[Nico13]

Hi all.
Recently finished, well almost finished CNC router, see my Build Blog HERE was travelling a bit slow, not what you would call rapids at 500 mm/min.
My initial setup was with the ubiquitus Toshiba chipped TB6560 24V system single control board for the three axis and it worked ok and quiet and smooth at 800 or 1600 steps per rev no problems but a full X axis travell took about 90 seconds you know grass grows faster type of thing.

So the upgrade after talking to a mate and a bit of reading etc I purchased three M542H controllers in a kit with a five axis BOB and then a 48V 10A supply to go with as I was told more volts = more speed.
Connected it all upand could not get the motors, NEMA 23 280 in oz, to turn, lots of grumbling and groaning but not a turn.
Finally stumbled upon the fact that I needed to set the maximum pulse width in Mach3 of 5us and away they went, but not all three axis wanted to run the same all just a little different but they ran.

Tried for ages to get them all the same, X ran the best and I managed to double my initial speed to 1000 mm/min but not smooth or quiet, turning but growling not singing like they should.
The PC driving this is an old XP 1Ghz machine, I had tried the different Kernal speeds and over a couple of days of trying different tuning of the motors etc, all of a sudden the Z axis stopped followed by the Y and the X over the next couple of days.
All the time I was swapping contollers arround to see if they had died but they all still worked individually up until the last so I put it down to the BOB or the LPT port.

Next step put all the old gear back to see if that worked and it did as before with no problems.

What now?
It was suggested that I needed a "Smooth Stepper" or another option was a "UC100" I have also considered running it from one of my Laptops running Win 7 and USB interface.
Suggestions welcome on what next with these controllers etc.
Thanks for taking the time to read this.

The steppers had been disconnected in the middle picture to reconnet to the old board.

Nico.

[Nico13]

A couple of notes re the BOB.
None of the limit switch connections functioned or the Estop either, tried active high and active low, not sure why they didn't work.
All the pins and ports were configured for the new BOB as per the supplied info as it did work initially.

I did have a couple of odd Estop activations when trying to jog the X axis via the keyboard and it took a couple of goes to reset it.

[aarggh]

I can't comment on the BOB you have, but I have seen many on the market with pretty bad layout, poor signal conditioning, and very dubious opto-coupled or buffered inputs. It's a shame you've already bought new drivers as you would have been far better with a Gecko G540. But the 542's are generally pretty rock solid I've found. My large machine uses Leadshine 542's and has performed flawlessly for years.

When you have disconnected/connected the BOB and drivers, etc, you never did this under power did you, as if you don't allow enough time after switching off for the currents to decay, you can blow the drivers.

Apart from that, unless you have a flaky BOB which sounds likely to me if the whole lot has stopped working when it was working previously, it must be Mach3 tuning and settings. Also, in the PC BIOS, what have you got the LPT port set to, ECP+EPP can cause problems, standard LPT is what it should be set to.

I do think it sounds like your BOB though. Can you borrow one off a mate to try?

cheers, Ian

[Nico13]

Thanks for the comments Ian.
I have yet to check the bios setting for LPT 1 but I cant recall altering the settings although it is an old PC and I may have changed things some time back.
No connection changes were made under power as I'm aware of the back EMF surge problems possible so always powered right down.

[ger21]

Try upping the pulse width to 10-15.

If your motors have dual shafts, a damper may make them run smoother, and give you more speed.

If you run Drivertest.exe in the Mach3 folder, do you see a very flat line with a steady pulse rate? If not, you may need to change your PC from ACPI to Standard.

I'm not so sure that a Smoothstepper or UC100 will help that much, unless the problem is with your PC.

I use a 1Ghz PIII, with a Xylotex drive, and can get up to 190ipm (~4800mm/min) with 250oz motors and 24V. I need to use dampers to achieve these speeds and get smooth running.

[john-100]

Hi Nico

If it works with any 2 drivers connected to the breakout board but fails with 3
it could be the USB port is limiting the 5V supply to the B.O.B

if I remember correctly a USB2 port limits the current to 100mA
until it detects a USB device then the limit is increased to 500mA

a while ago I traced the circuit of the BOB you have
Attachment 261104





note the missing supply to the 7414 used for the 5 inputs ( 4 limit switches & E-stop)



assuming the 542 driver opto-isolator input is a 1.8V LED and 200 ohm resistor

with the addition of the 56 ohm resistor connecting the IC to the BOB output terminals
the current for 3 direction inputs and 3 step inputs = 3.2/256 x 6 = 75mA

If the supply current for the BOB and any 542 driver enable input is more than 25 mA

the USB current limit could be part of the problem

try retesting with a 5V power supply instead of the USB port


John

PS
added picture of Chinese BOB PCB
re saved 2nd picture that's too low a resolution "ATTACH=CONFIG]261110[/ATTACH"
web site will not let me delete duplicate !

