purchased very cheaply on eBay and it had a brake, so I hacked the motor up and used the brake and the motor case, see pic below.
Hood
-
I asked about the VFD because like the other drives I am skeptical of what works, so I want to know if it would be possible to just replace it wait any other VFD. There is already a rotary encoder run off of a belt to the spindle drive, and I assume that the large box hanging off of the motor is a tachometer or something to monitor spindle RPM. I'm just trying to determine the best way to figure out what does and doesn't work, I have 9 axis drives (5 on machines) , 3 spindle drives (1 on a machine), 6 X/Y motors (4 on machines), 3 Z motors (2 on machines) and 3 spindle motors (2 on machines), but the machines are in storage while I'm relocating.
Like you said earlier, a car battery should be able to turn the axis motor and verify it works easy enough, can I just use a regular 120v wall outlet to test the drive? It wont be the power the motor needs but I'm also not trying to turn it at 2000 RPM...
I am ordering the IP/A along with the ENC and an additional I/O module so I will be able to get spindle positioning working...somehow...
My machines are small (~20" square envelope) so a handheld pendant seems unnecessary, is there some other functionality that comes out of the MPG module? I also have seen wireless USB Mach3 pendants for $160 which is a lot cheaper than the CSMIO package so I feel like I must be missing something.
I like the idea of a brake, at work we have a servo on/off function that automatically triggers the brake (our Z axis mast is also something like 20,000lbs so its kind of necessary...) with a button, or any time we lose power, it would be nice to set it up like that so anytime the machine powers down the brake is automatically triggered, I could easily accomplish this with a relay right?
-
I dont use a hand held pendant but I use the MPG module to connect a MPG (Hand wheel) , feed and spindle override pots, axis select rotary switch and panel buttons etc. I have that mounted in my control panel (see pic) and I just have a long cable that goes back to the electrical cabinet to connect the MPG module to the main CSMIO.
Regarding the 120v, not sure if that could be used or not, I am not really the person to ask as I dont have any formal electrical/electronics training ::)
Most servo drives have a function in them for brakes, the power is sent to the brake as soon as the drive is enabled and taken away when disabled. If your drives do not have that feature then you most likely could set it up via the CSMIO enable function and use a relay as you are thinking.
Hood
-
I have the IP/A, MPG, ENC and 2 I/Os on order, the IP/A is back ordered so I have an extra week to sort out my drives. I'm going to see if I can get the axis drives running on single phase 220, if that works out then I will just get a cheapo 5-10hp phase converter to run while I save up for something nice.
That control console looks super cool-I have to admit after my first experience with Mach3 I never really thought people would/could use it beyond silly hobby machines but this project is looking more and more promising.
-
All of the machines I have done with Mach are a bit more than hobby machines. I dont do mass production on any but all would be capable.
Hood
-
I got a response from CS-Labs about using an induction motor to drive the spindle:
"CSMIO/IP-A can home the spindle only when the spindle is driven by +/- 10V servo drive.
If you have a regular 3-phase induction motor with an encoder then you have to buy a VFD which has spindle homing function with encoder."
Are there VFDs that come with an encoder, or just ones that are compatible with encoders?
If I want to try and run the axis drives on single phase what kind of documentation do I need to find? I have parts lists, operation and service manuals, controller manuals and a folder full of wiring diagrams-from what I can tell at least. What do I need to look for to determine if its feasible to even do this?
-
That is good to know regards spindle orientation and the CSMIO.
Regarding the VFD, what they are meaning is some VFD's can take an encoder input for orienting the spindle, amongst other things, most of the cheaper ones cant but most of the ones on machine tools can or at least have the option to add a card/module for it.
I am not really qualified to say about what you will need to connect up to get an axis drive running from single phase, or indeed if it is possible.
Hood
-
Well hopefully none of this will be needed as the stock drives will work...
How does the VFD work with the IP/A controller, is it still +/-10V signals? Does the analog controller limit they type of VFD that I could run?
-
There are two options for spindle control in the IP-A.
You can use the Spindle Axis option and that will output +/-10v
You can just use one of the normal Analogue outputs for the spindle and that will output 0-10v and you would use relays for for reversal.
Hood
-
Just a quick browse turned up a few VFDs capable of accepting analog inputs with encoders, but they all say the feedback is for better speed control, would I need to find one that has an encoder just for homing?
-
Yes you would have to find one capable of orientation.
