Mirac - Help needed to HOME lathe
Moderators: Martin, Steve, Mr Magoo
Mirac - Help needed to HOME lathe
I'm running Mirac software V2.25, on Mirac Lathe, but it won't HOME, I've tryed all buttons marked X and Z on keypad, and X and Z on PC keyboard, but it doesn't move.
Have I got the right Mirac software ?
The machine will home using J; and K; in an RS232 TTY program
Also, is there any free CAM software I can use to generate parts program, without dongles etc.
Have I got the right Mirac software ?
The machine will home using J; and K; in an RS232 TTY program
Also, is there any free CAM software I can use to generate parts program, without dongles etc.
MIRAC s/n-MR19041A, May 1991, s/n with electric 40344
Perhaps the Desk Top Tutor keypad is faulty.
Try changing the software to use the QWERTY keyboard instead, by editing your MIRAC.GO file. Edit it using notepad, changing the lines that reads...
DENFORD1.KBD 1
18.2 $2F8 3
to read...
QDENFORD.KBD 1
18.2 0
(make sure you have the QDENFORd.KBD file in your mirac directory. If not it is inculded in the DRIVERS download from the denford website)
Try changing the software to use the QWERTY keyboard instead, by editing your MIRAC.GO file. Edit it using notepad, changing the lines that reads...
DENFORD1.KBD 1
18.2 $2F8 3
to read...
QDENFORD.KBD 1
18.2 0
(make sure you have the QDENFORd.KBD file in your mirac directory. If not it is inculded in the DRIVERS download from the denford website)
Mirac lathe Help
Thanks Mr Magoo,
There is no Desktop tutor keypad, the keyboard is fixed to the machine on the left of the headstock, under a built-in monitor. This may be the difference between a Mirac and a Mirac PC ?. There are other keys that appear to work as I can page up & down, scroll up & down, turn coolant on/off etc. I will try and use a comms sniffer program to make sure all the keys are working
We have also tryed the Qwerty keyboard, to no avail, but I will check the Mirac.go file again
I'm still wondering if I've got the right software version (Mirac.exe 2.25)
Bradders was talking about version 1.42 in some of his posts
There is one other thing I haven't tryed, and that is to link pins 1,4 & 6 on the 9 pin D connector for the PC, this may be some kind of loop back connection, so I'll give that a try too
Thank again
There is no Desktop tutor keypad, the keyboard is fixed to the machine on the left of the headstock, under a built-in monitor. This may be the difference between a Mirac and a Mirac PC ?. There are other keys that appear to work as I can page up & down, scroll up & down, turn coolant on/off etc. I will try and use a comms sniffer program to make sure all the keys are working
We have also tryed the Qwerty keyboard, to no avail, but I will check the Mirac.go file again
I'm still wondering if I've got the right software version (Mirac.exe 2.25)
Bradders was talking about version 1.42 in some of his posts
There is one other thing I haven't tryed, and that is to link pins 1,4 & 6 on the 9 pin D connector for the PC, this may be some kind of loop back connection, so I'll give that a try too
Thank again
MIRAC s/n-MR19041A, May 1991, s/n with electric 40344
Mr Magoo was right on the money!
We checked the keyboard was fuctioning, and all the keys worked, using a TTY program we got feedback from each key
Next we checked the interlocks, there was no oil in the oiler, so we filled it up, but it still wouldn't home
The Mirac PC drawing showed the Guard switch (Input 13), the Door fully open (Input 14) and the Air pressure switch (Input 15). Our Mirac has no automatic door opening mechanism or any compressed air connection, however the Oil low switch gave me a clue as to how the interlocks work. If the Oil is low you get a signal on Input 9 (Not present on Mirac PC), if this logic is applied to the Door fully open and the air pressure switch, then if they are not connected, they will not stop the machine from working. The guard switch is present on both machines, so this should be on when the Guard door is closed. The trouble was the door was closed and the input bit was on, but it still wouldn't home !
In desperation, I made a note of all the connections and wire numbers so I could make a wiring drawing, hopefully this might make fault finding easier
As I sat at home till the early hours, like a man with a mission, doing a rough wiring drawing, it suddenly occurred to me that the Input terminal blocks PI 3 and PI 4 may have got crossed over by previous owners. (Earlier this year we had to get a small tranformer replaced on the LCB3 slave board, as it kept blowing fuses, but we were very careful to mark all the connections before the board was removed !)