[Nico13]

Thanks John and Gerry,
The USB power is from a plug pack not a USB port but I will check its current limit to make sure it can supply enough.
The driver test is not as smooth and flat as it could be so I will also check out ACPI to Std mode etc.

Gerry those are impressive speeds from your system and yes I do have dual shaft motors but am unsure of the dampers you refer to, will have to look those up and their implimentation.
Mach3 mentions that the max pulse width was 5us, does it actually accept number greater than that? I did try entering 6 or 7 without effect as I was unsure if it would accept it.

Lots to go on with here and play with over the Christmas break thanks guys.

[ger21]

Yes, Mach3 will accept pulse width settings up to 15.

http://www.cnczone.com/forums/steppe...er-damper.html

[Nico13]

Yes, Mach3 will accept pulse width settings up to 15.

http://www.cnczone.com/forums/steppe...er-damper.html

Ah yes, commonly known in the automotive business as a harmonic balancer, (a damper). I'm a mechanic you see.
Thanks for that and as I do have dual shaft motors I may try the simple fly wheel for a start and move to the damper style if further damping is required.

[bhurts]

I think I remember reading about people using hockey pucks to make dampers with success. There also cheaper. I'm not sure but I think they just drilled a hole slightly off center and pressed them on the rear shaft. You should be able to search it on the site.

Ben

[victorofga]

nico


your card possible makes midband resonance.. you can fix it probably.. but im not sure if it worth...

I just looked for your machine.. I know it sounds bad, but in my opinion just throw away the whole board..
for this machine worth to match a better control..


I uploaded over here your pictures, because it is a very nicely built one..

im using ncstudio..

it is a pci card in your computer, connecting with a BOB to your router.. you can hook onto a vfd and control the spindle, on off, and rpm
you also have automatic Z zero , restart from a breakpoint next day and a lot more stuff..
it makes the 160 kHz

with 10 mm lead it makes abot 8000 mm per min.. with 5 mm screw still you can reach the 4000 mm per min , when 0.0125 mm the step... if step larger then speed also larger accordingly..
so if steps are 0.025 mm then you reach the 8000 mm per min

all cost about 55 dollar..

Wholesale NCstudio Controlling Card for Wood Cutting Machine - Alibaba.com

you get the card, get a BOB also get a 15 pin cable in the package and of course the ncstudio program


for this you can buy a remote controlled handheld unit

Wholesale Wireless Handwheel Ncstudio 3 axis controller HB03 - Alibaba.com


so you can setup easy without reaching keyboard...


also theres a kind of usb control for mach 3 that cost about 160, the dsp system about 300 you can buy.. that fast..

[Nico13]

Thanks for help and suggestions of ways to remedy my problems. Some of these may be looked at again if I have more trouble.

Below is a copy of the post I put up on my build blog. DIY CNC Router | Ken Nicholas Nico13 Photography Astro Photo & DIY CNC

Well, miracle, mystery and intrigue, definitely the unexplained.
The bottom line is though that for some unknown reason when connected up to the new kit including the break out board and the three individual stepper controllers it suddenly decided to fire up and all three axis worked as advertised.
So you tell me, if you’ve read the post previous you will have read about the efforts I’ve been to trying to rectify the drive failures only to have it suddenly work in the configuration that had failed so miserably before. The big test now is will it keep on working?
I did return to the shed after tea and booted it all up again just be sure I wasn’t seeing things as this had been a mystery for about two months now and yes it’s still working so some tests to come to check normal operations, some test cuts of items I had originally made to compare, so tomorrow if time allows before work it will be put to the test so again wish me luck, reports to follow.

[Larken]

Are you using a Linear or Switching [power supply for the 48 volts ? If its a switching, it may be whats making the growling. Especially if its common to all 3 axis.

[Nico13]

Are you using a Linear or Switching [power supply for the 48 volts ? If its a switching, it may be whats making the growling. Especially if its common to all 3 axis.

