Easiturn3 conversion to Mach - some observations
Moderators: Martin, Steve, Mr Magoo
-
- Posts: 26
- Joined: Wed 02 Mar , 2011 9:48 am
- Hardware/Software: Easiturn 3
Easiturn3 conversion to Mach - some observations
A couple of people have asked me about converting my Easiturn3 to Mach - which I am in the middle of doing - so I am appending some observations and comments here in the hope that they might be of use to anybody going down the same route.
Rick
-----------------------------------------------------------------------------------
+ The Combivert inverter that was fitted as standard cannot be used with modern spindle control boards - it has a resistive (0-10k ohm) input control mechanism, not an analogue (0-10v) one. If you do want to keep it, I believe that the new Soundlogic breakout board can drive resistive controls, but I've never tried it. I am going to replace my combivert - which is misbehaving - with a Teco inverter but have not done this yet (the new one wont arrive here till mid next week).
+ One problem is that the combivert board also has the main power breakers and the +24v output that drives all the control relays (coolant and lube pumps, etc etc). Not sure yet how best to deal with that issue - may be simplest to leave the combivert wired in just to provide power for this board. Alternatives I can think of are to use a separate 24v PSU, or to replace the board entirely, or wire the pumps directly into a 240v supply. If you do the latter, you'll need to provide +12v for the low voltage worklight. This lot is something I plan to tackle next week, when the new inverter arrives.
+ Getting the original electronics box / control unit off the top is a challenge - these lathes were built to last ( ). You'll need to remove the front and top panels, then the keyboard circuit boards, then remove the PNC circuit board on the right hand end - this has multiple plugs/cables/bolts that run thru the steel frame that prevent you going any further. Carefully disconnect (and label!) the three sets of armoured cables that come up through the hole in the baseplate - these connect to plugs at the back of the control unit power compartment. Once this is all done, you can unscrew the right half of the frame - mine had one screw at front centre (behind the keyboard), two on the top of the right hand panel (one each on the front and back pillars), and two underneath the baseplate, which take a bit of getting to. At this point, the right half of the box will slide off, exposing the turntable. On mine, this was like a pipe flange that had been pressure fit into the main casing of the lathe. If you have this setup, take a 4lb hammer and a cold chisel, put the chisel into one of the slots in the flange and wallop it sideways a few times until the flange starts to rotate in the housing. This breaks the seal, and then you can lever the flange vertically out with a crowbar. Then you can take the frame of the control unit off.
+ The wiring of my lathe did not match any of the diagrams that are in the manuals that the Denford folks have kindly placed into this forum. However, the schematic at the back of the PNC3 installation/maintenance manual was accurate as long as I ignored cable numbers and colours, which did not match the actuals. It therefore took me two days to trace all the wiring through, and then about two hours to wire in a breakout board, power supplies, switches, and new stepper drivers and get things working under Mach (it works fine now).
+ On my lathe, the armoured cables coming from the lathe body were grouped into three sets, each going into multiway plugs on the original control unit power compartment. The largest plug (C01) on the left had all the power connections (care: 240v here as well as low voltage) and cables for switching axis power, coolant pump, lights, etc on and off. These operate relays on the combivert board. The middle plug (C02) had the connections for the X-stepper motor at the top, the remainder of the connections being for limit and homing switches. These all connect to the saddle box, at the back of the cross slide. The switch connections all ran through a single 12-way shielded cable. The third plug (C03) on the right had the connections for the Z-stepper motor at the top, and for the spindle encoder at the bottom.
+ I have retained the existing stepper motors but if you need to replace them, NEMA34 motors will fit directly onto the mounting points. The steppers on my machine seem to be basically unused, so I wont replace them unless they have a problem. There is reduction gearing in both drives (small to large pulley on the belts). If your drivers are set up like mine (set with 1600 steps on the driver) Mach will be 800 'steps per' in the motor tuning.
+ I replaced the existing stepper drivers with new 4.2A ones - type PMR542. You can get these from Gary Higgins at Zapp Automation (and if you do, let him know that I told you to call!). If you get the PSU from him, choose the 48v 350w model (S350-48, I think it is). You can power this from the 240v feed that comes up one of the armoured cables from the machine, no need for another external mains plug and lead.
+ I've used a Uniport breakout board for my conversion, and you can get that from Zapp too (or directly from Roy at diycnc.co.uk). I've got a separate small PSU in my controller box to feed this 5v rather than using the USB power connection simply so that I dont have yet another external cable to get in the way.
