I bought a 25L drum of rocol ultraglide, usually something like £35 for 5L!
I almost bought a 20L drum of the rocol coolant.
The inverter looks an OK price, I spose I should start looking for one.
So your almost finished?

Almost finished? Don't be daft! I'd like to think i'm getting close but I'm not so naieve to actually believe it yet

Spindle motor arrived yesterday, looks very shiney indeed. Only 20kg too so I can actually lift this one. I need to take the pulley off the old motor and fit it to the new one, and ideally I will be able to fit the new motor to the old cast metal flange which sets the height, mounts to the machine and provides an anchor point for the DC brake. However I cant get the old flange off (removed the nuts but the bolts wont budge) so I'm getting a local engineer chap from our village to have a look at that on tuesday. Theyre used to dealing with tractors and combines so a motor that seems heavy to me and difficult to handle will be no problem to them with all their forklifts and agricultural gear.

What i'm probably going to have to end up doing is making up some sort of adaptor to fit the new motor to the old flange and then will have to extend the shaft as well as well modifying the diameter to fit the old pulley.

Sounds close to finished to me;-)

I made good progress yesterday, got all the axis motor drives connected and wired up to the main PSU, ran the PID tuning setup and ironed out a few other minor faults. (Had to disable the charge pump for the time being due to noise issues)

Everything was connected and ready to go so I installed mach 3 and then BAM! computer died.

I actually cant belive of all the things to go wrong and hold me back, the lack of a working pc would be it. I spent nearly all day rebuilding this thing, pretty much swapped out and replaced every part and it wont even install windows.

I think I have dodgy ram but new memory would cost more than a new computer so that looks like the way things might be heading.


More Expense ! *sigh*

Found some sensibly priced Rambus memory on ebay so i'm going to give that a go before shelling out on a new pc base unit.

This morning Alan Soaper, a very skilled local engineer came over to my workshop to help me separate the old DC spindle motor from a heavy steel casting which is used to mount the motor to the frame and tension the belt. As well as this we also removed the pulley from the old motor to go onto the new one.

For the first time in a long time I actually had some luck on my side and by some pure fluke, the old casting that the old DC motor attached to also fitted perfectly to the new motor! I was expecting to have to make up some sort of adaptor but it just dropped straigh in - both the bolt holes and the centre circlular flange.

The shaft on the old motor was slightly larger both in length and diameter so that wasnt going to be a direct fit to the new motor. I've left the motor and pulley in alans capable hands and I trust he will have a solution for me in a week or two from now.

I've also had some time to ponder over the maths for controling the axes accurately. Again I was lucky and a quick look through the brigdeport maintainence manual gave me a lot of the numbers I needed to start off with so that was a big help. I havent tested any of this yet so i dont know if its right or even nearly right but the numbers I came up with seemed sensible so I am quietly confident.

Bridgeport Series II Interact 4 Ballscrews:

5mm pitch

Z axis - no reduction

X + Y axis - 2:1 belt reduction

1/0.2 = 5mm

thus Z Axis motor Revs per mm = 0.2

X + Y motor axis revs per unit =

0.2*2/1 = 0.4

Motor Steps per revolution: 500 (optical encoder lines) x 4 (interpolated lines - quadrature encoder) = 2000

Mach 3 Steps Per unit = 2000 x 0.4 = 800 (x+ y axis)

Mach 3 Steps Per unit = 2000 x 0.2 = 400 (z axis)


The mach 3 manual explained something about a difference between actuall ballscrew pitch and effective ballscrew pitch - it was related to the type of thread - whether is was a multistart type or not and how many starts it had. To be honest I didnt really understand it so I'm going to assume for the time being that the number quoted in the maintainence manual for pitch is the right one and go from there.

Mach 3 includes a self calibrating routine which (using a dial guage) is able to work out the exact number of mach 3 steps per unit by working out the difference in error between commanded distance moved and measured distance moved.

Not much happening ATM, mainly due to lack of computer. I've given up trying to fix the other one and having replaced everything else the fault must now lie with either the processor or motherboard, this being the case theres no point repairing so I'm now on the look out for a second hand pc in the region of 2.4ghz with at least 512mb ram to use as a replacement.

