Messages

211

General Mach Discussion / Re: z axis driven pump: pressure regulation /logic input question

« on: July 27, 2010, 11:24:36 AM »

I don't have an Arduino at the moment, but they sound like fun. Why use the Arduino for THC if Mach already has it?
If smooth steper doesn't do THC, probably the ncPod doesn't either, but I'll ask them.

Rich B

212

General Mach Discussion / Re: z axis driven pump: pressure regulation /logic input question

« on: July 26, 2010, 05:44:56 PM »

Hi Hood,

I guess the important point is, how critical is the pump... not sure. If I hadn't already got a linear stepper on there that works well enough (but just needs a bit more control), then I would be pleased with the simplicity of your idea, and do it like that. Also it would free up another axis, which I may find useful at a later point. (I won't go into that here, yet). The extruder pump needs to withdraw fully on tool change ('syringe full of paint change' in my case), so needs to be under Mach control anyway. I could trip that from inside my tool change macro, or I may go with the THC or even just Z axis control, for now. Hopefully enough people out here you can guide me to the right resources.

Thanks for your advice,

Richard B

213

General Mach Discussion / Re: z axis driven pump: pressure regulation /logic input question

« on: July 26, 2010, 04:43:04 PM »

Okay, been offline for a while, lots to reply to... good!

Hi Hood:-

I thought of using a relay to 'dump' the step/dir signal so it never reaches the motor; thought it might be a problem because you can't guarantee to switch it if afterrather than during a pulse. Shame about not being able to switch an axis on/off when machine is running, like you say, but I was thinking of there being a macropump (thanks Tweakie) which just generates "G1 Z1" every time an input to mach 3 receives a logic 1. Would a pulse generator mean using a dc motor? Or do you mean altering mark/space ratio on a switch mode power supply to an AC motor? It on;y needs to be a tiny motor driving a screw thread. It has a motion control 35NCLA linear stepper at the moment, does very well.( 5V, 0.46A per phase, 4 wire)

Poppabear:-

I see your point on acceleration/ deceleration, but there is another 'axis' (A axis) that runs during plotting, that worka as an 'applicator'  by taking a dab of paint extruded from the syringe driven by the z axis, and 'tattooing' it onto the page. It runs at about 2000 cycles per minute; this is a real 'drag' on the X axis,: I wonder if a tiny Z axis move may not make any difference? as it involves the least hardware changes, I'm tempted to keep my fingers crossed and try it. The code below is want I have been running the machine off to test it. It works as my final code will, except that in this case THE Z COMMANDS ARE WRITTEN IN, rather than generated by a macro. It explains how the 'axes' A and Z work for me, though.




G01 F35000 X0.80 Z0.4 A-4                    (plotting begins here; A-4 means 'lift applicator to it's maximum height, so it picks up some paint from the extruder/syringe)
G01 X0.85 A4                                 (A4 means 'drop applicator down onto cpaper, so it leaves a dot of paint there)

G01 X0.9 Z0.7 A-4                            (Note the Z position increases by .03; this value gives good results, but still need constant pressure for best performance)
G01 X0.95 A4                   

G01 X1 Z1 A-4                 
G01 X1.05 A4                   

G01 X1.1 Z1.3 A-4             
G01 X1.15 A4                   

G01 X1.2 Z1.6 A-4             
G01 X1.25 A4                   

G01 X1.3 Z1.9 A-4             
G01 X1.35 A4                                 
G01 X1.4 Z2.2 A-4                            (plotting ends here)



Hood again:-

"Scott, yes that was my concern doing it that way, you would have to have the feed set so that switching on.off the pulse didnt stall the stepper which in turn would impact on the other axis."

This sounds important; but I don't quite know what you mean by "having the feed set...". If the stepper is only getting clean step/dir signals from the macropump when an input is present, does that conform to what you mean?

Tweakie:-

The plotter has no Z axis as such, I just happened to wire the extruder pump to that output on the ncPod controller; similarly, the A axis doesn't work as a positioning axis either, it just drives a little applicator thingy.

Stirling:-

Hadn't heard of THC before, sounds exactly what I'm looking for. Will it adjust my Z axis position in response to a logic input to Mach3, without interrupting other axe's moves? If so, great, assuming my ncPod supports this...

I was wondering if, although it sounds weird, I should just run my plotter in Mach3 plasma rather than Mach 3 mill; but then I noticed there are only 3 axes in Mach 3 plasma. Which is easier, adding another axis to Mach 3 Plasma, or adding the THC button to Mach 3 Mill? Also, where should I look for information on THC? Not much in this forum, only a little in the basic documentation, and no 'guide to using Mach3 plasma' either.

Kf2qd:-

The extrusion is controlled by Z axis. In my code, no plotting occurs when there is a "Y" axis move, all it does is straight horizontal lines. As you can see from the code, the A axis movement is 'tied' to the X axis movement. Currently the Z is, too but this is the bit that needs to change.

Thanks all,

Rich B

214

General Mach Discussion / Re: z axis driven pump: pressure regulation /logic input question

« on: July 24, 2010, 12:34:56 PM »

I seem to have mistyped; should be:

Do until input1 =0  'ie do until the pressure reaches minimum value, triggering logic 0 from pressure sensor circuit
          G01 Z(Z+1)  ' ie increment Z (pump)

that's better!

