guess the wires

[Agroshield]

I think something like this would be a close relative of the part spanning the slotted disk:

OPB365T55 Optek, Screw Mount Slotted Optical Switch, Phototransistor Output | RS

uk.rs-online.com

See if it has a part number or manufacturer on it and you will find a datasheet telling you what its output should look like.

The wiring is bottom right of diagram on p.44 of manual above.

It should be simple to verify that one: give it power and when the light beam is through a hole, it will register an output. When the beam is blocked, it won't.

To make that work exactly as is, you'd need a tacho that has a configurable number of pulses per revolution. A lot of the cheap ones only work on single ppr and that makes the display a bit vague at slow speeds because it is a long time (in computer terms) between seeing pulses

At very worst, you might have to make up another holey disk with fewer holes in it.

 

[m_c]

It's not a complicated PCB.
I can't quite make out the numbers, but the on the front photo you posted, I'd say the top right is the voltage regulator.
It's a Siemens L something. It looks likes L9245H, but that doesn't bring up any data sheets, and I suspect the 4 isn't a 4. If you can find the datasheet for that, you should be able to trace what is the input power, and what power it's outputting.
The opto will likely need DC voltage, so if you can figure what pins are outputting DC voltage, combined with where the 24VAC is coming in, that should only leave figuring out what

I'd guess the big resistor on the back is a crude voltage dropper pre-regulator IC, and the diode on the right hand side is being used as a half wave rectifier to get DC.

It's a Microchip PIC doing the processing on the top left, and I'd guess the lower left chip is most likely a register shifting IC for driving the LCD.


However, having a look at the wiring diagrams.

Starting at the left (the labelling isn't very clear, so I'll simply start at the left)
1 & 2 - Power in. Going by the notes, this should be 24VAC. Going by the fact the second wire is labelled F1, I'm going to assume that means Fuse 1, so i'd guess F1 should be the 24VAC, and 2 0V.
3 - Red to sensor (assuming conventional wiring, I'd assume this is going to be 5VDC or 24 VDC)
4 - Green from sensor
5 - White from sensor
6 - Black to sensor (0V for sensor power)
7 - CN15, wire 60, which appears to go here, but I can't fully make out the writing -

And also having checked the 24VAC supply on that same diagram, my assumption above appears to be correct-

 

[Brad93]

Yes F1 is a 0.75A fuse back on the control panel in the electrical cabinet.

I don't know if these files are any clearer.

 

[Brad93]

Opticoupler

 

[Brad93]

Other chip

 

[m_c]

The top right is an opto-isolator, which I'm guessing is providing some input protection from the opto-sensor, in case something goes wrong.
And the top left is a what you've just posted (I mistook the Motorola logo for Microchip), and uses the AC input as a time clock to generate a comparison with the opto sensor.

The lower left chip might be a micro-controller, and may explain why it's been defaced.

I'd start with the basics, like making sure you're getting 24VAC.
However I'm wondering about that other wire, that we can't make out the writing for. Should the tacho be permanently on with the lathe on, or does that extra wire switch the tacho on?

 

[Brad93]

Yes i have 24V AC coming up the red wire.

 

[Brad93]

My lathe is not using the CSS despite the wiring being there.
You needed the very expensive heidenhain dro to use the constant spindle speed function.

 

[Brad93]

I've got the main wiring diagram from my lathe but not the one showing the tacho, however it may be of some assistance.

 

[Brad93]

That's the only bit on my wiring diagram that relates to the tacho

 

[sako243]

So what I could do with is a tacho capable of working from this existing sensor and ring, otherwise i've got a project and i don't need anymore projects.

Can you get a part number off the photo sensor? Or some better photos of it.

I'm away currently for a work trip tomorrow and then some deer stalking but once I'm back / in the lab next week I can knock you up a circuit board with display to do what you want.

To make this easy at my end and yours all I'd need is:
1. Ideally a part number or pin out for the photodiode (we could work it out if need be).
2. Confirmation you have a 24VAC / 750mA supply to the display box
3. Number of holes in the disc
4. Ratio of spindle rpm to disc rpm

If you want it made to suit the same holder / box some dimensions of the display would be ideal.

 

[Brad93]

Can you get a part number off the photo sensor? Or some better photos of it.

I'm away currently for a work trip tomorrow and then some deer stalking but once I'm back / in the lab next week I can knock you up a circuit board with display to do what you want.

To make this easy at my end and yours all I'd need is:
1. Ideally a part number or pin out for the photodiode (we could work it out if need be).
2. Confirmation you have a 24VAC / 750mA supply to the display box
3. Number of holes in the disc
4. Ratio of spindle rpm to disc rpm

If you want it made to suit the same holder / box some dimensions of the display would be ideal.

Blimey Sako that's a blinding offer.

I couldn't see any markings on the sensor, however i've not yet been in touch with 600 group to see if they have any other schematics. They may well come back with a drawing or part number.
I will remove the sensor and have a good look at it in the meantime.

