View Full Version : Building My First Router

06-21-2003, 09:16 PM
OK Im a Machining guy. I run the Engineering department in a machine shop and do a lot of the programming. Ive wanted a small machine for home for a while and well after joining this forum, Ive gotten the hair brained idea that I want to build one. Ive read the warnings to start small and cheap, so the 4 foot by 8 foot design is out :D Obviously I have access to machine tools so a nice sturdy aluminum gantry design is what I have in mind.

I am hoping You folks will be willing to help me along the way. I am comfortable with the mechanical and programming parts of this endevour, but when it comes to the electronics and motors ... well lets just say I learned more in the last month reading these threads and links than I knew before. Translation, I know enough to be dangerous ;)

Any ways, Im thinking travels:

X= 2 ft.
Y= 3-4 ft.
Z= Approx. 10 in.

I want to be able to cut aluminum and wood.

I came up with some Aluminum Extrusion that was scrapped due to not meeting specifications. It is a custom extrusion that measures 4.5" square with top and bottom walls .75" thick and side walls that are .19" thick. I figure to build the frame and some of the gantry/column from this.

I am in the process of accquiring pairs of linear guides for XY and Z.

I have an old 486 /66 PC that 3 years ago just needed a hard drive (hopefully thats still the case). I intend to run TurboCNC on this.

I have done a preliminary arrangement layout, but Im waiting for the guides to lock in the dimensions. When I get them I will post the plans.

Ive done some reading and as of now Ive concluded Servos are the way to go if I want some power and speed.

Given my schedule (and budget), I expect this project to last into the fall.

I would like to thank those who have posted their designs and experiences as they have really helped me so far.

06-21-2003, 09:23 PM
Hey sounds good. You will do fine. keep posting and we will try to help as much as we can!


06-22-2003, 01:06 AM
Were behind you 100 %

06-23-2003, 12:13 AM

SO Im working on getting the linear guides and my preliminary arrangement is in order. I think Im going to decrease the Z axis target travel a bit probably 6-8 inches. in favor of rigity and weight considerations.

Ive done some more research on the drives and motors and Ive concluded that I will try to find some used servo's and use Gecko 320's. From what Ive read so far, it sounds like these are pretty flexible and will do what I want.

Ive located some used Servos that unfortunately I have limited info on. What I have so far:

They were manufactured in 1989.
They are rated 5 amps / 43 Volts / 200 Watts
5V DC 200 c/t resolution Encoders Are Attached.


1. I'm assuming these are brushed motors (is this a mistake ?)
2. Do these sound reasonable for the machine I am planning.
3. Do you see any problems with integrating these components ?

I am trying to get more specs on the motor, but no luck yet.


06-23-2003, 07:24 AM
If they are indeed brushed I think they should work great.


06-23-2003, 10:54 AM
If you plan to use the motors with a G320 then you need to make SURE that they ARE brushed! The motor specs sound great! Good luck finding used Gecko drives. I have not seen any but I never really looked before :)

06-23-2003, 12:01 PM
Hi Dave,

I run the engineering department in a machine shop in Dayton, Ohio. I too am familiar with machining and programming, but needed some help on the electronics (still need that help).

I started my machine a year ago. Got my steppers running first because I knew I could do the design/machining on the mill.

Servo's are something I hope to move up to soon.

Good luck.

06-23-2003, 12:16 PM
Thanks Bill,


I probably wasnt clear in my post. My thoughts were to get used mechanical parts and servos, but buy the geckos and all the stuff that will go in the control box new. I figure everything in that box is where I would likely have the most problem, so new components with documentation will give me the best start. On the otherh and if I blow up $400 worth of Geckos, I wont be a happy camper either. :eek:

Actually, your website convinced me the Geckos were the way to go and I think that will be money well spent since they sound like they leave some easy options for up grading the servos or the Turbo CNC if that becomes neccessary down the road.

Still unable to get additional specs, but spoke with a technician at a repair shop who knows the motors and he confirmed they are brush motors. Looks like the price may be right.

06-24-2003, 06:51 PM
I have another question. :)

I wasnt going to worry about the electronics stuff for a while, but I came across a Power supply and Im wondering if it can be made to work with the motors Ive found.

Again the motors are rated 43 VDC / 5 amps each.

The Power Supply is rated at 48 VDC / 31 Amps with an input voltage of 115/120 VAC. It is enclosed in a case and appears to be from some sort of telecommunications system. I was thinking I could leave this as it is and were it into a homade control box that just has the Geckos in it.

1. Is this rediculously over sized for the application ? The price is right. And you cant have too much power ... right ?? Dave;)

2. I havent seen Transformers or powersupplies rated specifically at 43 volts. Is 48 V. OK ? I thought I saw +/- 5% in the motor specs (dont have them with me at the moment) Do I need to drop the voltage down from 48 V ? If so, What is the easiest way to do that ?