It is a switching supply and I had noticed the possibility of some harmonics creeping in as my motor speed is beeing set in mm per min small changes of say 30-50mm min would see the motors become noisy and a further shift of another 30-50mm min and they would quieten down again so it could well be related to the power supply.
With my older win XP PC running the machine I can only attain 1000mm per min max reliably but with my otherDesk top machine running Win 7, a dual core AMD processor I can get another 200 mm per min out of it easily and quite smooth and quiet.
So there are a couple of factors influencing it I guess, so I will look around for a second hand later model PC to upgrade to, should help make it a little better over all.

[Nico13]

Well my good luck ran out today and the unreliability and missing steps has crept back in again and quite badly at times on all three axis. Had a play with the motor tuning but no joy.
Some code I run tends to be worse that other code, all code is coming from Vectrics Cut2D. When it works it works realy well but when it doesn't it's, well it just doesn't.
I had run the sample Vectric code for the little train Loco wheel and it cut perfectly two days before.
Was a hot day in the shed, dont know if there is any relationship there as I can't seem to get this thing going long enough to prove anything.

So next step is a different BOB and I'm going to get the Homann MB-02-V6 Bidirectional Breakout Board, it's Aussie made and just up the road from me so will give them a try and that fails they say that board is compatible with the smooth stepper or other motion control boards. They are unfortunately sold out at the moment but will let me know when they are back in stock.

[James Newton]

That sounds like a bad power supply connection... or a bad supply. Check with a voltmeter to ensure that ground is ground everywhere it should be (e.g. at the bob, at each driver as compared to the power supply ground). Then check the power supply voltage while running. And I mean the motor power supply as well as the logic supply.

My next guess would be corrosion or other bad connection in the data path. Especially since the original problem was about one axis after the other. And because it went away after you had disconnected and then reconnected everything. Disconnect and reconnect each end of the cable 20 times.

Then I'd guess the opto-isolators were hinky... I hate opto-isolators. Just me I guess, but they cause so much hell...

[H500]

It is a switching supply and I had noticed the possibility of some harmonics creeping in as my motor speed is beeing set in mm per min small changes of say 30-50mm min would see the motors become noisy and a further shift of another 30-50mm min and they would quieten down again so it could well be related to the power supply.

If the motors vibrate at certain speeds, then it's resonance. A damper should get rid of it.

As for the missing steps, try inverting the polarity of the step pulses and see if it helps.

[Nico13]

Thanks James I think you may have struck upon something with the possible port corrosion as like you say it is an old PC that has sat around with nothing connected to the printer port.
Checked it out today and after a few wriggles in and out it was almost running perfectly again so I will do abit more work on that.

I also checked the voltages at the individual controllers as Larken had suggested and that was dead steady also checked for any AC from the supply but it is a very clean DC.
What I have found is the step pulses measured on the axis that plays up drop off from a steady 3.7kHz to erratic numbers and the LED on the BOB for that axis also reflects this as it blinks rather than being a steady light when running ok.
This has always been the case with the axis that plays up, the step pulses go erratic.

H500 your idea of inverting the step pulse polarity was not some that I would have tried as I wouldn't have thought the stepper controllers would have accepted it that way around.

Here is the test I cut the other day when it first came back to life, it's a good one to use for testing.
I just need to get it to keep doing it, I'll get there in the end so thanks again for all the input.

Nico.

[James Newton]

You make steam trains. Or are making a steam train. How cool is that? Pictures?

[Nico13]

You make steam trains. Or are making a steam train. How cool is that? Pictures?

Yes and no James, this is from a selection of sample Gcode/CRV file from the Vectric Home page and it's really good to use as a test for the machine but I have almost made enough of them for a complete set and so I may continue and make the rest of it as a little project.
A note on the positive side is your suggestion re the corroded printer port socket was on the ball as I've given it a good working over like you said and the machine is up and running without missing a beat at this stage.
I will make some more train wheels as a further reliability test so that Loco may be made sooner rather than later.

The only other issue is some Gcode I produce from an old CAD program I have that exports DXF formatted files that I use in Vectrics Cut2D still miss steps for some reason, I may need to get some later software for that but can't afford it at the moment. Do you know of a good free Drawing packege that I could start with?

What I really made the machine for was to make Perspex templates for my wife for her Patchwork Quilts and also because I just like making things.
Thanks for your help.

Nico.