Messages

101

General Mach Discussion / Re: Motor Tuning with Gecko 201 and 202 Drives

« on: March 12, 2009, 02:38:30 PM »

This has been covered many times in this forum.

On the motor tuning page, you must enter the number of steps per unit. It depends on whether your friend requires his standard to be in inches or millimeters. It does not matter to the system, and Mach can deal with inches or millimeters, no matter what units it has been initially set up in.

It all seems to rest on the ptich of the leadscrew - whether this is in inches, or millimeters. I initially had a 10 turns per inch leadscrew, therefore I set my lathe up in inches. I have just changed my leadscrew to a ballscrew (more accurate) and this is 5mm pitch - so I have now set my machine up in millimeters.

Once the machine is set, all the standard measurements to set up tool tables and offsets must be entered in inches or millimteres. It then makes no difference to the operation of the machine since a code of G20 will make the machine operate in inches, and a code of G21 will make it operate in millimeters, and all values entered in the tables will be changed correspondingly.
When you say you have an 8 thread pitch - what do you mean - this must be accurate as you will see, and be careful that someone has not sold you a metric screw as an imperial. If he is using  a fixed machine, he will know.

To calculate the number of pulses - and note the word calculate - it is an follows - and all number multiply the previous answer.

The vast majority of motors now work on 200 pulses per revolution - check that these motors do - or alter the number accordingly.
Gecko 201 drives give a 10 micro pulse output - 200 x 10 = 2000
Your gear down appeasr to be 2 to 1 - 2000 x 2 = 4000
Finally we come to the pitch of the leadscrew  - on my original leadscrew it would have been 10 turns to the inch - 4000 x 10 = 40,000 (units in inches)
on my new leadscrews it would have been 4000 / 5 = 800 (units in millimeters)

That is how you do it . you will have to calculate the final  line yourself.

After you have done that enter the result in your x and y (and z) axis, and test it. Move right, stop and zero your DROs (by pressing the zero x, zero y), set up your measuring instrument ( I use a digital caliper). Then, using the MDI line enter G0 X1 (inches) of G0 X25(mm's). This should move right again (to avoid backlash) and measure the distance the table has moved.

Do not expect to be absolutly accurate (obviously the more accurate the better) because it is your measuring that is at fault, not the machine. If you are within 2 thou or even 0.01 (mm) then that is pretty good and shows that you calculations are accurate.

Have a go - come back if you get in bother.

102

General Mach Discussion / Re: calibration question for the experts

« on: March 10, 2009, 05:17:31 AM »

Spot on Hood - divide by 409 and multiply by 409.6 and the length of the pocket comes out at 120.03mm

103

General Mach Discussion / Re: Can "Direction" &"Steps" signals be saved and stored in physical "memory"?

« on: March 10, 2009, 05:05:38 AM »

I suppose the answer must be yes, but as Tweakie says, with my system producing 60000 pulses per inch, I would end up with a lot of pulses.

A far better system would be a brain, of a short Macro which would produce the same result.

104

General Mach Discussion / Re: Mill just wanders off

« on: March 08, 2009, 06:23:19 AM »

I suppose I have the best of both worlds, in that I have a Warco lathe with a milling head attached. The lathe can turn 9 inches, although at that size you get problems with the spindle motor power since the VFD cuts down the power when running at under 50% speed. I have fitted a set of six pulleys, so I can get down to 200 rpm at full motor speed.

The other drawback is that the milling head isn't really rigid enough for big milling jobs, but thanks to cnc, I can keep the cuts down to about 0.5mm (20thou) and it seems happy with that. If you are cnc, it doesn't matter how many cuts the machine takes. I had it chewing through 10 mm plate last week making some suspension parts for a train I am building. Each part took about 45 minutes - but I went off and had a cup of coffee. I had 16 to make, but I had them finished inside a week.

The lathe is simpler to set up. since it is not likely you will bother with homing switches. I do have one on the cross slide, because I have a centre drilling attachment on the cross-slide to take the place of the normal centre drill. I home the cross slide, and then put in the appropriate offset, and the drill comes across and centre drills the workpiece.

Other that that, work tends to be zeroed in the chuck. X is zeroed by taking a small cut off the workpiece and measuring the diameter of the bar . This is typed into the X DRO (or 1/2 the value if you are on radius). The Z is zero'd by takeing a facing cut across the workpiece and setting the DRO to 0. It is as simple as that.

In other words x0 is the centreline of the lathe, and z0 is the workpiece face.

Get going and get cutting - great fun.

105

General Mach Discussion / Re: setting up a bridgeport mill to cnc

« on: March 07, 2009, 04:15:14 PM »

I would set up home switches. On my lathe/mill I have two lasers on the table (cheap from Maplin at £10 each) they are on adjustable mounts to get them parralel with the table.

They shine onto detectors that are mounted on the body of the lathe. The detectors are four wire devices, Schmidt trigger, and give a very fast, clean signal. The big advantage is there is no physical obstruction in the machines path. so if this goes astray, it doesn't wipe out all your switches. A further advantage is you can mount as many detectors as you wish for each laser to stop the table in a number of different places if you want to.

