New Machine Build- Kelly's Machine Tool Warehouse - MD001

[kregan]

OK guys here we go.

The MD001 should arrive in a few days and I need to start thinking about the servo's drives. I have been really pleased with the DMM Tech AC servo's and I would like to go with them again.

The motors are rated at 3000 RPM and I would like to go with a 2 to 1 gear ratio on the X and Y and maybe 3 or 4 to 1 on the Z if it is needed.

Ball Screws would be from Linearmotinbearings on eBay.
items in linearmotionbearings store on eBay!
RM1605 (16mm) on X and Y
RM2005 (20mm) on the Z
Lead = 5... I assume that is 5mm or .20" per revolution?

Please correct my math if it is wrong.

I am thinking of using:

400W AC brushless servo motor (60EM-DHT-36)
Max 3000rpm
Rated torque: 1.27Nm(179 oz-in)
Peak torque: 3.6 Nm(507.6 oz-in)
Motor

At a 2:1 belt reduction that = 358 oz-in Rated and 1015 oz-in Peak at 1500 RPM. 1500 RPM x .20" = 300ipm

OR

Direct drive
900 Watt brushless AC Servo motor, NEMA34
Max:1080 rpm
Rated torque: 2.9 Nm(409 Oz-in)
Peak torque: 7.2Nm(1015 Oz-in)

1000 rpm x .20" = 200ipm

What do you think?

Edit:
My current machine uses the 300 watt motors at a 2:1 reduction. I can get up to 188ipm in rapids and that is plenty for me and I think it would be fine for the larger table of the MD001.

[kregan]

From the IH web site:

Our Turnkey 3 axis CNC Mill uses 410 in/oz peak motors on the X and Y axis.
The drive ratio is 4:1. So the ballscrew sees 1640 in/oz of torque.
Our math: 410 in/oz x 4 = 1640 in/oz

The Z-Axis uses a 648 in/oz x 4 = 2592 in/oz

So the 507 in-oz motors from DMM at a 3:1 ratio would give me 1521 in-oz.
With a 5mm (.20") lead screw that equals 3000 rpm / 3 = 1000 rpm.

1000 rpm x .20 = 200ipm rapids.

[Starleper1]

It's been discussed and argued about a million times on this forum, but me personally, I'd prefer a slower machine that is reliable, rather than a really fast machine that faults all the time.

I'd take torque over speed any day.

Your math seems right though.

[cyclestart]

Lead = 5... I assume that is 5mm or .20" per revolution?

5mm or roughly 5.08 rev/inch if setting up in standard measure. Always willing to tackle the easy questions LOL.

I have the same machine and may take it cnc someday. Best of luck.

[kregan]

To clarify the motor I listed from DMM is the most powerfull they list. So while I am not looking for 200ipm rapids it looks like with the math I did I have some overhead in case I need to go to 3.5:1 or 4:1 to get the torque I need.

If they had 800 in-oz I would just get those and not sweat it!

[sdubfid]

Ive got the 300w dmms and my friend has the bigger ones with internal gearing. He is really happy with them. So powerful he said he bent his router a bit with them haha.

I was at hui's house last weekend picking up some extra lmit switches and called him with questions a few times and he has always helped.

I used to have an md001 and I would just run the nema 34 1000w motors direct drive. That would be a nice simple setup.

[ninefinger]

To give you an idea of the minimum required to make the MD001 move, I used (and still am using) the 50oz-in continuous rated Nema 23 servos (350oz-in peak) on the X and Y axis with 4:1 reduction (ie 200 oz-in at the screw neglecting friction loss) and 5mm lead ballscrews form linearmotionbearing2000 on ebay. For the Z axis I started out with the same nema 23 servo and a 5:1 reduction (ie 250 oz-in at the screw). I was able to run it this way for a while but I had to have the acceleration dialed back on the Z, mostly due to tight spots at the top and bottom of the travel. I've since changed to a 119 oz-in (600oz-in peak) and I gained a little bit of acceleration.

Personally the 400w servos are what I'd choose, unless going with direct drive and larger servos can save enough money to pay for themselves (1 belt and 2 pulleys per axis plus additional brackets versus a coupler for direct drive). Looks like the encoder is suitable for direct drive (12 bit accuracy ~ 4000 counts/rev @ .2" rev = 0.00005" )
Checking out the pricing it looks like it can - the high torque servo & drives are only $48 more than the 400w servos & drives.

Mike

[kregan]

Mike,
I have been following your conversion. Great job!!!

