DDCSV1.1 4 Axis controller

The Shop > CNC

<< < (135/202) > >>

chriscnc:
Messed with it all day all G02 moves regardless of the code are breaking the arc's made 3 different parts/G-Code and all have the same issue I've run the code through simulators and it works fine. Ive run a few different post processors and same results. One thing may be my firmware is newer, so maybe its a problem with the firmware. if some one could please run the code and look on the draw screen and see if the outer line is broken on the end of the arcs. thanks much if it's my firmware I can just load an old version from the link in this forum

(Program: 1001)
(Toolpaths: 2D Contour3, 2D Contour4)
(Units: millimeters)
(Tool: flat end mill, D=3.175 CR=0.000)
(Workpiece: 0.000 <= X <= 152.000 | 0.000 <= Y <= 70.829 | -10.000 <= Z <= 0.000)
(Tool travel: -0.587 <= X <= 152.587 | -0.587 <= Y <= 72.369 | -10.000 <= Z <= 15.000)
(Safe Z: 15.000)
G90
G21
G64 P0.0254
G00 Z15.000
S 10000
M03
(--- 2D Contour3 ---)
S 10000
G00 X76.027 Y49.141 Z15.000
G00 X76.027 Y49.141 Z5.000
G01 X76.027 Y49.141 Z1.000 F762
G01 X76.027 Y49.141 Z-4.683
G01 X76.048 Y49.129 Z-4.804
G01 X76.106 Y49.093 Z-4.907
G01 X76.194 Y49.039 Z-4.976
G01 X76.298 Y48.975 Z-5.000
G01 X76.569 Y48.810 Z-5.000
G03 X77.005 Y48.915 Z-5.000 I0.166 J0.271
F2500
G02 X78.360 Y49.674 Z-5.000 I1.354 J-0.828
G01 X88.613 Y49.674 Z-5.000
G01 X88.613 Y50.723 Z-5.000
G01 X63.419 Y50.723 Z-5.000
G01 X63.419 Y49.674 Z-5.000
G01 X73.623 Y49.674 Z-5.000
G02 X75.211 Y48.087 Z-5.000 I0.000 J-1.587
G01 X75.211 Y18.107 Z-5.000
G01 X76.772 Y18.107 Z-5.000
G01 X76.772 Y48.087 Z-5.000
G02 X77.005 Y48.915 Z-5.000 I1.587 J0.000
F762
G03 X76.900 Y49.352 Z-5.000 I-0.271 J0.166
G01 X76.629 Y49.517 Z-5.000
G01 X76.525 Y49.581 Z-4.976
G01 X76.438 Y49.634 Z-4.907
G01 X76.379 Y49.670 Z-4.804
G01 X76.358 Y49.683 Z-4.683
G00 X76.358 Y49.683 Z15.000
(--- 2D Contour4 ---)
S 10000
G00 X17.238 Y72.369 Z15.000
G00 X17.238 Y72.369 Z5.000
G01 X17.238 Y72.369 Z1.000
G01 X17.238 Y72.369 Z-9.683
G01 X17.238 Y72.345 Z-9.804
G01 X17.238 Y72.276 Z-9.907
G01 X17.238 Y72.173 Z-9.976
G01 X17.238 Y72.052 Z-10.000
G01 X17.238 Y71.734 Z-10.000
G03 X17.556 Y71.417 Z-10.000 I0.318 J0.000
G01 X136.000 Y71.417 Z-10.000 F2550
G02 X152.587 Y54.829 Z-10.000 I0.000 J-16.588
G01 X152.587 Y43.500 Z-10.000
G02 X108.500 Y-0.587 Z-10.000 I-44.087 J0.000
G01 X43.500 Y-0.587 Z-10.000
G02 X-0.587 Y43.500 Z-10.000 I0.000 J44.087
G01 X-0.587 Y54.829 Z-10.000
G02 X16.000 Y71.417 Z-10.000 I16.587 J0.000
G01 X17.556 Y71.417 Z-10.000
F762
G03 X17.873 Y71.734 Z-10.000 I0.000 J0.317
G01 X17.873 Y72.052 Z-10.000
G01 X17.873 Y72.173 Z-9.976
G01 X17.873 Y72.276 Z-9.907
G01 X17.873 Y72.345 Z-9.804
G01 X17.873 Y72.369 Z-9.683
G00 X17.873 Y72.369 Z15.000
G00 Z15.000
M5
M2

Ryan7777:
Sorry Chris! I don't know enough about Gcode yet to give you an answer to your problem. But I can run your code in a couple of different bits of software and see if it's broken in simulation. But as you say, it may be more of a hardware issue. Someone with an actual DDCSV might be better help. I'm still waiting for my first one to arrive in a day or two.

I just wanted to toss in a request for Ethernet connectivity to Benedict on the new firmware front.

I know some of you don't like the idea of having your beloved mill attached to the wild world web for sneaky hackers to start turning out fidgit spinners using your mill bed for material, but I plan on simply setting my home routers firewall to block my mill from ever seeing the wider web! I just want local connectivity so I don't have to swap USB sticks every time I need to make a little change to my gcode or do a simple test. It's nice they give you a sacrificial connector on the controller, but I'm more worried about remembering to eject my drives safely than that. And I have bricked a stick or two by just that in my day.

Oh, and as far as reading the FPGA flash, that's probably where the axis difference resides. And it wouldn't be useless to read it, the FPGA is essentially Frankenstein's monster without a brain until it is powered up. It only knows what it is once it reads the flash for configuration, you don't burn anything to the FPGA, it can't hold anything on its own once powered down.

