Puruvesi MyCNC-ET7 Controller first look.

[Neale]

No idea what your problem is. All I can say is that I plugged it into the ethernet port, configured the laptop with a 192.168.0.address, and away it went. Can't say that I did anything clever that I can boast about! It's an ET6 rather than 7, of course, so might be slightly different firmware on the board? I think that I saw something about having to plug into a computer port rather than hub, presumably to avoid having to use a cross-over cable, but even if you are using a hub or switch, most of those are auto-switching these days anyway.

I have the limit switches configured and working, at least as far as the diagnostic LEDs going on and off as required. However, I've lost the printed copy of my IP/M manual where I noted down all the wire colours, connections, etc, which makes things slightly more difficult. Also trying to find out why the e-stop system isn't working. I have e-stop switches and fault outputs from the EM806 all in series with the safety relay so the relay trips in an emergency including driver fault. Can't get it to latch on now, so I suspect a wire has been dislodged somewhere but lost in the cable guide. However, the whole e-stop/relay stuff is independent of the controller - just switches a connection to the board's e-stop input - so the problem shouldn't be anything to do with the ET6.

Struggled a bit with motor configuration. There is a reference in the manual for the basic motor setup that says that speeds are units/sec (not units/min, like Mach3) but it's not defined for the "motor tuning" page. I've just set very conservative values as I would rather the motors creep than overspeed and trip the drivers.There is also a min, max, and "mid" speed to define - but I can't find an explanation of the "mid" speed. But I can't test any of that until I have the e-stop/safety relay working. E-stop handling in this software looks a bit more complicated than IP/M and Mach3 so that's going to need a bit of decoding as well.

I'll do a bit more reading tonight and have another go, fresh, in the morning. Wish me luck! I expect this will all be clear in time - I had similar issues first time round with the IP/M and Mach3.

[JAZZCNC]

No idea what your problem is. All I can say is that I plugged it into the ethernet port, configured the laptop with a 192.168.0.address, and away it went. Can't say that I did anything clever that I can boast about! It's an ET6 rather than 7, of course, so might be slightly different firmware on the board? I think that I saw something about having to plug into a computer port rather than hub, presumably to avoid having to use a cross-over cable, but even if you are using a hub or switch, most of those are auto-switching these days anyway.

No not connected through any switches or hubs etc, I never do. I've fitted 100's of ethernet controllers so it's not exactly my first rodeo battling with stubborn cards who don't want to talk to each other, so I did all the usual checks, cables, etc of which I've got dozens of various types, lengths, etc. However, this thing behaved oddly from the very first plug-in because it wouldn't show at all even when I pinged the board.
Only after I run some network trace software did it appear. To be honest I thought the IP on the board was wrong at first and was just about to dig into changing that when I spoke to Clive S who suggested running IP trace software and sure enough the IP trace found it. From then on it appeared when pinged and connected every time, very odd.?

The adapter and card connect ok and talk to each other no problems and the software loads up fine then it just closes the software.? Nothing I do can keep it from closing. This is all on Win10 with a fresh install and fully updated on a nice fast PC with SSD drive etc.!

The Win7 Laptop found it straight away stays connected, software loads, and stays loaded but it crashes randomly. All of this is with nothing connected to the board but an Ethernet cable and power.

I even tried to install it on a Linux PC that I have for Linux CNC but give up on that because Linux is complete F@#$ing mystery to me.!!

After all this, I was ready to smash the card into a billion pieces but I knew it was time to walk away. So I contacted Ivan at Puruvesi and he was great but nothing he suggested helped so we were going to arrange for an engineer to remote into my system but at the same time my grandaughter got sick which turned out to be Covid19 and all hell broke loose so not got back to it.
I'm sure when I do they will sort me out, or I'll put it in the 90ton press we are building and turn it into Tea coaster....


Watch this for the motor settings.

[Neale]

Wouldn't dare try teaching granny to suck eggs! But sometimes talking through a problem with someone who doesn't really know what you are talking about kicks off that Eureka moment. Didn't realise that the problem wasn't basic network connection (although my software found the board without me doing any config apart from setting laptop IP address in the right subnet) but myCNC talking to it. Starts to sound a bit like a hardware/firmware problem at board level? I'm also using the latest download of myCNC but haven't (yet) seen anything like that.

