emco pc mill 50 & similar

[jerogsxr]

Hello! i have the software and every disk , send me intructions please.

[cwalker1960]

very informative thread. I have recently acquired a pc mill 50 and was hoping to use the original stepper board and motors , it seems that the bipolar stepper conversion is the way to go . I thank everyone that has treaded these waters before me for sharing their experience and saving me the heartache from beating a dead horse. I got it ,, gut it and start fresh.

[gavztheouch]

Hello I am looking at buying an Emco mill 50 . I would like to know an accurate measure of the Y axis travel, there seems to be a lot of variation online from 90mm up to 140mm?

Many thanks
Gav

[hutchison]

Hello I am looking at buying an Emco mill 50 . I would like to know an accurate measure of the Y axis travel, there seems to be a lot of variation online from 90mm up to 140mm?

Many thanks
Gav

basic unmodified specs as follows , (have edited into first page also)

working area
X:190 Y:90 Z:190
Table load: 10kg
Table dimensions: 420 x 125
3 phase reversible asynchronous Spindle motor power 100%/60% = 370/550W
RPM: 200-2500
Steppers: 5 phase AC synchronous closed loop axis drive system (berger lahr i believe)
Rapid: 750mm/min
Feedrate: 750mm/min
SAFETY: axis overtravel switches and door switch
machine dims : LXDXH 840x865x710
machine weight : 160kg (using two long metal poles :6ft @ 30mm O.D slotted through the base ) two strong people can lift this machine, OR, use an engine crane etc!.
(be warned its not for the weak, if your not used to lifting heavy weights off the ground dont bother use an engine crane and strap it using the bars).
Spindle taper: BT30 (sk30 / iso30 fits but sometimes change the pullstuds etc)
Standard collet system: ER25
power consumption: ? (runs off 110-240vac standard 13A single phase mains in UK)

[cwalker1960]

Hutchison's measurements seem about in line with the one I have..If you don't get the computer that goes]
with it , you may have some problems. I have absolutely no idea of the workings of the original software as the one I'm working on didn't have it ,I spent a few days trying to get the original stepper motor boards to take step and direction commands with very limited success. It really looks to be a substantial little machine. only time will tell as I have just begun on the conversion. I looked into 5 phase drivers and that didn't seem practical so i'm converting it to 2 phase motors and drivers and probably mach3 but possibly linuxcnc. It's for a friend and i think mach3 will serve him better. personally i like linuxcnc , but when the whole cnc thing is new to you the last thing you need is to throw a new operating system into the mix.

Hello I am looking at buying an Emco mill 50 . I would like to know an accurate measure of the Y axis travel, there seems to be a lot of variation online from 90mm up to 140mm?

Many thanks
Gav

[hutchison]

These machines are rock solid Castings, you wont be dissapointed with the cuts. I took mine out of storage after 3 years fired up no issues yet. I am still running the original Software (wincam) i create dxf's and just import them over, and process them through the cam package, fairly easy.
Limitations with the original software (wincam) the basic cad cam package, and WinNc (control emulator for O.E controls like fanuc , siemens etc) are unless you have a post processor fr the mill50 for 3d cad cam like Licom alphacam(emco provides free post processor for this), then you would be limited to 2.5 axis milling G code output, as that is all OEM control package software outputs.

But, of course if your going to retrofit it out you can have full 3+ axes control with whatever control / cam you choose.
My friend in holland runs usbcnc and 4 axis on his 50, he also fitted new ballscrews.

I like the wincam, it works right out of the box for me no fuss, no 1,000,000 settings to mess with its a coined package, you can focus on the machining.
I also have a 2 axis lathe which i retrofitted with mach3 and new ballscrews/spindle electronics.
I got the lathe out of storage just recently and the pc is just crashing with mach , its been instorage and nothing was changed!, its very frustrating i'm having to spend yet more hours reconfiguring the PC in the hope its going to work again, otherwise im looking at other options, maybe ess.

i made a bracket to hold a kress 1kw 30krpm spindle a few years back, its removable easily, its a split collar with a simple shc screw,works well for engraving or high speed drilling/micro machining, just move the vice under it. I was careful when i fitted it to examine the main spindle housing for suitability of the fixing holes.







here is the safe machining points if you want to make a similar bracket


sketches:
I dont gaurantee the centreline of the spindle axis to the Kress bracket axis but it sould be hwithin a mm. General toolmaking principles apply! square ends of blocks up, align part 1 to part 2, bolt it, run a clock over the assembled plates in a known 90 degree axis to check for perpendicularity, ream dowel holes in situ. hope it helps someone. dont forget to make sure your spindle axis is true to the table, the emco spindle housing rotates like a regular turret mill upto 90 degrees.







i apologise if there any missed dimensions , its late, as usual.