If, when you turn it on, you then press "RefAllHome" on Mach the machine will move to each switch in turn, and stop there. It then resets the Machine Co-ordinates DRO's to 0,0,0. It is doubtful this will be in  such a position that you can machine from it (although with laser detectors you could position them in such a way as to do so).

This is where offsets come into play. Each program will be written  with a 0.0.0 position convenient for that program, and therefore, you must put an offset command - G55 to G59 (+250) to bring the table to the correct position for the program, from it's home position.

This sounds a bit daunting, but actually takes less time to do than to describe.

106

General Mach Discussion / Re: Mill just wanders off

« on: March 07, 2009, 03:08:36 PM »

I was going to answer this earlier today, but my server went down.

You must be sure, when you start, that your machine is properly zeroed. This means the machine co-ordinates, and the program co-ordinates. If you do not, any offsets introduced will not necessarily result in the right position.

The machine keeps position by using machine co-ordinates - displayed on the DROs when the Machine Co-ordinates button is pressed and the surround is illuminted. If you have home switches fitted, then the only way to alter these is to "home" the machine. If you do not have them fitted, or have them disabled, then pressing "RefAllHome" with zero the DROs.

So - the way to start (if you have no home switches) is to jog to the X0Y0Z0 position of the program. Press the Machine Co-ords button, and the "RefAllHome" the DRO's will all zero. If you then press the Machine Co-ords button so the led surround goes out, you can now zero the DRO's (if they are not zero) by pressing the zeroX zeroY and zeroZ buttons. This zeros the program co-ordinates. If you check the Config/Fixtures table you will see that G54 (the default offset) is now zero.

If G54 is not zero, then some other offset is selected, and you should select G54 and repeat the procedure. Unless you use offsets, all the table should be zero, and you can clear any figures that show in there.

Your machine-co-ords and your program co-ords now coincide with each other, and you can run your program, without any stray offsets value interfering. The problem is that Mach3 alters its Machine-Co-ordinates to reflect the actual machine position, with all the offset values added to it, whereas the program co-ords reflect the value of X,Y and Z in the program. The two are often miles apart.

107

General Mach Discussion / Re: Newbie needs help

« on: March 07, 2009, 02:52:20 PM »

If you want to run more than 500 lines of code, then you must use sub-routines, and an editor, like Microsoft Word, to split your code into 500 line chunks. Word is best at this, becasue it automatically numbers the lines, and as you chop 450 - 500 off, it renumbers.

I did this on a 5000 line code for drill and milling a printed circuit board, and it worked a treat. The main prgram just called the 10 subroutines in order.

When you  feed it into mach 3 at the beginning, then, as Hood said, it goes through all the lines, and the subroutines, and paints a full picture.

A bit messy, but it works. Thank goodness I don't use long program.

108

General Mach Discussion / Re: need mach motor tuning explained better

« on: March 06, 2009, 12:10:34 PM »

Yes - I agree with you, the ptich circle is a crucial measurement - but, if you open it out into a straight rack. if the pitch is 16 teeth per inch, then the distance between the tips of 16 teeth is 1 inch irrespective of where the pitch circle is (usually about 1/2 way down the tooth). This is because the tooth is roughly trianglar and therefore the centre point of the tooth will be in the same position for the whole of its depth.

The distance between the tip of the teeth when it is  in a circle or pinion will be in excess of this, but as you "straighten it out" into a rack, all what were the spokes of the wheel become vertical and parallel, and therefore all the same distance apart.

In the diagram you have posted this shows the distance between the teeth is regular - and is equal to the pitch circle of your cog. What is crucial in this debate is - what is the distance between the 16 teeth (on the rack)

109

General Mach Discussion / Re: Mill just wanders off

« on: March 06, 2009, 06:03:25 AM »

Clive - does this just "happen" or have you done something just before like "start from here" or similar.

It sounds like Mach 3 looses it's memory, which it can do if you "start from here" without backing up enough for Mach to have full details of it's position

110

General Mach Discussion / Re: need mach motor tuning explained better

« on: March 06, 2009, 05:55:09 AM »

Bill, you typed your last post while I was replying to John -

You are getting confused with pitch diameter, and all the other measurments there are for cogs. They are very simple. The number of teeth. The other measurements are included for us so that, for example, I make gearboxes for my trains. I need to know the ptich diameters of gears so that I can drill the centres for positioning the gears the correct distance apart. I need to know the overall diameter of the gears so I can machine clearances. However, the working part of the gear is the number of teeth.

If you have  two gears, one with 10 teeth and one with 20, and mesh them, the small one has to turn twice to drive the large one once. If you convert the large one to a flat rack, the small gear, if it turns twice, will move along the rack by 20 teeth - it doesn't matter what physical size the gear is. If you then measure the distance on the rack between 20 teeth, that is the distance travelled. If we assume the teeth are uniform, which most are, then the centre of one tooth to the centre of the other will be the same along the rack, and through all the teeth, so wear or shape does not matter.