Hello Guest it is August 08, 2020, 12:59:03 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.

Topics - eggplant

Pages: 1
General Mach Discussion / alignment
« on: March 21, 2011, 08:24:49 PM »
Hi all,

Just a quick question. If say I put a piece of material to cut onto my machines table, how do I go about making sure its exactly aligned along the x and y planes? I could of course line up the workpiece by making it level with the edge of the table but if the workpiece was smaller and was to be positioned in the middle of the table how would I ensure it was straight? For eg if I was to cut a square 60mm square onto a piece of 100mm material how would I ensure the square I cut was parralell to the 100mm material (assuming Im not cutting right through of course !)


General Mach Discussion / feedrate/motor tuning
« on: March 21, 2011, 08:19:45 PM »
Hi all,

I found the answer a while ago whilst browsing the forum but now cant for the life of me find the post.

Im up and running now and have gone through the motor tuning and calibration. I found the maximum smooth motor velocity for eaxch axis and then knocked a bit off for safety. Whilst I can get to high speeds, the motor sounds rough. At the moment the maximum I am able to get is about 800mm/min. I will be mainly cutting mdf, plywood, hdpe and some aluminium. The machine is small and uses a proxxon (like a dremel) so only small bits. Is this speed ok? I know the feedrate varies depending on the material, tool, etc but from looking at videos on youtibe mine seems quite slow and whilst a few test cuts have been sort of ok, I want to get the machine and software set right before I start going any further.


General Mach Discussion / mad noise and vibrations at slow speed
« on: March 21, 2011, 08:01:08 PM »
Hi all,

My machine is now functional and calibrated as much as I can with a ruler and magnifying glass - waiting on a DTI arriving.

It runs smoothly but if I move X and Y at once, with a small amount on one axis but large on the other it makes a horrible noise and vibrates.
For eg, from 0,0 I can do G1 X50 F500  and it works fine as does G1 Y50 F500 but, if I have say G1 X50 Y1 F100 it hates it and sounds a bit like when you have the step timing wrong on a stepper or try to use too low a voltage for a set speed - not exactly the same but similar.

Hard to explain, I know but Im sure someone must have come accross this problem too. If anyone can offer any assistance? All 3 axis work fine and can move together at all speeds except when one has a small distance to go and one has a lot more to go.


FAQs / Assembly question
« on: March 12, 2011, 11:55:30 AM »
Hi all,

Yet another question !

The main assembly of my machine was bought from Ebay, new but unused and unassembled with no instructions. I managed to find a couple of diagrams of it (each diagram is slightly different) and managed put it together however Im at a loss as to how the threaded drive is supposed to be supported at the opposite end to the motor.
I have attatched a photo showing how I have assembled it. The locking nuts and collet each side of the bearing at the motor end are obvious, but the locking nuts at the other end (lose at the moment) dont seem to serve any purpose but Im sure they must somehow help support the end of the threaded bar. The brass coloured thing has a bearing inside, the 2 nuts will just about fit into it.
The "best" diagram I found is here
Im sure this will be obvious to someone !



Larger pic here

Hi all,

Project is making slow but sure progress. 3 axis moving table machine.

 I have fitted temporary limit/home switches- X,Y,Z have their own home switch made from optical sensors from old printers, the home switches are also configured as limit switches. The other ends of each axis have a mechanical switch, these are connected in series to one input. Wiring is a mess as you can see but will obviously be sorted once I figure everything out and get it working properly. Motor couplings etc are also temporary !

Couple of questions here, my X and Y home/limit switches seem to work fine, but the Z switch pretty much always triggers a limit before it triggers a home. I'll try to explain - I press REF ALL HOME and Z moves up as per the first photo, but then more often than not, instead of stopping and setting Z to zero as the other axis do, it triggers a limit and therefore makes me do a reset and jog away. This happens 99% of the time, on occasion it will actually set the home position without setting off the limit warning. What I did as a temporary measure to get the machine sort of zeroed was to set the switch to just a home switch rather than a limit switch.
I noticed that each time I homed, the actual position was slightly different - only fractions - Im presuming this is down to the accuracy of my machine. What I think may be happening is that when the Z axis has homed, it isnt moving back away from the switch quite enough, is there anywhere I can alter how much it moves back off the switch once home has been found? Also is there a way to home an individual axis rather than having to do all 3 at once?

