TOS FN20 Upgrade to CNC

[Ark1]

About 18 months ago I decided to upgrade my Tos FN20 milling machine to CNC. I have watched others upgrade their machines and always wondered why does it take so long. Well in short order it does, there is a lot to consider,buy and actually do. This is my first exposure to CNC although being in the electrical trade I have seen machines working I have never been the person working on the drives, electronics or programing. All of this is new to me.

The Tos FN20 weighs about 2000 pounds and I am the second owner. The machine new in 1980 was purchased for $27,000 and was used in a small machine shop to produce a few parts. The job was a fillin for apprentices. The machine was used for about 5 years and then sat for 7 more years when it went up for sale and I purchased it for $1750. The machine has power drive on all 3 axis and has a 3 phase cooling pump motor. The spindle is 2.5 hp the axis drive is 1.5 HP and the coolant pump is 1/6th of a HP.

One the considerations was to find a used CNC machine but I soon realized that I would have to upgrade the computer and motors and drives. That a used CNC machine that I could afford would most likely need a lot of work to bring into tolerance. The machine I have is very versitile as the table comes off of the machine and a small engine block can be mounted directly to the T slot face of the machine.

When purchased a large dividing head with chuck and a 5" vise came with the machine. Working tolerances ( lash ) with this machine were good on the Y and Z axis of less than .003" the X would need work as the lead screw was producing a lash of .020" with the backlash nut turned in. The screw was worn in the middle working position and when driven to the ends of the lead screw binding would occur.

The first part of this build is the tear down and repair. Although not having anything to do with CNC a good foundation and what is required must be addressed in any conversion.

The Tos brand was not imported into the USA because of the states ban on importation of tools from communist countries. Canada did not have this import restriction.

I will start with some tear down pictures.

[Ark1]

Taking the machine apart was a challenge. When I purchased it I got the electrical EWD and CWD prints ( Elementry Wiring Diagram and Connection Wiring Diagrams) but the shop had lost the manual. I have looked on the internet but a machine of this vintage was before the internet. I did find 2 new machines with DRO's never used in the Europe in my seach.

A lot of the machine is put togehter with dowel pins and hidden screws. The machine was also painted with a brush by somebody and that covered alot too. My dad retired and at 83 would spend an hour or so each day scraping and cleaning the parts that I took off the night before. The Gibbs and ways were good, covered in gunk and sticky old oil but there were no scratches or rust spots to contend with. The table that comes off weighs about 70 pounds and the traveling Tee slot face is over 100 pounds. Each sub assembly's small parts was put in a plastic bucket and I used old paper coffee cups to hold the bolts, washers and bearings and keep them in order. I wrote on the inside of the cup what part of the machine the parts came from. I took lots of pictures as I went, and used 2 gallons of varsol and paper towels to clean the old oil and dirt from the mill. I have a wood stove that is brick lined and that is where the old oily rags and paper towels ended up.

[Ark1]

The task of cleaning the machine just takes a lot of time. I used 4 spray cans of paint stripper to remove the old paint that was applied with a brush. Then there was the filling Bog that was flaking off in some places, it was over 1/8" thick, what would pry off ended up on the floor.

With the X lead screw removed, I took measurements and made a drawing of the ball screw I needed to replace the worn Acme lead screw. The information was sent to Zap Automation and Gary Higgins sent me a quote for 25MM x 5MM pitch C5 ballscrew with a Zero lash ball nut. I was a little leary because the machine had a 1" screw with .200" lead. and my scales on the machine are impearial. For that reason I asked a few questions and was told that Mach3 can make calculations with a mix of both metric and impearial screws. I sure hope that is correct. Also I purchased a Rodgers DRO board so that I could use the DRO instead of the scales on the machine if I wanted to machine a part still by hand. Being new to CNC I wanted to keep my options open just incase I have to do a milling operation by hand. The Quote for the new ball screw worked out to $650 with shipping to Canada, I had to pay another $75 sales tax when it arrived. The main issue was that the ball nut would have to fit into the same mounting location as the Lead screw nut. I was prepared to machine an adapter to mount the ballscrew nut in it if it was required.

Here are some more pictures.

Cleaning the sump is a terrible job the coolant goes rancid and gooey. I purchase 5 gallons of Circle Cut 735 and will hope the days of smelly sumps are over.

The small parts were sand blasted. In restorations I perfer to sand blast, but each part must come off the machine and be cleaned of sand afterward. I would have loved to take the rest of the machine out behind the barn and sand blasted it. The problem was that this machine has 3 gear boxes for the drive axis full of oil and bearings and gears. To do this right I would have had to take this machine down to the last nut and bolt. Looking at the condition of the machine I really didn't think it needed it and it would have taken me another year to do it.

I used epoxy 2 part primer as the base coat on all bear metal and on existing paint that was not removed to seal it.

Ark1

TonyK.