215

General Mach Discussion / Re: z axis driven pump: pressure regulation /logic input question

« on: July 24, 2010, 12:30:54 PM »

I've just thought:-

If I write a VB module that works like:-

Do until input1 =0  'ie do until the pressure reaches minimum value, triggering logic1 from pressure sensor circuit
          G01 Z(Z+1)  ' ie increment Z (pump) until correct pressure is reached

can I call it in the Gcode and have it run in the background for as long as the machine is on? I'm sure the syntax is way off; this is just really to describe how it might function.
I haven't used VB with Mach before, in fact I only started to learn VB about 2 weeks ago, so if someone could tell me if I'm on the right track or not, that would be great

Thanks,

Richard B

216

General Mach Discussion / z axis driven pump: pressure regulation /logic input question

« on: July 24, 2010, 10:08:55 AM »

Hi,

I have a plotter that uses incremental movement of the 'z' axis to drive a material deposition pump. (Another axis (A) causes a small piston to oscillate and print the material onto a flat surface.) I had originally wanted to use the spindle drive for driving the pump, but my ncPod controller does not do spindle control, apparently.

There is a slight problem with pressure regulation, which I intend to solve by fitting a load cell, with circuitry to generate a logic 1 when the pressure falls below a predetermined value, and logic 0 when pressure is satisfactory. I hope to use the logic 1 signal to start the pump (ie cause the z axis to step), and logic 0 to shut it off again, thereby achieving pressure regulation.

Is there a way in Mach 3 to start/stop an individual axis in response to a logic input? It might be a bit like giving an Estop signal to just the one axis.

Does anyone know if Mach can do this? Or should I think about putting an Arduino on the machine to control the z axis seperately to Mach 3? Bear in mind that the z axis s not being used to position anything, just to provide pump presssure. The actual printing is done by the A axis, which must remain under mach control.

Any ideas?

Rich

217

General Mach Discussion / Re: speed control of stepper used as spindle motor

« on: April 24, 2010, 05:23:20 AM »

Okay, I've realised I still have part of the original problem, and in order to explain this I need to tell you some more about the machine:-

It's basically a 2.5d plotter, working from artwork in photoshop that's had a grid applied to it to turn it into a series of thin parallel lines. The jpeg file with these lines are then vectorised and turned into hgpl in the program I currently run (R2V). This involves x, y and z axes, as the z axis controls pen up/ pen down From there it should be easy to send that through a program that converts the hgpl into g code (not tried to yet), where, as I said earlier in this post, the paint is only pumped during horizontal strokes (x axis only, no y or z to interpolate.) Why? it's just easier that way, I know that it works from running an old Roland plotter for this  a few years ago.

The problem is, of course, that my g code will then only contain code for the x, y, and z axes. I still have to generate the code that tells the a axis (pump) how many steps to take while the x axis does it's thing. Spindle contorl would have been so much easier, I could just tel Cam Bam or whatever to only switch the spindle on when the z axis is in 'pen down' position. Oh well.

So it's back to my question about how to write an app that does that. (See earlier post for my crude program idea!)

Should I save the g code as .txt, and make a macro somehow to do that in microsoft word or some other text editor? Wild guess...

Can anyone give me some pointers on how to do this properly?

Ta,

Richard B

218

General Mach Discussion / Re: speed control of stepper used as spindle motor

« on: April 23, 2010, 04:47:18 PM »

Thanks, Simpson, this sounds like a good way around the problem, especially as I don't use the max speed as it stands. I'll get back to you in a couple of days and say how it went.

Richard B

219

General Mach Discussion / Re: speed control of stepper used as spindle motor

« on: April 23, 2010, 01:29:53 PM »

You are correct, Simpson36, that is what I'm trying to achieve, and yes, the controller doesn't seem to want to know. The developers of the ncPod don't know if it handles step/dir for spindles, but assume it does. they don't know Mach3 well, apparently. I'm surprised they haven't tested this feature, but there.

I take heed of what you say about speed problems in combining/ interpolating rotary and linear axes together, in fact I'd considered using the z axis to drive the pump for that reason. Only trouble is, that might make generating the g code difficult,as I already have a physical z axis, so I'm hoping there's a way to configure the a axis as another linear axis.

Do you know if this is the case?

The problem remains in either case of 'fixing' the ratio between x axis movement and pump axis movement during 'painting' (Assuming the axes can work well enough simultaneously.) This is where I may need to write an app to edit my gcode that works like

 'read x axis value
 divide by the number needed to get correct ratio
 add this new value into the code and precede it with the designated axis letter'

This might be easier than trying to generate the pump axis code in, say, Cam Bam or whatever g code generator end up I using. This simple approach won't allow for non simultaneous starting of both axes, of course, but it's a start. Once I get the principles involved in getting to this stage, I should be able to have a go at more sophisticated stuff

I have no idea how to do this yet; it feels like it should be simple, but I bet it won't be!

Any suggestions?

Thanks,

Richard B

220

General Mach Discussion / Re: speed control of stepper used as spindle motor

« on: April 22, 2010, 06:08:02 PM »

That's what I suspected. An alternative approach to solving this problem might be this:

What I want to do is to operate a linear stepper to deposit paint in straight lines (ie movement on the x axis only; no depositing of paint during movements involving any other axes), so I could 'link' the number of steps made by the x axis to those taken by the axis driving the pump, and play around with the ratio between the 2 rates to get a rate of deposit as required.

Would I need to write some code to achieve this? Or is there some way of slaving the a axis (pump) to the x axis, but specifying different steps per, or speeds, to achieve the desired ratio(s)?

This is for anyone to chip in on...Thanks Hood

Richard B