[mcv300]

Hi all,

I am building a table for plasma, electronic is candcnc.

Today I tested the gantry ( made of 1,8mm, 50x50mm steel tube, 1200mm long ) and found that the gantry is not rigid, during moving it sways a little

I am using single motor, can it be the cause ? How to increase rigidity ?

Any suggestion is appreciated , thank you

[jcc3inc]

Sir,

Any chance of including a picture of your present setup?

Regards,
Jack C.

[mcv300]

Hi Jack,

This is my setup picture.

I tested some Gcode today (with a pen attached) , it can draw perfect circle and boxes at 1500 mm/m , but when i change feedrate to 4000 mm/m the circle is not round anymore / jaggy , the box is okay at feedrate < 3000 mm/m

[jcc3inc]

Sir,

How do you provide drives to both sides of your crossbeam? It is not clear from the picture. More pictures of drive details would be helpful as well as a picture of your pen tracing so that we can see which direction is the cross axis and which is the long axis, and maybe then have a better idea of where the problem is. Are you using rack and pinion for the drives? I like R & P since it is quite solid.

Regards,
Jack C.

[mcv300]

Hi Jack,

The x axis is driven from 1 side only, it is belt driven.
The y axis is also belt driven,
The belt is quite stable, not vibrating during travel
All axis use 15mm linear guide rail with blocks , should be stable

Picture 1: the active end ( driven side ) is okay, no movement when pushed by hand
Picture 2: the passive end ( non driven side ) move +/- 1 mm when pushed by hand

Jon

[jcc3inc]

Sir,

Your machine appears to have good craftsmanship on making and assembling the parts.

You could improve the stiffness by using the approach we used on machines with single side drives. Attached is a picture of a 32" x 36" test bed machine that I made. It uses "delta bracing" as it is sometimes called. The philosophy is that the left side is the master side; it determines the straightness of travel. The carriage drive is applied to that side only (for economy and simplicity). The right side is the slaved or non-driven side; its function is ONLY to support the carriage on that side. Spacing between master and slave rail is NOT critical.

If you want to use the approach you now have, you will need either a cross shaft to allow the belt drive from BOTH sides, or you can add a second drive on your right hand side.

Regards,
JCC

[Torchhead]

Hi all,

I am building a table for plasma, electronic is candcnc.

Today I tested the gantry ( made of 1,8mm, 50x50mm steel tube, 1200mm long ) and found that the gantry is not rigid, during moving it sways a little

I am using single motor, can it be the cause ? How to increase rigidity ?

Any suggestion is appreciated , thank you

We always recommend driving both sides of any gantry over 1 meter.

With a single sided drive, The problem is inertia. Take a 10 lb weight and tie it to a 1.5 meter arm and sling it back and forth at 2500 to 4000 mm/minute. You must have a very rigid arm structure (no flex) and a very rigid mount for the single side. Imagine how much force you can apply with a 1.5 meter long lever.

One solution is to drive the gantry from the center under the table, but that does not work well with plasma. You can put a belt or rack on each side and a long axle across the gantyr to drive both sides from the same motor, but it makes it harder to adjust the "mesh" on a rack and pinion if the pinion is fixed.

The amount of time and expense you will need to make the single side drive work will probably exceed what one more motor and drive train will cost.

TOM caudle
www.CandCNC.com

[mcv300]

Thank you for your advices , gentlemen

Yes, I think the inertia is the problem, so I will be incorporating dual drive gantry to next design.

I made some test this morning and find that a crossbeam can help . I am now adding a crossbar to the gantry as suggested, hopefully the vibration can be reduced, but unfortunately it reduces the x travel .

I will keep you updated tomorrow when it is finished . Have a good day

[mcv300]

I have added cross brace today. It seems that rigidity increase.
Tomorrow I will make some test, hopefully everything will be okay

[jcc3inc]

Sir,

It would be better if you could extend the distance between front and rear bearing sets on the master side. One might consider a ratio of 3:1, 3 = width of cross axis, 1 = distance between master side bearings. While it does require more rail on the master side, it is a simple solution.

Regards,
JCC

[mcv300]

Tested the table today...

Please advice, why are the circle and the diagonal line not perfect (chair).

it was drawn at 1750 mm / m ( around 70 ipm )

[jcc3inc]

Sir,

I assume that you are using stepper motors.

I would have to say that you have some springiness in your drive system. I prefer rack and pinion as it is quite stiff. Try programming a series of 45 degree zig zags along a line. Then as the machine travels while generating the slight oscillation after the change of direction, apply a small force by hand to see if you can stop the overshoot. With careful observation and testing you should be able to find the looseness/springiness. The path I am refering to would look something like this:
->
/\/\/\/\/\/\/\/\/\
l
/\/\/\/\/\/\/\/\/\l
<-
Also, check to see that your torch mounting system is not at fault because of overhang and a support that is not stiff enough.

Regards,
JC

[mcv300]

Thanks for your tip JC,

I made other tests today, it seems that the X drive motor is not strong enough . I use 100w servo ( 120 oz in peak ) with 1:4 reduction

I am planning to use dual drive this time, shall I use stepper or servo ? how much torque do I need ? ( my gantry weight around 10kg )

Thanks

[jcc3inc]

Sir,

When designing something like this, one starts with the total weight of the system that is to be moved, the max speed needed, the acceleration needed, the friction associated with moving, and then add extra torque, maybe 25% or more.

Next one either selects a motor and gear reduction system. Since you have a motor and drive amplifier, you should try to determine a good reduction ratio and the means to get there. Motors vary much in that they can be small, high speed or larger and lower speed. Likely you are NOT near the optimum gear reduction for your particular motor.

What speed and torque ratings does your motor have? That data should be on the motor spec sheets.

How do you get your position feedback to tell the CNC what your position and speed are?

What max cutting speed and what rapid traverse do you need?

If your motor specifications are available online, maybe I could look then up and comment further.

Regards,
JC

[mcv300]

Problem solved :

1. The belt on X axis is not properly tensioned.
2. The Y axis reduction pulley is broken
3. The accel rate is too slow on Mach setting

I have succesfully made square and circle .

Thank you for your kind informations

[4R8]

Hi there, Just a theory but is it possible your linear rails are not 100% parallel on all planes?

This may cause the slave end of your gantry to slightly bind (stall) then catch up. You won't see this on a square but it will definitely show up with a circle or diagonal line.

You could possibly try connecting your carriage to the gantry loosely on the slave side to allow some movement in case the rails are not true.

[mcv300]

4RB : I will try that tomorrow , thanks

[4R8]

hopefully from your lack of update, you are up and cutting well.