Thanks Again, Dave

06-24-2003, 07:45 PM
Yes, Your motors will accept a range of voltage. They were just speced at 43VDC. That supply sounds good to me.

06-24-2003, 07:59 PM
I think at the gecko site it says the power supply can be 5 volts over the rated motor voltage.


06-24-2003, 08:06 PM
Wow !

You guys are fast.

I really, realllllllly apreciate the help.



06-24-2003, 09:56 PM
Wee aim to please ... You aim to ... PLEASE.


07-14-2003, 09:00 PM
OK Im back ...

Got my linear bearings purchased just making final decision on X axis length since the rails I was able to afford wont give me my originally planned X axis travel. Im thinking maybe 24" x 24" is more suitably sized for my work area anyway.

My original motor deal fell through, but I have found some more and I have a question similar to my previous question.

The motor is rated 32V 7 amps 180 Watts.

Can I safely run this at 48 Volts ? Or is that pushing it ?

And regarding resolution, is the theoretical resolution (mechanical play aside) as simple as: screw pitch / encoder count per rev ?

so a 200 count / rev encoder on a .2" has a resolution of .001"

Or is there another factor Im missing ?

Thanks Again


07-14-2003, 09:21 PM
There is no need to run them at 48 volts. I would use a 120 to 24 volt transformer. Rectified it will give you 33.6 volts and you will have plenty of power at 7 amps.

The resolution is actually 4 times the encoder PPR devided by the pitch. The gecko320 multiplies the pulses from the encoder by 4 (I think it counts the edges of the pulses and there are two channels)


07-14-2003, 09:33 PM
Thanks Balsa Man.

I bought that 48 V. Power Supply (dirt cheap) before I actually bought the other motors and well as it turns out, I wound up not getting the other motors so I have this really "Spiffy" 48 volt power supply that I wanted to make sure I really cant use before I donate it (or almost donate it) to another CNC Zone'ers project :)

Thanks for the quick response, and I guess I will not be too concerned with the powersupply for now.

BTW, How do you like your linear guide set up. How wide are they ? I have the THK 20 MM wide guides for my X and Y axis.

07-14-2003, 09:41 PM
Well, if you have the powerwsupply already, I would try it. Monitor the situation for heat tho.

My linear rail system works great. 20 mm sounds about right. I need to order new seals (wipers?) for mine as mine are used and the seals have seen better days. I put all the bearing blocks with the better seals on the long axis where most of the dust ends up.


07-17-2003, 12:42 PM
I figure i better get some picture going so heres the major components I was able to russle up. So far Im on budget :D

I decided to go with the Round shafts and Ball bushings on the Z to save $$. I figure dflection on the Z guides wont be an issue with only 6-8" of travel. I also figure with the design Im considering, adding Z travel will be easy if the machine seems rigid enough once its up and running.

Next Pic you see will be my Cad Design which I can finish now that i have most of the major components nailed down.

Ive Also decided to pick up a little dril / mill / lathe for home. Since I have virtually no machine tools at home. I will do some of the machining at work after hours, but I really want to keep that to a minimum.


07-21-2003, 07:24 PM
Received my motors today, I was a little nervous because I couldnt get detailed specs on them aparently they are OEM spec motors and the manufacturer cant release any data on them. Looks like they are brushed with 250 pulse per rev encoders. I had to take the encoder apart to determine this. I need to find someone local who can give these a quick test for me, but they look real good. Back to the CAD station.


07-28-2003, 10:25 PM
Well I had an electric shop that does alot of motor work give my motors a spin and they run real smooth. So I guess the ebay thing wos a good choice. Certainkly saved me some money.

I am abandoning the 48V power supply and I will build one as Balsa man suggested. I am embarrased at how poor my electronics knowlege is. I did have several electronics classes in Engineering school, but that was a long time (relatively) ago. I dont remember much more than V=IR. So I went and picked up "The Art of Electronics" at borders yesterday. I think its written in greek (just kidding). This router project is really teaching me alot. Much of what Im learning is from this forum. My concience got the better part of me so I did pony up a few bucs for the cause.

Other than that Im just trying to find time to complete the design which I will post here for critiquing (sp?) when it is further along.

I am currently working out the ball screw mounting details. I have decided to go dirrect drive but keeping a timing belt arrangement in mind as a possible option down the road.

07-29-2003, 12:27 AM
What pitch ball screws?


07-29-2003, 09:07 AM
Pitch is .203.

Youre gonna tell me Im way too far down on the power curve arenít you ?

I was trying to save a few bucs by not going with the reduction, but Im looking again and 20 IPM is only 100 RPM direct so maybe I need to rethink that. Or consider higher pitch Acme screws.