+ I suggest that you forget trying to use the homing switches - they are seriously weird and I could not get them to work. Mach does not need them anyway - you can use the limit switches for homing, which is what I am doing. If you want/need to wire each limit switch separately, you'll need a second breakout board - use a PCPPS card (also from Zapp / diycnc) and configure Mach accordingly. If you go this route, you'll need an add-on parallel card in a PC that already has one parallel port. Configure the PC so that the addon card is LPT1, and the onboard port is LPT2 or you'll risk running into issues with how the PC/Windows assign pins to inputs and outputs. If however you wire all four limit switches in series, it only requires one input port on the breakout board so you dont need the extra parallel port, which is both easier and cheaper, and still works fine with Mach. In this case, you'll need to rewire the switches in series - I have done this inside the saddle box on the back of the cross-slide.
+ The lube points under my cross-slide were clogged with muck - probably worth checking yours and cleaning or replacing them if need be.
+ I have not tackled spindle speed control yet, but will do this when my new inverter arrives. I'll append my observations on this and on combivert replacement when I've done it.
Rick
-----------------------------------------------------------------------------------
+ The Combivert inverter that was fitted as standard cannot be used with modern spindle control boards - it has a resistive (0-10k ohm) input control mechanism, not an analogue (0-10v) one. If you do want to keep it, I believe that the new Soundlogic breakout board can drive resistive controls, but I've never tried it. I am going to replace my combivert - which is misbehaving - with a Teco inverter but have not done this yet (the new one wont arrive here till mid next week).
+ One problem is that the combivert board also has the main power breakers and the +24v output that drives all the control relays (coolant and lube pumps, etc etc). Not sure yet how best to deal with that issue - may be simplest to leave the combivert wired in just to provide power for this board. Alternatives I can think of are to use a separate 24v PSU, or to replace the board entirely, or wire the pumps directly into a 240v supply. If you do the latter, you'll need to provide +12v for the low voltage worklight. This lot is something I plan to tackle next week, when the new inverter arrives.
+ Getting the original electronics box / control unit off the top is a challenge - these lathes were built to last ( ). You'll need to remove the front and top panels, then the keyboard circuit boards, then remove the PNC circuit board on the right hand end - this has multiple plugs/cables/bolts that run thru the steel frame that prevent you going any further. Carefully disconnect (and label!) the three sets of armoured cables that come up through the hole in the baseplate - these connect to plugs at the back of the control unit power compartment. Once this is all done, you can unscrew the right half of the frame - mine had one screw at front centre (behind the keyboard), two on the top of the right hand panel (one each on the front and back pillars), and two underneath the baseplate, which take a bit of getting to. At this point, the right half of the box will slide off, exposing the turntable. On mine, this was like a pipe flange that had been pressure fit into the main casing of the lathe. If you have this setup, take a 4lb hammer and a cold chisel, put the chisel into one of the slots in the flange and wallop it sideways a few times until the flange starts to rotate in the housing. This breaks the seal, and then you can lever the flange vertically out with a crowbar. Then you can take the frame of the control unit off.
+ The wiring of my lathe did not match any of the diagrams that are in the manuals that the Denford folks have kindly placed into this forum. However, the schematic at the back of the PNC3 installation/maintenance manual was accurate as long as I ignored cable numbers and colours, which did not match the actuals. It therefore took me two days to trace all the wiring through, and then about two hours to wire in a breakout board, power supplies, switches, and new stepper drivers and get things working under Mach (it works fine now).
+ On my lathe, the armoured cables coming from the lathe body were grouped into three sets, each going into multiway plugs on the original control unit power compartment. The largest plug (C01) on the left had all the power connections (care: 240v here as well as low voltage) and cables for switching axis power, coolant pump, lights, etc on and off. These operate relays on the combivert board. The middle plug (C02) had the connections for the X-stepper motor at the top, the remainder of the connections being for limit and homing switches. These all connect to the saddle box, at the back of the cross slide. The switch connections all ran through a single 12-way shielded cable. The third plug (C03) on the right had the connections for the Z-stepper motor at the top, and for the spindle encoder at the bottom.
+ I have retained the existing stepper motors but if you need to replace them, NEMA34 motors will fit directly onto the mounting points. The steppers on my machine seem to be basically unused, so I wont replace them unless they have a problem. There is reduction gearing in both drives (small to large pulley on the belts). If your drivers are set up like mine (set with 1600 steps on the driver) Mach will be 800 'steps per' in the motor tuning.
