How accurate?

[bearwen]

I have downloaded the plans for the JGRO. I have a few questions 1. how do i determin what size motors I need. 2. how accurate is the jgro machine? I am not sure what material I will build it out of MDF or ply or solid lumber maybe angle iron and weld it up not sure I am wanting to build this JGRO about say 2'x3'x1' I will use this machine to do wood carvings, foam carving and some metal work( the metal work mainly will be maching parts for a bigger and better CNC) Please break it down to me I have no parts as of yet need to know how to figure what size motors are needed and what control board to use ( I would like to purchase the control board and larger motors for the bigger machine and reuse these parts once I get the hang of this). Ok the long winded question is done.

Thank You JGRO for the plans!

[ger21]

1. I'd say use no smaller than 200oz motors, and a HobbyCNC or Xylotex as the minimum. Both sell complete kits with motors and drives. Bigger motors can get you more speed, but when you need bigger drives, the price starts going up fast.

2. As accurate as you build it. Accuracy can be tough to define. IT depends on what screws you use to drive it, how accurate (straight and parallel) the bearings are, how much flex does the machine have, .....


The JGRO is really not suitable for cutting metal, imo. It has too much flex due to the unsupported rails.

No one can tell you what size motors you'll need for a bigger machine, without having a lot of info about that machine. More than likely, if you buy Gecko drives now, you can use them later. But, I'd go with the low cost option mentioned above first, and sell it when you decide to upgrade.

[bearwen]

Ger21 what would you recomend I dont really understand the motor size I keep thinking oz/in but I cant put this together in my head as to how much torque or holding power you need for a cnc Im not quit sure how it all plays in. As for crews again recomendations please I would like to build this thing as accurate as possible but a reasonable price I am thinking about building the jgro at about a 2'x3' size and build it out of wood with extra supports for stability or metal as I have welding equipment and plenty of scrap angle iron recomendations please. and bearings again recomendations. I know I am asking for everyone to build it for me but I figure you guys have been doing this for a while and there have been many built and built different ways so I figure learn from other people.


What are the average accuracys that people are getting with the jgro?


Berry

[spalm]

Berry,

You keep mentioning accuracy. The JGRO is a starter machine; accuracy was not its intended purpose. Go ahead and build it, it is easy to build and cheap. It is not accurate and not fast, but you will learn a lot. You can actually cut stuff with it, though. Buy some 200oz/in motors and controllers as Gerry suggested. These will be big enough to re-use when you get around to using your welding skills for a more stable machine. There is still a lot of software and other concepts to learn. Gotta start somewhere.

Steve

[Mcgyver]

since these guys are giving you hard time on accuracy, I'll help you out. the answer is 42.

Steve and Gerry are right; asking about accuracy is like asking how long is a piece of string...and there advice is good. build away and by end of the project you will understand why there is not a simple answer and you'll start thinking of accuracy in terms of repeatability, resolution, backlash compensation, feed screw quality, bearing quality, deflection under load, etc etc.

As someone else who is also on the cnc learning curve, don't get stressed figuring it all out in one night. to be really really good at this from diy perspective requires fairly advance electronics and machining/mechanical knowledge and abilities, these take time to develop as its a lucky few that already have both nailed down. a project's the best way accelerate up the curve imo.

having said that, if there is some parameter that you just can't live without, try to better define it, but remember that from cutting foam to metal you've just about covered the spectrum so be reasonable with your first machine expectations

[ger21]

How good are you're building skills, working with wood? That can make a big difference. I personally wouldn't build the JGRO, but something like Joe's machine. It fixes some of the issues the JGRO has. But you're still going to have trouble cutting metal with a machine made of wood. Joes CNC Model 2006 R-1

[Jason Marsha]

Berry,

As far as accuracy is concerned I have never had a problem with my JGRO and accuracy. I have cut the finest fonts on my dogs tags with engraving bits and never had a problem with the fine lines meeting on the spot.

The fastest rapid move I have achieved with 5/16-18 threaded rod, 127oz/inch motors with mach3 set to 35000 pulses without whipping of the threaded rod is 70 inches/min.

With the stronger 200 oz/inch motors from Hobbycnc or 280 oz/inch from other suppliers running about 30-36 volts and 1/2-10 acme rod rapid speeds of 120 inches/min is possible. At this fast speed however the machine must be very well aligned to eliminate binding that would occur, for example, if shaft bearings were out of line.

Cutting speeds is a different story. I cut pine at 40 inches/min with a 1/4" upcut
spiral and 1/4" deep per pass.

With the present motors I have thats the fastest reliable cutting speed I would get without losing steps. The 200oz/inch motors would allow me to cut much faster.
As stated by the others flexing is a problem and will be compounded as you push the envelope of cutting faster.

I made some changes to the machine such as stronger rails on the long axis as well as supporting the long rails and adding sides to the machine.

Check out my thread, its not a build log but a pictorial overview of the ongoing use of the machine.
JGRO Pics Finally

As the guys said , its cheap and easy to build.
You will learn volumes from building and operating this machine, in time a bigger better machine can be built.

Warning : This hobby is very addictive.

Jason

[posix]

Berry,

Warning : This hobby is very addictive.

Jason

I didn't know sleepless nights drawing parts, near-miss car accidents (due to mind wandering off on a tangent thinking whether that steel pipe you just bought will deflect) and nightmares of huge falling leadnuts are called "hobby"

[Jason Marsha]

Posix,

I am not sure about the nightmares but as far as car accidents are concerned I had some near misses (running of fthe road) driving home late at night after working on the machine at my friend's woodwork shop.

Jason

[rweatherly]

I think the unloaded accuracy is very good -- if you don't load up the tool, it should be as accurate as the leadscrews, alignment, backlash, etc. My machine was all made by hand, to the drawing, and no real pains were taken to align it. I tested my machine by machining 2" squares out of 1/4" hardboard. Depending on the direction I went around the square, the dimension was out about .020" (+.020" in both dimensions when cutting CW and -.020" in both dimensions when cutting CCW -- I could have the CW and CCW reversed, but you get the idea). There is no doubt in my mind, that if the tool had little or no load, the squares would be within a couple of thousands, so if I needed an accurate final dimension, I would save a light cut for the last.

That said, I have engraved some 4" high text into wood that look perfect.

I like the machines that have the pipes fully supported, but on my machine when I manually pull the spindle perpendicular to the rails the spindle deflection is about 10x the movment of the rail, so a rock solid rail is not going to cure all problems. The supported rail is going to help, so if I were going to make another machine I would do that.