July 6th, 2006

It all starts now, and It’s all going to be documented here. While I have already done a fair bit of research already I wont bore you with too many details from the past.

A few months ago I purchased a Bridgeport Series II Interact 4 at auction. The plan is to eventually start my own CNC machining business but for the time being this is just a hobby project. At the time of auction the Bridgeport had a spindle drive fault which was why the company that owned it decided it was time for a new machine. They didnt use it for about 6months and during that time it developed a power supply fault in the TNC 151 Controller. I won the auction, and aware of these faults in advance, had plans in place to do a full electronics retrofit.

I did briefly look into the possibility of repairing it as is, but the price of repairs/replacements of old equipment was just too prohibitive. Heidenhain wouldnt sell you replacement parts even if your life depended on it and have contracts with their suppliers to prevent them selling to anyone else either. The best they could offer me was a replacement (like for like) in the region of £1,600. While I could have got it for as little as £400 from some private repair companies , if the spindle drive needs replacing too, that’ll be in excess of £1000 too. Considering I estimated a full retrofit of all the electronics (except motors) would be in the region of £2,500 it soon became apparent that a retrofit would be the only sensible way to go.

My Aims/Goals for this project:

* Restore this CNC Mill to full working order
* Gain knoweldge in the process of retrofitting CNC machines
* Increase my working knowledge of electrics & electronics
* Learn about 3D CAD/CAM
* Focus the knowledge I have gained into a viable CNC based business plan
* To have fun!

Since winning the auction and having the machine moved over to my barn, very little progress has been made. Appart from removing a few panels and having a poke arround inside, everything else is largely unchanged. Today the plan is to visit the machine with a handfull of tools and make a start removing all the old electronics cabinets. While I currently have no intention of keeping/using the old stuff in the long run, everything will be carefully dissassembled and documented in case I change my mind later on in the project.

Here are a couple of photos of the machine in the factory before it was moved to my workshop:

July 8th, 2006

As a general rule, destruction precedes construction and this project is no different. I spent the best part of yesterday trying to take the machine appart as carefully as possible in case I needed to re-use anything at a later date. This thing is very well built and incredibly heavy. Hopefully most of what has been removed wont be going back, considering how much smaller modern electronic components are, a single small box on the back should provide ample space for the new servodrives.

Below are some photos of the Bridgeport mill and the main electronics cabinet:

July 14th, 2006

Yesterday I spent a few hours at the workshop wondering just what I have let myself in for. I have now completely removed all the old electronics cabinets and stripped the machine down the the bare minimum components needed to operate.

The next thing to do will be to suss out what all the wires do so I can build an electronics system to match it. Once I can figure out all the relevant voltage and signal inputs/outputs the rest should be quite easy.

Finding out the voltage of various components already attached to the machine is proving to be a bit of a challenge because it really is a mixed bag. You have everything from motors running at 415v AC 3 phase to solonoids runing at 24VDC.

There is what looks like a suds pump mounted inside the main casting with two wires comming from it. This suggests single phase, but it could quite easily be 110v rather than 240. I’m tempted to buy the maintenence manuals from Hardinge but so far I have been put off. The spares people I spoke to were not particularly helpful; they couldnt tell me how many pages the manual consists of, if it would contain the information I needed, nor could they send me some sample pages and considering on person admited that its an old photocopy with writing almost impossible to read in places, i’m very reluctant to shell out the £70 asking price. They dont even accept credit/debit card - I’d have to send a cheque!

I did have some photos of the latest progress but for some unknown reason my digital camera decided to eat them. I’ve re-formatted the memory card so hopefully I’ll have more luck next time.

Yesterday I spent a few more hours up at the workshop. After giving the place a long needed tidy up, I managed to further my understanding of the machine a little bit more. Previously I discovered a large cylindrical object at the base of the machine which looked a little like a hydraulic ram but with an air supply going into it; this was connected to a small actuator at the top of the drawbar.

I have since discovered that this is a Pneumatic/Hydraulic convertor with a built in mechanical advantage. The ratio is printed on the side as 25:1 so if it takes a 100psi air supply input then the output is 2500psi oil to the hydraulic ram in the drawbar. - Clever stuff!

I have just fired off an email to Rutex (http://www.rutex.com) to find out if their servo drives will be compatible with the encoders on my servo motors. If they are then I can breathe a big sigh of relief though if not then I’ll have a whole new (and expensive) challenge on my hands.

Having reached a stage where I dont think I will be able to identify much more on my own, I have given in and decided to cough up for the official bridgeport maintence manual. While a lot of it is self explaintory, a lot of it isnt. For example, there are limit switches with 6 wires comming from them! From my limited knowledge a switch should have a max of 3 contacts NO, NC, C and perhps a 4th being an earth. Where the other two come from I have no idea.

