MechMate CNC Router Forum

MechMate CNC Router Forum (http://www.mechmate.com/forums/index.php)
-   MechMates already cutting (http://www.mechmate.com/forums/forumdisplay.php?f=78)
-   -   Here's You Sign #98 - Pennsylvania USA (http://www.mechmate.com/forums/showthread.php?t=3566)

pblackburn Sun 30 October 2011 08:45

Here's You Sign #98 - Pennsylvania USA
 
I have been laying out and planning for about a year now. Steel is ordered and will be here on Wednesday. This will be a winter project. Finally have the 3d cad design laid out. The machine will be bolt together with alignment pins to make ease of moving the machine at a later date between locations (about 5 years away). Have machined all the parts that need machined. Made most on manual mill and lathe but did some on a cnc lathe. Heat treated all that needed to be heat treated.
On the longitudinal rails I am opting for V rail instead of grinding. My thoughts are I can make the wheels by the dozens at a RC40 hardness which is much less than the RC60 hardness of the V rails. Should provide a long lasting and smooth traverse. However it does add to build phase mounting the v rails..
My plans on the electrical are to use the PK296A2A-SG7.2 servos with the Gecko 203V connected to the PDMX-126 + PDMX-134. Mostly so I can use a solo 24VDC power supply to supply the inputs and outputs on the machines since the 126 can use 24vdc. I plan on Mach3 for the HMI. Plans are for using inputs for homing, soft limits, overtravel estop and tool height touch probe. My original plan was to use a Antek PS-5N38 power supply but I am leaning now toward the PS-5N35.
I have had the laser cut steel for about a year now. The electrical panel is a 24x36 Hoffman with disconnect. All of the remotes stations,parts, cabling and most hardware have been acquired. Have a spare workstation computer and am purchasing a touchscreen monitor for the HMI. I still need to make the aluminum heat sink for the drives.
As I do with most things I start. I have read, pondered and hesitated until I feel the most appropriate way to proceed. I hope to keep you posted as time permits. I appreciated any comments and criticisms that you my have.

Richards Sun 30 October 2011 09:37

Pete,

With the PK296A2A-SG7.2 motors, you can use a power supply up to about 45VDC. (Gerald has reported running those motors at higher than 35VDC with excellent results.) The reason that THOSE motors can be run at higher than normal voltages is because Oriental Motor artificially reduces their current rating from 4.5A to 3A on those motors/gearboxes because of the gearbox.

Rough calculations show that the motors can handle 150 Watts. (The electrically identical PK296-3A motor is rated at 4.5A and 35VCD X 4.5A = 157 Watts.) The A2A motor would only produce about 100 Watts when run at 35 Volts and 3A. At 45 VDC, the motor would produce about 135 Watts.

The old rule of thumb is that the higher the voltage, the faster the speed and the higher the Amps, the greater the torque. At 35 VDC, you will have all the speed that you need on a CNC machine. At 3A, you will provide all the torque that the gearboxes can withstand, so a 35VDC power supply is adequate to do the job you need done; however, if a 35VDC power supply were not available, then a 40VDC or a 45VDC power supply could be substituted.

pblackburn Sun 30 October 2011 16:30

Mike,
I am still trying to grasp the servo math. Not that I don't understand math but I am a person who has to fully understand how something works for it to make sense. I have worked since 97 as an electro-mechanical technician but have had limited exposure to servos and their drives. Most times it is only just replacing the motor itself but never having to adjust the design. Most instances the servos and the drives are set at optimum unlike other processes. In a maintenance position it is important to have a firm grasp on the technology you are working with or it is almost impossible to troubleshoot it. It kind of reminds me of a variable frequency drive and a inverter duty motor. I just need more time to study it but it would also help to have the unit set up so I could troubleshoot the process.

smreish Sun 30 October 2011 17:43

Pete,
We use stepper motors - not servo's.
Steppers move with a dedicated resolution without the need of feedback - unlike inverter drives or VFD's you might be used to.

Good reading might be here

pblackburn Tue 01 November 2011 14:56

Thanks for the link, I have also read this as well on the Gecko website. I am more looking for how the step occurs and how the field controls and stops the step. I well disassemble one when it fails to better understand it. Thank you both for the information.

bradm Tue 01 November 2011 16:23

Just don't disassemble a modern stepper motor that you want to use again; you'll ruin it.

pblackburn Tue 01 November 2011 17:13

I understand that. I did refer to when one fails so no harm will be done since it is already broken.

Gerald D Tue 01 November 2011 23:51

Old type magnets are potentially ruined when opening. I have replaced failed bearings in a modern stepper without apparent loss of torque.