As I checked the Mirac PC drawing the penny dropped, I remember saying to my friend that it is normal practice to fill up inputs and outputs consecutively, but the guard switch was connected on the PI 4 connector, which at the time I thought was a bit odd, because it meant PI 3 would be spare (5 inputs), but if PI 4 should be in PI 3 instead, the Guard switch input would be on when the door is closed and the software may then work !
I had an opportunity to test this theory today, I swapped the connection blocks over, pressed the Home key, then X and hey presto, it homed the X axis (Yippee !!!), next pressed the Z key and the Z axis was homed, the readouts went to zero, and we were ready for take-off !!!
So it's all working, hardware, leads, drives, and software, now all we have to do is learn to use the software.
We put in the test CNC program (in the generic turning manual), and used the simulate function, it worked, but, (there's always a but !) when we went into Machine control, pressed F9 and selected Execute CNC, it read a few lines, altered the toolpost, started the spindle then came up with "Overrun" error, so we need to study the manual to setup Tool offsets and zero points etc. before we can start machining
A big thanks to everyone who have contributed to the resurrection of this little CNC lathe
In the last 3 months I have learned a lot about this machine, so if I can help anyone else, please e-mail me
I now have a set of drawings that I will convert to bmp for any one else to use if required, they are pretty basic but may help with fault finding
We checked the keyboard was fuctioning, and all the keys worked, using a TTY program we got feedback from each key
Next we checked the interlocks, there was no oil in the oiler, so we filled it up, but it still wouldn't home
The Mirac PC drawing showed the Guard switch (Input 13), the Door fully open (Input 14) and the Air pressure switch (Input 15). Our Mirac has no automatic door opening mechanism or any compressed air connection, however the Oil low switch gave me a clue as to how the interlocks work. If the Oil is low you get a signal on Input 9 (Not present on Mirac PC), if this logic is applied to the Door fully open and the air pressure switch, then if they are not connected, they will not stop the machine from working. The guard switch is present on both machines, so this should be on when the Guard door is closed. The trouble was the door was closed and the input bit was on, but it still wouldn't home !
In desperation, I made a note of all the connections and wire numbers so I could make a wiring drawing, hopefully this might make fault finding easier
As I sat at home till the early hours, like a man with a mission, doing a rough wiring drawing, it suddenly occurred to me that the Input terminal blocks PI 3 and PI 4 may have got crossed over by previous owners. (Earlier this year we had to get a small tranformer replaced on the LCB3 slave board, as it kept blowing fuses, but we were very careful to mark all the connections before the board was removed !)
As I checked the Mirac PC drawing the penny dropped, I remember saying to my friend that it is normal practice to fill up inputs and outputs consecutively, but the guard switch was connected on the PI 4 connector, which at the time I thought was a bit odd, because it meant PI 3 would be spare (5 inputs), but if PI 4 should be in PI 3 instead, the Guard switch input would be on when the door is closed and the software may then work !
I had an opportunity to test this theory today, I swapped the connection blocks over, pressed the Home key, then X and hey presto, it homed the X axis (Yippee !!!), next pressed the Z key and the Z axis was homed, the readouts went to zero, and we were ready for take-off !!!
So it's all working, hardware, leads, drives, and software, now all we have to do is learn to use the software.
We put in the test CNC program (in the generic turning manual), and used the simulate function, it worked, but, (there's always a but !) when we went into Machine control, pressed F9 and selected Execute CNC, it read a few lines, altered the toolpost, started the spindle then came up with "Overrun" error, so we need to study the manual to setup Tool offsets and zero points etc. before we can start machining
A big thanks to everyone who have contributed to the resurrection of this little CNC lathe
In the last 3 months I have learned a lot about this machine, so if I can help anyone else, please e-mail me
I now have a set of drawings that I will convert to bmp for any one else to use if required, they are pretty basic but may help with fault finding
MIRAC s/n-MR19041A, May 1991, s/n with electric 40344
- Denford Admin
- Site Admin
- Posts: 3649
- Joined: Fri 10 Feb , 2006 12:40 pm
- Hardware/Software: Go to User Control Panel > Profile
Enter as much information about your CNC hardware and software as you can - it makes it easier for everyone to know what you're talking about then. - Location: Sunny Brighouse
- Contact:
Not sure if you have already seen this manual for download: ?
viewtopic.php?t=953
Thought it may help with the offsets.
Stick with it - once you've got to grips with one CNC, then they are all pretty much the same after that
viewtopic.php?t=953
Thought it may help with the offsets.
