Mach3 control of spindle and coolant

I'm just finishing a design for a control board to interface Mach3 to my CNC project.

formatting link
(impression of the pcb design if anyone's interested).

I notice that Mach3 allows for output signals for CW/CCW rotation of spindle and Flood/Mist on the coolant front.

I have implemented three relays and I was idly pondering on how this will go.

I can forsee one spindle relay (CW) and one coolant relay for a mill and two spindle relays (CW/CCW) and one coolant relay for a lathe.

Do mills go backwards?

Now I can't envisage a situation where I would want to program Mach3 to switch between Flood and Mist on the coolant front, so I'm not too worried about only having three relays. I'm thinking that I will set the coolant up before I run the G-code and it will be sufficient to use Mach3 to merely switch it on and off.

The fourth axis could (if not required as an axis) be re-purposed to give a further 3 relay outputs (or indeed speed control of the spindle but I haven't investigated this further).

I would appreciate any comments or observations the group may have.

Reply to
Dave Osborne
Loading thread data ...

SNIP

What are you using as the 'opto'? Have you checked that your computer port can actually drive it? Even with the PCI cards I'm still finding that while the slow opto's can be driven fairly reliably, getting enough power to drive a fast one directly is a bit hit and miss. REALLY need to buffer the input side, but with relays on output, and stepper drivers that already have isolated inputs, do you really need a second layer of opto's anyway?

Reply to
Lester Caine

Good point.

I've done extensive research on the web and it seems that you can only rely on the parallel port to provide a maximum of 4mA. The KB814 looks like it will fire reliably at less than this. I can always try Sharp PC817s later if I have problems (if I can find a supply at a good price).

REALLY need to

Well, the point of the Opto's as close to the PC as possible is to protect the PC from problems with your electronics; to protect your electronics from noise coming from the PC and to provide galvanic isolation between the grounds, so I would have thought that *this* layer of Opto's is the one you really need!

In fact, the only vulnerable part of my design is the slotted opto switches I'm using for home detection. I'm not prepared to buy industrial units, so a fault on one of them may interfere with the 5V supply. Easily identified and fixed, though.

Cheers.

Reply to
Dave Osborne

I'm having similar thoughts about my own project, for a mill. But I'm planning on commercial boards, I just have to decide how to set them up. Most of these only have two relays available. One of my designs (I have several) contemplates one output controlling the start (and stop) of the spindle, but having a mechanical disabled, forward and reverse switch to control direction. I can't see using reverse that often, but if I miss it off, I'll find a use - tapping?. The off position is useful (essential?) to allow a dry run of any routine, without a tool, and with the spindle stopped, just to check clearances and to look for program errors. And it's an extra comfort when you have your head in the cabinet. The spindle is three-phase, but I think I have enough outputs to have 4th axis as well as a speed signal for the inverter.

Alternative design has manual coolant control, allowing two relays for forward and reverse.

I see that some commercial spindle control boards don't have a stop- start relay. They turn speed to zero so that the stop is implied. They only need one relay to switch to reverse. I've tried this with my inverter, and yes, the motor does stop completely, and doesn't hum, so the inverter is clever enough not to waste energy. But I'm nervous seeing the "run" light flashing away: I'd like the comfort of a postitive stop.

Wilfrid Underwood

Reply to
Wilfrid Underwood

Thanks Wilfrid. I also would be a little nervous about implementing spindle control without a positive "stop". I have a conventional run/stop circuit in my design which drops out the 24V to the stepper drivers and to the (external) spindle/coolant relays when a limit is hit or estop is actuated. All commercial inverters I have come across have the facility (somewhere) for a conventional run/stop circuit, so it would be a case of manually hitting stop if you want to poke your head in the machine. My design is finished now. I've taken the brave (!) step of sending off to China to have boards fabricated.

Earlier in your post you mentioned tapping as a reason to want to reverse the spindle on a mill. Are there any other reasons why you might want to run the spindle in reverse?