Still waiting for Alan to finish modifying the new 3 phase motor to fit the old pulley, hopefully this will be done next week.

3kw VFD has arrived from china, looks good and instructions arent too difficult to understand, of course I cant test it now until I get the motor back from alan.

Recieved a phone call yesterday from Mark at Forco/Erics to settle a payment disupute where they did a load of work that I never asked them to do and then expected me to pay for that work rather than just the brush change which I asked for and was quoted on. After nearly a year they have finally agreed to accept the intially quoted price as full payment.

While on the phone I explained about my problem with the spindle motor I gave them and payed £1000 for them to refurb only for it to die after 5 mins of total use. They have agreed to take a look at that for me as well, and while I dont plan to go back to that motor now as I'm liking the 3phase setup quite a lot, at least if I can get it working I should be able to sell it for at least as much as I paid for the refurb job, maybe more.

Next week I plan to obtain a 'new' second hand pc and then I can continue with my testing of the axes. Having done the maths to work out the mach 3 steps per unit I should be able to get reasonable accuracy straight away , at least enough to stick a pen in the spindle and draw some recognaisable shapes with gcode.

I still need to wire in the coolant pump, the oil pump/oil level sensor, the drawbar solonoid, the drawbar proximity sensor, the drive disable switch and the switched external sockets. Ovbiously it'll make sense to test this as I go along so I'll need the computer before I can start that.

Once I have the axes moving under computer control I will run the calibration sequence and (using a borrowed dial indicatior) I'll calibrate mach 3 to be as accurate as possible. When I get the spindle motor back from Alan that will slide straight into the machine and the wiring shouldnt be too complex either.

Could have the machine making its first cuts in just a few weeks from now!.

Once the machine is operational I will need to use the machine to make new encoder mounts which will be sealed to keep the dirt out. I will also manufacture a MPG handweel control and an engraved freestanding control panel which will house a number of switches, touch screen display, keyboard, mouse and internal keyboard emulator.

With that done, it will litrally be a case of adding a final lick of paint, checking the lube system is oiling every thing its meant to and tidying up a few loose ends before the project will finally be considered completed and a success.

A likley follow on project would be the addition of a rotary table 4th axis as this potentially would be both cheap and easy to implement whilst vastly increasing the capability of the machine. For the electronics I would need to add another servo card and change the PSU transformer to match the motor voltage. The 4th axis could then be bolted to the table and wired in through a removable connector in a matter of hours.


Another possible (but unlikley) follow on project would be the addition of an ATC module. While I feel this is not beyond my ability, unless I find a desperate need for auto tool change then I'll probably run this machine as is with the power drawbar and then buy a VMC when I have the money.

Sounds even closer to being finished.
I spent most of the last month doing other stuff. Got to make an effort and get something done in MAY.

New PC arrived, only £70 inlcuding shipping for a reconditioned dell unit with a 2.8ghz progcessor, half gig of ram and legit copy of xp. Didnt take long to install my parrallel port cards + drivers + cnc software & windows updates so I was up and running again in no time.

Took the pc up to the barn, plugged everything in and was quietly confident I'd be up and running pretty soon,however no matter what I tried I just couldnt get mach 3 to talk to the pc properly. Some limit switches would work but some wouldnt and I didnt want to try anything else before getting the limits sorted.

After a lot of head scratching, swapping ports and port numbers and pin numbers I finally discovered that there were a few dry joints on one of my home made breakout boards. Made a quick trip to my local computer repair shop to borrow their fine tipped soldering iron and then I was back in business.

With the limits working I powered up the axis drives and moved onto the motor tuning. Having previosuly worked out my mach 3 steps per unit and knowing the rated speed of the axes (5000mm/min), all I had to do was type in the values and then choose a sensible acceleration value.

