[Tophness]

Hey guys, first time doing anything with CNC. Just ordered a chinese 6040Z from eBay, but I heard the controllers were crap and I didn't want to run it on an old PC with parralel, so I've ordered the recommended upgrades in advance.
48V PSU, 3x DM556 drivers, and now I'm wondering what motion controller breakout board I need.
I can only find 24V ones. Is it possible to use a step-down converter to power the board? Or will this mean the drivers are only getting 24 delivered?
Are there any other than the all-in-one geckos that can handle 48V? Preferably with a native USB interface

[john-100]

The DM556 stepper drivers have opto-isolator step , direction & enable inputs are designed for 5V logic signals
( by adding external 2K resistors they can be connected to 24V outputs from a PLC )

which motion controller would you like to use ?

As long as it outputs 5V step ,direction & enable signals you only need to provide a low power DC supply for the motion controller
( the DM556 stepper drivers being powered the 48V DC supply )

John

[ger21]

What software do you want to use? That will determine what controllers you can use.

The 48V powers the stepper motors only, and does not connect to your breakout board/controller.

[Tophness]

Ahh I see. So VDC+ is connected directly to the PSU and the motion controller just sends pulses to Signal.

I guess Mach3 will do for software since it's going to come with the 6040 and apparently there are preloaded config files for it.
It'd be good to just go straight from Fusion 360 to sending the gcode to a board that's isolated enough to connect it to my main windows 10 PC.
This one looks like it outputs the 5V signals. I guess it doesn't matter if it uses an extra 12-24 power supply since it's isolated from the motors' PSU anyway.
https://www.aliexpress.com/item/32977965996.html

[john-100]

looking at photos found on line of the motion controller

Attachment 424396

Attachment 424398

I expect the limit switch common GND and PWM 0-10V analogue GND will be connected together

as on this parallel printerport breakout board

Attachment 424400

if you connect the 0-10V analogue output to a VFD with an isolated input you have no problem but

spindle speed control boards like the XMI-2360 with live input terminals that are usually connected to a 5K potentiometer
you would end up with the limit switch wiring being live !!!

XMT-2360 sieg mill speed control


John

[Tophness]

Yeah I was looking at that breakout board too. Do the 6040 VFDs have isolated inputs?
If there's an easier option I'd rather go with that. If I can just switch over the wires with the new components that would be ideal. Does it help if I power the motion controller with a step down converter coming from the VFD?
I'm not sure how the spindle board even gets it's power or connects to the motion controller, I just saw so many people doing it without any details about how to upgrade that I assumed everything must be labelled in a standard way

[john-100]

I would not like to stake my life on all 6040 VFDs having isolated inputs

photos from various posts show different manufacturers use different electronic boards
so the control boxes may look the same on the outside but the inside being very different

for power the VFD will be connected directly to the mains supply
the spindle speed controlled by the 0 to 10V analogue control voltage from the breakout board or motion controller


John

[Tophness]

Oh so VFD and the power supply both have their own mains supply. I would have thought it would be connected to the PSU in some way since it was one of the VFD/controller box combos that only has 1 IEC jack. I guess they're both just connect to the same one.
I have a little electrical engineering background, but this is a bit foreign to me and I can't find any schematics of anyone else who's done the upgrade so I'm just trying to picture it all in my head. Hopefully it will be self-explanatory when it gets here.

If the gecko g540 is just a matter of switching cables over, I'd rather do that instead and just reuse the new DM556 drivers for something else.
Last thing I wanna be doing with a new several kW machine I'm unfamiliar with is soldering new components and hoping everything's grounded properly

[Tophness]

It looks like this guy has done it with the motion controller you were talking about.
https://www.youtube.com/watch?v=ID8v34JxGnU
It looks like every EN+ DIR+ and PUL+ are all just supposed to be wired permanently to 5V

I've attached how I think it's supposed to be wired up. Did I fry anything?

[Tophness]

Oh I would have fried it. I get what you're saying now. The PWM and GND is supposed to be connected to the pins where a pot is on that speed controller board for the spindle, not the VFD.
And you would get a nasty shock if you added limit switches and touched them? Including the automatic zeroing tool?

Ok so what about an optoisolator like this? https://www.aliexpress.com/item/32952549335.html
Or should I make sure they're powered from the same source? Maybe use the PWM to control a digital pot? Isolate the limit switches?
If it's a flaw in the design of the motion control card, maybe a higher quality board like one of these would be properly isolated? https://www.aliexpress.com/item/32910734344.html https://www.aliexpress.com/item/33003165433.html

[ger21]

I guess Mach3 will do for software since it's going to come with the 6040

Mach3 is $175. Your 6040 will not come with a legal license. It'll either have a pirated copy, or you'll be running in demo mode.

