Hello Guest it is March 28, 2024, 03:36:55 PM

Author Topic: Boxford 160 TCL - running as is?  (Read 14157 times)

0 Members and 1 Guest are viewing this topic.

melee

*
Boxford 160 TCL - running as is?
« on: August 21, 2008, 02:57:30 AM »
Hi
I have recently bought one of these, which of course is missing the PC, manuals and all software.

It is approx 2000 vintage and has almost had no use.

Given that it works perfectly in manual mode and is a fairly new machine with 8 position tool turret and Mitsubishi inverter, I can see no compelling argument to throw away half the machine and replace the steppers, motors etc.

Does anyone have experience of getting one running as is??

Does anyone have manuals for the 160?

Does anyone have a wiring diagram for the 160?

Previous posts suggest the paperwork is out there, but I can see nothing about conversion using original equipment

All help gratefully received

Melee

Offline SimonD

*
  •  143 143
    • View Profile
Re: Boxford 160 TCL - running as is?
« Reply #1 on: August 22, 2008, 03:05:50 PM »
I have just converted my Boxford 160. There is a long post about it on here somewhere.
I replaced the electronics but kept the spindle drives and steppers. All I threw away was the processor board and the stepper board. These were placed with a breakout board and Geckos.

I cannot help with getting it to work using the original software or even connecting it up directly as I do not have the information, however, I have the Boxford CD with all the manuals in PDF format. So I have both the programming and User manual if you want to contact me off list. (simon at byte-sized dot com)

Simon

melee

*
Re: Boxford 160 TCL - running as is?
« Reply #2 on: August 23, 2008, 06:53:58 AM »
Hi Simon

I have read your various entries as you converted your 160.  That partly gave me the interest in getting and converting one myself.

The 160 is about the smallest CNC lathe which is still capable of producing something useful.

Mine is quite a modern one, 2000 - 2001, the stepper board is clearly marked and there is a connector block between the stepper and processor boards which shows every pin out you should need to control it via an opto board.
I will just need to identify where the limit switches, eStop etc connect in

I am looking towards patching into the relevant wires, perhaps using diodes in some cases to prevent unwanted reverse current feeds to the processor board, so that I retain a manual capability from the front panel of the machine, as well as any software jog controls, plus full and MDI auto operation from software.

I will contact re the manuals, thank you for your response

Melee

melee

*
Update
« Reply #3 on: August 28, 2008, 03:27:50 AM »
Hi

I have had partial sucess, but not in a way that will please Mach 3 fans

Had a complete blank under Mach 3

I cut the 12 pin plug on the stepper board, which includes the X Y and Z step and dir connections, in half so that the other feeds remained between stepper and processor board.
I then replaced the top half of the plug with another linked into relevant connections on my opto board (DIYCNC optoport).

All I got was a slight clunk when changing direction but no stepping.  Swap the plug back over and the lathe moves fine in manual mode so nothing wrong with steppers etc.

I increased the step pulse time to max, swapped drives from active low to active high, re-checked wiring until blue in the face - nothing, just a slight clunk when changing direction from + to -

Then I did what I should have done in the first place, I went to Linux and started EMC2

The configuration is pretty detailed, but even with default settings I had something trying to happen on both axis.

I took off the stepper motors, counted teeth, measured how far the carriage moved in one revolution of the screw etc to get the motor tuning figures I needed.

Motors are 400 step, step drivers are GS-D200S which data sheet says can drive are step or half step, so max micro steps are 2
X axis motor has 16 tooth sprocket - screw has 20 tooth sprocket

Z axis motor has 12 tooth sprocket - screw has 20 tooth sprocket

Both screws appear to have a 2mm pitch, so 200 steps per mm

The max speed of travel from data sheet is 1200mm/min which = 20mm/sec
Acceleration is mm/sec squared so calculated at 400 mm/sec

Having entered these figures into the configuration screen of EMC2, I went to the AXIS_turn screen, brought the lathe on line and had control of both axis straight away!