Similarly the axis servos have far more wires than I think they ought. A permenant magnet DC motor usually has 2 connections + and -. This on the other hand has about 8 in total; 2 x Red, 2x Black, 1 x Earth, and then a thick grey wire which looks like it has another 4 smaller coloured wires inside. Now the ovbious thought is those are signal wires for the encoder… but theyre not - thats another cable alltogether. My guess is that the double red and black wires are for multiple sets of brushes and perhaps the other wires are for an internal temp sensor. Either way its far too much guesswork so I will wait eagarly for the manual to arrive before putting power to anything.
Once I have sussed out a bit more of how the machine works I will have to make a proper plan detailing just what work needs to be done each day, otherwise I can see time escaping me and before I realise it, it’ll be the end of the summer and I’ll have got nothing done.

As always, any news will be posted on here as soon as I have a chance.

July 19th, 2006

Turns out that the heidenhain encoders on my Bridgeport Series II Interact 4 are completely incompatible with the new Rutex servo drives I was intending to use. Apparently the signal is modulated by current rather than by voltage which is where the problem lies. It looks like I may have to end up buying new encoders which is something i’d rather have avoided.

There is one glimmer of hope in that somone has offered to swap my old motors, encoders and drives for new. I was wondering ‘where is the catch’ as no doubt you are too, but he explained that because the new replacement parts are incompatible with old systems, people with old machines in need of spares in a hurry will pay good money for them. I will find out more about this tomorrow.

I’m not sure if I actually have anyone reading this blog other than myself at the moment, but in case there are, I apologise for the recent dissapearence of posts and downtime. The other day I tried out some desktop blogging software only to find that none would work. To cut a long story short i ended up having to reinstall wordpress twice and its taken me until now to fully restore the database. No doubt this wont be the last blog software related problem as I intend to change the skin/template soon to something more exciting.

On the Bridgeport side of things there have been many developments. Yesterday I moved the two electronics cabinets from my workshop up the road to my workshop at home. My car didn’t think much of it, and nor did my back … those old boxes are heavy! Today have totally gutted the boxes and taken a good look at what was inside. I will try to re-use as much as possible as there is no point wasting money on new parts unnecessarily. Anything which I don’t use will be sold, a lot of it could be very valuable to those needing urgent repairs.

The offer for the motor swap still stands but I’m not yet sure if it is the right thing to do. While shiny new brushless motors are very tempting, the extra work involved and cost of brushless drives along with a suitable powersupply may prove to be just too much work. Afterall, realistically I only have until the end of the summer holiday to get this thing working; if it drags on much longer than that i’m going to run into further financial problems.

While I eagerly await the arrival of my Bridgeport maintenance manual, I have managed to find out how more parts work just through my own investigation. I now know a lot more about the wiring for the servo motors and limits. There are two limit switches bunched together on each end of the X axis and two limit switches positioned somewhere near the middle of the Y axis. My guess is that the two limits side by side is a safety mechanism so if one fails the other will take over after an additional 5 - 10mm of movement.

I have also discovered what all the wires for the axis servos are for. Two are for a taco, another two for a temp. sensor and the other two are + and - for actually turning the motor. - All very simple once you know how!

Hopefully I’ll have more exciting news tomorrow, but in the mean time here are some photographs to wet your appetite:

The Bridgeport with electrical cabinets removed:

July 25th, 2006

….well almost.

Having spent the last few days doing boring bridgeport related paperwork (scribbling down PSU schematics off the top of my head and obtaining datasheets) I was determained to make some tangible progress. Armed with a large hacksaw and a backpack full of tools I walked up to the workshop with one sole aim - to see the beast move. With datasheets in hand I was now able to work out which wires were which, and after liberating them from the armour plated outer shell I was ready to power it up.

This was a rather nerve racking moment. Until now I’d never seen the machine operate under power as the controls were fried when I bought it. Until now the only proof I had that the machine was mechanically sound was the word of the guy who sold it to me. I was about to find out wether I had bought a money making monster or a 2 ton paper weight. As I connected the two motor wires to a 24v drill battery there was a small blue spark as the Y axis whirred into life!

I cycled the table to the full extent of travel in each direction and all was running smoothly, now I was able to see the linear slide covers and was relieved to see no rust and a liberal coating of oil. I repeated the test on the X and Z axis and found everything there was good too. The X axis did sound a little noisy but I think that was just the motor brushes and nothing more serious than that.

I was also very supprised at how fast the table moved even at 24v. Considering that once finished it’ll be running at 140v, its going to fly!.

