[hab82]

I have an X3 that I have converted to CNC. While I am still working on the Ball Screw upgrade, I am more interested in improving Spindle performance. I have an X2 with a belt drive conversion kit that gave some benefits but also drawbacks. Improved speed, quieter, but less torque.

I would like to get the benefits without the drawbacks, plus some additional benefits. I want program speed control and synchronization for rigid tapping.

I think I would like a top speed of about 5000 rpm but also be able to use a 1 inch drill at reasonable feed and speed rates.

I am thinking probably a top mount 3 phase motor with vfd control and cogged belts. I would appreciate comments and if available links to proven examples.

Hubert

[Jeff-Birt]

A 1" drill on an X3? Drilling a 1" hole with a 1" bit is quite a chore for a small machine, not to mention that a 1" bit is L-O-N-G. Do doubt you would want/need to predrill the hole to a smaller size first. I would suspect that you would be better off to make your 1" hole with a smaller end mill with a pocketing or profiling operation.

[hab82]

The 1" drill capacity is from the Spec sheet for the X3 and was used as an example to say I don't want to give up low speed torque to gain high speed. I know how to circle cut with an end mill and do it routinely rather than make tool changes. Yes the X3 is still a small mill even though it is much larger than the X2.

I have seen references on the zone to a conversion to a 2hp 3phase VFD conversion and am trying to track it down. I would like to understand the pro's and Con's of this much power on the X3. Part of this comes from using CAM software developed for the bigger machines that expects to plow through at full cutting tool capability and assumes the power is available. On my X2 the spindle power was very much a limiting factor as steppers could feed fast enough to stall the spindle.

All of my 1/2" shank Silver and Demming bits mounted in 1/2" R8 collet are shorter than 1/4" bit mounted in a chuck, and faster to change bit sizes 1/2" to 33/32". So bit length is not a problem.

[Jeff-Birt]

I suspect that the 1" spec come from a spindle motor power standpoint (there is enough power to use up to that size drill) rather than the praticle application of a drill that large.

All of my 1/2" shank Silver and Demming bits mounted in 1/2" R8 collet are shorter than 1/4" bit mounted in a chuck, and faster to change bit sizes 1/2" to 33/32". So bit length is not a problem.

Well compare apples to apples, a 1/4" bit mounted in a collet will be much shorter than your 1" drill.

Take a look at Automationdirect.com they have some good low horsepower 3-phase drives and motors. Take a look at the spec sheets they have, that will give you an idea of how much torque you will get over the speed range you want.

[hab82]

The comparison to the 1/4" drill, is that we don't think anything of the length of the 1/4" drill in a chuck I possibly should have used 7/32 bit as one that we normally do not have a collet for. Once you move above 1/2" using Silver and Demmings you really tend to use the collet any way since you have a common shank size. The length hasn't been the problem. Yes, I can get the Torque with the current noisy gears. I gave that up with the X2 by going to the belt drive. Getting variable speed to high speed and retaining torque involves trade offs. Hopefully, I can learn from others experiences. Probably, to gain the experience personally, I will initially move to a two range Belt drive, and modify the current DC control with a spindle sensor and use a digital feedback loop to adjust the speed. I have 30 years professional engineering experience, just not in this particular neighborhood. Math makes things sound possible, but what does practical experience show as smart.

Reviewing my recent experience after I had problems using the larger bits and having full 4 axis now available, I have opted to use the same size cutter for almost all work above a 1/8" radius. Below that I use drills. However, when I get to deeper cuts, out comes the bigger cutters or drills. Earlier when I still only had manual Z axis I preferred fixed size plunge cutting thus the use of the silver and demming bits.

Isn't automation great. It gives you additional options that were a pain before.

How much Horsepower do I need, How much torque do I need? I can do the math but what fudge factors should I add? When does the power become wasted because something else is the limiting factor?

On VFD for hobby use do you really need the VFD rated motors $$$? How far above the base frequency can you push them 2:1, 3:1? What derating factor should use? Is it better to use a two pole motor for speed? or a 4 or 6 pole motor for torque? Can you use the VFD's drives at full ratings or do they need to be derated?

