My First Router

[Cold Fusion]

Great looking machine and great looking cuts.

[Hobbiest]

Awesome! Nothing else to say...awesome.

[Mr.Chips]

Good job.
This is what it's all about.
Hager

[cbcnc]

Mark,

That is a fine looking machine! It's an inspiration for us all. Congratualations!
Of course now you will have to live up to what it's capable of. Please post further pictures of your work for us to see.

Chris Barker

[kong]

Good work! nice to see it has all come together for you.
What did you do about the table-top in the end?

[rmtucker]

Kong

Table top has just arrived from Metalfast this morning.
Thanks for the info my friend.

Mark

[Carver]

Mark, Know you get tired of hearing what a fantastic job you've done, But you have kept many of us sitting on the sidelines smiling fiercly for some time.

Congrats on a really cool machine and on your attitude retention after the crash. I have some commercially produced routers that can only wish they had your attention to detail.

Keep us up on your advances in cutting and the new axis progress.

Best of days, and thanks for sharing,

Phill Pittman
www.masterwerkes.com
digicarve@verizon.net

[rmtucker]

Hi guys

Well i have spent about a week trying to tune the motors in but i am still losing steps.Can not tell if it is mechanical or the electronics but i tend to sway towards the electronics.It seems like a very gray area this,How to wire the inside of your controller.The reason i feel it's electronic is that it does not matter what i set acc/vel to in mach2 i am still losing steps.Or am i just expecting too much.Any advice that anybody could give me at this point would be much appreciated because right now i got £2000 of scrap metal.Can you tell i'm pulling my hair out?.

Mark

[balsaman]

It's very important that your step/direction signal cables are shielded, especially inside the controller box. The shield ground on my cable inside my controller came off and I had all kinds of trouble.

Eric

[rmtucker]

hey balsaman

is there any pics of your control,maybe i will be able to see where i have gone wrong?

mark

[balsaman]

Yeah, it's not going to help you much tho. Too full.

Eric

[balsaman]

The cable from the breakout board (little board right in the middle) to the geckodrives (on the door) is the one that the shield wasn't grounded.

Also make sure you set the software to "active low" on the step pin for the geckodrives. Otherwise you loose one step each new gcode move for that axis.

Eric

[rmtucker]

balsaman

should the dir pin also be active low ?

mark

[balsaman]

Dir doesn't matter, as it only changes when the direction changes, and doesn't affect the steps. Changing it to active low should change the axis direction.

[daytrader]

Great looking machine.

What is its repeatability in a small area, such as 2 in square? I'm looking to build a machine to do inlay work in pool cues, which I think would require a repeatability of .001in (within a small area). *Edit* If you were able to measure this with your loss of step problem.

How did you ensure the linear rails were parallel in both directions? How flat was the surface you mounted the linear rails to? Did that surface irregularity transfer to the rails? Did you level the machine before adjusting any of the rails?

Why did you decide to use two leadscrews to move the gantry? Before the leadscrews were attached, did you notice any racking of the gantry, which would dictate the use of two screws? Do the screws seem to rotate simultaneously, or do you think one may start to rotate before the other, contributing to some racking of the gantry?

Thanks for the inspiration.

[turmite]

Mark do you frequent the Mach2 user board over on Yahoo? Seems like a good bunch like here but I hate the Yahoo format.
There has been a long thread about some lost steps in one version of the Mach2 releases and you might want to check which release it is.

BTW when I sw your first cutting pic I though you had taked the kitchen table and engraved it a little!!

Mike

[rmtucker]

Hi guys

I am afraid i have neglected everybody and everything over the last 2 weeks.I had got the machine up to the final stages of building and hit some major software problems.So i am afraid the remaining metalwork and hardware has come to a standstill.The software problems involved lost steps and constant velocity in Mach2,And has still not been resolved.The trouble is that there are no other pieces of software out there offering constant velocity at such a small price.So it is a matter of hanging on to see if we can resolve this matter.

As far as repeatability is concerned (excluding lost steps),It is repeating to within .01mm which is all that can be measured with a standard clock.Gantry racking is something i thought of in the design stage and is really reliant on the type of application and width of table.In my instance the gantry is about 800mm long which would cause a problem when taking heavy cuts so i went for 2 leadscrews and they work perfectly and in tandem.Getting the linear rails parallel was not a problem beacaus the frame work they are built on is parallel to .001".Squarenes is something that can be adjusted by adjusting the twin leadscrews etc.

As far as the thread on mach2 is concerned (it is me!).

Regards
Mark

[turmite]

Hey Mark I have been reading the thread over on Mach2 and you guys are way too deep in for me anyway. I will say though that I believe Art will get it fixed. I am curious, and not having read every post on this thread here, are you running servos or steppers? The reason I ask is I am running in cv also and was wanting to know how you will know if you have lost steps if there is no absolute linear feedback. I posted a question down on the electronic's forum here and so far balsaman is the only one that has replied the post but still not to my satisfaction.

I'll ask you and if you want you can have the mods to delete this.

If a machine runs servos without linear scales for a closed loop feedback system, isn't it just like a open loop stepper system except more powerful?? Which brings me back to my question, how cal you tell if you are losing steps?? I have a need to know!

Mike

[rmtucker]

Turmite

Measuring lost steps is easy.Mark either the pulley on the motor shaft or leadscrew and mark an adjacent point on the framework.
Set the machine to "0" and run your program which at the end should go back to 0,0,0.If the marks line up again then you aint lost steps.I find in exact stop mode i can tune the motors to any settings i like and the machine will return to 0,0,0 but if i run a program in cv mode ie 3d machining it loses lots of steps.
I am using steppers and gecko 210's

Regards
Mark

[turmite]

I am using steppers with the 201's but have not checked for lost steps like that. I did do it with a dial indicator but not on very long moves. I think I will do that this afternoon before I run my program I was just about to run. What rapids and moves are you running, as well as which version of Mach2. I am still using release 1.90 and will not move up till the problems are solved and the program has been stable for a while. I am just now getting back to work full time because this retrofit was not as simple a deal as I though it would be, but then again I also install a 4th axis this time and that has been a bear. Let us know how things go.

Mike