Stick with it - once you've got to grips with one CNC, then they are all pretty much the same after that

- Denford Admin
- Site Admin
- Posts: 3649
- Joined: Fri 10 Feb , 2006 12:40 pm
- Hardware/Software: Go to User Control Panel > Profile
Enter as much information about your CNC hardware and software as you can - it makes it easier for everyone to know what you're talking about then. - Location: Sunny Brighouse
- Contact:
Any information you have will always be appreciated - if you do upload it, then please ZIP it all together and post it into the downloads section:
viewforum.php?f=15
viewforum.php?f=15
OK Guys, need some more help, here's where we are,
1. Machine now homes
2. Have set tool offsets for tool 0101, in X and Z axis
3. Have written short CNC program, (9 lines) cut down from Generic Tuning manual:
N1G21 Set to metric
[BILLET X2Ø Z4Ø Set sim billet size
N2G98 Set feed per minute
N3G28UØWØ Ref. point return
N4M6TØ1Ø1 Change Tool to 1
N5M3S191Ø Set spindle speed
N6GØX2ØZ2 Position Rapid traverse X,Z (Not touching billet !!)
N7M5 Stop Spindle
N8G28UØWØ Ref. point return
N9M3Ø Prog. reset ang rewind
4. Saved the CNC program, and tool offsets
5. Run program check, "Checked program is OK"
6. Run Simulation, results as expected
7. Execute CNC
8. Get "OVERTRAVEL" error on first move command, N6
Is there anything else we need to SET before we can start machining?,
or have we missed something in the tool offsetting ?
Please help !!!!
1. Machine now homes

2. Have set tool offsets for tool 0101, in X and Z axis
3. Have written short CNC program, (9 lines) cut down from Generic Tuning manual:
N1G21 Set to metric
[BILLET X2Ø Z4Ø Set sim billet size
N2G98 Set feed per minute
N3G28UØWØ Ref. point return
N4M6TØ1Ø1 Change Tool to 1
N5M3S191Ø Set spindle speed
N6GØX2ØZ2 Position Rapid traverse X,Z (Not touching billet !!)
N7M5 Stop Spindle
N8G28UØWØ Ref. point return
N9M3Ø Prog. reset ang rewind
4. Saved the CNC program, and tool offsets
5. Run program check, "Checked program is OK"

6. Run Simulation, results as expected

7. Execute CNC
8. Get "OVERTRAVEL" error on first move command, N6

Is there anything else we need to SET before we can start machining?,
or have we missed something in the tool offsetting ?
Please help !!!!
MIRAC s/n-MR19041A, May 1991, s/n with electric 40344
- Lone_Ranger
- CNC Expert
- Posts: 221
- Joined: Mon 01 Oct , 2007 15:23 pm
- Contact:
.
I noticed in your code there is no G90 or G91 for Absolute or Incremental movement, possible the machine control is defaulted to Absolute (Most are!!) so you may not need it in but check in you manual.
If it is on default to G90 (Absolute) then your code will take the tool post/turret to 20mm from Machine Zero on the X and 2mm from Machine Zero in the Z and would probably crash into the chuck and you have it at full rapid on a G0 so I would be double checking EVERYTHING if I were you before pressing any buttons and/or putting a G91 in the code so that is the distance it will travel, and/or a G01 and a feed rate for safety!! Oh and if you have an override for the traverse then turn it down to minimum!!
With a billet set at X20 and Z40 then to be rapiding to X20 and Z2 the tool would be 38mm past the face of the billet!!
Could possibly be how you have your "Work shift" and "Machine Zero" set.
Normally all movement towards the chuck/workpiece are - values eg -X20-Z20 therefore to use positive values in your code requires Machine Zero to be set at the chuck end of the machine.
If the machine zero and work shifts are not set right then the tool post/turret will try to move in a postitive direction which is away from the chuck so you will go "overtravel" as soon as the control reads the line of code for the movement.
For example I have my Zero set to the face of my chuck as it is a solid constant I can actually measure from, so when my lathes are "homed" the numbers shown on the screen are actual distance from my Machine Zero, my Denford Cyclones show as X107.56 and Z201.58.
Once "homed" the Absolute values ie X and Z should show the same as the Relative values U and W.
All my programs use Absolute movements so the machine is told to go to a specific XZ coordinate in positive numbers ie no + or - signs in the code.
Here are a few lines of code from one of my programs for a Denford Cyclone with Fanuc OTC control which I don`t think will be a million miles from your machine.
Also usually a good idea to number your code in 5s or 10s as I have so that if you need to alter the code at the machine it`s easy to do, otherwise it means doing it away from the machine and reloading!!