KenC Wed 02 November 2011 01:09

There really isn't much to see inside of the stepper motor unless you undo the windings... modern hybrid stepper motor can be open-up & go-back without much hassle, just watch out the clearance between the stator & rotor, they are real close...

pblackburn Thu 03 November 2011 15:36

I would just like a better understanding of how the steps are made and locked by the motor. The theory is there but the sight is not. I have spoken with some of the engineers I work with about them but they have not used steppers in so long because servos are mostly use on the items we work with and build for inhouse projects. They tend to like the encoder reference for absolute position where as with a stepper it is an open system mostly used without feedback on position but not always. Some of the machines we have, have steppers with an encoder. Obviously you do not want to tear down a perfectly good motor for understanding of how it works but for the most part in industrial maintenance it is the repairing that teaches you how something works and not the theory because theory and practicality do not always sync up. The problem is that you do not repair a motor as it is not cost effective unless is a specialty motor or is large enough that the cost of new is more than a repair. This is not the case with servos and steppers so we have always bought new. I would like to know for my piece of mind and understanding. I hope this explains my post a little better.

bradm Thu 03 November 2011 19:32

Oh, we understand and appreciate engineering curiosity around here. Check this out for a photographic stepper teardown. Note the ridges on both the stator and the rotor. These define the step/lock geometry.

Gerald D Thu 03 November 2011 23:49

Pete, your logic is sound.

In the old magnet type steppers (aluminum/nickel/cobalt "alnico") the motors were magnetized after assemby, and could be re-magnetized after repair. . . . . if you could find the right equipment. A device that whacked hundreds of Amps for a couple of milliseconds - long enough to align the magnetic particles, but not so long as to melt everything.

From what I understand, newer steppers have rare earth magnets and I don't know if these are magnetized after assembly, but I doubt it.

The world's stock of "rare earth" is rapidy running out. It is mostly located in China and for obvious reasons they are protecting their resource. So, the type of magnets in our steppers tomorrow, and their handling, can be different again to what we have today.

KenC Fri 04 November 2011 02:43

Actually, Australia has the biggest deposit of rare earth, since Aussie government discourage refining of rare earth due to the handling of radioactive rasidue, they are trying to send their raw rare earth to my country for refining & dump the radioactive left-over in my backyard... & They say 1mm thk PU can contain all the leach... LMAO!!!

pblackburn Sat 05 November 2011 14:23

Thanks Ken, Gerald, Brad, Sean and Mike for your input. It is greatly appreciated.

pblackburn Sat 26 November 2011 19:59

Base almost complete
 
I am almost done with the base (table portion). Starting to organize everything for machining down the rails to the correct height for the V-rails to be mounted. Induction hardened V-rails are ordered and will be pre-drilled and tapped. Have some shimming that will be required on the X rails to set them level/plumb. I checked all the steel suppliers and had to opt for rails that have a lazy C. Since my build is bolt together, the channels are straight but the flats are not parallel. The base has been set with a machinist level. Is it better to (1) shim or (2)drill the rail in multiple places on the trailing edge and use fine thread set screws with jam nuts to set the plane. Access to a 10 foot surface grinder would be great but I do not like the added cost of that.

smreish Sun 27 November 2011 04:24

Pete,
I have experienced this before on one of my builds. Since that table was bolt together, I opted to align it like a car. I actually "toed" the leg assembly to cause the top flange to rotate about the x axis and get into plane. The end view would look like a trapezoid /__\ or \__/ when finished, but kept me from re-preparing the mating surface of the top of the C channel for the angle iron rails. I think my flanges were out about 3 degrees. The other very easy option is shims. Adding them to the outside edges to get it all in plane is pretty easy too....I just preferred to alter the table.

....
I did read on other person here on the forum actually putting a skim coat of Auto Body filler on the top of the channel and skimming a level layer with screed across both members to get a uniform datum to mate to.