My next question is also regarding switches, but more to do with positioning. I am getting very confused with setting up the home position. I have been through quite a few tutorials/giodes etc but Im still lost. I have taken a couple of pics which will hopefully explain things better than I can in words.

I have indicated on the bottom two photos where my switches are. The pic without the table top removed to show the switches is taken from the end of the Y axis, the other from the front of the machine.
Hopefully Im right in my X,Y zero position, but no matter what I try in the configuration I cant get it to move to the rigth place, for eg if I put G0X10Y20 I would expect the machine to move to roughly where it is on the pic (i.e. the table towards the front of the machine by 10, and to the left by 20 therefore "moving" the tool to 10,20 in relation to my 0,0 position). It does move, but I cant for the life of me get it to move the right way ! Im sure this has to be something simple in the settings, can anyone offer any advice?



Z axis home/limit switch Bigger Pic

Table coordinates Bigger Pic

Switch layout Bigger Pic

FAQs / power supply capacitors
« on: March 04, 2011, 06:52:44 AM »
Hi all,

Working on my power supply, using the transformer and bridge rectifier from an old audio amp (Im also going to use the case to house the supply and my driver boards etc)

I have identifed the windings and got myself 5V DC regulated using a 7805 regulator with the standard couple of capacitors (from 9V windings on the transformer) and also 36V (from 25 VAC windings).
The 36V goes through a bridge rectifier, with a smoothing capacitor which was temporarily 4700uF 50V, this wokred fine in a test driving a motor. I read up and calculated the capacitor value for my system which came to about 17000uF. The nearest I could source without paying the earth was 2 x 10000uF 100V capacitors (new) from ebay. My idea was to parallel these giving me 20000uF. They arrived, and are MASSIVE ! They measure about 4 1/2 inches high by 2 1/2 inches accross. Have I got the right ones? I wasnt expecting anything this big. The spec of them is:

10000uF -10 +50 %
MAx ripple 14.9A at 100Hz @ 85C
55/085/56 (not sure what these numbers mean)

Here is a pic of one. Can anyone please tell me if these are ok to use ?




FAQs / Grounding question
« on: March 04, 2011, 06:34:06 AM »
Hi all,

Putting together my DIY machine and am just working on the power supply. I have built 3 driver modules - 1 per motor, and a BOB. I have just had a concerning thought whilst planning the mounting of the PSU and other boards.

My BOB is a simple affair, basically a pcb with a DB25 connector on which routes the port pins to screw terminals with LED indicators. The screw terminals from the pins connect via cables to the driver modules. I used screw terminals rather than plugs/sockets at the moment as it is obviously easier to change wiring if need be. Once Im 100% happy I will probably change the terminals for molex connectors or similar.

I have 2 power supplies at the moment, a 5V to supply the logic, and a 32V, both standard mains power supplies that come with most low voltage appliances - black box that plugs into with a cable and typical DC socket on the end. My new supply (hacked from an old audio amp) now gives out 5V regulated using an LM7805 and 36V.

The parallel port pins 18-25 are also connected to the db25 outer metal casing, which is of course then connected to the PC ground when plugged in, and therefore also connected to my mains earth. I have connected the grounds of both supplies and my BOB, and it all works fine.

Is this correct so far? should the pc ground/earth be connected to my 5V/32V ground? My new power supply will obviously be earthed, and at the moment will be sharing the common ground too, i.e. everything that is  0V will be connected, and also connected to mains earth. Should they all be connected like this or should I seperate the mains earth and DC grounds?

Hope that makes sense !



General Mach Discussion / pulse frequency (motors slow and jerky)
« on: February 24, 2011, 10:28:12 PM »
Hi all,

Very new to the world of cnc. In the process of building a small machine and Im up to the stage where Im ready to start putting it all together. I've done plenty of research and whilst there is no 64 bit support for Mach3, it does seem a very popular and reccomended package so I have downloaded the demo version for now and hopefully I can get things working.