This was all going well until I tried to run the motors and I would get a very small amount of movemnt and then it would suddenly stop. After exiting from the motor tuning screen it became apparent why. I was getting messages saying limits and e-stops had been triggered and reseting/clearing the fault would only last a few seconds before it happened again. Disabling all the inputs solved the problem and allowed me to run the motors but of course now I had no limits or estop,

It seems that now my biggest problem is that of interference; seemingly generated by the servodrives and/or motors. Energising the logic side of the drives causes no problems but as soon as I switch on the main power input to energise the motors I get random trips on all my limits every few seconds, regardless of whether i'm running the motors or not.

I'm not sure where to go with this really. I've tried the input debounce option in mach3 with no real success so now i'm thinking the answer might be in shielded cables, suppression caps/chokes, or optoisolated boards.

Either way I cant make much more progress until this is solved cause the limits are also the home switches so I need them to work for referencing purposes as well as safety.

Sent an email to Tom @ Rutex regarding my interference problem and this is the reply I got:



Following Tom's advice I moved all my inputs over to the cnc4pc breakout board on the second parrallel port and all my problems went away like magic. There were still a few glitches but I hope to get rid of them with some 0.1uf caps accross the terminals.

With the software not triping every few seconds I was able to have a better play and even had the axes referencing to their home positions. Typed in a bit of gcode and saw the axis moved and even ran a short gcode program.
I think my maths for the mach3 steps per mm is a bit out, maybe by a factor of two for the z axis but this is no problem as I will run the calibration routine next week and get it spot on.

I then turned my attention to the power drawbar and after unjamming the solonoid valve which had become a bit stiff from inactivity, I was able to hook up the compressor and operate the drawbar. It worked faultlessly and for the first time ever I was able to put in a toolholder.

With 80psi of air in the system (as per maintainence manual) I was also able to see that the power assisted knee was working which was good to know.

Alan also tells me my spindle motor will be ready for collection soon.


Next priorities

I need to purchase a few bags of 0.1uf supression caps. I will try putting these accross the limit switches where they connect to the break out boards in an attempt to kill the occasional voltage spike. If this doesnt solve the problem entirely I will consider ferrite beads, shielded cable, re-routing the cables or a combination of all three.

I also need to buy or find a diode to put accross the connection of the drawbard solonoid to protect the solidstate relay from back emf.

I need to wire up the drawbar solonoid to a switched output and I need to do the same for the lube pump and suds pump. I don't even know if the suds pump still works as I have never used it in the time i've had it and it all looks a bit gungy in there. However I have 2 litres of fresh soluable cutting fluid so i'm hoping it will flush everything out and run well.

The spindle motor and associated electonics is ovbiously a priority and will be something to do next week when I get it back from Alan.

I think I also really need to spend a few days learning about mach 3. It is an incredibly complex piece of software and my knowledge at the moment is only minimal. In order to set up this machine and use it properly I'm going to need a much better understanding of how the software side works. I suppose this research should also extend to cad/cam software as well. I need to learn how everyting all fits together in practice.

Like I said before;-).....................

Sounds good. I remember seeing a post about interference on CNC zone,
the bloke had great problems, shielded everything and still had problems.I think US digital sent him more enncoders, and he didn't have any more
problems; he did try running the newencoders on plain wire, coiled around an inverter and it still worked perfectly!

My machine had shielded wires, on everything, the encoders are individually shielded twisted pairs. almosy everything seemed to have .1uf caps to ground.

Graffian: hopefully the encoders are not the problem (i think is unlikley as they do not cause problems when the encoders are powered up but the drives arent). I cant afford new encoders!



Havent updated this for a while so here's some more.

Got the spindle motor back from Alan who did a sterling job of modifying the old pulley to fit the new motor. The inverter drive already arrived a few weeks previously so I was very keen to wire everything up and see things spin.

After printing the 70page manual for the VFD and working my way through several hundred parameters I was ready to power up. With great excitment I pressed run only to hear a high frequency noise ramp up and then cut out followed by a cryptic error message.

I exchanged several emails with the supplier and to cut a long story short I was facing having to send the thing back to china for replacement. As a result of that I went to work on other areas of the mill.

Spent a few hours watching tutorial videos on mach3 and got my head around how the co-oordinate systems work as well as lazy cam and a few other bits.