I fixed my e-stop/safety relay problem last night (and I really don't like to admit publicly to official Muppet status, but really I should have checked that *all" the e-stop buttons weren't locked in...). I also remembered that I took the fault output from the drivers to the "servo fault" input on the IP/M so I need to check how to configure that on the ET6. I think it's on the "Alarms" config page. About 2am, I realised the answer to an associated problem which comes from a difference between IP/M and ET6. I had hoped that I would need minimal wiring changes (if any) in the swapover - same 24V signalling, differential outputs, etc. However, the IP/M exposes both ends of each opto-isolated input so you can choose how to connect on an input-by-input basis. The ET6 groups input into blocks of 4. Each opto input has back-to-back LEDs so they are polarity-insensitive but each group has one end of all LEDs tied together to go to + or gnd according to requirements. That's fine for the limit switches which is one block of four, but then I have e-stop input from relay, touchplate, and driver fault. My previous logic needed different connections for touchplate (shorts to ground, of course) and the other two switch to +24V. However, it's farily trivial to modify those so that I can use a group of inputs with the common end to +24V and switch to ground for the inputs. One little puzzle out of the way.

Would all have been a touch easier if I had not lost my printed copy of the IP/M manual where I had listed all connections, wire colours, etc. Never did have a proper wiring diagram, just a few sketches and did the rest off the top of my head. Both unreliable record-keeping mechanisms...

[Kitwn]

[QUOTE=Neale;117501]I fixed my e-stop/safety relay problem last night (and I really don't like to admit publicly to official Muppet status, but really I should have checked that *all" the e-stop buttons weren't locked in...)./QUOTE]

I call this 'Finger Trouble'. Often embarrassing, but they're the simplest faults to fix.

[Neale]

...and at least I went to bed having solved one problem! Even if I was the one who created it in the first place...

You learn from mistakes (although in my case, very slowly) but it's better to learn from other peoples'. Plenty of learning opportunities from me, then

[Neale]

Well, some small progress this afternoon. Just in from the workshop for a coffee and a think!

Machine is now moving; homing works fine after going through the homing macro creation business. Actually much more straightforward than you might think from the manuals as the macro wizard does most of the work for you. Need to sort out "units/sec or min" issue as everything moving much too slowly but I think I know what that is. However, there's a teensy weensy little problemette with the gantry squarring. I've currently configured things so that Z/Y/X home in that order and I run the gantry squaring macro as a separate function afterwards. The idea is that once all is tested, I shall have a single button that homes then squares in one user operation. Homing and squaring look fine (although not yet calibrated, the gantry movements look like it's doing what it should). However, after the squaring process, the A motor direction is inverted so that when you jog off the homed position, each end of the gantry moves in a different direction. E-stop/reset gets things back in order, but looks like a bit more work needed here!

E-stop signal from safety relay now being seen by the ET6 although I haven't gone through the e-stop macro/PLC yet to check what it's doing. Not sure how to trip a drive fault to check operation there - drive one axis into a hard stop? And can't test the touchplate or start getting tool length macros in place until I resolder the connection back to its pin on the external socket where it broke inside the heatshrink. That had me head-scratching for a while.

All good fun and no real show-stoppers yet. But there's a way to go yet...

[Neale]

...and a bit more progress to report. In case anyone's listening!

First job was to sort out machine speeds. The only reference to this that I had found in the manual was that you used "units/sec" in the equivalent of the "motor tuning" page. Oh no you don't! It's mm/min (as mine is configured to use mm). Once I sorted that, movement speed was fine.

Although the homing macros worked OK, the problem mentioned above (A motor direction reverses after squaring routine) was just a bit of a problem. Well, show-stopper in fact. However, I magicked it away by not selecting an option in the macro wizard about "reset work position after alignment". Partly because I realised that I didn't know what this did, and partly because it seemed to be generating a line in the macro that didn't make any sense. After unticking the box, the code worked fine - apart from the fact that it did not set the X machine coordinate to zero after the process completed. I tracked this down to another odd-looking line in the generated macro that called a function that I couldn't find in the manual, and given the parameters passed to it, I couldn't see what it was doing that was useful So I removed that line as well. In fact, whatever that line was supposed to do, it seemed to crash the macro at that point so the rest was being ignored, including the machine zero set. All now working fine, and my "home all" button runs through the Z/Y/gantry zero-and-square sequence very nicely. I shall be reporting these findings in the support forum and see what they say.