Good luck....Show us a picture with reference (like a machinist square) to show the affected area for consideration.

pblackburn Sun 27 November 2011 08:09

Sean, will do. please give me a bit of time.

pblackburn Mon 28 November 2011 10:42

1 Attachment(s)
Sorry for the wait, Sunday is about the only day I have with the kids and they come first. The table (crossmembers) are about 0.003" (0.0762mm) out from corner to corner. The longitudinal channels are within 0.0005" (0.0127mm) over the length. I was going to use a machinist square to show you the angle but that would not represent the problem accurately. The bottom of the C channel is mated and holding the table within an acceptable tolerance I believe. Being out so little will help when I have to surface the table. The C channel has a U appearance to it so both the top and bottom faces are toed in. I have compensated for this to ensure I have the proper travel on the Y axis. I have never seen precision C channel from a forge but I know you can order precision ground channel but the cost is outrageous. Anyways, the best way for me to correct this is to have it surface ground, problem is the closest machine shop to do this is 200 miles from here (we use them where I work for some of our machining), next option is to rent a linear mill (about 700 a week), next option is to shim or use leveling screws (only problem I see with that is the screws may tend to grind themselves in over time, I don't know because I do not know the vibration that is generated by the machine). That is where the question came from. Here is a picture with a straight edge. In this picture the gap is 0.0735" (1.86mm) and the other rail with the same problem is 0.122 (3.1mm) at the very edge. I have precision shim banding so shimming can be done. I know the amount of shim will be less since the X-rails are offset on the channel, I am looking for alternatives to make the best decision. Thanks for your time.
Attachment 12717

JasonC Mon 28 November 2011 10:52

With the level of precision your going after what are you going to be using it for?


Jas

pblackburn Mon 28 November 2011 11:02

Jason,
I am just used to precision with my line of work. Trust me when I say it is hard not to be, when I built my home I was way to precise and drove a lot of people insane including my father who is a carpenter. But that is wood and it is forgiving, this is steel, the more accurate the base the more accurate the machine. This will be a router and mostly 3D items. I have seen how a slight variance can destroy a bearing so my thinking is the closer all the components are to true, the longer the longevity of the machine. Less downtime equals more profit. I hope that answers you question

JasonC Mon 28 November 2011 12:25

I got ya. :) I use to be that way until my college instrutor told me "time is money". I think he was trying to get me to think in terms of being realistic to the project. I have figured that in this case of building a homemade cnc machine there will be a little give in percision just keeping the costs down. With so many pieces to be welded and the movment involved with the heat of welding, and with the materials being used for a affordable build...it will be a challenge.

But good luck.

Jas

pblackburn Mon 28 November 2011 14:53

Thanks for the best wishes, I am lucking enough to have access to a lot of things the average person would not so that helps with cost. This helps me in a lot of ways. The use of industrial equipment and machinery has saved me a lot. As far as a time constraint, I do not have that. I started planning last December, started buying and building select parts over the past year. Now I am at the phase that I can build non-stop until it is complete.

pblackburn Fri 03 February 2012 18:18

V rail
 
Well it took more than 2 months but the linear v rail finally arrived. Have been in a hold status because of it.

JasonC Fri 03 February 2012 20:53

ugh I feel your pain. On my V-rails that im mounting to... the L channel... I put them on my mill and cut a flat surface for then to sit on, a the top and in the radius inside corner. They sit on there beautifly.

I have been picking at the project but have been on hold with work driving us all into the ground.

Jason

pblackburn Sun 05 February 2012 13:23

I did something similar. I took them to the machine shop where I work and they trimmed down the 2.5" angle to the proper height for the v-rail to sit on them. They did this for me while I was on vacation. My v-rail came pre-drilled for 1/4-28 every 3 inches but not tapped to make easy transfer of the holes to the mounting surface. I got them through a local bearing supplier and the cost of the 10 foot rail was less than a 6 foot from McMaster Carr (which is not pre-drilled). They are induction hardened 58RC and the wheels I made I had them hardened to 40RC. I would rather the wheels wear than the rails. I am thinking of mounting brass brushes to clean the rails of debris.

pblackburn Tue 03 April 2012 19:20

Well X is finally done, now starting the gantry assembly. Off to do some research

smreish Tue 03 April 2012 19:57

What kind of research? Instructions on gantry fabrication?

pblackburn Wed 04 April 2012 14:54

Yes. The angle that needed corrected changed the original planned distance so now I need to figure out how to make the gantry. I feel like I need to start with the distance between the rollers (center to center) and work backward to find the length of the 2 x 4 tubing. I am sure there is a crucial distance for the mounting of the steppers that have to be maintained.

baseball43v3r Wed 04 April 2012 17:43

10 20 440, length is y + 500 mm. Is that what you are looking for? if your y length changed, just change that number. the mounting of the rails to the beams allows for some flexibility in the end length, if you are off slightly.

Gerald D Wed 04 April 2012 23:28

Pete

From old drawings, tube length is distance between centers of rollers plus 33mm. Let us know if your research turns up anything much different.


All times are GMT -6. The time now is 17:49.

Powered by vBulletin® Version 3.8.3
Copyright ©2000 - 2024, Jelsoft Enterprises Ltd.