[cwalker1960]

We're going to have all the electronics sans computer for a pc 50 left over so if anyone is looking for anything , pm me.
Hutchison,
I got the motor swapped out on the z axis over the weekend and temporarily wired it up.I'm very excited about the results. Needs a little more fine tuning but just quickly I had it within 0.010 over 2 inches. Checked with dial indicator. I haven't timed it yet but Mach tells me I'm jogging at 100 IPM . Also haven't measured force yet but a friend had both hands on top of the spindle mount and couldn't stall it Impatiently waiting for the weekend to get the other 2 axis converted , I'm ready to take this thing for a test drive..

[hutchison]

i think you'll find the original ballscrews are 8 x 2.5pitch. that may help.

[cwalker1960]

i think you'll find the original ballscrews are 8 x 2.5pitch. that may help.

Thanks for the screw pitch , that'll make setup easier.. I tried to pm you back but I got an error saying your mailbox was full
I found some documentation on the original steppers . they seem to be around 35 oz/inch ,, that sound right?

Y4A 400 000 (control board for spindle)
Y4A 015 000 ( i'm guessing position feedback as all the prox switches went to it )
Y4A 030 000 ( Stepper driver board)
3 Berger Lahr stepper motors VRM 564/50 LNB
Ge Fanuc control panel

[Arjand]

Hi, I've bought a PC Mill 55 and Turn 55 recently and working with WinCAM at the moment. The software is reliable and works well at the moment, but when I want to extend to a 4th or 5th axis, on the Mill 55, I have to retrofit.

I read here that the steppers of the Mill 55 can be reused when retrofitting. I don't like the idea of not being able to use my mill when working on a retrofit. So my question is:

Can I install both Mach 3 and the original electronics where I can switch between them by moving some connectors? That would be ideal.

[cwalker1960]

I can't be much help with a pc55 , I converted a pc50 that uses 5 phase steppers. I believe I've read that the 55 uses 3 phase steppers. If that is the case leadshine makes an affordable 3 phase driver. I couldn't find any 5 phase drivers at a reasonable price so I done a complete conversion with new 2 phase motors and drivers. Some have tried to tap into the emco boards to get it to accept pulse/direction commands. i don't know how successful they have been. I started out trying that and finally did get one motor to move but couldn't get it behave right so I gave that up. I think it would be a real challenge to have it run the original setup with Wincam and mach3 .
as for my conversion , I bought an AMD ITX motherboard and mounted everything inside the emco case the only external components are the monitor , keyboard and mouse. i used 240 oz/in steppers and if I had to do it again the only I would do different is do away with teh belt reduction and direct drive the motors to the ballscrews. but even with the belt reduction I still managed 90 IPM rapids.

[richjones]

Hi - I have a PC mill 50 that I've got working with Linuxcnc - the steppers, drivers, spindle and VFD are all as standard - obviously the main controller card is replaced by the computer running linuxcnc (mini-itx inside the control box). I put together a simple circuit to convert the 3.3v logic input/output of the computer to the differential drive required by the mill. The differential inputs to the drivers and VFD are 0v/5v - I used 26ls31 and 26ls32 chips to do the conversion for me.
There are pinouts earlier in this thread - to complete the story:
Axis:
3x axis are driven by differential 0v/5v logic.
E-achse doesn't seem to do anything (maybe it is enable, but leaving it floating is sufficient. Don't know, but works without it)
SR X,Y,Z is a rotation sync pulse, but I'm not using it
The connector doesn't have a ground, so you have to make sure the connection is grounded somewhere (grounding is important anyway)

Spindle:
Spindle runs as PDM, with 0v/5v differential clock input driving speed
Differential 0v/5v dir drives direction
Differential 0v/5v phase and index outputs (input to the computer, SR+/- and NI +/-) can be used for rpm - I've only used the index pulse for rpm so far, but it works okay like this.
0v/5v Differential enable (E/AMD +/-) used to start/stop spindle. IMPORTANT - this triggers on the rising/falling edge, so if you just put a circuit together with these driven from the voltage rails it won't work if the circuit is on before the mill is switched on. But you might as well drive it from the computer using spindle enable. Also, I think this is different from the other inputs, in that you can enable the spindle with a rising edge on the +ve input OR a falling edge on the -ve input - but I'm not sure and I'm driving it using a differential signal anyway.
Don't know what IAMD does - guess it is a current feedback, but all works okay without it connected.

limit switches - normal open switches, so connect in using normal circuitry (ie, drive high using a resistor to 5v rail, then activating switch takes input low).