Once I was happy with the movement of the servo motors and axis mechanics, I turned my attention to the motor encoders. I tried to remove them but my attempts were quickly thwarted by the lack of an appropriately sized spanner. Yesterday I discovered that a company called Encoders UK Ltd were able to supply me with a unit that would convert the siniousodial signals from the heidenhain encoders to 5v TTL. While at first glance this may seem like a great idea and the solution to all your problems, at £200 each I will be seeking an alternative.

My current plan is to use the tough waterproof metal housing from the old encoder and house a modern high resolution encoder within. This way I should be able to get away with using ‘budget’ encoders which are about £30 each without worrying about coolant or metal chips getting inside.

While playing arround with the Z axis, I managed to find out what some of the switches ‘up top’ were for. Things which I previously thought were to do with the power draw bar turned out to be simple limit switches; in fact for some reason one of them seems to be missing. My experimentation today also seemed to confirm the previous idea of the dual limit switches as a safety backup. They are positioned in such a way that the second one will activate about 5-10mm after the first. Unless the maintainaince manual suggests a better idea, I will wire up the first as limits and the second to e-stop ensuring that the machine wont go on running with the problem un noticed if the first limit fails.

At some stage I will add a page of all the relevant datasheets i’ve collected whilst doing this project but for now I will just post some values here. In a previous post I mentioned about all the wires comming from the servomotor and not knowing what they were for, now I do and they are as follows:

Tacho (T1, T2) - Red
Thermal Overload (K1, K2) - Red & Blue in grey sheath
Motor (A1, A2) - Black
Earth - Green/Yellow

Optional Features - Brake (B1, B2)
This can either be a DC input or may have an attached rectifier to enable it to be operated from an AC Source.

July 31st, 2006

Progress has been rather slow recently, but the latest developments are as follows:

* Bridgeport Maintaince Manual still hasnt arrived. Apparently it should have been delivered on Thursday but somthing has gone amiss
* I have come to the conclusion that using the existing spindle motor is going to be too difficult/expensive thus I plan to replace it with a 5hp 3 phase 240v motor which can then be run off a 1ph - 3ph 240v VFD.
* I have watched through every single instructional video on Mach 3. My mind is now numb, but I think I have a better understanding of how the software works.
* The encoders that I wanted to buy to replace the old ones wont physically fit on the axis motor shaft. Removing & turning down the motor shaft on each axis is somthing I’d really rather avoid so this is causing more problems.
* I still havent finalised my main PSU design, nor do I know where i’m going to be able to source the necessary components.
* Despite the mill not actually running yet, I already have ambitious plans for a 4th axis, handheld MPG and automatic tool changer.
* I’m concerned that I wont have enough pins on a parrellel port (or two!) to use all the inputs and outputs that I require.
* With the inention of purchasing components this week, and realising just how much it’s all going to cost I have reluctantly put my small CNC mill on ebay to help pay for the bridgeport retrofit.

…Thats all for now.

August 1st, 2006

Yet another day has passed and no more tangible progress has been made, yet I have been sowing seeds of knowledge which will hopefully speed up the later stages.

Todays achievements:

* Came to the conclusion that my original limit switch plan was a little over ambitious with the limited number of inputs and outputs easily available to interface with a PC. Now there will be a single pin for each axis which shares x+, x- and home. E-stop will share a pin with the secondary limitswitches on each axis which will automatically be triggered if the main limits fail for any reason.
* Still waiting for the bridgeport manual (get your act together Hardinge!)
* Phoned a company to enquire about a 5hp 3phase spindle motor. The motor specs seemed promising but they never returned my call to confirm prices and upgrade options - *sigh* another company to chase up tomorrow!
* Found out a little more about the G-Rex motion controller (www.geckodrive.com) and established that while it looks like a fantastic bit of kit, with Mach 3/4 support so limited at this time I think i’ll give it a miss.
* Did a bit more encoder research and found a company that might be able to help. www.motioncontrolproducts.co.uk have a couple of encoder options that sound promising but I will have to wait until tomorrow to find out exact details.

I really need to start putting my ideas down on paper before I start buying components but I’m reluctant to go into too much detail just yet while I still know so little about things such as the drawbar. Hopefully all will become clearer once the maintaince manual arrives.

August 2nd, 2006

This morning I got up early and walked up to the workshop armed with a small screwdriver and a digital vernier. After narrowly avoiding an attack by a heard of 30 stampeding cows I managed to get the measurements I needed and returned home. To my great excitement, the motor shaft turned out to be 10mm diameter rather than 12mm. This meant that the encoders I had in mind would fit first time without any modification.

Upon returning home I made a phone call and spoke to Gary at Motion Control Products about the Encoders that I had in mind. He confirmed that they did indeed sell something suitable - US Digital 10mm 500 line hollow shaft optical rotary quadrature encoders with the option of single ended or differential ouput in the region of £30 plus VAT.