I only started with Mach3 because I thought it might be easier until I found my way around things!
As I program in Linux and use it for everything else, I should have stuck with it.

I tried entering the same settings in Mach 3 (albeit Mach3 wants them in a slightly different format) and still nothing whatsoever happened!

By assigning the tool changer turret to the Y axis, I was able to jog it under EMC2.
I will have to figure out how to link it to 'M6Tn' G codes and get the stepper to move in 50 step increment (8 tools / 400 steps per rev).
That is of course assuming there is no reduction gear in the tool changer, more stripping down possibly required.

How have other people got their turrets working properly?

cheers

Melee

Offline da21

*
  •  87 87
    • View Profile
    • M-Machine
Re: Boxford 160 TCL - running as is?
« Reply #4 on: August 28, 2008, 03:35:34 AM »
the one thing to check is that you have the port set to 1 for LPT1 , and did you reboot after installing Mach3
the port driver needs to be installed , and the pc needs a reboot for this to succeed correctly

are you using any Enable pins to your stepper drivers
if you attach your xml config file , no doubt someone will check it , just in case 

 

melee

*
Re: Boxford 160 TCL - running as is?
« Reply #5 on: August 28, 2008, 06:06:41 AM »
"the one thing to check is that you have the port set to 1 for LPT1 , and did you reboot after installing Mach3
the port driver needs to be installed , and the pc needs a reboot for this to succeed correctly "


Yes I used 0x378 default par port and installed Mach 3 weeks ago, many reboots and sucessfull system and port checks since.

"are you using any Enable pins to your stepper drivers "
 


I am patching into the step and dir pins only, the analogue signal and enabling feed is still coming from the processor board.
It works perfectly with EMC2, someone is sending me screen shots of their setup config with the same machine under Mach3 so I can compare, albeit theirs is an older machine and some of the hardware is different.

melee

*
Re: Boxford 160 TCL - running as is?
« Reply #6 on: April 29, 2009, 04:56:08 AM »
Hi

I completely converted my 160 a long time back.
The problem detailed previously was caused by a parallel port lead which was wrongly wired (probably came from a scanner or may even have been a 25 pin serial lead)

Since than I have done exactly what most other people do, got on and used the lathe and never passed on the information that I myself was searching for initially!

So the below is to redress the balance and hopefully leave a starting point for others embarking on the same journey.

Whilst initially leaning towards maintaining manual control from the front panel, I decided that it would involve a lot more wiring complexity and everything I might want to do manually could be easily achieved by using the keyboard to jog axis or using the MDI line.

My experience using the lathe afterwards confirmed this, the only time I would have used manual would be to say face up a billet prior to setting up for a job run, MDI line or jogging is fine for this.

So here go the basics. (It covers conversion to run under Mach3, it will run under EMC2 but the setting described are Mach specific)

Assuming that you have full manual control from the front panel, the first 2 things you need to order are a Opto isolated breakout board and an opto isolated Spindle board.

I got mine from www.diycnc.co.uk. They work, are made in UK and are cheaper than US boards, especially with exchange rate dropping
They take about a week to arrive, the guy Roy makes them up to order.
What you need is
http://www.diycnc.co.uk/html/opto_board.html
and
http://www.diycnc.co.uk/html/spindle_2.html

I have attached a picture of the finished conversion showing these boards mounted under the cabinet fan

Once you have the boards it is easy to get the axis running quickly.
I just cut the 12 pin plug, which incorporates X Z and A (ATC) in the top 6 pins, in half and replaced the top half with a 6 pin plug from an old computer PSU and ran the cables straight to the stepper outputs on the opto port board.

I set up the ATC as the B axis under Mach and set the shortcut keys to jog it on Page Up and Page Down on the keyboard.
I have written a macro for M6 which automatically indexes the ATC to the correct tool position.
If all your screw pitches etc are the same as mine it should work straight off, if not you may have to tweak the settings.