+ I replaced the existing stepper drivers with new 4.2A ones - type PMR542. You can get these from Gary Higgins at Zapp Automation (and if you do, let him know that I told you to call!). If you get the PSU from him, choose the 48v 350w model (S350-48, I think it is). You can power this from the 240v feed that comes up one of the armoured cables from the machine, no need for another external mains plug and lead.
+ I've used a Uniport breakout board for my conversion, and you can get that from Zapp too (or directly from Roy at diycnc.co.uk). I've got a separate small PSU in my controller box to feed this 5v rather than using the USB power connection simply so that I dont have yet another external cable to get in the way.
+ I suggest that you forget trying to use the homing switches - they are seriously weird and I could not get them to work. Mach does not need them anyway - you can use the limit switches for homing, which is what I am doing. If you want/need to wire each limit switch separately, you'll need a second breakout board - use a PCPPS card (also from Zapp / diycnc) and configure Mach accordingly. If you go this route, you'll need an add-on parallel card in a PC that already has one parallel port. Configure the PC so that the addon card is LPT1, and the onboard port is LPT2 or you'll risk running into issues with how the PC/Windows assign pins to inputs and outputs. If however you wire all four limit switches in series, it only requires one input port on the breakout board so you dont need the extra parallel port, which is both easier and cheaper, and still works fine with Mach. In this case, you'll need to rewire the switches in series - I have done this inside the saddle box on the back of the cross-slide.
+ The lube points under my cross-slide were clogged with muck - probably worth checking yours and cleaning or replacing them if need be.
+ I have not tackled spindle speed control yet, but will do this when my new inverter arrives. I'll append my observations on this and on combivert replacement when I've done it.
- Denford Admin
- Site Admin
- Posts: 3642
- 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:
Re: Easiturn3 conversion to Mach - some observations
Thanks Rick, I'll make this sticky so it stays near the top of this section.
-
- Posts: 31
- Joined: Wed 23 Feb , 2011 9:16 am
- Hardware/Software: Denford easyturn pnc2
Converted to mach3
Denford triac servo /G simple/quickcam - Location: Dumbarton scotland
Re: Easiturn3 conversion to Mach - some observations
Rick your a star m8 thanks for taking the time to put this info
Up
Up
Re: Easiturn3 conversion to Mach - some observations
rickturner646 wrote:The Combivert inverter that was fitted as standard cannot be used with modern spindle control boards - it has a resistive (0-10k ohm) input control mechanism, not an analogue (0-10v) one. If you do want to keep it.....
... you can use the C18 board by CNC4PC. This has a real potentiometer on it.
https://www.cnc4pc.com/Store/osc/product ... cts_id=179
Expensive, but a similar thing could be made using a motorised potentiometer (ALPS).
I would beg to disagree. They are absolutely bog-standard inductive proximity sensors. If the switching characteristics of the standard ones don't suit, other ones in the same package (M8) with different switching characteristics (e.g. logic level) are available.rickturner646 wrote:I suggest that you forget trying to use the homing switches - they are seriously weird and I could not get them to work. Mach does not need them anyway - you can use the limit switches for homing
Before you abandon them altogether, you should perhaps do some tests with a dial indicator to check the repeatability and/or hysteresis of your limit switches. Microswitches (especially the ones on Z) are not really appropriate technology for home switches.
-
- Posts: 26
- Joined: Wed 02 Mar , 2011 9:48 am
- Hardware/Software: Easiturn 3
Re: Easiturn3 conversion to Mach - some observations
Hi Darryll,
I cant disagree with you - there is more than one way of doing each of these tasks, and thus it comes down to a matter of personal choice.
Undoubtedly there are ways to make the resistive controller on the combivert work - the Soundlogic card is one that I mentioned, and machines such as the Wabeco CC-D6000 cnc lathe are still being built this way. However, the combivert on my lathe is misbehaving (as I have said in previous posts elsewhere on this forum) and so I need to replace it anyhow.
The homing switches are another case in point - it is possible to use them with Mach, but then again, they are not mandatory as it will work just as well without them. Not using the homing switches - and connecting the limit switches in serial fashion - means that you only need one parallel connection (thus only one breakout board, one pc parallel port, etc). This makes life a bit simpler on that front.
The homing switches fitted to my lathe are not the originals and dont match any of the documentation (a phrase that is getting to be a very familiar one!) and I could not get them to work. So, for now, I have chosen not to use them. I've only disconnected the wiring, not removed it, so I can always change my mind later if I want to.
Rick
I cant disagree with you - there is more than one way of doing each of these tasks, and thus it comes down to a matter of personal choice.