Later on in the evening I made another phone call and had a long conversation with Tom Eldredge from Rutex He explained in great detail everything that I could possibly have needed or wanted to know about the Rutex servo drives and how to wire them up. Until now I had very little idea as to how to proceed with the electronics build. As a result of the conversation with Tom I was able to take all the notes I had frantically scribbled down and turn it into a diagram detailing how everything should go together in order to get the axes moving.

While this may not sound particularly exciting, to me this is a big milestone and a huge relief. Until now I had a very good but vague idea of what parts I needed, but I had no idea of which bits connected to what, or how many power supplies I needed.

Here is the diagram I scribbled together in a few minutes before going to work tonight. Please bear in mind that it is not 100% correct or accurate so use it at your own risk. It is more of an aide memoir than anything else. A correctly drawn detailed wiring diagram will follow in the future.



If I get a chance tomorrow I intend to start drawing up a CAD design for a mounting plate to attach the new encoders to the SEM servo motors. Over the following week I have many more tasks to do including designing the main high voltage, high current power supply and sourcing all the components, removing the old spindle motor, sourcing a new spindle motor and suitable spindle drive, and designing all the other bits which haven’t yet been properly considered.

My maintenance manual from Bridgeport still hasn’t arrived so I still don’t have a clue how the drawbar works. I have send them an email to ask what is going on so hopefully I’ll get a better explanation tomorrow.

August 4th, 2006

Such an eventful moment it really needed its own heading. Finally at long last the manual has arrived. Hopefully now I will be able to piece together the missing bits of the jigsaw to understand how the machine works.

In other news, unfortunately my UK supplier of optical encoders didn’t have enough 500 lines in stock so I’m having to order direct from US Digital instead. Having said that, with the exchange rate being so favorable at the moment its not such a bad thing.

Hopefully this weekend I will get a chance to remove the old spindle motor and design a mounting plate for the new one.

August 19th, 2006

Despite the lack of updates I have been hard at work. With the exception of just a few things, I have now sussed out what all the sensors do, where all the wires go, and what voltage everything operates at.

Motion Control Products Ltd didnt have enough of the encoders I needed so I ended up purchasing direct from US Digital instead. I purchased 3 x E5 type encoders which are TTL quadrature, have a differential output and have 500lines of resolution. I also purchased a single 300line E4P encoder to use as a MPG for the handwheel control. In total the order came to about £160.

I have also put in a big order with Rutex for 3 servo drives, a motherboard, chassis and all the other relevant components which i’m hoping should arrive in the next few days.

I still need to order a signifcant quantity of parts from CNC4PC.com but they have been uncontactable for quite a while now so i’m guessing that they are on holiday at the moment.

Most of my efforts recently have been concentrated on design work. Using a simple 2D CAD package, I’m in the process off drawing a highly detailed & complete diagram of the entire electrical system. The theory is to perfect everything on paper first to save a lot of time when it comes to the build.

Once I’ve finished the general electronic wiring diagram, I will draw up another diagram specifically for the main power supply in the system. This is such an important part I feel it justifies the attention of its own individual diagram. Once I have that sussed I can order the components for that too.

The saga of what to do with the spindle motor is perhaps the single biggest challenge right now. I have so many options and all of them are very expensive and have numerous drawbacks. Options include:

1: Keep the current 400v series wound DC motor. This is all well and good from a mechannical perspective, but generating 400vdc from a 240v single phase supply and then getting vairable speed control on top would be rediciously expensive.

2: Re-wind the current motor to 200vdc. This has the huge advantage of being able to bolt it straight back in afterwards without having to machine up any new mounting plates and for this reason it is my preferred method at the moment. Dissadvantages include the price of re-winding and the possibility that it might not even be able to be rewound to a different voltage at all.

3: Replace the motor with a new 240v 3 phase motor then run through a VFD. This sounds simple enough but has some hidden problems. For a start i would have to machine up some special mounting plates to make it fit in place of the old motor, and in addition to that I would also need to find a way of attaching the pulley belt and spindle brake. To further confuse matters I’ve heard a lot about having to de-rate a 3phase VFD running on single phase. If I understand this correctly then a 4kw motor would need about 7.5kw power input to the VFD. Seing as I only have 7kw to play with for the entire machine, going with this option would mean I’d have to get a seriously underpowered motor arround only 2hp.

4: Replace the motor with a single phase ac motor. This will have the same mechanical problems as the 3 phase motor but avoids the de-rating issue. Though I think there are also issues with speed control for single phase motors.

5: Replace the motor with a 200vdc motor from a smaller bridgeport machine. Hopefully this would be the same size and shape elimiting the need for mechanical modifications. The only real issue with this option is availability. I might not be able to get one even if I have the money.

