Machsupport Forum
Mach Discussion => General Mach Discussion => Topic started by: ViKiNG on August 18, 2017, 04:27:18 AM
-
Hi
I have recently purchased a 1984 Denford Triac. This was marketed as a "desktop" mill. It is larger and heavier than I would put on my desktop. It has 300mm X travel and weighs 250kg
I am in the process of replacing the existing under powered (1/2 Hp) spindle motor with something with more power and torque. The original motor is AC and I am replacing it with a DC servo motor manufactured by callantechnology.com I am looking at the 746W (1Hp) M4-2950 B which uses 0-10VDC for speed control. I will also need a DC Drive - Speed Controller, something like the Parker 507 (http://ph.parker.com/us/en/dc-motor-speed-controller-dc506-507-508-series).
I have a C11 NC board (cnc4pc.com) and a USB SmoothStepper BOB (warp9td.com). I am planning on using either Mach3 or Mach4.
My question is
How does Mach3/4 control the motor speed. I am guessing that Mach3/4 sends 0-10VDC to the DC drive which then sends it on to the motor. I would appreciate some detail of how this works. I have solid mechanical background but limited electrical experience.
I would appreciate some advice on a suitable DC drive for the motor. I have not found one that matches the motor specs which are: 115V, 10A, 0-500RPM. Power input is single phase 230V 15A
Thanks
Keith
-
Hi Keith,
Welcome to the forum.
For spindle speed control Mach3 sends a 0% to 100% PWM signal at TTL level (5V or 3.3V peak). The PWM duty cycle is controlled by the Gcode S### command and turned on / off with the Gcode M3,M4 / M5 commands.
It is up to a connected board (BoB or motion controller) to convert this PWM signal into the 0 to 10 Volt (or whatever) signal required for a VFD.
Hope this helps.
Tweakie.
-
Hi,
if you require speed control only then all you require is a variable DC voltage. If you wish to have an indexable spindle for rigid tapping say
then you will require a DC servodrive that can accommodate the encoder fitted to the servo. The latter option while desirable for future operations
or bragging rights does severely limit your choices and possibly makes such a drive expensive by virtue of rarity.
Is there a reason you've chosen this particular servo for a spindle? The over whelming majority of servos these days are AC servos, no pesky brushes
to wear out. I suspect that due to the abundance of choices they would be cheaper too....I would recommend a matched servo and drive.
I adopted this strategy when I bought a servo for a spindle, I bought secondhand a 1.8kW Allen Bradley servo and matched it to an Ebay Allen Bradley
drive. I did have to buy the servo setup software from Rockwell Automation (current owners of Allen Bradley) but it made combining the two was easy.
I now have (at least) three choices for driving the spindle:
1) Set up 8 distinct speeds in a table and select a speed entry by asserting three servodrive input pins in binary fashion
2) Have the servo follow an analogue voltage applied to one input pin, the gain and slope are programmable
3) Use position control, with or without electronic gearing which allows superb resolution and accuracy for rigid tapping
Despite having such a variety of control modes I find that in practice most milling jobs require the spindle speed set once at then left alone for the
rest of the job. Unless you use a wide range of tool diameters like fly cutters and facing tools don't get to hung up on speed control, you may find you've
invested a lot of time, effort and money which in practice don't get used much.
Craig
-
Hi Craig
The reason that I chose this particular motor is that the machine manufacturer dropped the AC motor and replaced it with an SEM www.sem.co.uk branded motor. This has now been taken over by Callan callantechnology.com . The two big draw cards for me are
1. The physical size (Ø108mm) is a good fit for the space in the machine head
2. Motor speed is high at 5000 RPM, most of what I have found on the net has a max RPM of 3000
I had not found Allen Bradley, so will have a look into this brand.
I had not though of using a range of preset speeds, something I will also have a look at. My day job is NC programming and am used to the luxury of outputting any speed that is calculated from the CAM software. It is also valuable to be able to over ride the speed at the machine to tune the cutting conditions
Thanks
-
Hi,
I think the frame is a 'standard' size, just what standard...its the great thing about standards is that there is so many to choose from!
You are right, 3000 rpm is a common rated speed for servos in this power output. There are a number of manufacturers who regularly produce
fast servos, Baldor comes to mind. They tend to be expensive tho. While Baldor servos can be had on Ebay pretty cheap the drives are
eye-wateringly expensive, Baldor is owned by ABB and they traditionally are really expensive, I avoid ABB on principle!
The Allen Bradley servo I bought is rated to 3500 rpm. Its listed max is 5000 rpm but without trying to hoodwink the servodrive software I cant
get it to go beyond 3500 rpm. The extra speed would be nice for 1/4 inch tools but I'm not stressing, this is a hobby machine, if a job takes an extra
5 min then so be it.