Undoubtedly there are ways to make the resistive controller on the combivert work - the Soundlogic card is one that I mentioned, and machines such as the Wabeco CC-D6000 cnc lathe are still being built this way. However, the combivert on my lathe is misbehaving (as I have said in previous posts elsewhere on this forum) and so I need to replace it anyhow.
The homing switches are another case in point - it is possible to use them with Mach, but then again, they are not mandatory as it will work just as well without them. Not using the homing switches - and connecting the limit switches in serial fashion - means that you only need one parallel connection (thus only one breakout board, one pc parallel port, etc). This makes life a bit simpler on that front.
The homing switches fitted to my lathe are not the originals and dont match any of the documentation (a phrase that is getting to be a very familiar one!) and I could not get them to work. So, for now, I have chosen not to use them. I've only disconnected the wiring, not removed it, so I can always change my mind later if I want to.
Rick
-
- CNC Guru
- Posts: 525
- Joined: Tue 14 Nov , 2006 12:12 pm
- Hardware/Software: .....
4 Quatros
3 Oracs
2 Easiturn atc
12 station lathe atc
Mach 3 - Location: Quatroland
Re: Easiturn3 conversion to Mach - some observations
i've used the standard teco inverters on orac conversions and early triacs, they work well but don't have a lot of bottom end grunt. Teco make a "vector" version which has better low end power.
-
- Posts: 26
- Joined: Wed 02 Mar , 2011 9:48 am
- Hardware/Software: Easiturn 3
Re: Easiturn3 conversion to Mach - an update
A quick update on progress so far.
As of tonight, I have the spindle working under Mach control. I have a Spindle2 card (also from diycnc.co.uk) driving the original Combivert inverter through its resistive controller. To do this, connect the Combivert COM connection (yellow, numbered '55' on the schematics) to the 0v output terminal of the Spindle2 card and the REF connection (green, numbered '56') to the adjacent V-OUT terminal on the Spindle2 card.
The input DIR/STEP/GND connections of the Spindle2 card are wired to the a-axis connections on the Uniport breakout board, and Mach has been configured to use this to drive the spindle. All that remains is to feed the Spindle2 card 12v power.
I found that the problems that I was seeing with the Combivert were due to a loose internal connection - having found and fixed this, it now seems to be working ok (touch wood!) so the shiny new Teco inverter can stay in its box for now.
I still need to sort out reversing the direction of rotation and the encoder, which isnt working for some reason. Tasks for the next couple of days. I'll post here anything of note that I find.
Regards to all,
Rick
As of tonight, I have the spindle working under Mach control. I have a Spindle2 card (also from diycnc.co.uk) driving the original Combivert inverter through its resistive controller. To do this, connect the Combivert COM connection (yellow, numbered '55' on the schematics) to the 0v output terminal of the Spindle2 card and the REF connection (green, numbered '56') to the adjacent V-OUT terminal on the Spindle2 card.
The input DIR/STEP/GND connections of the Spindle2 card are wired to the a-axis connections on the Uniport breakout board, and Mach has been configured to use this to drive the spindle. All that remains is to feed the Spindle2 card 12v power.
I found that the problems that I was seeing with the Combivert were due to a loose internal connection - having found and fixed this, it now seems to be working ok (touch wood!) so the shiny new Teco inverter can stay in its box for now.
I still need to sort out reversing the direction of rotation and the encoder, which isnt working for some reason. Tasks for the next couple of days. I'll post here anything of note that I find.
Regards to all,
Rick
-
- Posts: 26
- Joined: Wed 02 Mar , 2011 9:48 am
- Hardware/Software: Easiturn 3
Re: Easiturn3 conversion to Mach - some observations
Good news - I now have the lathe working completely under Mach 3.
In the hope that it will perhaps make life easier for somebody else, I will attach to this post an electrical schematic, the XML file for Mach (either rename it to 'mach3turn.xml' or use a 'profile' shortcut), as well as a diagram showing the pin-outs on the three cables/plugs that come up from the lathe body to the original electronics unit - all the control cables you need are in these bundles. Note that I've had to upload the XML file in ZIP format as the forum does not allow XML uploads, so you'll have to unpack this first.
I am however not happy with the speed control being provided by the Spindle3 card, which seems to be fairly inaccurate - commanding S200 for example gets me 221 rpm, and the difference between the two is not linear, increasing with rpm. (And to forestall the obvious questions - yes, I have used Mach's autocalibrate, and this is *after* that has been done, and yes, the 'actual' speed is accurate).
So, I have a Digispeed card on order, and when that arrives I'll fit that and see how it goes - I'll append the results to this forum in due course.