[Ark1]

As I worked along I finally came to the end of the cleaning. Primer was next on all parts and then the single stage Ureathane from Pro Spray. Single stage automotive paints have been phased out in Canada. Industial colours still remain, custom colors are a bit difficult to get. Most car paints now are water base coat with clear Ureathane top coat.

As Much as I looked on the internet I was not able to locate a manual for this machine. I do have orginal electrical drawings but no exploded views of any of the parts of the machine to take it apart or even look for bearings.


Here are a few more progress pictures.

Ark1

TonyK.

[Ark1]

As I stated the Acme Lead screw was worn and lash was about .012". The screw on this machine was 1" X .200". After looking on the Net and considering pricing I emailed Gary Higgins of Zapp Automation and provided him with a drawing for a ball screw replacement and ball nut. The ball nut had to be small enough to fit inside the lead screw mount as I didn't want to remove this section of the machine and have it sent out for machining work. Gary suggested a C5 ball screw and zero lash nut of 25 mm x 5 mm although close the imperial scales would no longer be accurate on this machine. The ruler would be, but the hand wheel with .200" graduations would not. This is where the encoder would have to take over with the DRO in Mach 3 to resolve this difference.

It took about a month between the time I ordered the new ball screw and it was delivered to Canada from the UK. With this in hand I took measurements and proceeded to turn an adapter to house the new ball nut and mount in the Tos FN20 X axis. I turned the adapter on my lathe and changed the gear head reduction drive to metric. A 3" section of 1008 carbon steel was used to turn the adapter. The adapter was screwed into the housing of the Mill and cinched with the set screw the Acme lead screw nut was held with.

While looking at the new ball screw I realized that the nut flange was on the wrong end of the screw. I did specifiy this in my order but a quick email to Gary Higgins of Zapp Automation provided me with information on how to reverse it. I had to turn a sleeve that was less than the minor diameter of the ball screw and unscrew the nut on to the sleeve to prevent the balls from dropping on the floor. See pictures.

The new ball screw and nut was slid into the machine and I used a centre punch to mark and drill then tap each hole of the ball nut to fasten to the new adapter. The ends of the X axis with bearing to hold the ball nut were installed. ( just to let you know this work was done around May of 2010 )

I was also ordering encoders, power supply transformer and a control cabinet at this time. With this project there are a lot of things that are required to make a functional machine and at times it takes parts over a month to come in. So having more than 1 part of the project on the go isn't a bad thing, but a very good use of time.

Ark1

TonyK.

[Ark1]

By June of 2010 I had the mill back together. The design of the machine allowed me to still use it with the power feed on the Y and Z axis but the new ball screw could not be used with the machines power feed as the drive coupling was a keyway cut the entire length of the X lead screw. ( The balls in the ball screw wouldn't like the keyway) Over the Winter I purchased from Ebay the 3 DC servo motors from a company in Toronto for $150 each. The Motors were 72 volts DC. I followed the suggestion of TotallyRC of this site to use DC servo motors and the Gecko 320X Servo drives. Because I am new this I really didn't have a starting point between steppers or servos nor any preference to either one. He also recomended the AMT 102 Encoders from Didgi Key for $33 each and I ordered them as well.

While the Mill was not working I searched the Internet for TOS FN20 milling machines to see what else was being used for Chip and coolant containment. I purchased a sheet of Stainless Steel and made the enclosure around the TEE Slot Table. A poly carbonate front window is also used.

The Mill then sat for 2 months as I ordered more material for the electrical portion of the installation. I have 3 Acres of land and the summer months require more out door maintenace and projects like these go on the back burner.

I purchased a used control cabinet 12" deep x 24" wide x 36" tall with a back board and also at the time a control wiring harness from a used machine with 100 terminal blocks for $100. If you are in the Hamilton Area, Electrical Machine Corp. ( EMC ) on Brampton Street sells used electrical equipment for a fraction of the cost of new. I cleaned the cabinet and spray painted it the same color as the milling machine. The enclosure has a gasketed door seal with 2 turn screw latches and a piano hinge door.

The computer I will be using is an old 1MHZ Pentum PC that is about 6 years old. I cut the frame up to fit inside the enclosure and mounted it to the back board. All of the electrical components would be mounted to the back board, wired then tested before it would be bolted into the enclosure in the shop.
The finger wire way was purchased locally from an electrical supply house.

The enclosure would be kept presurized by 2 5" 12 volt fans that draws air through a furnace filter. My shop is multipurpose and to leave a computer sitting in there will only shorten it's life in the dust and dirt. The wiring of the back board was done in my basement out of the dust and dirt of my shop as well. I also purchased a Rogers DRO board that allows Mach3 to use the encoders mounted on the DC servos to show machine position if I want to still run the Mill in manual mode. To use this board a 2 port parellel port card was ordered and installed in the computer when it was on the bench.