If anyone is interested, I'll put some pictures together, example circuit, etc.

I've also got a couple of interface PCBs spare as I had to order a minimum quantity - if anyone is interested I could send one for a contribution to the pcb printing fund - it isn't particularly elegant in layout, but works fine.

[Arjand]

Nice find Rich Jones. So I understand you figured out with signals are used on the rs485 cable en reverseengineered you own circuit? Do you think it would be possible to link this with something like USBCNC (Planet)? I have the Mill 55, and I really like the idea of leaving the original electronics in tact, as they work well.

[cwalker1960]

Don't know for sure but I think someone said the pc55 has three phase motors ,, if that is the case leadshine has fairly reasonable 3 phase drivers, I find it hard to trust an analogue input without real encoders.

[richjones]

Hi Arjand - not quite...

The standard mill has an external computer that sends rs485 to the mill, which uses an internal computer (the Y4A015000 module) to generate steps (for the three axis controller) and a pdm signal (for the spindle VFD controller). In my setup the linuxcnc computer replaces the Y4A015000 module to generate the steps and pdm signal.

Practically, this involves unplugging two 16 way idc cables between the Y4A015000 and the stepper-driver and VFD-driver, and instead plugging them into my circuit. This circuit then plugs into the parallel port of the linuxcnc computer. I leave the Y4A015000 in and connected - it could be returned to standard just by replacing the two idc cables back into the Y4A015000.

The circuit isn't complex - just 3x 26ls31 and 1x 26ls32 to sort out all the differential signals, and a single resistor to pull-up 5v for the axis limit switches. The circuit needs 5v which I take from a computer USB port (you can take it from the mill itself, but the 5v looked a bit dirty, so I used the computer 5v line).

I imagine it would be simple to connect to a usbcnc as that is the same - normal ttl outputs that require interfacing to the 0v/5v differential signal needed by the stepper-controller and the vfd in the mill. I don't know about how it would deal with the limit switches and the rpm feedback signal, but I can't imagine it would have much difficulty - once the signal is converted into normal 5v ttl there isn't anything special about it at all (apart from that the vfd wants a pdm rather than pwm signal - easy to set up in linuxcnc (and I think mach), but I don't know about usbcnc).

I really don't know about the mill 55 - I'll post up pictures of the back of my system and you can see if the drive modules have similar part numbers, or at least look similar. I can imagine that Emco left much of the circuitry intact between the two mill types (no point in reinventing the wheel), and just changed the rs485 protocol slightly and moved to 3 phase stepper drivers in the output stage of the stepper driving circuit. Don't know though...

Just to make it explicit - all that is required is for the normal ttl signal to be made 0v/5v differential. This means that when the computer output is at 5v, the normal input (eg, for spindle enable E/AMD) to the mill is at 5v, and the -neg input to the mill (for the example, E/AMD-neg) is at 0v. When the normal output is at 0v, the normal input is at 0v and the -neg input is at 5v. All this should take is a simple 'not gate' on the 'neg' inputs, but the actual circuitry on the mill uses 26ls31 and 26ls32 pairs to do the differential encoding/decoding, so I used the same circuit. Shouldn't make any difference, but using the 26ls31/32 makes sure that each line changes at exactly the same time - if you use a not gate then the not line will be fractionally slower because it has gone through additional logic. But as I say, I can't imagine that the Emco circuitry is that sensitive to the lines changing at very slightly different times. It is easy enough to get 26ls31 and 32 chips, so not worth experimenting with.

[I will own up to not doing it 'properly' - the official circuitry uses resistor pairs on both ends of the connection to make it a beautifully controlled impedance, and I guess to make the system very tolerant of a short circuit in the cable - I didn't as it worked fine without it and if I blow up a 26ls31 via an unlikely cable short circuit I can get a new one for <£1]

The big gotcha that I think has caught people out on the VFD controller is that the spindle enable/disable runs on a rising/falling edge, rather than on the logic state of the line. Thus if you set it up using hard-wired 0v/5v enable it won't work - it needs to see the enable line be driven high in order to operate. This caught me out for a bit. Practically this doesn't make any difference, as setting spindle enable in linuxcnc creates the necessary rising edge. As I mentioned in the previous post, however, I think that the + and - enable lines do not both need to be driven to get it to work.

I'll post up some pictures of the circuit and of my set-up as it will make things a bit easier to understand.

[Willywonka321]

Great work, I feel like you have discovered the "Rosetta Stone" of EMCO machines. I just picked up a 50 and a pc 50 turn for my classroom and would love to do a conversion like this so leave the machines stock. If you had some more documentation and parts left over I am very interested in getting to this. My students would be way better off (in my opinion) using Linuxcnc instead of the Wincam we are running off of. Please let me know what my costs would be.

