Hello Guest it is April 23, 2024, 05:22:43 PM

Show Posts

This section allows you to view all posts made by this member. Note that you can only see posts made in areas you currently have access to.


Messages - jimpinder

981
General Mach Discussion / Re: testing a reflective object sensor??
« on: December 28, 2007, 05:24:45 AM »
According to my maths, you can change the resistor in the supply circuit to a 147K and this will do. However, I still think your easiest way is to wire a 5v regulator to your 0v - 12v pins. This will give a regulated 5 v supply for all your sensors. I don't know why Bob Campbell has gone for 12 volts - since the computer works at 5v - it may be so that it is compatible with CMOS chips as well. (Does Campbells board demand a 12v signal back ??)

The 5v regulators are cheap - a matter of pence (or cents). They have three leads - one to 0v, one to supply volts and the other is the output at 5v (usually slightly under) they are as simple as that. They do not consume any current in themselves. A 1 amp version, or 0.5amp version should be more than adequate.

They are completely transparent to the supply voltage, in that even if that varies, the output remains steady.

You can then use the circuit posted (but don't forget to remove the pull up resistor on the output, you don't need it.
You can test the sensor with a volt meter before connecting it to the computer or fitting it to the mill.

Jim

982
General Mach Discussion / Re: testing a reflective object sensor??
« on: December 27, 2007, 06:46:33 AM »
I am using the Fairchild sensor, looks like the same one.

If you look back in the posts, you will see a wiring digram for this. Basically the transmitter is wired to a 5 volt supply thorugh a current limiting resistor. This shines down and then reflects back to the receiver.

The diagram shown ( and I am very sorry, but I have forgotten who submitted it, although I enclose a copy here) shows on the input a pull up resistor. The inputs lines to the computer already have pull up resistors, and two proved too much, so I removed it and the sensor works very well. It is mounted above the lathe chuck mounting plate (which I painted matt black) the reflector was from my grand-daughters "stick on" shiny things, and I picked one that was like a small mirror (about 6mm by 6 mm).

I picked off the power supply from my stepper motor drive cards (24 volt) I used a 5 volt regulator on a small piece of circuit card. This is quite adequate and saves any other input to my box of tricks. The regulator runs up to 1 amp, so there is plenty in hand for other sensors or applications.

I found the sensor very sensitive to being the correct distance from the reflector (although if you study it you can see why) - it has to be at 90 o to the spindle, and the beam has to go down the transmitter, then bounce back up the receiver.
Once it is installed and working it is faultless. I have several others ( I think I got four) and I am thinking a replacing my limit/home switches with them, if I can figure out how to mounted them. I think they will be a lot less susceptible to swarf and other little metals which short the terminals of my micro switches.

Jim

983
I am having much the same problem with mine -  I set x home, and it homes correctly, but as it backs off the limit/home switch it triggers the e-stop - I have to reset the e-stop before setting z home.(which then does the same thing)  I will try the debounce solution as suggested.

Brett - yes - I had Skype, (to talk with my son in Iraq) but then wiped it off after he came home. I will download it again and get in touch.

Back to this thread though - how did you go on downloading your web-cam - did windows automaticall install it, and then did it run first time with Mach 3 ???  Did you have to adjust any settings or drivers ???

984
General Mach Discussion / Re: OK that does it!
« on: December 25, 2007, 08:30:46 AM »
You stopped a bit early - it wasn't even Christmas day :D

985
General Mach Discussion / Re: Help, motors only humming but don't rotate
« on: December 25, 2007, 08:24:03 AM »
My wife says that my CNC with Mach3 will not work on Christmas day !!! I don't know how she knows.

Having said that - you haven't really said a lot - what are you expecting. Have you got your motor attached to a driver card etc. Has this worked before ??? In other words what have you done.

New PC - download Mach3 - connect cable to printer port - (connect other end to a breakout board) - connect leads from pins 2 and 3 to step and direction pins on the driver card (and also a 0v return wire) - (if you have eight wire motors (four segments) then cross couple the wires to give four wires (two segments)) - connect driver card outputs to the four wires of the motor. Connect power supply to the card.