I would suggest that you consider the price of a replacement spindle motor be the cost of the combined motor and drive. Depending on the
sophistication of a DC drive it may tip the total cost to be more than an AC servo and drive. An asynchronous AC motor (ie induction motor) and
matching VFD will be cheaper than any servo but probably 50% bigger for the same power.
As far as servos go... there are servos and servos! Some brands like Fanuc and Seimens are just over the top price wise and then there are brands like
Baldor and Allen Bradley and many other US or European brands which are just expensive. Delta servos from Taiwan have a good reputation and are
well priced, not bargain basement but not bad. There are cheaper Chinese brands and a lot of those have been rebranded to sound US/European and
for hobbyists probably good enough.
The further you get from known brands the more online research you should do BEFORE you stump up your cash, once they've got your money
its too late for you to decide that the unit is not good enuf!
The range of control strategies is so broad that its hard to specify beforehand. I personally have found that very simple speed control is adequate.
I will at some stage use position control for indexable operations like rigid tapping, that capability is built in and just waiting to be used.
I note that in your original post that you have a USS. At the current time the ESS has a Mach4 plugin but the USS does not. Warp9 are working on
producing a Mach4 plugin for the USS but you may be frustrated waiting. I use an ESS/Mach4 and VERY happy with it.
Craig
-
Thanks for the reply Craig, that's all really good sound advice. I was guessing that a new AC motor with a new VFD would be a more expensive option and like you say research is the sound way to find out. Interesting that it may now give DC a run for its money. I have another reason to go to DC and that is I am remaking the spindle to go from R8 taper to BT30 taper. This will also enable me to add a support bearing on the tail of the spindle. The original spindle only has a pair of angular contact bearings at the nose end of the spindle. This is an improvement that the manufactured also made to the more recent models. With this modification there is now limited space for the spindle motor and the form factor of the DC motor is a perfect fit for the machine. So I am committed to using a DC motor. I tried to embed an image of my machine in my first post but it did not work for me. I see that there are very few images in the forum threads which is a shame as a picture is worth a least a thousand ...
Thanks
Keith
-
Hi Keith,
pics often need to be compressed somewhat before they can be attached or alternately linked to a suitable site like photobucket.
I used a RegoFix ER25 cylindrical tool holder and turned up my own housing to accommodate matched ABEC7 angular contact bearings at the nose.
I used plain deep groove bearings at the tail, mainly due to budget....ABEC7 bearings are bloody expensive! It worked out pretty well.
I am quite obsessive and want much better than I can afford or have the skill/machinery to make. In this case I just had to buckle to the fact I couldn't
afford what I really wanted and so made the best I could from what I could afford. I'm glad I did, its by no means perfect but its certainly good enuf to
use and I take satisfaction from its making. That is after all what the hobby is about.
Craig
-
Yes it is always nice to have the good gear, its hard to beat quality tools. When I buy tools I like to buy quality as I know I will have use out of if for a long time. Great that you have made parts to improve your machine. It always feel good when a project works out well.
I am lucky that I work in a toolroom and have access to use the machine tools. This is coming in handy to make bits for my project.
Keith
-
Hi,
years ago I was working in a workshop and we had an old lathe, big sucker but pretty worn. I had no problems turning bushes and things like that
provided I didn't need to get closer than a couple of thou.
When a boat propshaft came in the boss would call Bob, a local guy, retired fitter and turner trained in the UK old school. He had a little notebook that
he had made for this lathe and he would get it out and turn tapers to PERFECTION! It was my job to blue and lap the prop onto the shaft and I never
had to work hard for it... I asked Bob any number of times how he did it but the old bastard would just smile and tap his head with his finger.
I've tried for years ever since trying to improve on the basic accuracy of the machine tool I'm using at the moment. The real trick is not just the quality
of the tool but the smarts that you need to add to get really good results like old Bob did.