My plan is to have ALL the design work finished by the end of the month. I’,m also reluctantly selling my small hobby cnc mill to pay for the rest of the parts needed to complete the retrofit.

August 22nd, 2006

Today I called up Forco Electrical Services Ltd and had a long discussion about solving my spindle motor problem. Initially I tried to find out about re-winding the old motor to a lower voltage but after explaining things in a bit more detail, they suggested a different solution which involved keeping the current motor as is, and using a complex arrangement of 2-phase step-up transformers along with some form of VFD.

Hopefully in the next few days I will recieve an email from them explaining their solution in more detail, along with some prices.

In addition to motors, I discovered that Forco also sell transformers of almost any size and type for very reasonable prices so I’ll probably end up purchasing my main Axis PSU transformer from there too.

Once again, I realise updates have been in short supply but I have been exceedingly busy. On top of that I have finally hit upon the realisation that the blogging platform just wasnt for me.

I want this to be a community site, and the traditional blog format just doesnt fit in with that model. On top of that I just found the interface incredibly fiddly and irritating which put me off posting at the best of times.

In the past few weeks there have been many developments so I'll try to cover as many as possible briefly here:

Firstly, I have ordered the encoders for my motors. I purchased 3 x US Digital 'E5 Type' 500 Line Differential Quadrature TTL Encoders. I Ordered these direct from www.usdigital.com and after shipping, duty, cables and connectors they work out at about £40 Each. I Also purchased a single ended 'E4 Type' 300 Line TTL Encoder which I plan to use for an MPG/Manual Pulse Generator/Handwheel Control



I have also purchased the Rutex Servo drives. This included

3 x R2020 200v/40A Perm Magnet DC Servo Drives
3 x R2211 Differential Encoder interfaces
1 x R2110p 4 Axis Motherboard
1 x Black Annodised Aluminium Chassis

All this cost just under £460 inclding shipping, but those great people at UK Customs added another £100 to that price



The Rutex chassis is designed to hold two 60mm cooling fans and the motherboard has a 24v screw terminal to accomodate this. Many companys sell such things but by far the cheapest I could find was Ebuyer which sells 12v 60mm fans for just over £1. Two of these wired in series and youre sorted.

I also bought two parrallel cables at the same time.

The Rutex motherboard has a DB15 connector to connect things such as limit switches, e-stop and a charge pump. Rather than soldering all that to a DB15 Connector, I designed a PCB breakout board to do the job. I havent built this yet, though I will probably end up building it on veroboard to save time/money




Ovbiously the new encoders were not going to fit directly in place of the old ones so an adaptor had to be made up. My plan was to machine up a new end plate for the motor, fix the encoder to that, and then use the old encoder housing to keep everything sealed. I could have probably done this myself but I needed it to be perfectly accurate and the old motor cover was proving very difficult to measure, especially when the datasheet from SEM didnt seem to match up with what I had in front of me.

With this in mind and the ovbious lack of time available, I spoke to Martin over at MS Precision Engineering in Paignton, Devon. They have done work for me in the past which has always been of a very high standard and fairly priced too. This was no exception and last friday I was able to collect my new encoder mounts which are very shiney indeed. It cost me £80 for the three identical components.




After speaking with a few other people about the proximity sensor on the power drawbar unit, I think I finally understand how it should work. It has a + and - supply and a signal/switch wire ... basically just like a transistor. When it detects metal in front of it, the signal wire goes high or low (not sure which yet). The sensor operates at 12v so I will need a 1.5Mohm resistor to reduce this to 5v to match up with the 5v PC parrallel port.

In order to progress further with the motor, I removed it from the machine and took it up to Forco for them to have a look at. This was a tremendous challenge in itself due to the sheer size and weight of the thing. It weighs more than I do and the only way to remove it was with a very large agricultural forklift.




A few days later they got back to me with some more info. The brushes were worn away to nothing and needed to be replaced. Also the brushes had never been replaced in the life of the machine and all the useage over time had worn away the commutator in the motor. I have been told that the brushes need replacing with new, the commutator need skimming on a lathe and the bearings may have to be replaced too. All this is likley to cost in excess of £250 before you even start thinking about controllers.

On the plus side, if it'll last another 20 years like that then its not a bad investment, and as it would cost £4k for a new motor, repair seems the most cost effective option. We have already ruled out modern 3ph motors as being too expensive after all the extras such as cooling fans and spindle brakes have been taken into consideration.

There is even a possibility that the old motor will run ok with a 240v armature voltage but we are waiting for data from contraves to confirm this.

The only other unforseen potential problem is current draw. Mark seems to be quite certain that this motor will draw vast ammounts of current. Over double my estimates. Because of this I really need to go and check how much power will be availalble to me at the site where my workshop is located.