Cheers, DaveyOz

Reply to
Dave Osborne

I've not actually used a mill spindle in reverse, so can only invent hypothetical situtations. I've not found any left-handed cutters. I suppose I could mount horizontal cutters onto the arbour backwards, but then I never seem to use horizontal. But a friend has an aluminium plate on his Taig mill, on top of the bed, that is peppered with threaded holes. This makes workholding, particularly of thin material, very easy. So project 1 when I get the new mill going, is to make a similar plate for myself. Lots of holes to tap!

I've been looking to the States for inspiration - particularly pmdx.com and warp9td.com. One has a board that can handle two parallel connections and/or generate more input/output signals from one link. The stepper signals require a whole pin to themsleves, but less time-critical links use a time-sliced approach to get more than one signal down (or up) one pin. They've re-developed the Mach3 driver to support this approach. The other firm has developed a USB link to Mach3, opening up the future when the parallel port is a thing of the past. I've no elecronics expertise so will defer to you and Lester when it comes to building it yourself.

Wilfrid

Reply to
Wilfrid Underwood

See, I would just get a steel plate lasered and tap it on a pillar drill... What's the advantage of aluminium in this instance?

I couldn't think of a reason to run the spindle in reverse at all. I never thought of machine tapping on a mill. I can only assume you are talking about using a machine tap, rather than a flying cutter otherwise you would not need to run the spindle in reverse?

I'm not sure how using a machine tap would work with motorised Z feed - Would you synchronise the feed with the pitch of the tap or would you have some kind of special tapping head that drives the tap but lets the tap move vertically of its own accord once it has started cutting?

In terms of control, I'm waiting for the printed circuit boards from China. I've ordered four boards, so there's an opportunity for someone to have a play if they want. I would be looking to recover the cost of the board, though. I'll post a photo if anyone's interested.

Cheers. DaveyOz

Reply to
Dave Osborne

On my friend's mill, I suspect the aluminium is used to reduce weight, both because the mill is lightly constructed, and to reduce delivery costs! My mill is heftier, so steel is a sensible option

Yes, I'd need some kind of tapping head to allow the tap some vertical float, as I cannot exactly synchronise the spindle with the z-axis. I have seen these with a manual guidance - I envisage adding a spring for auto-feed. But - here's the amusement - I know just where I can get (borrow/use) one that can be used on a pillar drill. It doesn't need the spindle to reverse, as it contains a gearbox that auto- reverses as the tap is withdrawn. So even though I shall build in a reverse feature in the mill, I'm not likely to use it.

One step at a time - I've got the spindle/inverter set-up running OK. I've now ordered some control boards and have a PC with Mach3 standing by. I need to sort out the drilling of the stepper motor mounting plates, and the fixing of the belt pullies to the shafts (there's a 2 to 1 reduction here). Then I can lash-up the wiring so the machine can work, and I can begin to use it to make the other parts I'd like.

Wilfrid

Reply to
Wilfrid Underwood

Yay! PCB's have finally arrived. I can get down to some serious mucking about in the shed!

formatting link

Reply to
Dave Osborne

Lester, you didn't speak a moment too soon. This parallel port business is turning into a right PITA.

The Parallel port on the PC is fine. The additional IEE1284 port PCI card I bought is doing my head in. If I use it as port 1, the outputs

2~9 are not man enough to drive the opto's and if I use it for port 2, Mach3 doesn't seem to be able to make pins 2~9 as inputs. To make things worse, I can't find the CD that came with the PCI card that has the drivers on... I need to see if there's a utility that allows one to change the "mode" of the port.

Cheers, DaveyOz

Reply to
Dave Osborne

Been there - got the t-shirt .... Plug the chip number into google and it should point you to generic drivers. I'm using them rather than the disk that cam with the board.

I scrapped the opto-input board here simply because it was not going to work. You need good buffers on the incoming signals. It IS almost a case for a buffer dongle on the computer, locally powered from the USB simply to drive a remote opto board ...

Reply to
Lester Caine

Oh, I haven't given up on this design cycle yet. I've just had a frustrating afternoon ;-)

Thanks for the heads-up about generic drivers.

Cheers.

Reply to
Dave Osborne

PolyTech Forum website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.