New Machine Build Custom Spindle (Custom Motor!), Very Gradual Progress Planned - Page 2


Page 2 of 2 FirstFirst 12
Results 21 to 32 of 32

Thread: Custom Spindle (Custom Motor!), Very Gradual Progress Planned

  1. #21
    Member
    Join Date
    Mar 2014
    Posts
    735
    Downloads
    0
    Uploads
    0

    Default Re: Custom Spindle (Custom Motor!), Very Gradual Progress Planned

    Following just to learn. I really cant contribute much to this discussion but I'm interested in what you come up with.
    I built a 3 axis router that used a 2.2KW, 24K RPM spindle. The machine was plenty stiff and using it I found that my limiting factor in how fast I could cut was the HP of my spindle. I never approached max VFD current until I used a spoilboard surfacing bit. Then I had to be careful with my feedrate and step size or it would trip out on overcurrent.



  2. #22
    Member StrawberryBoi's Avatar
    Join Date
    Aug 2021
    Posts
    199
    Downloads
    0
    Uploads
    0

    Default

    Quote Originally Posted by 1Jumper10 View Post
    Following just to learn. I really cant contribute much to this discussion but I'm interested in what you come up with.
    I built a 3 axis router that used a 2.2KW, 24K RPM spindle. The machine was plenty stiff and using it I found that my limiting factor in how fast I could cut was the HP of my spindle. I never approached max VFD current until I used a spoilboard surfacing bit. Then I had to be careful with my feedrate and step size or it would trip out on overcurrent.
    Hmmmmmm

    You could trip an overcurrent without hitting 2.2kW if you were going under the rated motor speed. The current is proportional to (torque - losses), and the rating on a motor is not a hard limit. A 0.6Nm motor could produce 0.7Nm for example, but the current rise from 0.6-0.7Nm might be double that from 0.5-0.6Nm due to saturation effects. Basically the iron is starting to become less of a magnetic conductor and so your magnetic resistance (called Reluctance) goes up and more magnetic field strength (like the magnetic voltage) is required to push more total flux (like the magnetic current) through the iron.

    Torque is exactly the cross product of the current through the field times the radius about center of rotation in the plane of rotation this occurs at. That's the portion of the electrical current that is perpendicular to the field AND tangent to the circle of rotation.


    Long story short, you can always pump more current into a motor and trip your overcurrent. If you are looking for HP you have to run at or just under the rated speed to max it out. If you have a VFD with vector control and field weakening in most but not all motors, about 5% past the rated speed will allow for the highest power due to higher efficiency. A 3kW VFD might be able to run that 2.2kW spindle at say, 2.3kW at that point.

    The practical issue there is if you hit your torque limit on the spindle it will slow down and likely overcurrent, so you have to run it under max power.

    Also keep in mind that for short durations the spindle can run at something like 200% rated torque and power, but it will overheat.



  3. #23
    Member
    Join Date
    Nov 2013
    Posts
    4282
    Downloads
    0
    Uploads
    0

    Default Re: Custom Spindle (Custom Motor!), Very Gradual Progress Planned

    Hi,

    Then I had to be careful with my feedrate and step size or it would trip out on overcurrent.
    VFD's draw a lot of reactive power that does not get translated into useful power at the spindle, this is called PowerFactor (PF).
    It means that a VFD will draw a great deal more current relative to its output.

    A 2.2kW VFD will have a single phase input current rating of 23A or there abouts. That's a lot of current and will very likely cause the circuit breaker to open.

    There is a solution....a line reactor. It is a coil wound on a transformer core and inserted between the incoming AC supply and the VFD. For a 2.2kW installation
    anywhere between $50 and $100 should do it. Then you'd be able to run right up to 2.2kW WITHOUT the circuit breaker opening.

    Craig



  4. #24
    Member StrawberryBoi's Avatar
    Join Date
    Aug 2021
    Posts
    199
    Downloads
    0
    Uploads
    0

    Default Re: Custom Spindle (Custom Motor!), Very Gradual Progress Planned

    I'm starting to think you are a lobbyist for the line reactor industry... LOL

    Not all VFDs will do this. Those with active rectifiers can implement Power Factor Correction (PFC) and eliminate this issue almost entirely.



  5. #25
    Member
    Join Date
    Nov 2013
    Posts
    4282
    Downloads
    0
    Uploads
    0

    Default Re: Custom Spindle (Custom Motor!), Very Gradual Progress Planned

    Hi,

    Those with active rectifiers can implement Power Factor Correction (PFC) and eliminate this issue almost entirely.
    Agreed, but can you name and point out models that do....in hobbyist power ranges?

    It's very common to have standalone PF correction units that stand alongside the VFD, or banks of VFDs, in the tens and hundreds of kW class
    but not for 10kW and below.

    The reason I'm so pro reactors is that at my previous home my garage, where my machine was installed had a 2.5mm2 cable providing
    the single phase supply. Despite having only a 800W spindle I was having trouble with nuisance tripping of the breaker and even more concerning
    was the small but noticeable increase in the cable temperature.