Hood
-
Would it be possible to replace the existing spindle driver with one of these VFDs, or if that drive is toast am I at the point where I would need to replace the entire unit? The spindle also already has an external rotary encoder driven by a belt, would it be possible to connect that to CSMIO or the VFD to gain some control?
-
Not sure if the motor would be suitable for a normal VFD, it likely would but you would have to check up on the frequency/voltage etc that it needs.
The encoder should be fine as long as it is 5v square wave.
Hood
-
OK, so the current task is to test the Servos/Amps and figure out how to get them running, Ill deal with the spindle once I know how big of a phase converter I need... I took a good look inside the cabinet yesterday and followed the cables around to the master PCB and to the large transformer. Unfortunately I couldn't tell what cables hooked up where so I don't know what the input voltage is for any of the Amps yet, but I got a part number so hopefully I can find some sort of wiring diagram. If I can get these set up so they don't require an initial 3 phase input before hitting the transformer it will hopefully save me from having to buy a bigger phase converter.
To test an Amp I just need to connect it to the Servo, and hook up a (*********v?) battery to power the motor right? Then put a little battery over the +/-10v to get it turning? Obviously I can't be connected to the controller or anything, because there is no power but is it likely there are any other connections I need to have? I haven't touched any of the wires inside the cabinet so I assume everything else should be connected however it was intended.
-
To test an amp you have it powered and the motor connected then apply a voltage of between -10 and 10v to the analogue inputs.
To test a DC servo motor you disconnect the motor and connect something like a car battery to the power wires of the motor and it will rotate, reversing the polarity on these wires will reverse the motor.
Hood
-
Do I need to have the exact right power for the amp, or can I apply something low to prevent any damage but enough that it will give the motor some juice? I will keep looking through the manuals for the voltage applied by the transformer...
-
Anything up to 10v would be what you are wanting but preferably a variable source as I suggested earlier as that will allow you to see how it behaves through the range of speeds.
A 9v battery connected via a pot will give you 0-9v and then reversing the polarity will give you 0 to -9v.
Hood
-
Sorry I meant to power the Amp/Motor, not to send a signal to the motor.
-
I would think you will need to supply fairly close to the original, whatever that was.
I know they are DC amps but I have no idea if they take AC straight into them or whether they use an external supply?
Hood
-
That is what I was looking for the other day. I am pretty sure that the Amps run back to one of the transformers so I am thinking that they run off DC, I just can't quite tell which wires go where because its all a huge rats nest at the moment. Once the controller arrives and I will start pulling some of the bits and pieces out and hopefully have a better idea of what goes where.
-
It looks like the wires go direct to the transformer, if thats correct then it will be AC going into the drives. If it was DC then there would be at the very least a rectifier and most likely some big capacitors.
So assuming it is AC then you will likely have to feed very close to what meant to be going in which looks to be 190v?
Hood
-
Its been a while, hopefully someone will still be able to give me some more advice.
I ordered up replacement AC servo axis motors and drivers that will run off of the CSMIO IP/A controller, now I am trying to source a replacement for the antiquated spindle motor...
I believe that the last time I posted information about the spindle motor I may have missed a few details, but here is what I have, copied straight off the label:
FANUC AC Spindle Motor Model 3 (I do not believe that this motor is a servo any more, but is actually an induction motor)
RATING CONT./30MIN 3.7/5.5 kw
AMP (~) CONT./30MIN 17/22 A
VOLT (~) 200 V
RPM 1500/6000 rpm
POLES 4 P
INSULATION CLASS F
PHASE 3
AMB. TEMP 40 C
My question is, what kind of motor and driver can replace this? With the IP/A I assume I will need +/-10v analog control, and to use the ENC module it says I need an incremental encoder, are there any other features I should look for besides matching voltage, rpm, and power rating? I know you have experience with the IP/A on a VMC Hood, do you have anything to add?
A replacement Baldor motor with these specs and matching driver (from Fadal) will be around $6k, I also found one from an automation company we use at work for closer to $10k... I'm hoping to keep this in the <$2k range... The best thing I did find was that CNCmakers.com had AC spindle servos, with incremental encoders and analog speed commands.
Thanks all!
-
Cant recall where we were, do you have 3 phase power? If so is it just 230v 3 Phase?
Is the spindle drive you have dead?
It does look like it is not a servo but rather just an AC spindle motor, it would still be a decent spindle motor if you have the drive for it.