[OlfCNC]

Preferably with a native USB interface

That was a big mistake, because you can then only use their junk USB card and pirated mach3.
If you were to order the machine with printer port instead then you could use any motion control cards like mach3, mach4, UCCNC, LinuxCNC, PlanetCNC etc.cards.
Also you could then extend the machine if using a more I/O motion card.

Now you are limited to a buggy pirated mach3 and the number of I/Os are what you have on that motion card. Or if you want to change or extend then you will have to rebuild the control box.

[Tophness]

That was a big mistake, because you can then only use their junk USB card and pirated mach3.
If you were to order the machine with printer port instead then you could use any motion control cards like mach3, mach4, UCCNC, LinuxCNC, PlanetCNC etc.cards.

I'd rather not invest in software at this point. The demo version should even be fine since I won't be doing anything advanced, and MACH3 seems to be popular.
Isn't the parallel port board essentially just allowing the desktop PC to control I/O pins to the drivers with an old shift register? I'd rather free up the PC and not rely on a dedicated PC's timing to do it.
If I'm misunderstanding I'd love to know.

[Tophness]

After watching some youtube videos, I think I was right the first time and the PWM signal goes to the VFD?
Is this connection right?

Also, what limiter switches do I need?
I was looking at these SPDT ones, but is there any way to make them safer? Should I get normally closed or normally open? https://www.aliexpress.com/item/32848486852.html
Surely they can't all have this design failure where the switches become live. Are there any other boards under $150 that would be recommended?

[Tophness]

It just got here. Looks like mine is a little different?
There doesn't seem to be any spindle control board, the VFD is just a single board and takes AVI COM and FWD wires from the parallel breakout board, as if it were already set up for controlling the spindle with MACH3 via breakout board.
It looks like it has opto-isolators though. I guess it should be safe to just switch out the parallel breakout board with a USB one, add in new drivers and 48V PSU?

[damo1023]

Hello,

just want to do the same - upgrade to DM556 and upgrade motors.

Now for a year, original machine with TB6560 works ok. But now I have some other project where I need 2 motors and drivers, so it is a chance to upgrade my 6040 and use existing motors and drivers in new project.

I plan to:

- install supported rails and 10mm aluminum backplate on X axis
- change power supply to 48V 10A
- add another supply 24V 1-2A or step down 48/24V for supplying controller etc.
- buy 2 Nema 23 3Nm motors
- buy DM556N drivers

- Y axis: install 3Nm motor
- X axis: leave original 76mm (I guess 1.9Nm) motor
- Z axis: replace original 56mm (1.3Nm) with 76motor from Y
- A axis: install 3Nm motor

- 2 motors 1.3Nm (from Z and A) use on another project


Now, I've found 2 similar Nema 23 3Nm motors, difference is one is 3.5Amp, other is 4.2Amp. I guess I'd go for 3.5A, since I will drive them with 48V.

Will 48V 10A psu be sufficient?

Any comments appreciated!

@Tophness - do you feel any difference when you changed the drivers?


regards, Damo

[Tophness]

@Tophness - do you feel any difference when you changed the drivers?

I kept the original (unknown/unmarked) motors and used a 48V 8.3A PSU for the motors and a 24V wall wart adapter for the BOB and they run fine.
From what I understand the microstepping of the DM556 is more dynamic and efficient with more protection, so it shouldn't need a huge PSU, but if you're going to use 4.2A motors, I would probably go for atleast a 10A PSU.
My one wouldn't fit in the same slot as the original PSU. Still trying to work out how to ground/shield everything. I was going to connect an earth wire from the new PSU to the earth on the IEC power jack instead of using the wire from the jack to the chassis, but after read conflicting information about ground loops, I've decided to leave it as is without any connection to mains earth and just try to shield the PSU in another case.

[damo1023]

Hello,

I've made all the modifications. Now I am losing steps in Z axis. Does anyone know current specification of original X and Y motors?

Below is the picture of new setup (not full wiring yet done on this pic)

[damo1023]

Now I am losing steps in Z axis.

Actually I am losing steps on ALL axes. I've heard this might be a timing problem with step/dir and DM556. Did anyone bumped on this one? How did you solve it?