Looking at my machine for ways to reduce noise and just generally tidy up I decided to ditch the encoder wire break out boards and purchased some shielded rj45 connectors allowing me to plug the encoders directly into the servo drives.

I also spent some time wiring up all the remaining inputs and outputs and was able to test it all though mach 3 with successful results. By pressing buttons on the computer I was able to switch on and off the external 240v socket, power up the suds pump (for the first time since owning the machine) - it soudned ok but i had no water to test proeprly, also operated the drawbar from the computer and enabled the oil pump however i was not able to verify if that was working as it is so slow and small. - something for another day!

Now that i'm not using the limit inputs on the rutex drive I can abandon the custom db15 breakout board entirely (another potential source of noise ingress) and wire the power directly from the 24v supply to the rutex board through a custom db15 cable/connector. This does however create the problem that i'm now lacking 4 inputs which I really needed. However I think I can solve this by creating a LPT breakout board which cuts off those four input pins from the drives.

Today I had another look at my spindle and VFD setup, I thought I'd try reversing the motor wiring in case for some reason that would make it work. I wasnt even taking it too seriously, just having a tinker for 10mins before work. Only then did I realise why it hadnt been working, one of the phase wires wasnt properly connected so it was trying to start the motor on just two phases.

Sorting out the misplaced wire suddenly solved all my problems and the 4hp motor roared into life! Of course then I had to go to work so I didnt really get the chance to do much more. Had just enough time to play with the accel/decel ramps a bit and add an external enable switch. Tomorrow I will have to connect a POT and also run the motor for a much longer period of time to make sure everything is running ok. If it can run for more than 5 mins without leaking magic blue smoke it'll be an improvement on the setup this is replacing!

you have the spindle speed controled by mach?

Will do. Might have that part wired up by the end of the day.

Started playing about with customisation of Mach 3.

Using Screen4 & MachScreen I added some extra LED's and buttons to my screen and had a go at coding them.

To make editing them easier within mach 3 I set each button to call an M code and then created custom M codes for each function.

M12 is now my drawbar control.

Pressing the drawbar button runs the following code:

If GetOEMLED(11) Then
Message("Spindle Must Be OFF to Change Tool!!")
ElseIf GetOEMLED(852) Then
DeActivateSignal(OUTPUT1)
Message("Tool Locked")
Else
ActivateSignal(OUTPUT1)
Message("Tool Released")
End If

In english this first checks the spindle isnt running (might add a further check to ensure spindle is homed).

Then depending on the current condition it either activates or deactivates the drawbar output.

I was going to do all the spindle stuff in hardware.
basically it won't switch the solenoid to push the drawbar down if the heads not in up, there are a pair of limitswitches for toolchange position and if the spindle is running. there is a spindle orientation unit, the spindle was always put in a certain orientation for toolchanges.

As I am going to be downing the changes manually so I will just use the switch on the pendant.

Its probably quicker/easier to do it in hardware but I prefer the flexability that software provides. Plus if I ever decide to build an ATC then half the functionality will already be there.

It would be quite easy in mach 3 to have the spindle park in the same orientation every time you stop it.

Well I got the POT wired up to the VFD ok so I had variable speed and I played about with a few parameters to tweak accel/decel etc. Then I thought i'd give the motor a long run for an hour with no load to see how it coped.

Measuring the temperature every now and then I became concerned how warm the shaft of the motor was getting and expressed these concerns to the supplier. The supplier suggested I dissassemble the motor and check the bearings etc but there was nothing visibly wrong and they were all turning freely. With this in mind I was advised not to worry and that it is normal for these motors to run quite warm.

- I think i'm just getting overly paranoid now cause of all the problems i've had in the past.

Yesterday I took all the software and screen modifications I had done at home to the workshop to test on the machine. All worked flawlessly. While I was there I also removed the spray mist coolant unit which i had no intention of using which has cleared up some space in the side cabinet. Now of course I have to find a blanking plug for the t-joint it was plumbed into and my local engineering shop doesnt stock that particular thread. First job for tomorrow will be modify the old connection, boring it out to 8.5mm and tapping with an M10 thread. Then I can plug it with a 10mm bolt and some PTFE - sorted!