I would think if one wanted to really dive deeply, you could Frankenstein in socket for flash chips or just read from it on board, take the resulting bin or hex and fidgit with it bit by bit. It would be a slowwww process, but it's been done with other devices...

Respectfully,

Ryan

chriscnc:
Thanks Ryan7777,

The code is correct, I've tested it on other simulators and my tormach... the controller is just having issues with transitioning from a G02 move to a G01 move the straight line is always shifted off and broken. My firmware is stock as it was newer then the latest revision here so I didn't change it as there were notes from benedict he was porting the newer firmware, I just wonder if it is my firmware, I would roll it back to the modded one here if I knew it worked.

Ryan7777:

--- Quote from: chriscnc on August 30, 2017, 03:31:17 PM ---Hey group, so some advice on the internal vs external power, im converting a Multi-cam cnc router to our loved ddcsv. to original control is from 1998 and has to be drip feed... and sucks with fast 3d moves. I've built many CNC mills etc from the ground up but this one im doing more of a conversion reusing 90% of whats there it has nice IMS drives and power supplies and magnetic home switches etc.

So here is my question the drives have a nice 5v supply already tie'd to drive for the opto (3 wire 5v+,step,dir) can I leave the 5v from the PSU and not use the 5v from the DDCSV, as the grounds will still be the same so the signal will be driven low either way. second is the home switches, they are magnetic 3 wire but are tied to the 24v psu will the home input take a 24v v.s the 12V as noted and supplied from the ddcsv, I would hate to need to swap these out for 12v ones, or I can try and source a level converter. Basically, there is a real nice5/12/24v PSU all wired up already in there and am wondering if I can/should use it v.s the DDCSV power outputs and only tie the signals in, my concern is some of the loads such as these 1998 magnetic switches and opto drives may pull more than the DDCSV step down regulator is putting out.

--- End quote ---

Reading back a bit here about your mill and the power supply question.

Often AC to DC power supplies are fine sharing house keeping duties and will live together just fine sharing grounds. It's done all the time and shouldn't be a problem.

But I see a lot of people that tend to over think level shifting and resort to more trouble than they need. For example, it sounds like you need to shift 24VDC out of your magnetic sensor down to 12VDC?

I would advise you not to try to put 24VDC into a port that expects a maximum of 12VDC. It's likely optoisolated but you'll probably smoke the opto. Just don't do it.

That said, a simple resistor voltage divider is what you need! Two resistors is all that you need in fact. I've seen people suggest opamps at half gain and weird stuff like that which is totally overkill. A lot of parts that only end up going to waste.

I'll let you research the math :smart: but there is a quicker shortcut...

There are voltage divider calculators on the web that will let you input your starting voltage, your desired voltage and whatever resistor you choose for one leg, and it will give your the value for the other. The better calculators will tell you how much power your resistors will dissipate and how much current it will be drawing. Since you are dividing your voltage in two, it's really easy. Just choose two resistors of the same value! But since you are dropping half of 24 volts, you need to pick higher value resistors! Two 100k ohm 1/4 watts would probably be just fine. This will lessen the current draw on your sensor output and dissipate less heat, use less power and save more trees.

Bowden's hobby circuits hosts my favorite divider calculator because it calculates power dissipation and current. And it does the math!

Just use the 24 volt signal output out of your your magnetic sensor and tie it to the top side of the divider, the other end of the divider to ground, and the middle of the divider is now your new 12 volt DC magnetic switch signal. Easy. The grounds should already be shared.

Shifting up would require active circuitry or relying on the higher voltage supply and whatnot of course, but there are chips made just for that, opamps are great for amplifying AC, but a lot of hassle for D.C., just use a purpose made chip for DC shifting, no extra components and oscillation issues.

Sorry if you know all this already, but I see people who could stand to learn that you don't always need a complex circuit to do a simple job that a couple of simple resistors can do!

-Ryan

PPRINT:
Chris, sorry you are having problems with this. I know too well how stressfull these things can be.

There are a couple/few different ways to output an arc move in a cnc. Some controls are finicky about what they like...and I have already found out this is a finicky one.

Here is some code I run fine in my machine. Please give it a try and let me know if it works correctly for you. It is in metric units.

Set the G54 anywhere you want on the table and set the z height anywhere above the table. The program cuts down to Z0.0 so it should not hit anything.

edit:
The comments in the code below indicates a large material size, but just disregard that. The program only profiles a 6" x 6" box that has .5" corners.

(ARC_TEST)
(EnRoute Software)
(www.enroutesoftware.com)
(9/13/2017 7:05:06 PM)
(Material Size)
(X= 431.8000, Y= 762.0000, Z= 19.05)
G49 G80 G17 G90 G0 G94 G40 G98
T1 M6
(1/8" End Mill )
S18500 M3
G54 G90 X153.9875 Y76.2000
G43 Z2.50 H1
G1 Z0.0 F635.0
G1 Y139.6118
G3 X139.6118 Y153.9875 I-14.2875 J0.0882
G1 X12.7882
G3 X0.0000 Y146.2460 I-0.0882 J-14.2875
G1 Z9.5250 F20000.0
G1 Y6.1540
G1 Z-2.5400 F635.0
G3 X6.1540 Y0.0000 I12.7000 J6.5460
G1 Z9.5250 F20000.0
G1 X146.2460
G1 Z-2.5400 F635.0
G3 X153.9875 Y12.7882 I-6.5460 J12.7000
G1 Y76.2000
G00 Z2.5
M5
G0 Z25.4
M30

Thanks,
Rob

Navigation

 Message Index

Go to full version