[richjones]

Sorry to not reply - I've been called away with work - I'll update things when I get back, probably mid week. Cheers.

[richjones]

Hi.

For Axis - Need 2x 26ls31 - this is the same chip Emco use. Pinout is as:



Each axis uses two gates, the input comes from the parallel port, the outputs go to the dir and dir/neg lines for each axis, so for example, using the following pinout for axis z (on the 16 way IDC connector input to the axis controller)



Parallel port DirZ line (whichever one is set up in mach / linuxcnc) -> 26ls31 pin 1
26ls31 pin 2 -> Emco axis idc connector pin 7
26ls31 pin 3 -> Emco axis idc connector pin 8
Parallel port StepZ line -> 26ls31 pin 7
26ls31 pin 6 -> Emco axis idc connector pin 9
26ls31 pin 5 -> Emco axis idc connector pin 10.

Note that as well as the power to the 26ls31 you need to connect the enables - just connect pin 4 to +5v and pin 12 to gnd.

The other lines on the Emco idc connector are:
1, 2, 3, 4, 5 = SRX, DirX, DirX/Neg, CkX, CkX/Neg

11,12,13,14,15 = SRY, DirY, DirY/Neg, CkY, CkY/Neg.

The other two axis are connected in exactly the same way. Each 26ls31 can do two axis, hence the requirement for two chips (but can use some of the lines for the spindle connector, hopefully see next post)

The SR lines are for positional feedback from the axis controller to the computer - I've just left these unconnected.

I don't know what the E-Achse - Some Emco mills use a brake on the Z-axis, but it doesn't seem to do anything on the mill 50.

The Emco axis idc connector is numbered as follows:


With the key/cutout on the odd pins (up side on the image above).

Note that the Emco axis connector doesn't include a ground, so you have to make sure this is connected at least somewhere.

[richjones]

For Spindle:

Bit more complicated than the axis, as use:
26ls31 for signals computer->mill and
26ls32 for signals mill-> computer

The Emco 16 way connector is connected as:

(obviously thanks to the original suppliers of the info for these Emco pinouts - these were essential...)

DirMd and DirMd/Neg (pin3 and 4) are the spindle direction. Connected using a 26ls31 channel
ClkMd and ClkMd/Neg (pin 7 and 8) are the spindle speed line (encoded as PDM). Connected using a 26ls31 channel

The spindle Dir and Clk are connected in the same way as you would a stepper motor...

E/Amd and E/Amd-Neg (pin 1 and 2) is the spindle enable - connect using a 25ls31 channel

Ni and Ni-Neg is the spindle index. This is information from spindle-controller-> computer. This is connected using a 26ls32 chip (the partner chip to the 26ls31). Pinout is:

So for the spindle index (using first 26ls32 channel):
Ni -> 26ls32 pin 1
Ni-Neg -> 26ls32 pin 2
26ls32 pin 3 -> computer spindle index input.

Chip power and enables as per 26ls31.

Spindle SrM and SrM-neg (pins 13 & 14) is the spindle pulse line (phase A line in linuxCNC). This is connected in the same way as the spindle index line.

I've not worked out what IAMD (pin 11) does - I think it is a current feedback to tell when the spindle is under too much load, but I've just left it unconnected.

[richjones]

The axis limits are fairly easy to work with - each is a normally closed switch which opens when the axis -'ve limit is reached (they don't limit on the +'ve limit).

[I say easy to work with but somehow I thought they were normally open when I got my circuit made up, so I had to use a screw terminal block to get the limits working, rather than just plug them in - bit annoying!).

The easiest way to work with them is to connect them all up in series, connect one side to ground and the other to the 'all limits' line on the linuxCNC computer, along with a resistor to the +5v line (it is just a pull up resistor, so value not crucial. I used about 2Kohm, but it was just the one that came out of the box first. Probably 1Kohm -> 10kohm would be fine (TTL style pull up), and I assume most parallel ports are MOS, so even 200Kohm would presumably work fine as well)

The big problem I had with the limit lines is the impact of noise - I don't think the lines are shielded as they snake around the machine. I had to use a capacitor across the lines and a ferrite ring to remove the impact of the noise.

(I think Emco use 24V on the switch lines, and I only used 5V - this brought the noise problem to the fore - It is completely under control using 5V and the capacitor/ferrite, but I might go to 12V if I was doing it again)

I think that Mach allows software debouncing of the limit switches which would probably solve the problem as well. I don't think LinuxCNC does this, though.