From a math viewpoint until eddy loses become an overwhelming driver I should be able push the motor at a higher frequency maybe at a constant horsepower or slightly decreasing horsepower range. So if I could push a four pole motor to twice the frequency at full horsepower rating I would have double the torque at low range and I would not need as drastic difference in pulley sizes. These are the thoughts I have, but what does experience dictate.


Hubert

[wildwestpat]

Hi Hubert

If I have understood you correctly you are wishing to power your spindle with high torque at low speed say 50 rpm right upto 5,000 rpm plus. This is a tall order. The demands on the motor are going to limit you and or wreck your finances!

The electronics can be made to cover the required range but a three phase motor will have problems. At the higher speeds the frequency of the supply dictated by the number of poles will bring magnetic losses in to play that will limit the ability to generate a meaningful torque and insulation will limit the voltage that can be applied to try and force current into the winding to increase the magnetic forces. This results in heating due to the B/H curve of the iron used for the motors laminations. The heat will also further weakening the insulation.

The torque at low speed requires abnormal current to be pumped through the motors windings and the gauge of wire used and its resistance start to take a limiting role.

If you look at the integrated motor and electronics VFD available you will find that 10 to 1 is common and I am unaware of any that can achieve 1000 to 1. Please feel free anyone who knows of commercial products that have AC three phase motors.

Most of the commercial variable speed drives that require a wide speed range with useable torque use DC motors. There are circuits like the ones used by the mill maker to pulse the power supplied to the motor. There are various variations on this type of drive which try and compensate for the loss of torque at the low speed end with out introducing excessive cogging in the mechanical rotation.

If my reasoning is correct you should be looking to a similar spindle drive to your existing one with a mechanical gear change mechanism to multiply the torque for low speed work and increase the speed at the higher end if you really need high feed rates or wish to do engraving.

I also doubt the stiffness of the X3 or similar mill to provide chatter free use of large tools with high feed rates. Try putting a DTI on the table of your mill to measure the deflection of a tool mounted in the quill. With the mill adjusted for normal use apply pressure by backing the vice into the tool in both the X and Y axis. The deflection of the DTI being a measure of the ridgidity of your machine under static loading i.e. the steady state with a tool cutting properly. Under dynamic conditions the forces will be a lot higher even with normal direction milling. A factor of three is the assumption I would make on top of the calculated tool cutting torque. Applying this force by backing the vice into the tool will give you some idea of the forces and resulting deflections. If I am right then this simple test could save you a lot of expense and nugatory work beefing up the drive only to find it has no practical pay off othaert than being a fun exercise.

Hope this helps as the money might be better spent on some bigger ball screw motors alowing tighter gib settings to improve ridgidity rather than an exotic motor VFD electronics.

Good luck

Pat

[hab82]

Pat
Actually, I should have clarified this in my original post, but I am not adverse having multiple ranges as long as they have overlap. As VFD's are new to me I am trying to understand the practical limits. I realize that a motor designed for 400 Hz operation uses different magnetics than one designed for 60 hz. but what is the practical top limit of a 60 hz motor with derating. The vfd's I have looked at specifically the Hitachi X200 series show an output frequency range 0.5 to 400Hz. How much of that is usable. Just spotted a torque characteristic curve base on a Hitachi standard 4 pole motor that shows a continuous duty range of 45% rated torque from 6-120 hz. and a short time down to 1 hz. It also shows the peak continuous performance of 95%. So this looks like a reasonable 20 to one range for 45% torque and 4 to 1 range for 80% torque. With this in mind I could probably go to a dual range mode of 10-1000 assuming 10-50 are short periods. and 50 to 5000 assuming 50-250 are short periods. A 4 pole motor would require a step up ratio of 5000/3600 and a step down of 1000/3600. Does this make sense to others. Maybe a third mid range would be useful as well of 36 to 3600 or 3600/3600 with 600-2400 high torque range that would be good for steel. This would give low range for Drilling and Tapping, Mid Range for cutting Steel, High range for cutting Aluminum and with My Proxon with 5000 to 20000 spindle a super range for engraving.