The main reason I have not updated this site for so long is because I have been busy with 'Paperwork' I must have spent about 60 hours drawing up the wiring diagram in a 2D Cad Package. It seemed to be a never ending task but my theory is that now it's completed it'll save me a lot of time and money in the long run.




Windows Meta File
DXF

Now the wiring diagram is complete I need to write up a parts list and get all the remaining components ordered. I also need to design the 4kv PSU to run the axis drives. I have already sketched a rough design on paper but this needs to be done properly.

I think thats about it for now. At least with the new forum I have no excuse not to make more regular updates on the progress and with only 3 weeks left to get the machine running before I go back to uni, each day is going to be a hive of activity.

Well you know what they say..,"The best laid plans" and all.

Sadly the summer project didnt quite work out. Simply put it has turned in to a far bigger project than I could have anticipated with a very steep learning curve requring vast sums of money, time and knowledge.

Ok, so i'd be lying if I claimed not to have known this from the start, but I was hoping that I might get it finished for the summer.

Not to worry!

It will now be finished in the month I have home at christmas, and in the mean time I have been making more progress with the stuff I am able to do whilst at uni.

Before I left I made sure I left the mill in the best state I could. I topped up the oil tank, manually brushed way oil over every exposed metal moving part, and spread liberal ammounts of thick grease on other areas which could possibly rust in the winter damp conditions.

The spindle motor is still in exeter with the specalists Forco.

A week after arriving at uni, all the CNC4PC parts arrived from the US

While at uni I have finalised the design for the powersupply. I'm now going to have four separate 1Kva units enclosed in a single aluminium box.

I have ordered enough components to build one supply for testing purposes. Once I have assembled it in rough form to ensure it doesnt blow up on me, I will order the duplicate components for the other supplies.

With measurements for all the components I can CAD up a box to house it all in and then build/assemble it all in the uni workshops. This will include necessary heatsinking for the bridge rectifiers.

I've had an idea for monitoring/controlling over-temp conditions in the servo/spindle motors. While it will be a nice thing to have, I will probably leave it as a later addition as it is not crucial to getting the machine running.

Soon I will also order all the other electronic components required for the build.

Still to do:

Find out cable size/current requirements for powering the machine as a whole and get prices on a 50m length of cable.

Measure the exact cable length from fuse box to machine (if possible make use of existing setup)

Design in CAD, stainless box section housing for computer system.

Design control panel for computer

and many more things.


For the time being my priority is the powersupply. If I can get that finished before breaking up for the christmas holiday it'll save me some much needed time.

Happy new year to you all. My new years resolution for 2007 is to finish the bridgeport retrofit ...... no, seriously!

I know updates on my progress have been a little slow but this isnt because I havent been making progress, more so because I've just been so busy working on the mill itself.

I will *try* to do regular updates this year, at least 2 - 3 times a week as it only makes it harder when I try to cover a months progress in one post.

Here are a few photos and diagrams of the progression since my last post. I've tried to include a lot of photos so you dont have to read a short novel..



While I was at uni I did a rough design of a linear powersupply to run my machine. In the end I decided on making 4x 1kw supplies. One for each axis.



I went ahead and built a prototype based upon my initial design. With help from Reading Uni EE/Cyb department I was able to test and refine my design.



To give you an idea of scale, this is me holding one of the capacitors in my hand. There will be 12 of these in total.



Based upon the successful testing of my prototype, I went on to produce a refined design.



With all the main additional PSU components delivered from Farnell & Rapid Electronics, I was able to position the components on a board to make them fit in the smallest space possible whilst still allowing for easy maintainence.



It occoured to me that some things such as the e-stops, redundant limit switches and thermal overload switches exceeded the number of inputs I had on the computer.

All the switches are normally closed so the crude solution would have been to wire them all in series and use one input for each type. This would work, but means if there was a fault it would be very difficult to trace.

To solve this problem I went about designing a diagnostic board which would light an LED for each input that works correctly. My initial ideas were protyped in an electronics simulator program.



My machine has two sets of limit switches on each axis; the main limits used in every day use, and the backups positioned about 5mm behind the first. The idea being that if the main limits fail, the second limits will cut in before the machine can crash.

It would have been nicer to use logic gates rather than relays, but with time against me, relay logic was the quicker option.



Each axis motor and the spindle has a built in normally closed temperature cut off switch. This board allows all the switches to be routed through here before reaching the pc. If a motor overheats, the pc will shut down the entire machine and this circuit board will extinguish an LED to represent the failed motor.



Same princible here for the e-stops. This board allows 4 low voltage e-stop buttons to be connected. If any of the buttons are pressed it will send a signal to the computer to shut down the machine in a controlled manor. As an additional backup in case of computer faliure (of course computers never fail! ) the second relay output poweres up a time-delayed relay which after about 1.5 seconds, fires a big contactor which removes all power from the machine at source.