I have been talking with Mr. Li at DMM and he also agrees that the 400 watt system will work on all 3 axis.

I like the versatility of the belt drive and would rather spend the extra time building it then being locked into the direct drive. The numbers shake out the same anyway.

I plan on ordering them soon.

Originally Posted by ninefinger To give you an idea of the minimum required to make the MD001 move, I used (and still am using) the 50oz-in continuous rated Nema 23 servos (350oz-in peak) on the X and Y axis with 4:1 reduction (ie 200 oz-in at the screw neglecting friction loss) and 5mm lead ballscrews form linearmotionbearing2000 on ebay. For the Z axis I started out with the same nema 23 servo and a 5:1 reduction (ie 250 oz-in at the screw). I was able to run it this way for a while but I had to have the acceleration dialed back on the Z, mostly due to tight spots at the top and bottom of the travel. I've since changed to a 119 oz-in (600oz-in peak) and I gained a little bit of acceleration. Personally the 400w servos are what I'd choose, unless going with direct drive and larger servos can save enough money to pay for themselves (1 belt and 2 pulleys per axis plus additional brackets versus a coupler for direct drive). Looks like the encoder is suitable for direct drive (12 bit accuracy ~ 4000 counts/rev @ .2" rev = 0.00005" ) Checking out the pricing it looks like it can - the high torque servo & drives are only $48 more than the 400w servos & drives. Mike

[ninefinger]

I never noticed on your last build if you used his complete kit or just the servos and drives. Are you planning on the complete kit this time or just the drives and servos? I'm just curious about the rest of the parts. I recently ordered some proximity switches from them but have yet to hook them up. They seem quite nice for the price though - hard to beat $10 each for a prox switch.

Looking forward to your build - should be a whole lot easier with your cnc already.

BTW - if your planning on ordering from Linearmotionbearing on ebay - my experience (and a few others) has been that the ballscrews are great (and get the ends machined or at least the drive end) but make your own end supports / bearing blocks - theirs have too many issues to be worth the price.

Mike

[kregan]

I used the full kit on my first build and will do so again on this one. I was so new to CNC I thought it would make life easier since it comes with power supplies, and tune up software. The software is used to tune the individual drives and test the motors etc. via comm port. Very handy to have.

I also used the Liniarmotionbearing ball screws on my first build but I made my own bearing mounts and brackets.

Originally Posted by ninefinger I never noticed on your last build if you used his complete kit or just the servos and drives. Are you planning on the complete kit this time or just the drives and servos? I'm just curious about the rest of the parts. I recently ordered some proximity switches from them but have yet to hook them up. They seem quite nice for the price though - hard to beat $10 each for a prox switch. Looking forward to your build - should be a whole lot easier with your cnc already. BTW - if your planning on ordering from Linearmotionbearing on ebay - my experience (and a few others) has been that the ballscrews are great (and get the ends machined or at least the drive end) but make your own end supports / bearing blocks - theirs have too many issues to be worth the price. Mike

[dwalsh62]

Originally Posted by kregan From the IH web site: Our Turnkey 3 axis CNC Mill uses 410 in/oz peak motors on the X and Y axis. The drive ratio is 4:1. So the ballscrew sees 1640 in/oz of torque. Our math: 410 in/oz x 4 = 1640 in/oz The Z-Axis uses a 648 in/oz x 4 = 2592 in/oz So the 507 in-oz motors from DMM at a 3:1 ratio would give me 1521 in-oz. With a 5mm (.20") lead screw that equals 3000 rpm / 3 = 1000 rpm. 1000 rpm x .20 = 200ipm rapids.

Using the theoretical values is not considered good engineering practice, practical values are 0.92 of the theoretical value and is the preferred value that should be used when calculating power.

Also, the 5mm metric screws only have 0.196in lead, not 0.200in lead, a common mistake made by many it seems.


If you want a 0.200in lead screw then I recommend you buy an appropriate screw.

[kregan]

Very interesting... My current machine uses the 300 watt motors rated at 3000RPM and the same 5mm ball screw lead.

Max rapids are 187

200 IPM x.92 = 184

Originally Posted by dwalsh62 Using the theoretical values is not considered good engineering practice, practical values are 0.92 of the theoretical value and is the preferred value that should be used when calculating power. Also, the 5mm metric screws only have 0.196in lead, not 0.200in lead, a common mistake made by many it seems. If you want a 0.200in lead screw then I recommend you buy an appropriate screw.