To me this looks feasible. Does anybody have some recommended motors?

Hubert

[wildwestpat]

Hi Hubert

The selection of the motor will determine the actual range of frequencies that can be obtained. The Hitachi drive electronics have settings that are designed to protect the motor. The weasel words of the specification writer are evident in the simple statement about consulting the motor manufacturer on the selection of the VFD. In plain language speak this means select the motor then the drive electronics. - Hope there is someone out there that has the answer to your question about over frequency driving standard supply frequency motors and they respond to your plea for information..

However I suggest you test the X3 when you have completed the addition of the ball screws to see what the rigidity of the spindle is with respect to the table. I believe the limits on machining capability will be set by the stiffness of the frame and the rigidity of the various dovetail slides and quill rather than the torque / speed range available assuming the use is for shaping metal.

Good luck – I look forward to reading more about the modifications you are making and details of the improvement achieved.

Regards

Pat

[saltytri]

Hubert:

I'm not knowledgeable enough to suggest answers to all your questions. But perhaps I can offer you one bit of useful information. I've put a lot of time recently into selecting a motor and VFD for a lathe that I'm converting to CNC. I ended up with a 1.5 HP Baldor 3Ph TEFC in a 145T frame. That's half the HP of what you want and it weighs 41 lbs. Not a problem for me because it doesn't have to hang off the lathe. But I wonder whether it is practical to put a 3HP on an X3.

edit: I see you said 2 HP not 3 HP. Sorry about that!

[hab82]

Certainly everything is in the planning stage at this time. I saw a post where somebody had converted an X3 to a 2hp motor that weighed 56 lbs. While I have sent a private mail to him, I see from the stats that he hasn't been online since June. Doing some mental exercises and web browsing I have come up with a tentative plan. In this case I am looking at using 65 tooth Timing belt that measures to have 4 teeth per inch that comes with the X3. In this case I am looking at having a three stage spindle timing pulley of 51, 34, and 28 teeth. For the motor timing pulley having 14, 34 and 39 teeth. While the target ratios were 1000:3600, 3600:3600 and 5000:3600 based on the planned top speed in each range the actual choices were made to keep the minimum belt length approximately the same and approximately 2 inches shorter than the measured belt length. So the actual ratios are 988:3600, 3600:3600 and 5014:3600. I plan on cutting these myself. Since my stepper driven rotary table gives me a resolution of .002 Degrees I think I can come up with evenly spaced teeth.

I am looking at using the Hitachi X200-015NFU as the driver and either the North American PE145TC-2-4 or NA145TC-2-4 data sheets are here http://www.naemotors.com/premium/PE145TC-2-4.pdf and
http://www.naemotors.com/newt/na1800c/NA145TC-2-4.pdf
I would prefer the PE version as it is 6 pounds lighter and has better Inverter specs but I still don't have a price for it. The NA version has a price of $189.00 at
http://www.e-motorsonline.com/product_ACM03786.html
and as that appears to be in budget and priced in the same range as the Grizzley replacement 3/4 hp motor, I don't feel further design work will be wasted. Dimensions indicate that these motors will fit between the spindle and column with room for belt adjustment.

Thus my current plans are make the timing pulleys initially for the stock DC motor as its shaft is smaller and when upgrading to one of these it would just require re-boring and keying the motor pulley. I will build the motor mount to accommodate all three motors.

Then when either the DC motor fails or I get tired of the DC Motors performance I will upgrade to one of the 3 phase motors.

The initial results should give equal or greater performance at low speed as belts are supposed to be more efficient than gears, and give me the desired higher spindle speeds for cutting aluminum.
It also has the added benefit up being much quieter.
Hubert

[wildwestpat]

Hi Hubert

Glad to hear that the belt drive and existing DC motor are to remain in use for the time being.