My problem is, I cant get the motor to turn at anything faster than a few RPM, it is very grindy and misses about 10% of a revolution for each revolution (it is that slow it can easily be seen jumping)

At the moment things are still under development, and once everything is working I do plan to build a larger machine. My basic setup so far consists of the following:

Ebay purchased partial cnc kit, consisting of frame, bearings, lead screws - pretty much all the mechanical side which is all fine.

3 x bipolar nema 23's, 2.8A 260 oz-in/19kgf-cm

Driver modules for the motors are pacific scientific (pacsci) 5210's. I originally started with L297/298 drivers but had immediate problems with speed and came accross the 5210's on ebay for cheap so I am using those for nowl I appreciate they dont offer the microstepping that some more advanced modules do but for the time being the performance I get from them will be fine to get me up and running. Once I am operational, I will be able to see what needs changing when I build the larger one, I suspect the driver modules will be one of them.

Power supply is temporary, a 32C 1.5A supply that I happened to have. I know this isnt sufficient to power all 3 motors but is fine for testing a single motor.

I have a motor runnig fine from a small circuit with a 555 timer IC with the aforementioned power supply, and a small 5V supply so I know the motor and driver units are operating. I can get in excess of 1700 RPM with no load (guestimated with a diy rev counter but wont be far off). I suspect I could get more but I think that is more than enough, and since Im not on the power supply I will finally use I  havent spent time tweaking this. I can get dont to a snails pace too so I am confident that the motor and driver side of things are ok and my problem lies elsewhere.

 in order to test Mach3 with my system I have on the bench a direct connection to my driver module, basically the parallel port is replacing my timer circuit. I have a cable from the PP on the pc to a temporary breakout, all that is connected is gnd, and pins 2 and 3 which are configured in Mach3 as step and dir. I have a 5V supply and my 32V supply for the motors also connected, gnd's are commoned and the output & dir pins set up in Mach3.

The Mach3 driver test runs, the waveform is all over the place and fluctuates from about 400 to 1900 pulses.  The pulse rating goes from too slow, to excellent, to to fast. I suspect this may be the key to my problem.

When I fire up Mach3 and try to configure my motor as per the video tutorial I can get it to turn but as I mentioned at the start of my post it is horribly slow and rough. I roughly calculated the motor steps etc, and also tried values ranging from 200 to 1000's to try and get it to run somewhat better but I wasnt able to. I also varied the step pulse. I was getting nowhere so I read through the forum and I think the problem is that when I switch to the diagnostic screen I am getting a pulse frequency which fluctuates from about 1580 to 1900. From what I have read, I bellieve this ought to be somewhere near the kernel speed? This is currently at the default of 25000 and according to the documentation the demo version is locked at this although the other options appear to be available in the program. I have tried the sherline 1/2 pulse mode, and also 1/2 and full step on my drivers.

My main pc is a decent spec machine but is running windows 7 64 bit so unfortunately I cant use this one which I know would be more than powerful enough and would get the pc spec out of the equation.
The  basic spec of the machine it is running on is and AMD sempron (yes, I know - its all I have available!) 3000+ which runs at 2 ghz, 1 gig ram with a clean install of XP pro SP3 32bit. Nothing installed apart from drivers and Mach3. Ive read the optimisation guide and apart from installing as a standard pc its pretty much done. I work in IT and am happy that the machine is running pretty much as bare as it can with the recent XP install.

With only Mach3 open, and idle, cpu usage is 0%, available memory about 700. With Mach3 running the roadrunner gcode it takes about 2% cpu and 50 meg of ram so it doesnt appear to be a great load on the system.

The diagnostic page as I mentioned shows a pulse frequency of about 1580 to 1900 and the "time in int" from about 3.5 to 4.5. When running the roadrunner the pulse frequency initially starts off at about 8000 then repidly drops and varies from about 2500to 2800. What I also noticed is that if I do anything at all on the pc, i.e. launch task manager, then the pulse frequency will rise to about 16000 for a second or so.
As I have nothing else on the pc, I ran calculator in scientific mode and did a 423432432 (random large number) n! which is an old trick to hog the cpu. During the time it is trying to calculate, the cpu usage goes to 99% for calc but the pulse frequency on Mach3 steadies out at exactly 40605. When I stop the calculation then it drops back down to what it was.

Hopefully I have provided enough info without boring everyone to sleep ! and hopefully someone can point me in the right direction.


Pages: 1