%
o0030
N15G21G50X107.56Z201.58
(FAN ANGLE SIDE2 25-3-2008)
N25T0303M06
N35G97S1417M3
N40G0X82.01Z52.2M8
N45G50S4000
N50G96S365
N55G1Z48.0F0.2
N60X83.0
N65X84.41Z48.71
N70G0Z52.2
N75G1X81.02F0.2
Hope that`s of some help
Regards
Rob
PS Don`t be confused by the 2 G50s in the code, the first one is the same as your G28 and the second one with a speed after it is a max spindle speed for the CSS comand on the next line.
.
I noticed in your code there is no G90 or G91 for Absolute or Incremental movement, possible the machine control is defaulted to Absolute (Most are!!) so you may not need it in but check in you manual.
If it is on default to G90 (Absolute) then your code will take the tool post/turret to 20mm from Machine Zero on the X and 2mm from Machine Zero in the Z and would probably crash into the chuck and you have it at full rapid on a G0 so I would be double checking EVERYTHING if I were you before pressing any buttons and/or putting a G91 in the code so that is the distance it will travel, and/or a G01 and a feed rate for safety!! Oh and if you have an override for the traverse then turn it down to minimum!!
With a billet set at X20 and Z40 then to be rapiding to X20 and Z2 the tool would be 38mm past the face of the billet!!
Could possibly be how you have your "Work shift" and "Machine Zero" set.
Normally all movement towards the chuck/workpiece are - values eg -X20-Z20 therefore to use positive values in your code requires Machine Zero to be set at the chuck end of the machine.
If the machine zero and work shifts are not set right then the tool post/turret will try to move in a postitive direction which is away from the chuck so you will go "overtravel" as soon as the control reads the line of code for the movement.
For example I have my Zero set to the face of my chuck as it is a solid constant I can actually measure from, so when my lathes are "homed" the numbers shown on the screen are actual distance from my Machine Zero, my Denford Cyclones show as X107.56 and Z201.58.
Once "homed" the Absolute values ie X and Z should show the same as the Relative values U and W.
All my programs use Absolute movements so the machine is told to go to a specific XZ coordinate in positive numbers ie no + or - signs in the code.
Here are a few lines of code from one of my programs for a Denford Cyclone with Fanuc OTC control which I don`t think will be a million miles from your machine.
Also usually a good idea to number your code in 5s or 10s as I have so that if you need to alter the code at the machine it`s easy to do, otherwise it means doing it away from the machine and reloading!!
%
o0030
N15G21G50X107.56Z201.58
(FAN ANGLE SIDE2 25-3-2008)
N25T0303M06
N35G97S1417M3
N40G0X82.01Z52.2M8
N45G50S4000
N50G96S365
N55G1Z48.0F0.2
N60X83.0
N65X84.41Z48.71
N70G0Z52.2
N75G1X81.02F0.2
Hope that`s of some help
Regards
Rob
PS Don`t be confused by the 2 G50s in the code, the first one is the same as your G28 and the second one with a speed after it is a max spindle speed for the CSS comand on the next line.
.
Thanks for the feedback Bob,
We haven't put a billet in the chuck yet, or a tool in the toolpost, we're just playing around with the Codes to see what happens, we think we've allowed plenty of space between points so we shouldn't get any smashups!!
Some good points though, I remember programming in BASIC on a Sinclair ZX81 many years ago and quickly learned the number stepping trick, incase I had to add additional code, but I didn't know if in a CNC programe the instructions had to be sequential as demonstrated in the manual
My friend had another go at setting the offsets last night, with some success, as he got the program to run without displaying the"overtravel" error, but he did say the measurements were a little out, maybe that is because of what you have described above.
We copied the code from the generic turning manual, which describes a part program for a Novaturn, so there may be differences there. I tend to be a little hasty in trying things to make some progress and because of that I'm possibly skipping over important information in the manual
Time to sit down and read, I guess
Thank again
Derek
We haven't put a billet in the chuck yet, or a tool in the toolpost, we're just playing around with the Codes to see what happens, we think we've allowed plenty of space between points so we shouldn't get any smashups!!
Some good points though, I remember programming in BASIC on a Sinclair ZX81 many years ago and quickly learned the number stepping trick, incase I had to add additional code, but I didn't know if in a CNC programe the instructions had to be sequential as demonstrated in the manual
My friend had another go at setting the offsets last night, with some success, as he got the program to run without displaying the"overtravel" error, but he did say the measurements were a little out, maybe that is because of what you have described above.
We copied the code from the generic turning manual, which describes a part program for a Novaturn, so there may be differences there. I tend to be a little hasty in trying things to make some progress and because of that I'm possibly skipping over important information in the manual
Time to sit down and read, I guess
Thank again
Derek