    I was at that time employed as an electronics technician repairing welding equipment and knew full well the problem of poor, in fact rubbish,
    power factor. I fitted a modest home wound line reactor and the problem was solved.

    What I did not anticipate when I did this but it had a significant impact on the 'electrical noise' front as well. I have a USB connected pendant which is mostly
    quite reliable however would have issues with the USB connection dropping out when pushing the spindle. As it did not happen a great deal I just
    ignored it. After fitting the line reactor that problem went away. Since that time I have become an advocate both of line reactors for harmonic current control
    but two stage EMI filters as well. Very few of my machine cables are screened and without undue problems. I conclude then that a major and often overlooked
    means of electrical noise infiltrating sensitive control circuits comes from the AC supply, or more particularly the harmonics on that supply.

    I'm in the process of designing and building my own Field Oriented Control servo drive for a re-manufactured 3kW servo I already have. Amongst the features
    it will have is active PFC correction. The servo has a nominal peak current of 48A and my power supply will have to match it.

    Craig



  6. #26
    Member StrawberryBoi's Avatar
    Join Date
    Aug 2021
    Posts
    199
    Downloads
    0
    Uploads
    0

    Default Re: Custom Spindle (Custom Motor!), Very Gradual Progress Planned

    Spent another hour, pretty happy with how it's shaping up. Definitely a lot of little finagling that can be done, but the motor is already quite compact and power dense with great efficiency.

    Torque ripple is ~5% at rated load, which is pretty good. Good enough for most servo applications, though I'm going to try and reduce that. A small skew could be in order, but that adds cost at low volumes and I'd prefer to avoid it when messing with some tooth geometry may solve it neatly.

    The rated torque is around 1.9Nm and it can peak out well over 7.2kW. Could I make the motor smaller? HELL YES. Will I? Dunno, the efficiency is very nice. This may actually be more of a 6kW rated motor with liquid cooling. I'll have to start doing some thermal conductivity checks, the 21/4 arrangement provides a very nice low harmonic content and so iron losses are almost totally supply frequency and thus very low (given that they scale between freq^1.5 and freq^2.5 depending on the particular material and type of loss).

    Attached Thumbnails Attached Thumbnails Custom Spindle (Custom Motor!), Very Gradual Progress Planned-spindle3_001_generalresults16a15krpm-jpg   Custom Spindle (Custom Motor!), Very Gradual Progress Planned-spindle3_001_advanceangleplotting16a40a-jpg   Custom Spindle (Custom Motor!), Very Gradual Progress Planned-spindle3_001_generalresults16a15krpm_10degadv-jpg   Custom Spindle (Custom Motor!), Very Gradual Progress Planned-spindle3_001_motorconstants-jpg  

    Custom Spindle (Custom Motor!), Very Gradual Progress Planned-spindle3_001_generalresults40a15krpm_20degadv-jpg   Custom Spindle (Custom Motor!), Very Gradual Progress Planned-spindle3_001_statordims-png   Custom Spindle (Custom Motor!), Very Gradual Progress Planned-spindle3_001_rotordims-png   Custom Spindle (Custom Motor!), Very Gradual Progress Planned-spindle3_001_axialdims-png  



  7. #27
    Member
    Join Date
    Nov 2013
    Posts
    4282
    Downloads
    0
    Uploads
    0

    Default Re: Custom Spindle (Custom Motor!), Very Gradual Progress Planned

    Hi,
    could you explain the 'number of slots' in the stator design.

    21/ 3 phases = 7 slots per phase ??? and seven seems odd to me, I would have expected an even number??

    Craig



  8. #28
    Member StrawberryBoi's Avatar
    Join Date
    Aug 2021
    Posts
    199
    Downloads
    0
    Uploads
    0

    Default Re: Custom Spindle (Custom Motor!), Very Gradual Progress Planned

    Industry secret.










    Nah, just check it out here: https://www.emetor.com/windings/

    The coil spans 5 teeth/slots and the pitch is 5.25 slots. Wacky winding on this sucker, not going to be fun to wind.



  9. #29
    Member
    Join Date
    Nov 2013
    Posts
    4282
    Downloads
    0
    Uploads
    0

    Default Re: Custom Spindle (Custom Motor!), Very Gradual Progress Planned

    Hi,
    yes that is kind of whacky but I think I follow.

    I presume that by virtue of your work that you have the opportunity to have stator laminations cut?

    How about the winding itself? Do you do it by hand? How about insulation in the slots?
    Do you intend to impregnate the windings?

    Craig



  10. #30
    Member StrawberryBoi's Avatar
    Join Date
    Aug 2021
    Posts
    199
    Downloads
    0
    Uploads
    0

    Default

    Quote Originally Posted by joeavaerage View Post
    Hi,
    yes that is kind of whacky but I think I follow.

    I presume that by virtue of your work that you have the opportunity to have stator laminations cut?

    How about the winding itself? Do you do it by hand? How about insulation in the slots?
    Do you intend to impregnate the windings?

