First of, i must say i think this is my first post of hopefully many on this forum. i'm not sure if this is the right place to post this or not, but i'll give it a try . We recieved our phlatprinter Mk2 this summer, and have finally put it together, without too many glitches.

The only thing we did wrong, was that we managed to wire one of the phases wrong on the stepper motor for the Z axis, and by doing that we managed to burn fry it, loading it too long with one of the phases having opposite polarity of the other. So we ordered a new one of Ebay, A vexta 160oz inch thingy. Now the z gantry works somewhat ok. The motor runs smoothly and sounds like a motor should.

Firstly i set the motor up with switched polarity on both phases just to test it (i didnt bother to take out the card to look at the back of it, and i couldnt remember if it was A+A-B-B+ or B+B-A-A+), and at first glance it seemed to go really good. And after removing the inverted setting on z-axis it seemed to fit in to some of the g-code we got with the printer. So we where on to start our first real test cut.
And thats where the problem started, when we ran the script the z axis gradually increaset in cutting depth for some reason, not showing any increase in g-code or mach 3..
So we went on to switch the wires from the stepper motor, to the right setting. And before doing more cutting on our precious depron, we calibrated the axis.
Our test cut then seemed better at first, but then the cutting depth of the z axis gradually decreased without showing any signs of doing that in the g-code or the software.

So does anyone have any idea as to why this is happening?
And what can i do to fix it?

___
Tor Stonny L. Knutsen
Søgne Modellklubb
Norway

 

I don't know the solution, but here is some stuff you can read of similar problems:
viewtopic.php?f=135&t=993#p8686
viewtopic.php?f=140&t=1101

-Kwok

 

Y'know, I'm beginning to think it's a problem with the controller card, that it increments the position of the Z-axis by, say, .001", for every motion of the Z-axis. Like the chip has some counter value that's getting added (or subtracted) from the stepper motion command.
I wonder if there's a way to empirically test this, because the phlatcode values are the same for every motion in Z. :geek:

By the way Tor, welcome to the PhlatClub!

 

Dorsal, you might be on to something. All the hardware driver chips out there require a certain amount of time that the parallel port pin needs to be signaled to stabilize.

I don't use Mach or an ezcnc board so can't really do any useful testing. I have a feeling that the settings in Mach aren't holding the dir pin high long enough and you end up with a step in the wrong direction when the conditions (gcode line) are just right. I have seen posts where the problem was solved by switching to EMC2 (I think iflyos) and Frankrcfc went with a Hobby CNC board that uses a Sanken driver chip.

Is there an option in Mach to increase the hold time as required by the chip in the driver board?

 

My X was doing that. I made a test cnc file so the x would go an inch in either direction like a washing machine agitator. As it ran it kept getting wider and wider past my static marker taped on the top by the x roller.I worked for days trying different things before I decided it was the board. It was the controller section for the x that gave out. I got the new board and all has been fine since.

 

I had similar issues with the Z axis. I set the option to use the Sherline 1/2 pulse mode in the ports and pins configuration and solved my problem. You might want to try it.

Mine has been rock solid since doing this. I had tried every combination of velocity and acceleration setting in the motor tuning and could not eliminate the problem. I even changed controller boards and put in a new Gecko 540 controller but the Z axis problem persisted. It didn't happen often, but every now and then I would find the top skin of the Depron was not cut and it was because the Z axis had missed a step.

Once I set the Sherline mode, it has never missed a step since.

 

Ok,
Firstly, thanks alot for the detailed replies :-D

I think i'll try the Sherline mode first, if that don't work i'll try to change driver card and maybe start using emc2.


-tsknutsen

 

so this is what has happened since my last post.

We tried Sherline mode, didnt work. Then someone had an idea that we might have adjusted the vref wrong.... Well that we had not, and we managed to burn one more motor.
we bought 3 brand new vexta motors. Tried again and the problem persisted, ofcourse.

so we contacted easy-cnc, and behold, the opto-isolators on our board was from a bad batch. they sendt us new opto isolators. it took about 1 minute to replace them.

And the machine worked perfectly!
by now we have only have done one test cut, but the cut was completely and uttely awesome!