I'm not using the lathe currently, so what i can do is take the sensor and pcb/display off completely, and send that up to you when you have a spare moment?

Yes have 24v up at the board.

Disc is mounted directly to the end of the spindle so ratio would be 1:1?

Will count holes in the disc when i'm next at the workshop.

 

[sako243]

I'm not using the lathe currently, so what i can do is take the sensor and pcb/display off completely, and send that up to you when you have a spare moment?

Yeah if you want to post down the sensor and pcb down that'd be easiest. Do you need / want the CCS function because if its an output from the dro it's something that I might be able to synthesise.

 

[Agroshield]

Do you need / want the CSS function...

CSS is constant surface speed. It could be non-trivial to implement.

It would take an input from the DRO (so you know how far from centre you are) and would produce an output that goes the the VFD, so the machine speeds up as the tool approaches centre. It would also need some logic so the machine does not overspeed (as in theory at the centre your surface speed is zero, so your rotational speed needs to be infinity) and also as another input (maybe a pot/dial) some means of telling it what surface speed you actually want.

I think the lathe has low/high gearing between the motor and spindle, with a sensor (microswitch) so its brains know which gear it is in. That has to go into the CSS mix as well.

The DRO input would need to be in absolute coordinates rather than touching off on a piece of 50mm diameter stock and setting local zero or the system would go wrong. So you would need to know how whatever DRO is on the machine provides that service.

It is probably worth reading the manual and studying the circuit diagram to see how Colchester did it so at least you know what you have to reproduce if you did take on the task.

 

[Brad93]

Yeah if you want to post down the sensor and pcb down that'd be easiest. Do you need / want the CCS function because if its an output from the dro it's something that I might be able to synthesise.

No my lathe doesn't have the Heidenhain DRO for the CSS function.

 

[sako243]

What I meant / assume it was (given your display appears to have a pin for it) is that it's a feedback signal generated from the rev counter (the display) to feed back to the spindle drive to maintain a constant speed. Normally you'd have to fork out for a Heidenhain DRO to get this feedback signal but if we knew what it was then I could add it to my inexpensive display (with enough info).

 

[Brad93]

CSS is constant surface speed. It could be non-trivial to implement.

It would take an input from the DRO (so you know how far from centre you are) and would produce an output that goes the the VFD, so the machine speeds up as the tool approaches centre. It would also need some logic so the machine does not overspeed (as in theory at the centre your surface speed is zero, so your rotational speed needs to be infinity) and also as another input (maybe a pot/dial) some means of telling it what surface speed you actually want.

I think the lathe has low/high gearing between the motor and spindle, with a sensor (microswitch) so its brains know which gear it is in. That has to go into the CSS mix as well.

The DRO input would need to be in absolute coordinates rather than touching off on a piece of 50mm diameter stock and setting local zero or the system would go wrong. So you would need to know how whatever DRO is on the machine provides that service.

It is probably worth reading the manual and studying the circuit diagram to see how Colchester did it so at least you know what you have to reproduce if you did take on the task.

You're absolutely right.

The heidenhain system uses linear scales that have a system whereby you set the max rpm suitable for your job via the pot, and then tell it tool position etc and it does the rest.

 

[Agroshield]

...it's a feedback signal generated from the rev counter (the display) to feed back to the spindle drive to maintain a constant speed.

Constant speed (closed loop between VFD and rev counter) is not the same thing as CSS, which is constant SURFACE speed.

Constant surface speed implies variable rotation speed as the tool crosses the part - on something 12" diameter, it needs to turn at a much slower rpm to ensure 40m of material passes the tool in one minute (42 rpm). At 1" diameter, it needs to turn at 12 times that speed (500 rpm) so the same 40m of material passes in the same time.

In addition to preventing a divide-by-zero error at the center, a CSS system would need to have the facility for the user to set an absolute maximum rpm, even if that results in the surface speed dropping towards the middle. At some point on a large diameter workpiece, if the computer kept ramping the speed up, it would exceed what is safe for the situation. It is possible to see and hear CSS in action in a lot of Chris Maj's videos.

You could do what you are suggesting (closed loop, constant rpm) and it would be very good in itself because if the speed tries to drop when taking a good heavy cut, it will turn up the VFD to compensate. But it will be misleading to think of it as CSS.

 

[Brad93]

Here are the directions

 

[sako243]

Constant speed (closed loop between VFD and rev counter) is not the same thing as CSS, which is constant SURFACE speed.

Sorry I missed your post before Brad's detailing what it was. I thought it might constant spindle speed rather than surface speed so I get what it is now.

Still if the input signals are there and the output signal has somewhere to go designing the electronics won't be an issue. I'm looking into building a cnc controller for a Kingsland 150t press brake having already gone one for a Kingsland Hydraulic guillotine. The same boards could be repurposed for a lathe without much hassle (new software rather than having to redesign the hardware).