For additional saftey everything is normally closed so if anything were to fail it would shut down the machine rather than to continue un-noticed without the safeguards in place.



Components for the e-stop controller positioned on some stripboard ready for soldering. I wish the mill was running so i could machine my own PCBs!



The original bridgeport machine cabinet, emptied and ready for new components to be added. I think I'll give it a new lick of paint before putting it all back together.



Lots more shiny new components from farnell. Amongst other things are the 200w braking resistors for each axis, a 100amp digital electricity meter, three time delayed relays, five standard electro-mechanical relays, main power switch, 24v PSU, 50v lighting transformer, 120v suds/oil pump transformer, 2 x parrallel cables, 4 x shielded CAT5 encoder cables, and about 100metres of various wire.



Even more shiny components. This is all the stuff ordered from CNC4PC in the USA. Photographed are 4 x solid state DC relays, 4 x solid state AC relays, 3 x Encoder interface cards, a 0-10v analouge speed control board, an opto-isolated parrallel port break out board, a saftey charge pump (shuts down the mill when mach3 doesnt have control of the system), an index pulse board (optical taco) and a 12/5v PSU.




This is the stainless steel base I will use to mount the PSU. Originally I was going to use aluminium but it worked out much cheaper to use thin stainless welded to a box section frame than a single thick piece of aluminium.


I think thats enough for now! I will try to do more regular updates in future.
Targets for the next few days are to finish building the multi-input controller boards, assemble the psu components onto the stainless steel base and collect the spindle motor + controller parts from forco.

Today one of my housemates from uni announces that over the Christmas holidays he bought a PCB etching tank complete with chemicals. He showed me a few boards that he'd made up for an autonomous sumo robot and the results were very profesional indeed; certainly much nicer than anything you could do with stripboard.

I'm aware that there are various methods to etch PCBs but this one certainly seems easy.... Design the board in PCB design software, print the image onto gloss paper with a laser printer, iron the printed image onto a copper clad board to transfer the toner and put the copper board into the etchant. Anywhere where the toner is touching the copper remains, and everything else dissolves.

All I need to do now is learn how to use Eagle and i'll be away!

I'm really quite pleased that this oppertunity has presented itself as I was really struggling before. Trying to assmeble those circuits on non printed boards was turning into a bit of a nightmare.

Whilst doing a piece of coursework for uni I became aware of the importance of having a substantial plan in place as a reference guide to work to.

While so far I have lots of small plans and diagrams, a single overall plan to define this project as a whole is somewhat lacking. Without such a plan I'm never going to be able to progress at a decent rate nor will I even know when i've finished, if I havent first defined just that finished means.

Here i'm going to explore the grand plan in a little more detail; the main goal, and then specify the sub goals required to reach it. There is so much work involed in this project that there is too much information to keep and process in my head alone.

As a result it is easy to forget things that need to be done and progress slows as a result. What follows is more of a personal to do list/organisational tool but i'm posting it here so I don't loose it and so it's in the same place as all my other bridgeport related info.

It probably doesnt provide much value to anyone reading this but that is not the intention.

Main Goal: To get the machine running, put back together in one piece, in new electronic and mechanical condition, complete with integrated computer control system and a handwheel MPG control.

An additional goal for the future would be to add a 4th axis. This consideration has been acknowledged from the start but I will not concern myself with this as a main goal. The new electronic system has already been build with spare capacity to support a 4th axis, but I will consider the machine finished long before a get round to installing one.

Another goal is to build up this website to provide content rich information for other CNC enthsiasts. This really is more of a separate project than a contributing goal to the mechanical rebuild so I will not go into too much detail here. Suffice to say i will try to put in an hour of work each day but it is not a main priority.

Ultimatly once the main goal has been satisfied the new main goal will be to use this new asset I have to start a small engineering business. At that time I will need to give further consideration to customers & suppliers, employees and premesis. I hope that this time will come sooner or later but for now it is not a top priority. A CNC machining business without a CNC machine isnt going to go very far!.

What still needs doing:

• Print the 4 PCB designs by laser onto photo paper

• Etch the 4 PCB designs onto copper board

• Drill out the copper boards

• Solder in all the components to the copper boards

• Design/Build round terminal connectors to make internal connections in the power supply.

• Buy fuse holders and resistors for the power supply

• Build a mounting plate to hold the fuses, and diagnostic LED's on the power supply

• Drill mounting holes in PSU chassis

• Bolt all the components of the power supply to the stainless steel chassis.

• Drill/Weld new mounting holes for the powersupply in the electronics cabinet

• Reduce the size of the steel sheet that holds everything else in the electronics cabinet.