Rick
In the hope that it will perhaps make life easier for somebody else, I will attach to this post an electrical schematic, the XML file for Mach (either rename it to 'mach3turn.xml' or use a 'profile' shortcut), as well as a diagram showing the pin-outs on the three cables/plugs that come up from the lathe body to the original electronics unit - all the control cables you need are in these bundles. Note that I've had to upload the XML file in ZIP format as the forum does not allow XML uploads, so you'll have to unpack this first.
I am however not happy with the speed control being provided by the Spindle3 card, which seems to be fairly inaccurate - commanding S200 for example gets me 221 rpm, and the difference between the two is not linear, increasing with rpm. (And to forestall the obvious questions - yes, I have used Mach's autocalibrate, and this is *after* that has been done, and yes, the 'actual' speed is accurate).
So, I have a Digispeed card on order, and when that arrives I'll fit that and see how it goes - I'll append the results to this forum in due course.
Rick
- Attachments
-
- Updated version of the schematic - changed the e-stop symbol to a PTB switch, and modified the Spindle3 card power connection.
- mach schematic with spindle3.jpg (118.33 KiB) Viewed 21542 times
-
- Pinouts for the three armoured cable sets that come out of the lathe body into plugs C01, C02, and C03.
- Denford Easiturn pinouts.jpg (135.62 KiB) Viewed 21562 times
-
- Denford_with_spindle3.zip
- XML file for Mach3 - either rename this to 'mach3turn.xml' or use a Mach3 profile.
- (23.77 KiB) Downloaded 1032 times
Last edited by rickturner646 on Wed 23 Mar , 2011 20:46 pm, edited 1 time in total.
- bradders
- CNC Guru
- Posts: 1251
- Joined: Mon 13 Feb , 2006 12:35 pm
- Location: Brighouse, England
- Contact:
Re: Easiturn3 conversion to Mach - some observations
Rick,
Brilliant Diagrams
Brilliant Diagrams
-
- Posts: 26
- Joined: Wed 02 Mar , 2011 9:48 am
- Hardware/Software: Easiturn 3
Re: Easiturn3 conversion to Mach - some observations
I've just updated the wiring schematic diagramme above to make two important changes:
1) I've changed the symbol for the e-stop to that for a PTB switch, which is what you should have (fails safe in the event of a wire breaking).
2) I've changed the routing of the +12v power line for the Spindle3 card to go through RLA1 on the Uniport breakout card. MAKE SURE THAT YOU DO THIS, IT IS IMPORTANT. Whilst I am tinkering with things I have the Spindle3 card on a separate power supply (just to make my life easier - the card is not installed in the enclosure yet). I normally turn on the enclosure power first, then the spindle card, but for some reason this evening I powered up the Spindle3 card first, and got an interesting surprise when the lathe spindle took off at high speed without being commanded to do so....
Of course, this *should* not happen, but it did, and could lead to a nasty incident. So, to guard against it, I have routed the power supply to the Spindle3 card through the relay on the Uniport controller, which ensures that it cannot get power before the controller is up and running.
Rick
1) I've changed the symbol for the e-stop to that for a PTB switch, which is what you should have (fails safe in the event of a wire breaking).
2) I've changed the routing of the +12v power line for the Spindle3 card to go through RLA1 on the Uniport breakout card. MAKE SURE THAT YOU DO THIS, IT IS IMPORTANT. Whilst I am tinkering with things I have the Spindle3 card on a separate power supply (just to make my life easier - the card is not installed in the enclosure yet). I normally turn on the enclosure power first, then the spindle card, but for some reason this evening I powered up the Spindle3 card first, and got an interesting surprise when the lathe spindle took off at high speed without being commanded to do so....
Of course, this *should* not happen, but it did, and could lead to a nasty incident. So, to guard against it, I have routed the power supply to the Spindle3 card through the relay on the Uniport controller, which ensures that it cannot get power before the controller is up and running.
Rick
Re: Easiturn3 conversion to Mach - some observations
Hi Rick did you ever get the digispeed card? What were the results?rickturner646 wrote:I am however not happy with the speed control being provided by the Spindle3 card, which seems to be fairly inaccurate - commanding S200 for example gets me 221 rpm, and the difference between the two is not linear, increasing with rpm. (And to forestall the obvious questions - yes, I have used Mach's autocalibrate, and this is *after* that has been done, and yes, the 'actual' speed is accurate).
So, I have a Digispeed card on order, and when that arrives I'll fit that and see how it goes - I'll append the results to this forum in due course.
Rick
Very interested as I am looking to convert an easiturn myself.
thanks
Steve