Re: [lasersaur] Re: Machine Stop Error

[Ward Elder]

Make sure you only ground one end of the shield on shielded cables. Isolate the other end (tape it up)
Thank you
Ward Elder
Eldersoft
(204) 791-7754
 

From: laser team [mailto:la...@seeclab.com]
Sent: Monday, December 17, 2012 05:58 PM
To: lase...@googlegroups.com <lase...@googlegroups.com>
Subject: Re: [lasersaur] Re: Machine Stop Error
 

> I already have the latest version of the software v12.08f, and now all the
> wires are shielded and shields are connected to the ground of the aduino,
> which is connected to the ground of the power supplies of the laser and of
> the motors ..

Woahhhh....IIRC, the shield wires for the motors are not supposed to be
connected up (at least that's what the schematic shows).  I don't know
whether that's an oversight in the schematic or whether there is some
deeply cunning reason why it's wired that way (eg because the 
modules are opto-isolated from the rest of the system).

 

I'm not grounding the shield of the motor wires .. I'm grounding the power supply itself .. the power supply of the motors have a ground terminal that is connected to its chassis, this terminal is not electrically connected to its zero volt terminal ..
I'm using a stepper driver called Leadshine M880A instead of a Geckodrive .. it's also optically-isolated, and the zero volt terminal of the motors is not connected to any ground in the system ..

 

Also, you aren't supposed to ground the Arduino.  It's powered by the USB
cable which is grounded to your computer.  The Geckodrives are
opto-isolated for that very reason.  The schematic doesn't show *ANY*
ground from the Arduino board...possibly for this very reason.

If the ground in the lasersaur is floating relative to the ground in your
computer then much current could flow along the USB ground wire...that's
"A Very Bad Thing".

I'm sure Stefan will have something to say about that.

I not sure I understood this part .. what I think is that (correct me if I'm wrong) the ground of the Arduino is connected to the ground of LasaurShield (to which the shields of wires are connected), and this is the same ground of the USB (the ground of the computer) .. and this is the terminal that I connected to the water pipe with the ground of the laser tube and the ground of the motors power supply .. so what is the part I'm missing ??

> The problem is that in Egypt we have no absolute ground in residential
> buildings (the wall socket has only two terminals "neutral and line") ..
> grounding only exists in industrial plants and big commercial malls ..
> this is why I had to connect the ground wire of the whole system to
water pipe
> passing near the window :)) .. I don't know if this is an effective
> grounding technique, but I only wanted the ground wire to be connected to
> something that is buried in the earth ..

Well, so long as all of these shield wires are ultimately connected
together and to the frame of the machine, I don't think it matters whether
they are connected to "absolute ground" or not...but if it does matter
then I'm fairly certain that grounding it to a water pipe (assuming it's
copper, lead or iron) will do the job.

I could not connect this ground to the frame of the machine, because the aluminium T-slot I bought is coated by a thin isolating material .. I asked a fried and he told me that this material is due a process called anodizing .. however, I tried to make the continuity test on the surface of the aluminium with my multimeter, and the buzzer didn't beep .. even if I removed this coating with a sand paper or so, and connected the ground to this segment, it would not be connected to all the frame because the segments of the frame are not connected to each other because of this coating ..

How is your computer grounded?  If it's not - then I suggest tying the
ground of that to the same piece of copper pipe that the Lasersaur is
grounded to.

I'm running the machine on a laptop .. I don't know how to ground it, but as I mentioned above, the ground of the USB which is the ground of the laptop is connected to this pipe through LasaurShield ..

> One last thing I think is important to mention .. this error of the write
> timeout happens more frequently when I'm using the laser .. but when I run
> the machine with the laser switched off (only running the motors), this
> error decreases greatly .. I actually don't remember that it has ever
> happened without the laser .. so what do you think ???

Decrease...but doesn't go away entirely?  That's a very significant
statement!!

That means that the problem isn't *specifically* due to the laser itself -
but that the laser just makes it worse.

That *STRONGLY* indicates electrical noise issues.  The laser itself
produces lots of electrical noise (because it's high voltage, significant
current and undergoes high frequency switching)...but if you're seeing
occasional problems with it turned off - then your system must be
teetering on the bring of electrical noise failure all the time and the
laser just pushes it over the edge.

No, The problem does not happen without laser .. but I was not sure of this, this is why I said I don't remember that it has ever happened without the laser .. but today tried running the machine with a certain svg file (which was about 10 minutes long) more that six or seven times without laser and it worked fine .. and when I tried the same G-code with laser, the error happened after less than 10 seconds and the machine stopped .. so now I can say that the problem disappears without laser not only decreases greatly :)

 

What wiring are you using?  Was it the wire included in the BOM or did you
buy it elsewhere?  Some wiring has braided shielding, others have
foil...different conductor diameters...those things could make a
difference to the quality of the signals.

I'm using different wires than what are in the BOM .. I'm using for the motors 4*1 mm shielded cable (4 wires - 1 mm each) .. and for other signals 2*0.25 mm shielded cable .. Shields of both cables are foil covered with braided copper ..

Is the red power wire for the laser tube routed close to any of the other
electronics and cables?   I kept mine as far away as possible...maybe that
helps?

For this connection, I'm using a wire that I got out of an old TV flyback circuit .. it is the only place where I could get a high voltage wire .. It is very short (only 30 cm long) .. and the Arduino with all the electronics are placed about 1.7 m apart from it (they are placed in two diagonally opposite corners of the frame) ..

You could also try using a USB cable with ferrite bead...in case the one
you're using doesn't have one...or buy a ferrite ring and wrap the cord
around it a few times.

The cable I'm using already has this bead .. 

 

At the end I want to thank you very much for your time .. 

Regards, 

Ragy Samy ..