Depending on your answer regards the 3 phase I may be able to point you to some motors and drives on Ebay that may do you. I have not looked recently but there are nearly always ones on USA Ebay.
Hood
-
You got it, 3 phase 230v.
The drives I have a monstrous (about 2'x1.5'x1'), filthy, and in unknown working condition, but I could do a car battery test on the motors and verify their working condition. The encoder is also an issue because it is some wonky belt driven unit separate from the motor, and I am quite sure that they cannot be salvaged.
-
If I could keep the motors that would be ideal, but I assume that without the drive and functional encoder I could be SOL.
-
Thats a shame that its just 230v 3 phase as it makes it harder to find a servo that will be similar specs to the one you have.
You wont be able to test with a car battery as it is an AC motor.
Will have a search of USA Ebay and see if I can find anything that may be suitable for you but think the RPM will be the issue.
Hood
-
That was what I found when I called around, a motor capable of that wide of an RPM range gets pretty expensive.
The AC servo motors from china look physically identical unlike most of the induction motors I have seen, and have nearly the same specs as I found on my motor, but the information they sent me and the information on their website are conflicting, one says 220v 3 phase and the other shows 380v 3 phase...
-
I would say that is a dual winding AC spindle motor and not a servo. They are physically much bigger than a comparable servo.
Does your spindle fit via belt drive or is it direct, if belt then it may be possible to get a 3000rpm servo and gear 2:1
Hood
-
Just looking through the various pictures and I found a detail of the spindle driver "AC SPINDLE SERVO UNIT TYPE A06B-6044-H008" and the motor is "TYPE A06B-1003-B100"
I will look into those and see if it reveals anything else...
-
And of course to make everything super convenient, I have 2 different spindle controllers...
-
Scratch that, I have 3 different ones. I see the A06B-6044-H008 on fleaBay, but I'm skeptical at best about buying a used 20+ year old driver for $1000+
-
Ok if it is an AC servo then it may be possible to use a different drive although you may need a different encoder as I think Fanuc use proprietary ones. You mentioned the encoder was via a belt but if its an AC servo then I think there will be an enoder on the motor itself and the one on the belt will just be for the control for spindle position.
Hood
-
There is a "Optical Shaft Encoder, DC 5V, 1024 counts/turn" that is driven by a belt, that is the one I am assuming is garbage. When I get home I can check the wires coming out of the motor itself and see if there is any sign of another encoder. I wil also try looking around for 2500-3000rpm 7.7kW servos, I might be able to get one thrown onto my axis motor order before that ships.
I forgot to mention the belt reduction to the distributors I spoke with before, so they might be able to find me something for less than $6k as well...
-
How many power wires go to your spindle motor? I am still thinking it is an AC two winding spindle motor rather than a servo.
Why do you think the encoder is garbage?
Hood
-
There are 5 wire bundles, 1 encoder, 1 encoder power, 1 spindle power, 1 spindle encoder, and 1 spindle ground(?) wire.
I attached pictures of everything relevant.
-
And drive stuff
-
I am even more convinced that that is an induction motor. Not what you would call an ordinary induction motor but rather the type that is often used for spindles, ie usually dual windings so you can get the dual speed ranges by changing star/delta.
I have a similar style motor here, a Yaskawa and its drive, basically a very high end VFD with positional control.
Hood
-
Do you think it would be possible/worth while to find a replacement drive and mount up a new encoder? As I understand it the drive doesnt necessarily need to have positional control, but the encoder will feed to the CSMIO ENC module-I don't know if that makes it easier to find one. Also since its an AC motor the drive has to be a pretty specific match for the motor corect?
-
Also, like I said before I don't lnow if the drive is functional or not, if there was some way to test it without having to rebuild everything I could give that a shot. I think part of my hesitation is just how antiquated that drive is, it seems like it would be a huge pain to try and get that to interact with the controller.
-
I checked eBay, I can find replacement motors for under $1k, but the drives are $1.5k+
If there were alternative drives that would be perfect, I can save myself from buying a new motor/scrapping my old one and if my motor ended up being broken, a new one isn't going to cost me an arm and a leg.
I would be less inclined to buy a replacement of the same drive though, because then I am constrained to buying replacement Fanuc motors and drives to fix any issues.
-
I think the big problem is Fanuc stuff is usually more expensive than other makes and also documentation is much harder to come by, so dont really know much about how the motor and drive works. If it is an induction motor (I think it is ) then a VFD should be able to drive it, the issue would be you are likely wanting positioning and tight speed control and that means top end VFD. I know Control Techniques make all sorts of drives, maybe the UniDrive range would work but you would have to contact them and ask.