Then on to the spindle. Couldn't find how to specify which output pin to use for "spindle on" but I tracked down the one it was using by default and swapped wires to use that. Spent quite a bit of time trying to find how to set spindle parameters - there's so much that's configurable that it isn't always easy to find the appropriate config page - although once you find them, there's a lot you can do. Set min/max spindle speed, and somewhat to my surprise, the spindle starts and runs. Speed calibration wasn't that far out but I played with the calibration parameters enough to see that I can get it pretty close. Mach3 and the IP/M between them gave me speeds that were around 5% out which is perfectly acceptable but with a little bit of work I now have speeds within about 2% and can probably do better (reading speed from the VFD display which is probably fairly accurate as it's based on the supply frequency to the motor). Happy with where I am there.

Thought I'd then give the wireless MPG a go. The myCNC online docs tell you how to download a generic driver for the USB dongle, which I did. Turn on MPG, twiddle the dial, and the machine moved! That was quite a result as it wasn't entirely clear that my model MPG was supported. Went to the MPG config page - again, there's an awful lot that's configurable - and chose a "default" option that seems to most closely match my MPG button layout. What's particularly good about this config page is that there is a series of on-screen LEDs that show you which button has been pressed, and a table that lets you map button press to machine action. Things like spindle on/off and speed change from the pendant all work but I need to do a little bit more work here to sort out the rest. The MPG wheel works a bit differently to how it worked under Mach3. With Mach3, you had a choice of "step" or "continuous". There is a knob which selects step size (step mode, obviously) or speed (continuous). The step/cont button seems inoperative under myCNC, but the MPG wheel gives an approximation of "faster you turn, faster the machine moves", with slow speed giving single-step action. Step size switchable, of course, and the whole thing actually works better than under Mach3. In fact, it's all looking pretty good and I'm confident that once I have my tool-height setting macros in place (bit of development work needed there) I shall be able to assign them to my usual MPG buttons.

Pleased with how it's going so far. One of my biggest gripes is the difficulty of finding what you want in the online docs. Just about everything I've needed I have found eventually, but the search function seldom helps very much. For example, the docs on setting the gantry squaring is very comprehensive, but almost impossible to find.

[Clive S]

and a bit more progress to report. In case anyone's listening!

I'me listening

[JAZZCNC]

Looking good Neale keep it coming..

[Neale]

Fixed my broken wire and the ET6 can now see the touchplate. I've also tracked down a sample macro showing touchplate use (it was under "surface mapping" which is why I had problems finding it) and I think I can use it as a basis for my touchplate macros. The router has movable and fixed touchplates and I use a set of three macros for simple touch-off, touch-off with secondary reference to fixed touchplate, and tool height setting from fixed touchplate. However, once you can figure out how to use probing/touch-off techniques for a particular machine, record fixed touchplate position, etc, actually putting the macros together is relatively easy.

I also mentioned problems with identifying output connections. There's a bit of inconsistency here, although there might be good reasons for it. Sometimes you go to a defined page to set particular input pins. For example, this is how limit switches are defined. Sometimes you define a sensor input pin when creating a macro, as for the homing macros. But for the motor output, the pin number is defined in a separate "pins.h" file (which will be familiar to any programmers out there) which is then "included" in various bits of PLC code. As I say, I can see why there might be reasons to do things this way but it makes it all a bit tricky for a newcomer to the system. (Edit - the contents of this file can be edited within the myCNC environment under "Hardware PLC templates" - again, the name didn't really help find it but it is all there).

At that point, I couldn't think of any more good reasons not to cut something, so I downloaded the myCNC postprocessor for Vectric (there's one for Fusion 360 plus a couple of other CAM packages that I don't use) and created a quick bit of Vcarve engraving. I just used a bit of scrap ply that was lying around but here's evidence that it's working! Well, I could be faking it but trust me, I'm a doctor!



Image is slightly distorted but that's just camera angle.

Might take a little rest now before starting out on the tool height macros.