Craig
-
Hi Keith,
I have open in front of me the drawing of your DC servo M4-2950 -the main body of the motor 108mm dia and motor body excluding shaft
215mm long...correct?
http://www.ebay.com/itm/Allen-Bradley-MPL-A310P-HJ22AA-A-Servo-Motor-230V-AC-3-Phase-0-73kW-5000RPM/152457683587?ssPageName=STRK%3AMEBIDX%3AIT&_trksid=p2055119.m1438.l2649 (http://www.ebay.com/itm/Allen-Bradley-MPL-A310P-HJ22AA-A-Servo-Motor-230V-AC-3-Phase-0-73kW-5000RPM/152457683587?ssPageName=STRK%3AMEBIDX%3AIT&_trksid=p2055119.m1438.l2649)
This motor is shorter at 165mm and the body of the motor is 80mm square or fits in a 113mm circle. 730W at 5000rpm.
http://www.ebay.com/itm/ALLEN-BRADLEY-MPL-A330P-MJ22AA-KINETIX-AC-SERVO-MOTOR-7033-00-4402/192257239067?ssPageName=STRK%3AMEBIDX%3AIT&_trksid=p2055119.m1438.l2649 (http://www.ebay.com/itm/ALLEN-BRADLEY-MPL-A330P-MJ22AA-KINETIX-AC-SERVO-MOTOR-7033-00-4402/192257239067?ssPageName=STRK%3AMEBIDX%3AIT&_trksid=p2055119.m1438.l2649)
This motor is a bit longer at 215mm but the same frame size, ie 80mm square and 1.8kW at 5000 rpm. This is in fact a slightly faster but otherwise identical
servo that I used for my spindle.
http://www.ebay.com/itm/Allen-Bradley-Ultra-3000-Series-C-Digital-Servo-Drive-for-Servo-Motor-Used-Good/311641997971?ssPageName=STRK%3AMEBIDX%3AIT&_trksid=p2055119.m1438.l2649 (http://www.ebay.com/itm/Allen-Bradley-Ultra-3000-Series-C-Digital-Servo-Drive-for-Servo-Motor-Used-Good/311641997971?ssPageName=STRK%3AMEBIDX%3AIT&_trksid=p2055119.m1438.l2649)
This is the matching drive. Note that you still need the cables to join the drive and the servo, not trivial, one I Ebayed cheaply and one I made. I also bought
the Ultraware software required to program the drive at $200 NZD , about $130 US.
For your information.
Craig
-
Yes you have the correct drawing of the DC motor that I am looking at. Thanks for the eBay links for the AC Allen Bradley, that's really awesome information and looks like a really good alternative. I am very impressed with the compact size of the motor. The question that I have is how this compares for torque against the DC motor. Time for some research on the net...
Keith
-
Hi,
have a look on the label, the max sustained torque is on it. Aside from any thing else there is the power spec...power=torque x speed....easy.
The Allen Bradley MPL series are mid to low inertia.....the A in A310P stands for 230V, the 3 is the frame size, 80 mm square, the 10 refers to the length or
stack size and P refers to the speed, 5000 rpm is this case. The A310P in the first link is 730W. There is an A320P model of 1.3kW and the second motor I
linked is an A330P of 1.8kW.
The rated torque of the three models is:
1) A310P 1.58Nm
2) A320P 3.05Nm
3) A330P 4.18Nm
The motor I used is the A330H model which is rated to 3500 rpm but has a little more torque, 6Nm which suited me I want plenty of torque, but still 1.8kW.
You may note that the rated current of the small motor the A310P is just under 5A. You could use a smaller drive than the link I posted, the 2098-DSD-010
has rated current of 5A, overload 15A,(5/15A) and would run the A310P nicely.
The larger two will require more current, the 2098-DSD-020 is rated at 10/30 A. I used this to power my motor although if I could have found a 2098-DSD-030
rated at 15/30A I would've taken it. The larger the drive the more you pay. The 010 can be had for under $100 but $100-$200 is more realistic. The 020 can be
had for just over $100 but $200-$400 is more realistic. I missed out on an 030 for $280, $500-$800 is the norm.
The 730W motor and drive could be had for about $500 plus shipping. Whether you consider that good value for secondhand I don't know. Some of the
Chinese offerings are about the same but new. I ended up paying $700 NZD, about $500US for the motor, $250US for the drive and $150 for the cables,
near enuf to $1000 for a 1.8kW servo and drive. It kicks anus!!! I am happy that the value is there, the crappiest I was prepared to go was Delta and they
cost more in this power range.
Go to the Rockwell Automation site and get the MPL series drawings a see if you cant squeeze one into your machine, they are good.
Craig
-
Thanks Craig, the size in not a problem, I could go up to 110 square. The power of 700W and current at 5A is good as I need to run the complete machine off 230V 10A wall outlet. What you are suggesting looks good. The uncertainty is if the AC can deliver enough torque. I am not sure what it will delivery through the rev range. I have had a look for the performance curves and not been able to find them. The M4-2950 B DC option specification is continuous torque of 2Nm and peak of 14.9Nm. The DC option would definitely cost more, I am trying to get the best performance for around 5A current draw
Keith
-
Hi Keith,
the AC part of AC servo is a bit of a misnomer, it is really a 'brushless DC motor'. In order to spin the applied DC
has to comutate backwards and forwards, ie become AC.