The limit switches are all connected in series and attached to one input on the board.
It doesn't matter which one tripped, that will be obvious from where the carriage and slide are.

The interlocks ( cover and pulley cowl )are similarly connected in serial to the Estop circuit in Mach.
I did not connect the Estop button on the front panel, because it gave weird side effects with the 24v system being permanently latched.
Since hitting it will knock out the 24v which in turn will stop the charge pump on the opto board and prevent any movement, that was good enough

The index disc will need the slot opening out to 6mm.
If you have a milling machine, just a question of centering it on the hole above the slot and milling out a grove of 6mm through the existing slot and that hole.
If you want to do a lot of screwcutting you may need to cut 3 further slots and use the timing pulse input, but mine runs fine at present with just one slot and the pulse from the lower optical switch fed to Mach.
I have heavily shielded the pulse wire with screened cable which may well have helped greatly.

I found that the timing pulses only worked properly if I spliced into the cable to the serial board as opposed to cutting and replacing it.
Don't know why and don't much care so long as it works.
I have disconnected a little as possible from the original wiring.

If you download the wiring diagram you will see the connector block for the timing pulse.
I have attached a pic of the connection I made. The wires you need to splice into are GND and P.

The wiring of the speed board to the inverter is simple, look on the net for a manual for your particular inverter, there are just 2 wires to connect to govern spindle speed and 2 for spindle rotation plus a common GND
There is already a screened lead connecting the inverter to the Boxford boards.
I removed this and connected direct to the Spindle 2 board.
Red is V out, Yellow is 0V, Black is common on the relay with Blue being NC and Green being NO on the relay for spindle rotation.
The White lead is not used and is taped up.
A picture is attached.

That is pretty much it

Hope this is useful

regards

Melee

Offline SimonD

*
  •  143 143
    • View Profile
Re: Boxford 160 TCL - running as is?
« Reply #7 on: April 29, 2009, 05:07:25 PM »
Melee,

Do the front panel controls still work or did you bypass/disconnect them?

Thanks
Simon

melee

*
Re: Boxford 160 TCL - running as is?
« Reply #8 on: April 30, 2009, 03:43:08 AM »
Hi Simon,

I have effectively disconnected the steppers and the spindle control from the Boxford serial board.

Because I cut the 12 way connecter in half and replaced the axis half with my own plug, I could get jog control at the front panel just by swapping the plugs over. (you can see the second half of the plug loose in the first photo).
I have never had the need, jog and MDI line works fine for setting up.

I set out intending to keep the front panel.
I experimented with both connected via an array of diodes on each feed to the stepper board, so that the front panel should not interfere with Mach and vice versa.
This did not work for some reason, the front panel jogged , but Mach would not do so until the other feeds were removed.

An alternative would be a multi gang switch to manually switch from front panel to Mach.  I suppose you could do the same with the inverter connections.
As described, I decided that this would hugely increase the wiring complexity for little useful result and removed the connections from the Boxford boards.

The only exception was the index pulse.
For some reason the spindle speed reading was far more stable if I patched into the connector plug, than if I disconnected.

A chap called Graham whom I was corresponding with at the time, who was also converting a 160, had a lot of problems with erratic spindle speed.
He had cut the wires to everything and hard wired to a serial D plug.
It was not until he expanded the main slot by 50% and cut 3 further slots (ie set up as per the Mach manual for timing pulse that might be required for screwcutting) that he managed to get a stable speed reading.

I however managed a pretty stable reading with just one 6mm slot with my patched in connection.

Can't explain that bit, just know it works.

regards

Melee

Offline SimonD

*
  •  143 143
    • View Profile
Re: Boxford 160 TCL - running as is?
« Reply #9 on: April 30, 2009, 01:08:16 PM »
Melee,

Thanks, I was just wondering if there was an eaasy way...
On my conversion I spent ages connecting all the front panel switches to a PLC and connecting it to Mach3 via ModBus.
However I too don't use the front panel so my effort was wasted.

Thanks
Simon