• Buy some grey spray paint and re-spray the inside of the electronics cabinet.

• Get someone to build an aluminium enclosure for the braking resistors

• Colllect the spindle motor, and new controller components from forco.

• Check to see if the spindle controller components will fit in the same housing as the servo components.

• Update the main diagram to reflect the most recent changes

• Assemble all the component modules onto one metal board to be inserted into the main cabinet

• Make all the internal connections between the component modules

• Pressurewash & Steamclean the machine

• Remove the way covers, sand off rust, polish, grease and replace.

• Oil and grease all exposed surfaces and repaint areas where necessary. (Including servo covers and front splashguard etc)

• Put spindle motor back into machine, secure and retention drive belt.

• Attach the new encoders to the servo motors and wire up to the existing armoured cable.

• Re-attach the electrical cabinets to the machine.

• Re-attach auto-lubrication system and check for blockages at all the outlets.

• Extend all the wires to the sero motors through the armoured conduit so that they reach the appropriate place in the electronics cabinet.

• Make all the connections between the external systems on the machine and the internal connections in the electrical cabinet.

• Produce a wiring diagram to explain what colour does what

• Produce a fault finding diagram/flow chart indication likley causes of failure, how to test and possible solutions

• Speak to electrician about hooking up machine to electrical supply

• Build the PC controller and install operating system to the solid state hard disk.

• Test all the limits and saftey lockouts

• Connect up compressor.

• Test the power drawbar mechanism

• Run SPI tuning on each axis

• Install and configure mach 3

• Design the interface control panel for the computer control unit

• Design the main housing for the computer control unit

• Build the housing for the computer control unit

• Design MPG

• Build MPG

Despite a seemingly never ending workload I'm confident that good progress is still being made. I am spending at least 3 days a week at home just to work on the machine which is speeding things up considerably.



In return for some CNC milling in the future, my housemate very kindly produced some etchable circuit diagrams for the various interface boards. Hopefully he's going to get them etched for me this week leaving me to drill them out and solder in the components



At home this weekend I took a look at the main electronics cabinet, removed the door and ground away some of the internal raised sections to make room for the new PSU

I also found out that a friend of mine now has access to a full profesional paintshop facility. With a bit of luck I'll be able to get all the removable metal parts re-sprayed without breaking the bank.



The axis PSU is starting to take shape and I manged to mark, drill and tap all the holes before securing the components to the stainless steel base. I started to crimp the wires in place but I ran out of time so that'll have to be finished off next weekend.

It also doesnt look like the fuses and diagnostic LEDs will fit on this piece of stainless so the will have to be built onto a separate breakout board.



These are the braking resistors for the axis servos which prevent back EMF destroying the drives. I've been told that these will be very noisy and create a lot of EMI so I am having a metal box made up to shield them from everything else.

I also got a chance to speak to the company dealing with my spindle motor issues. They now have my transformers/controllers in stock and have adapted a system to make my spindle blower run off single phase. It also looks like the DC speed controller has a taco input so I can run my spindle with closed loop control.

After a long discussion they managed to convince me of how badly my spindle motor needed repair (striping, cleaning, skim/undercut commutator and brush replacement. This means it wont be ready for another two weeks and I had to part with even more money but in the long run its probably money well spent.

Next weekend I'm going to:

#Collect the aluminium box for the braking resistors

# Get the elec. cabinet internal steel backing sheet plasma cut to fit

#Drill/tap/weld mounting holes/brackets in the main cabinet for the axis PSU

#Do the same for the internal steel backing sheet

#Cut out any other holes for plug sockets, connectors and switches

#Sand/Steam Clean all the metal panels and take them to be painted

#Wire up the rest of the axis PSU

#Start building the PSU diagnostics break out board.

By then I should have also finished making all the other break out boards.

A few more updates for you...



Last weekend I finished off the power supply module fixing all the wires in place and including the time delayed starts.



Another view of the power supply showing the size of it.



Finally got round to making the breakout boards for the various limit and thermal overload and E-stop inputs.



Photo of the underside of the PCBs. It looks a little messy as it was done with iron-on rather than photo resist. The dark tinge is residue from the sanding process to 'finish' the tracks after etching.

There is one more custom printed break out board to complete which will hook up to the DB15 output from the Rutex motherboard. Due to the very thin tracks associated with the DB15 connector, the etching process was not perfect leaving some of the tracks broken.

I will try and strengthen the damaged tracks with wire and solder but if possible I'll get this board etched properly with a UV setup.

The one other board I need to make which wont be printed is to go from the output of ther servo power supply. This will connect to the supply and will contain diagnostic LED's, fuses and a connection point to the digital servo drivers themselves.