If you had 415v 3 phase you likely could have picked up some Allen Bradley HV drives and MPL motors, that is what I have on most of my machines, the Beaver Mill has a 5Kw/4000rpm motor , the Chiron a 9Kw 3800rpm motor and the Computurn Lathe a 12.5Kw 2000rpm motor, all are MPL B series motors and use DSD-HV drives.
If you can hook up some power to the Fanuc drives I think you should be able to test relatively easily, there should only be a few signals required, enable being one.
Hood
-
I can look into getting 440v, but I don't really see that happening anytime soon...
When they list a rated RPM for the motor is that the max RPM? Mine says 1500/6000 so would it be rated at 6000? Like you said before I could go for a belt reduction to increate my spindle speed, it looks like the MPL-A 230v motors come in about the right ratio, 5.5kW and 4000RPM.
Those Allen Bradley motors look like the AC spindle servo from China I mentioned before. I asked them about the 220 vs 380 and they said that they can rewire them for US customers with 220, so aside from finding a drive for less than $1500 I really think that may be my best bang for the buck.
-
It all depends on the motor and drive but usually the rated is the max although I have Berger Lahrs on the Chirons axes and their rated is 4000rpm and Max is 6000rpm.
The 5.5Kw may be too low on torque if you gear to get 6000rpm, suppose it depends on what you are doing.
You can usually pick up the low voltage DSD drives on eBay and they can be used with motors other than Allen Bradley as you can make up custom motor profiles. The only things you need are a suitable encoder on the motor and to know the motors specs.
Hood
-
My motor is only rated at 3.5kW so I would expect it to be close enough...
Those DSD drives are for servos right? Or could I potentially use one with my existing motor? I did notice that Fanuc literature is hard to find so just replacing what I have sounds more and more appealing, since I will no doubt have to set parameters and fix issues down the road.
I will look on eBay some more for a whole new motor/drive unit that is comparable as well. When looking for a motor I obviously need to match RPM and output, but for the IP/A+ENC what additional featureds should I be looking for? It can be open loop, but needs a differential encoder for the ENC, 0-10v speed control, anything else I'm missing?
-
Ah ok I was thinking it was 5.5 for some reason.
Yes the DSD's are for servos.
AC servos can be problematic as quite often you are tied to specific motors with specific drives. The Allen Bradly DDM and DSD drives however are quite versatile, the DSD's more so that the DDM's. You can, as I said earlier, set up a custom motor profile in them and as long as your encoder has hall signals as well as the normal 5v square wave they will be fine. The DSD can actually use encoders without Halls but each time you power the drive it will do a self sense and rotate about 1/4 turn. For an axis that could be problematic but for a spindle it would likely be ok.
If you find something on eBay give me a mail first with the details as I will hopefully be able to tell you whether the motors/drive combo will be compatable.
For the IP-A controller you will want a drive that can accept +/- 10v control signals, most AC servos can do that but there are a few out there that are only Step/Dir. The Allen Bradleys however can do both analogue and position control.
Hood
-
I'm waiting to hear back about the 230v Allen Bradley motors, but if it comes to the point of buying a new spindle servo I think that the GSK ZJY208 3.7 or 5.5 motors will probably be the route I take. They take analog signals, and have tge speed and power I need, and since they have closed loop control I know they should work with the CSMIO ENC module.
-
The GSK as with any other servo drive in speed or torque mode wont actually close the loop, or at least not fully, the position loop is closed within the CSMIO however so not a problem, was just trying to make that clear :)
Hood
-
Right, I meant that since the driver itself can make the loop it should be able to do the same vi the ENC module.
I did talk to CS-Labs about what I need though and they gave me a diagram for AC induction vs AC servo spindle control that showed encoder feedback going to both the ENC and the driver, I thought (like you just mentioned) that the feedback just goes to the ENC module, but according to this the loop goes to both? Or does one control speed and the other is position?
In any case those motors should do the job right? The 3.7kw specs are right on with my existing motor, and bumping up to the 5.5kW isn't much more money, so maybe I could turn this from a drill and tap machine to a more capable mill.