If you are going to spend real money on that VFD and associated motor then I would purchase them as a pair from an expert supplier. There are a number of factors that decide the actual performance delivered at the motor spindle. The specs you have pointed out, as far as I can see, do not give the full picture. For example the maximum speed and the lowest speed recommended. These parameters are used in the VFD along with others to define the motor protection features. The VFD will probably be capable of seeing off most motors by over driving it as the electronic parameter limits are designed for the widest possible range of applications. Motor manufacturers do not like to give these figures as they are rightly worried that their motors will be compared with those form other makers by users who do not necessarily appreciate the needs of matching the motor to the VFD for a given application. Both the VFD and motor manufacturer have application engineers to help potential customers. There are also outlets that specialise in providing proven pairings of motor and VFD and will tell you the performance that will be achieved and the settings.

A side issue is the large amounts of electrical noise these drives make and it would be a good idea to keep the leads very short and screened with metal conduit - flexible if you are not mounting the electronics on the head.

With the applications engineers time being free I suggest if and when you need to replace the DC motor you buy a matched pairing of VFD and motor.

Regards

Pat

[hab82]

Thanks for your concern. I am an electrical/electronic/computer engineer depending on which phase of my 40+ year career you reference. I have the necessary test equipment to "scope out" potential problems. Certainly, I will consult with the manufacturers representatives as appropriate. I appreciate hearing about others experience. The preferred motor the North American PE145TC-2-4 is available for less than $100 more than the NA145TC-2-4 and is in line with the cost of the replacement motor for the SX3 and KX3. In comparing the cost of the various motor sizes from 1hp to 2 hp and the appropriate drive electronics I have found these to reside on a fairly level plateau. The cost and weight do not double from 1hp to 2hp and seem to only vary about 20%. I would rather oversize than undersize as the life of the motor will be better in the latter. These mills have the reputation of having undersized motors, and only in the KX3 version do you get 1000W motor (aprox 1 1/3 hp) and the 5000 rpm range. I believe the industrially proven Hitachi VFD drive will provide better reliability than the electronics provided by Sieg at a relatively modest increase in cost ($25). From my limited experience with milling it doesn't present the same loading as many motor applications which run at full load almost continually. In most milling cases it is more of an intermittent type load of cut relocate cut, or drill clear chips drill type of operation. And as a hobby machine usually used for relatively short periods of time. Even in use the amount of time it would demand peak hp or torque is relatively low. My weakness here in not on the electrical side but the application side, what are the stresses driven by milling. Experience on the application side, I hope to gain from using the current motor and electronics, but my experience with the X2 encouraged me to budget for and plan for an upgrade. I feel the new belt drive will accommodate the various uses that I have encountered to date.

The purpose of this round of discussion was to determine feasibility and limitations. In my mind I have accomplished that, although I would appreciate input from anybody who has already attempted this. By making a tentative motor and VFD selection I can avoid making related desighn choices that would have to be discarded when the time came to make the upgrade. I expect the upgrade to occur but much later in my build cycle. I expect to complete the ball screw upgrade and addition of the fifth axis prior to taking this step.
The latest information I received on referenced motor is that it is manufactured in China and has basically a constant torque of 6.05 ft. lbs from 6 to 60 hz and then constant horsepower from 60 to 120 hz. These were figures were similar to the curves that were given in the Hitachi X200 manual, that I used for planning pulley ratios. Thus at low speed of Low range 4482 oz/in in stepper language at 47 rpm to 470 rpm dropping to 2282 at 940 rpm. Mid range would provide 1162 from 180 to 1800 rpm dropping to 581 at 3600. High range would provide 834 oz/in from 250 rpm to 2507 and drop to 417 oz/in at 5014 rpm. I haven't checked on alternatives yet but I believe I will use different configurations of my cam post processor for each pulley range and also for different configurations of the control software. I have also found out that the VFD can be controlled by the computer with the same techniques used for controlling the vsd drives of current DC motor, only requiring software parameter changes

Thanks All and I hope others found some benefit from this discussion.
Hubert