    Craig
    Yes, there are a number of laser cut lamination stack suppliers in the USA, mostly in Michigan. Though if I'm willing to wait it's usually much cheaper to go the Chinese supplier route even for prototyping.

    If I order the stack from a USA supplier then I would hand wind. I may do this either way.

    I did leave space for doing a powder coat insulation on the motor, a small edge perimeter near the ID and OD are masked so it's exposed stack there to prevent chipping and allow for mounting. You can't clamp down on the powder coat, it's brittle.

    Otherwise I would use a Nomex, typically a single layer 0.005" is rated for inverter duty insulation. I need to check, but I'd be looking for a Hipot test to pass something like >3500V iirc for inverter duty at this voltage. I've mostly done 650-1000VDC systems and 12-48VDC systems so this is an in-between voltage for me.

    As for varnish impregnation, I could do it, I have access to a suitable vacuum bucket err... chamber. I might alternatively use self-bonding wire. The varnish is for two purposes, the main of which is to prevent wire movement and thus abrasion, the other is to seal any thin spots and holes in the insulation of the wire itself. While they won't guarantee it, Essex magnet wire is so close to flawless on the spool that you can essentially assume there are no bad spots in the insulation. Lacing with nylon or kevlar can be used to secure the wires instead without needing to keep a bucket of varnish around.


    Now that I'm thinking about it though I'll probably varnish dip it, I'd like to employ direct on wire cooling so adding the extra protection is wise. Maybe I'll hand wind 1-2 for the experience and contract the winding and varnish for any future ones.

    I'll need to settle on lead wire, epoxy for anchoring the lead wire and the center tap as well. That's the other question, do I use a center tap or do I make a 6 lead motor...! Have to confirm this motor is balanced as a Delta and not only a Wye. If so I'll double the turn count and make it a two level voltage selectable motor.



  11. #31
    Member StrawberryBoi's Avatar
    Join Date
    Aug 2021
    Posts
    199
    Downloads
    0
    Uploads
    0

    Default Re: Custom Spindle (Custom Motor!), Very Gradual Progress Planned

    So I'll be doing a Hi-pot test at 2500V, which greatly exceeds the recommendations for safe testing, but I'll be designing around longevity so extra insulation is only good.



  12. #32
    Member StrawberryBoi's Avatar
    Join Date
    Aug 2021
    Posts
    199
    Downloads
    0
    Uploads
    0

    Default Re: Custom Spindle (Custom Motor!), Very Gradual Progress Planned

    It may have slipped the notice of the active fanbase for this design/build thread, but the demagnetization field was up to 94% at the 40A, 250% overload point. At this point, that means that some piece on the magnet is close to permanent demagnetization. I'm running the simulation assuming 'warm' magnets (at 80°C), which just so you know is truly very cool for magnets, so they are more prone to demagnetization than at room temperature. Given this is a more real operating condition though, it's wiser to back off from the demag point.

    What I did as well was drop from an N52 to an N42 magnet. The N42 is the most common grade of magnet outside of extreme performance motors, which usually have N48-N52. The N42 is a little cheaper and more readily available in small quantities, but most importantly modern N42 magnets are physically durable and available with full strength at ultra-high temp chemistries. This means that you could get an N42 magnet that can operate strongly at 200°C. Given that I'm targeting an <65°C touch temperature on the casing for the motor and I also want to reduce thermal expansion effects or heating of the bearings, I'll continue to limit the operating temperatures. This will also decrease the cost of the magnets. I'll likely opt for a 120°C or 150°C rated N42H or N42EH grade, rather than the N42UH grade. The temperature relative to the naming is totally dependent on the supplier, so N42H doesn't mean the same thing from one supplier to another.

    As I dial this in, I will lock down specific electrical steel grades as well as magnet suppliers and use the material data instead of the generic data. In general the generic data will get us within 10% of our final form. In order to make the final transformation we'll have to find the Crystal of... wait, that's a different thread. In order to finish out the design we'll incorporate 2-3 material specific B-H curve options, target 80% max demag field at the peak operating point, as well as >97% efficiency at the nameplate data point (knee point, rated power). This will give us a very cold running motor that would likely not even need active cooling excepting that we're concerned about maintaining dimensions.

    Anyways, dropping down to the N42 required some length increase, which I'm perfectly happy with. I also had to thicken up the magnets, the N42 requires about 3 to 3.5 times the airgap thickness to achieve these numbers, vs 2 to 2.5 times for the N52. This is one area where the high grade magnets sometimes come out a little cheaper (in high volume), due to smaller material usage for the same performance. (Typically they are used to increase performance, not maintain it). Since we aren't going to be ordering magnets by the thousands at this time, N42 is the butter for our bread.



Page 2 of 2 FirstFirst 12

Tags for this Thread

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •  


About CNCzone.com

    We are the largest and most active discussion forum for manufacturing industry. The site is 100% free to join and use, so join today!

Follow us on


Our Brands

Custom Spindle (Custom Motor!), Very Gradual Progress Planned

Custom Spindle (Custom Motor!), Very Gradual Progress Planned