-
Meant to have this attached:
-
If you are using a servo for the spindle with the CSMIO/IP-A you need to feed the encoder back to the IP-A so that it can close the loop to the spindle. In addition you also need to have feedback going to the Enc module if you want rigid tapping. You can do that with a separate encoder as in the diagram or you can also just feed the encoder to both the IP-A and the Enc. That is the way I have it on the Chiron and it works fine. Only thing that may be an issue is if you have the spindle geared, in that case it may be required that you have a separate encoder for the spindle, I will find that out soon as I have plans to increase my spindle speed from 3800rpm to 8500rpm using a belt and pulleys, just need time to get it done and doesnt seem to be much of that for me at the moment :D
Hood
-
My spindle is belt driven, but I am not sure that there is actually any gear reduction, I will have to check. Would it be better/would it matter where the integrated encoder feedback goes, vs the second encoder? I should be able to buy whatever kind of encoder I need off of McMaster-Carr or some industrial supply store right?
-
You will need one encoder feedback going to the CSMIO and the other going to the Enc module. As said you can likely just do as I have done and feed both with the same encoder signals.
Hood
-
My axis motors arrived earlier this week. As we just discussed, the positioning loop goes CSMIO->Driver->Motor/Encoder->CSMIO, my motors have all of the cables and plugs to connect the motor/encoder back to the driver, now I assume I could just lop off the connector and plug the wires into the CSMIO, but there are A LOT of wires, do all of these return to the CSMIO?
-
see pic, you have encoder outputs from the drive, thats what you feed to CSMIO.
Hood
-
The I/O under the CN2 label (left side) are all supposed to connect to the CSMIO anyways so that works, I just have to put those into the encoder A B and I +/- inputs?
-
Yes that is correct.
Hood
-
There is a note about some commands being for position control mode vs analog speed control mode. Are those to differntiate between digital and analog control, so I will not use the position mode I/O?
-
Not sure which you are talking about but most likely you will not use any position commands.
Hood
-
On the next page it has details about each pin. Some have a P or S mode which is for 'Position' or 'analog Speed.'
-
Yes ones marked P only are no use on an analogue command setup.
Hood
-
I get the Encoder Feedback ABZ+/- on the driver go to encoder feedback ABI+/- on the CSMIO. But loking as CSMIO vs the driver I see Coder Aout, Bout, Zout+/- on the driver but see Analog output XYZABC +/-10v and XYZABC ground on the CSMIO, thats 6 pins on the driver per axis, but only 2 on CSMIO. What am I missing there?
-
The drive outputs the ABZ differential signals which you input to the CSMIO, you will have that for each drive.
The Analogue command for the Drive comes from the analogue output of the CSMIO, again there will be one for each drive.
So Drive outputs A+ A- B+ B- I+ I- and they go to the CSMIO for each axis.
The CSMIO outputs +/-10v signal for each axis and they go to the respective drives for each axis.
Hood
-
I guess I dont get what those drive outputs connect to, because for my X axis I have A+, A-, B+, B-, C+ and C- pins. Those are noted as running in P and S modes, whatever that means...
I see where the CSMIO output goes now, the CSMIO has an X+/-10v and X ground pins, which connect to pins 12 and 13 on my driver, those are analog input +/-10v and analog input ground.
Will all of these pins be used (ex. alarms and servo enable, etc...)? Do the rest of them go to the Digital input pins on the CSMIO? All of my other projects have been with stepers and simple breakout boards so I'm sorry for asking what may be some really silly questions....
-
The A+/- etc are the pass through of the encoder signals. The motor has the encoder but the drive requires the encoder to be fed back to it for both position and speed modes. If using position mode, ie Step/Dir then normally you would not require the encoder outputs from the drive as the position loop is handled exclusively by the drive. With Speed mode however the drive still requires the encoder signals for Velocity but the control (CSMIO/IP-A) also requires encoder signals so that it knows where the motor is so it can command the drive with +/- 10v to keep it on track. So you use the encoder outputs from the drive and feed them to the CSMIO. As said it uses them to close the loop and also it updates Machs DROs direct from these encoder outputs so that even if you whack the E-Stop it will know where the axis is or if you have the drives disabled and manually turn the motors Mach will still get the DROs updated from the encoders and thus still know where it is.
Regarding Enables and Alarms etc, you connect them to the Inputs or outputs of the CSMIO, it actually has dedicated options for them. For example you can connect a fault output from the drive to any of the Inputs on the CSMIO and configure it as an axis fault signal in the plugin. If the drive faults the CSMIO will get that signal and halt all motion and also E-Stop Mach.
Hood