07-29-2003, 12:36 PM

It will work direct I am sure. This is how I run mine and it's fine. On the other hand if you want to fully utilise the motors (which probably go 2000 rpm) a belt reduction would be better. You should aim for 150 in/min at least. What is the diameter? I was thinking maybe you had a 1 or 2 pitch screw.

Try it direct, but plan ahead for posible reduction (as you said you were doing).

Keep plowing ahead! you are doing fine!


07-29-2003, 01:15 PM
Hey Eric thanks for the input.

Are your screys 5 TPI or 8 TPI ?

Im preliminary thinking 5/8 Screw dia X .203" pitch. If cost wasnt a consideration I would go with 3/4 or more probably, but the price (as was mentioned in another thread) of these screws from mcmaster is pretty darn good. Im going to run a few shaft calculations before I lock it in.

I intend to use thrust bearings at one end and radial bearings at the other end of the ball screw so I think I can minimize any eccentric loading on the screw. Using the 150 IPM baseline, thats about 740 RPM. Which Im thinking should be no problem on a 24" axis. I will see what the numbers say though. I dont want to buy the screws twice !!! they wont be that good of a deal then.


Speaking of which, did you isolate your screw vibration problem ?

07-29-2003, 02:01 PM
Mine are 3/4-6. Your screws will be great on a 24" axis.

My x axis still is noisy, as the scew is 48" long. It's much better now since I went back to the amco nuts with the saw cut antibacklash system I am using now. The 3/4"-6 acme tap made for very little backlash but there was probably 50 thousands "axial"? play in there which made for some nasty side to side whipping in the nut. Now that I can squeeze the nut a little its fine. I need to borrow a dial indicator because my screw is bent a little. I will put some masking tape over the threads in the center and use the dial indicator to see how much. I think I can tweek out this bend and it will be a lot better. Still working on the limit switches in between cutting some parts, so after the limit switches....

I was lucky with my direct drive setup because my motors are 115 volts. On 75 volts they go around 1000 rpm so that means 150" per minute is almost the motor maximum. If I would have gone belt drive my max speed would have been limited. Quite by accident it worked out for once! So far the motors stay very cool and have more than enough torque.



08-03-2003, 02:36 AM

I have had three bent 1/2" leadscrews that I have straightened. The first one I screwed up and now has a compound bend in it because I wasn't methodical about it. It's scrap now. The other two came out alright.
Here is how I did it:
Make sure that you know which way your leadscrew is bent and how much. I put them on v-blocks each end. Then I used some .010 brass shim stock that was attached to another block and sort of bent over the rod. I set the indicator on that. Then I took readings every inch and also marked the high points with a marker.
Make a graph of your readings. If it is a simple bend your readings will decrease consistantly as you get to the ends. Your marks of the high points should also be in a straight line. If the marks tend to go around the rod then it will be a harder thing to fix.
After you know how the rod is bent then you can plan on how to straighten it. I used the v-blocks again. I took a c-clamp and a block of wood. I tightened the clamp 2 turns and checked the reading again. Then 2 1/2 turns and checked it again.
The second rod bent at 3 1/2 turns of the clamp. That was my 10" z axis leadscrew. It was out about .008 and is now at about .002. That's close enough for now.
I know that this is the long answer but I found that it took some real analysis to get it right.

Good Luck,

08-03-2003, 03:58 AM
Part of the trick to straightening bent shafts is to not set your V blocks too far apart.

We do lots of hydraulic cylinder shaft straightening in my shop, using a hydraulic press. We use aluminum V-blocks to prevent marring of the surface. We use 3 V-blocks, two underneath the shaft on each side of the bend as support, and one under the press ram where the pressure is applied.

Anyway, the trick is that the spacing of the supporting V-blocks should not be any more than about 8 times the diameter of the shaft.

A bend in a shaft has a significantly increased stress strength in the long side of the bend. If your V-blocks are too widely spaced, the original bend will have enough strength of its own that it will rebound more than the less stretched areas farther from the bend. This will create a "dog-leg" bend, even if you are bending in the proper direction at the proper place.

Compound or dog-leg bends can be removed as well, but you will have to overbend one side of the bend, so that you get the shaft back to one simple bend. Then, moving your V-blocks back to the recommended seperation, and working your way both sides of center of the original bend will get you back to straight.

Use a straight-edge for guidance until you have it very close to straight, then use the dial indicator. Using the dial indicator too early in the process can be a source of great frustration.

08-03-2003, 01:19 PM

Thanks for the reply. I was hoping to get some professional feedback. It was quite an exercise for me to figure out how to straighten my leadscrews.
I was wondering as I did it about how far to space the support blocks. I guess I managed alright but I could see that there was alot involved.