Another huge and messy job yesterday was taking appart the auto oiler. I energised it via the pc but couldnt see or hear any movement. Not wanting to risk burning anything out if the motor had stalled or was polorised I ended up stripping the whole thing appart to see what was going on inside. Turns out everything was ok, it is just meant to run very very slowly. The motor only does 1/5 of an RPM so it was no surprise I didnt see immediate movement. At the moment this pump is switched with a button of its own but I will have to add some codes to the spindle macros so the oiler comes on when the spindle is running as per the original machine design.

Another thing that caught my interest yesterday was the drawbar proximity sensor. I'm currently trying to find out its exact purpose as I reacon it will be possible to run without it, using software to make the checks instead. i.e if the only thing its really checking is the position of the z-axis I can just put a line of code which checks Z = 0.000 before allowing the solonoid to be fired.

All my bits from rapid arrived yesterday, supression caps and all sorts. Included in this delivery was my shielded rj45 connectors for the encoders. However I was unable to wire these in as they were incredibly complicated consisting of 4 separate parts and myself nor two other local computer shops were able to work out how it was all supposed to go together. I know a network engineer who lives in my village who has probably come accross these before so its just a case of waiting until I see him down the local.

All of a sudden i've ended up with a lot of work to do at once. Should be able to get the motor in tomorrow and the VFD wired up. Then theres all the hundreds of little things, tweaking the servodrive tuning parameters, calibrating the servos in mach 3, learning how to use mach 3, learning 3d cad, learning 3d cam the list goes on! Also really need to clean out the base of the machine which forms the sump of the suds pump. I shone a light down there yesterday and theres all sorts of crap in there! A thick layer of oil/sludge as well as a few allen keys, dead insects etc. I think the only way to get to it will be to tie some rags to a long stick and try and get the worst of it like that. Theres no other easy way to get to it without stripping down the whole machine and taking off all the cabinets.

I know this is all too much reading for a lot of you so here are some long overdue photo updates:



The new spindle motor: 240v 3phase 3kw 4hp. Brand new and delivered for less than it would have cost me for someone to take the cover off the old one to see what might be wrong. While 4hp is about half the power of the old motor it should still be more than enough for what I want to use it for. Many of the manual bridgeports only have 2hp motors and are still more than capable of taking some big chunks out of steel.




This is the old yellow motor flange painted black to match the new motor. By some huge stroke of luck both the centre hole and the bolt holes all line up with the ones on the new motor so it will be just a case of dropping it in.




The new variable frequency inverter drive. Brand new with 12 month warrenty imported from china for about £200 including shipping. This takes a 240v single phase input and outputs 3 phase at a variable frequency of anything from 0 - 400hz. This is an incredibly customisable piece of kit with nearly 200 progammable parameters. Useful functions for this project include 0-10v analouge input for speed setting, digital inputs for forward / reverse, cusomisable accel/decel ramps and overcurrent protection.




This is the old pulley that has been modified to fit the new motor by local engineering legends Alan & Terry Soaper. Alan did a sterling job of this and saved me a lot of headaches as I didnt have the tools or experience to attempt it myself. Of course if the cnc mill was working it wouldnt be a problem ...catch 22 again

my drawbar has two limit switches, I would imagine if the lower one is on you can't start the spindle; or should that be if the upper one isn't on. Motors do get hot.

Made a lot of progress today:

# First went over to soapers and made up a blanking plug to close off the open t-piece where the old spray mist unit was connected.

#Went over to CEF and bought 10m of chunky 4mm twin and earth to use for the motor wiring.

# Fitted the custom made blanking plug

# Cleaned out all the gunk in the base of the machine using rags and pointed sticks.

# Tested the flood coolant pump to find it wasnt running so well (making funny noises and splashing water around on the intake side.

# Removed the coolant pump from the machine for refurbishment

# Fitted the new spindle motor to the machine

# Tensioned the spindle belt

I was going to comment on your lube pump, Gor that's a fast one mate;-)
Mine takes 17 minutes to do a revolution!