I will post a video as soon as we get to do more cuts, i just got to read up using phlatscript since all of our depron etc is in metric, and 3mm. And most of the planes etc i've found here is for depron 1/4" or something.

Again, thanks for replies and all.

-tsknutsen

 

I can't believe I missed this.
I had the same prob. I contacted easycnc and they told me to calibrate (for the umpteenth time ''that didn't work''). I switched to the hobbycnc board and steppers (problem solved). I'll have to get in touch with easycnc, if I can get the same service, I will have a ready to go set up for the ''Balsa Buddy''.

Jeff

 

Wow,,, wish I had found this sooner, I could have saved over $400 in replacement costs that still haven't fixed the problem. Guess I'll contact Easy-CNC next.

 

Too bad there has been more people with the same problem, and less without the same sollution.

When i contacted easy-cnc they told me to send a picture of our card, we did.. and we got the answer that we had a card containing bad opto-isolators. They also sendt me some pictures of how the cards with good opto-isolators should look.

So here are the pictures i got from easy-cnc. One is with the bad opto isolators, one is with the good.


-tsknutsen Attached files

 

tsknutsen, Thank you, Thank you, Thank you. Mine is like the 2nd pic on all 3 axis. I was about to give this board to anyone who wanted to check it out.

Jeff

 

Looks like they redid the pcb. The older ones look like this. And the two black chips have a slightly different part# and more legs:

Attached files

 

Well I will be...I just looked at my opto's and they are from the bad batch as well. Now the question is do I run to the parts house Monday and see if they have to good ones in stock, or do I call/email Easy-CNC?

I already checked my parts pins and I don't have anything that will cross to the new optos. Darn shame to, since I use them by the handful in my amateur radio interfaces...

Tim
AG4RZ

 

I wonder...should those of us with the problems post up when we bought our controllers, just so we can see the possible date range?

I ordered mine on the 5th of May, and got it on the 12th of May...

Tim
AG4RZ

 

Tim,
I emailed easy-cnc Wed night and they have not responded yet.

I'm not a sparky, so I'm wondering if I would have to change other parts, beyond the opto's, to ensure my board configuration will work properly.

Like you, I was one of the early MK-II buyers and received my electronics about the same time as you. my board has two matching chips on the opto socket, but the P/N is different than those previously pictured.,, mine 0749 over 815 over @y.

Dell

 

I have an invoice date of April, 20, 2009, I have the bad optos.

Jeff

 

I think easy-cnc should post the bad chip part #'s/date ranges on their site so their customers will know that they need a replacment part.

 

Dell, what the opto's do/are are Opto-isolators/Opto-couplers. What that means is pretty simple:

Basically, they consist of an LED and a photo-transistor that conducts current when it sees light.

Nothing downstream *should* have to be changed; the reason being the sole purpose of an opto is to isolate one electrical circuit from another, to prevent damages.

There are different types of opto's, some faster than others, some have more current handling ability, etc, etc. I need to look up data sheets of both the faulty and good opto's and see what the difference in specs are, if any.

Of course, it could be that this particular run of opto's were just bad...that happens, unfortunately!

Tim
AG4RZ

 

Tim, do you know what the numbers mean. Are the L# a lot number? I'm thinking the CNY17F-4 on mine is the product number? What about the @Y ?

 

Kwok, CNY17F-4 is indeed the part number for yours. L could very well indicate the lot #, and the @Y I have no clue.

It has been my experience that manufacturers will put all kinds of markings on components for their own info, as well as for a bulk purchaser, like an internal part number...

As soon as I get a couple honey-do projects finished, and cut a couple planes, I will see what i can dig up..

Tim
AG4RZ

 

The top line is the manufacture year and week. The @ symbol marks the anode (and pin 1) and the Y is the factory code.

 

LOL....Gas for the win! That will teach me not to think outside the "Motorola" box about components!

Tim
AG4RZ

 

I just sent easycnc an e-mail. Hector replied within an hour, asking for pics of my board, so he can ship the right parts.

PS; I am not a picture taker, I had to take over 20 shots to get 1 where you could read the numbers on the chips.