The torque of a DC motor comes from a comutating current in the armature in a fixed permanent magnetic feild
provided by the pole magnets.
The torque of an 'AC' motor is a comutating current in the 3 phase field windings operating on magnetised rotor.
The upshot is 'there is no damn difference'. An AC servo typically has an overload torque of three/four times its rated
torque, same as a DC motor. The equation power=torque x speed applies to both ie a 1Nm torque motor spinning
at 100 rad/sec produces 100W...end of story..no difference between AC an DC!
Craig
-
Hi Keith,
a picture being worth a thousand words....
https://www.youtube.com/watch?v=cdiZUszYLiA (https://www.youtube.com/watch?v=cdiZUszYLiA)
Craig
-
Hi Keith,
an AC servo can produce rated torque from zero rpm to rated rpm, and if your drive allows a technique called
'field weakening' steadily reducing torque at FASTER than rated speed. In terms of control flexibility they
are better than a brushed servo.
I run my drive on a single phase outlet rated at 10A also. It doesn't stop me from drawing as much as 30A
in bursts enuf to max my drive and servo....good...very good!
Craig
-
Hi Craig
I get what you are saying. I see that there are plenty of motors and drives for sale on eBay for reasonable prices. This definitely looks like a good option. Am I selling myself short @ 730W, should I look at the 1.3kW? How is it that you can run your A330P which is rated at 12A continuous from a 10A outlet? Are you running all the axis drives and other electrics for the machine off the same supply?
Thanks
Keith
-
Hi Keith,
yes the whole shooting match off on 10A circuit. Don't know what its like where you are but in New Zealand its not uncommon for a 10A circuit
to be a lot more heavily loaded. I work as a technician fixing welders and even little inverter welders regularly draw 30A. I know its not ideal
but that's what happens with very few if any ill consequences. The circuit feeding my mill is 2.5mm square and I'm not uncomfortable at running at
25A continuously. As it turns out you have to be doing something pretty extreme to draw that sort of juice.
I know the name plate says 12A but that is in fact the peak, not rms. It is however 12A peak per phase, and of course there are three of them! The
full load amps (FLA) referred to the 230V input is 28A! The rms doesn't sound quite so bad, 19.8A ...LOL
Of course the reality is that I seldom get anywhere close to max current and therefore max torque. Also lets imagine I'm driving a facing tool at high
torque but often modest speed, say 1000 rpm. Under that circumstance the voltage at the motor terminals will be about 1/3 rated ie 100V. The DC link
voltage is 320V (230VAC rectified and smoothed) and consequently the effective PWM of the link voltage to the terminals represents a transformation
of 3:1. If the phase current is 12A per phase the link current required to feed that phase is 4A. You can see where I'm going here, those occasions where
I require max torque often coincide with modest speeds and the effective transformer action of the PWM buck regulators means everythings kool.
I have three 5 phase 23 size steppers with genuine Vexta 230V drivers. According to the name plate they can draw heaps of current but in practice I've
never seen the draw more than an amp, thats roughly 230W...where could all that power go...little 23 size steppers couldn't absorb that much except briefly!
The 1/8 hp coolant pump draws more on average than one of the axis drivers.
The upshot is that I have to really bore into a piece of stainless with a 16mm tool to start drawing decent current from the plug and a that stage my mill
is 'a humming and'a jumpin' and can't really take any more, its not 250 kg but probably 150kg.
If you think I'm being a bit cavilier about it wait until you see someone plug in a single phase plasma cutter and start cutting, it makes my mill look liteweight!
Craig
-
Hi,
by the way go for the biggest and most powerful you can squeeze in there and/or afford. You may not use it all often, but you will use it
and the bigger it is the easier it will do it.
Craig
-
Hi Craig
Just read your bio and I see that like myself you are also in NZ, I'm in Hamilton, what city are you in?
Keith
-
Hi Keith,
I live in Akaroa, about 80k from Christchurch. I somehow imagined you lived in...I don't know...Sweden or somewhere.
Craig
-
I'm off to bed now, tomorrow I'll see if I can PM you and give you a call?
Keith
-
Hi Keith,
yeah that would be kool.
Craig
-
Thanks for all your help Craig
I have purchased the following Allen Bradley motor and drive on eBay;
AC SERVO MOTOR MPL-A320P-SJ22AA, 1.3kW, 5000RPM, and matched servo drive; Ultra 3000 2098-DSD-020. Am looking forward to getting my machine set up and running. The motor fits really nicely into my machine, its a very compact motor!
Keith