On the computer - mach3/ports and pins/port ********* selection - enable port 1
                                                          motor output - show port 1, pin 2 as the step for A axis, port 1 pin 2 as the dir for A axis
Make sure both are enabled in the first column, and you can tick active low on both pins (but you might have to reverse this)
Make sure there is no axis enable switch or pin on your driver card.

That should be it - away you go.

If you have done all that and it still doesn't work get back on the forum!!

Although how long I can stay on today is in the lap of the household authorities.

986
General Mach Discussion / Re: Help needed to get web camera going.
« on: December 23, 2007, 03:57:20 PM »
Brett - I don't know if you got time to look at this or not.

Anybody else got any input ???

Jim.

987
General Mach Discussion / Re: I really need some help!!!!!
« on: December 23, 2007, 03:40:43 PM »
Fred - You have said a lot, but nothing. You have a computer running Mach3, and you have (I assume) connected the printer port of the computer to a milling machine, which, you say, has the original driver boards in it - running the original motors.

Are they motors stepper motors or servos with a feedback readout ??
I assume you have set up the steps per unit correctly. and also tackled the feedback on each axis - and got that as accurate as possible.

How does the machine perform on straight cuts - is it OK - is it accurate.

The problem with cutting circles is the direction of each axis changes twice in the circle, that means Mach3 has to stop all movement on both axis, take up the feedback on the axis that is changing, then start both axies again - and has to do this four times. To keep accuracy, it has to decelerate and accelerate each axis together. If you listen to Mach3 doing feedback, it certainly sounds different to ordinary running. The speed is different and the motor seems hesitant - but it certainly seems to put it on correctly - mine is to within 1 thou.

What I would do is try running without feedback and see what happens. Your circle will be out, but see if the noise and odd behavious is any different.

As far as the zeroing problem is concerned - I take it this a a software zero, not zeroing to swithes. If you are stopping before your program is finished, have you cancelled all offsets before sending the machine to what it thinks is zero???

988
I am a bit late on this post - glad to see the answer was simple in the end.

And that is the answer - KEEP IT SIMPLE.

Each axis requires three wires from the computer - a step, a direction and the 0v signal ground - otherwise none of whats follows will see the signal. These need to be identified on the ports and pins and enabled as you now know.

The only problem is you only have 25 wires on the port and 7 of these are 0v signal ground. There are only three sets of usable pins - one with 8 outputs (2 - 9) one with another four outputs ( 1,14,16,17) and one with 5 inputs (10 - 13, 15). If you clutter up your board with nice looking, but useless attachments, you will run out of pins.

If you have four axis (you will probably certainly have three - and at some time want a fourth) that takes care of the 8 outputs in the first set. If you then want M3 M4 and coolant that takes three out of four of the second set and if you want 4 limit switches (with home switches combined) that takes four out of the five inputs - which leaves you one input and one output spare. I have had to compromise because I have a spindle index input, so I only have two limit/home switches. ( I have two spare inputs for something I might find I need)

You do not need charge pumps, and other fancy computerised switches to isolate your axies - this can be more easily done by having an on/off switch on your driver card box - I have one main on/off  and then switches to isolate each individual axis ( all illuminated) - and I do not switch the cards on until the computer has run up to speed (as per the instructional video).

About the only other switch I might consider fitting (connected to the computer) is an ALL STOP SWITCH which I shall mount on the machine. Although the computer is next to the lathe, a big push button is easier to hit than a mouse and icon.

The only trouble fitting all these extras ( and I don't decry them at all, I think Mach 3 is the best program around) they all need maintenance - and if something goes wrong in the (I don't know what happened, but everything stopped) category, it is just one more thing to check and test.

If you step over the number of wires on the standard port, then you are into expensive add ons, which, quite frankly I don't consider necessary, and are something to add on when you are up and running and turning out some useful product.

Keep it simple and have fun !!!

989
Hood - My apologies

Jim

990
General Mach Discussion / Re: Odd Startup Behavior
« on: December 21, 2007, 10:42:06 AM »
I don't know which country you are in.  Mine are from Maplin in the UK.

Part No  ULN 2003A (7 pairs) or ULN 2801A (8 pairs). They are not the ones I use, which are older and now out of production, but these are the present equivelant.

Jim.