The power supply was purchased from Antek. It is a 77 Volt DC power supply with 2 secondary windings for XYZ and ABC axis. It also has a 12 volt DC power supply for the vent fans and a 5 volt DC supply in case I need to run anything else. At 1500 Watts 8.5 amps per secondary is available. The power supply is neat and uncluttered with 6 15 amp secondary fuses.

When I ordered the Ball Screw I had the shaft lengthened by 1 inch to allow a mounting point for the toothed belt pulley. A hand and hold test was done and measurement taken to make adapter plates to mount the servo motors to the mill at each hand wheel location. I used the mill in hand mode to make the adapter plates. It is at this time I realized that the back splash was too high and had to remove it.

The motor mounting adapters were made from 1/2" x 6" aluminum flat stock, drilled, tapped and pined. I can weld Aluminium but was afraid that some warping may happen so I used the drill and bolt method for assembling the mounting plates.

Still more to come when I get the time.


Ark1

TonyK.

[Ark1]

One of the difficulties I had was that all of the pullies that I looked at had very small bore diameters. I did find a company in the USA that would make a pulley to my spec's from pulley stock, but the price was getting plain stupid.

I found these pullies from Motion Teck in Toronto. 20 tooth and a 60 tooth. My servo motors all have 1/2" shaft and the hand wheel shafts variy from 17 MM to 20 MM. I have attached the pictures of how I did the bore modifications by first removing one of the pressed on sides, increasing the bore diameter and then pressing on the side again with a C5 1" collet. The large pulley was chucked in the 4 jaw and dialed in. I also machined off the Hub. The set screw was relocated in one of the pulley teeth lands.

I am a Certified Electrician and started my apprenticeship with Ontario Hydro at the Nanticoke Generating Station. Back then I learned relay logic, then DTL ( Diode to transistor Logic, then TTL, Transistor to Transistor Logic and then dabled with some of the first PLC's. ( Apart from the PLC, the other types of Logic controls were all hard wire programed. Changing a program meant moving wires sometimes on soldered terminals.) A Generating Station is control wiring and lots of it, followed by trouble shooting and stacks of blue prints that at times seemed impossible to follow. CNC is new to me as well as G Code. Time as brought about smaller componets that do more for less money. What I have found is that each manufactor of Breakout boards, drives, power supplies and electronic relays all make a product that explains what they do, but not how they really go together. Even the software at times leaves gaps in the manual and what is happening in front of my eyes. The machine is now running, but getting there, well there is more to it than running some wires to a few terminals and standing back watching a machine do what I would like it to do.

The following will be from where I started doing the electrical portion of this project, with the mishaps, mistakes and unknown I faced. When I was an apprentice there was a saying, " At times it is the blind leading the Stupid" I was at times both of these and the surprises came just when I thought I was there only to realize moments latter I really wasn't. The method of how I wired my machine is based on the Ontario Hydro Standard that I was taught when I was an apprentice and then became a journeyman electrician. I am sure other's have differnt ways of doing wiring, but I am blinded as the way I build a control cabinet is the way I was taught back in the late 70's.

So first off every control componet must be in a steel box. The box keeps all of the parts out of the dirt of the enviroment. It keeps Live parts away from your fingers and it contains a fault or fire should one occur. When it comes to enclosure's bigger is normally better but before I could purchase one, I layed out all of the componets that I needed in the Box on the floor and obtained a rough measurement of the size I required. Everything in the box is mounted to a back board and if you need to get to both sides of the backboard it should be hinged to allow you to obtain access always from the front of the cabinet. Trouble shooting from 2 sides of a sheet of metal, will only lead to frustration.

So in the first Wiring picture my componets are laid out and fastened to the back board.

Connections for all of the motors, encoders and control external to the cabinet use military Amphenol Connectors. This way a motor can be removed for repair or swapped with another motor to test the encoder, motor or drive. If the machine needs to be moved 5 minutes of twisting connectors saves hours of wiring and rewiring with the chance of making a wiring error. Purchased out of China for $100 is cheap compared to the agravation of trouble shooting and terminating wires to move a machine. Pin #1 on all connectors is ground.

To hold the connector a C5 collet was gently chinched in the vise for soldering.

Electrical wire can be purchased in 100 foot coils from Nedco in different sizes and colors. Also used Electrical supply houses will give you a wire harness with good quality wire for a few dollars.

Start from the beginning with a drawing and write everything down. Even if it is on a blank page use the notes to make a large master drawing. Expect to modify what you have wired at some time in the future. All connections are done on a terminal strip, there are no twisted wire connectors or wire nuts.

At all times keep it neat, there is no fun trying to tug wires in a birds nest of conductors. Buy a label maker. I found this one on ebay for $49 and 3000 labels for $29. On the down side the battery pack failed. Well those garden solar lights when they turn into a dud, remove the batteries as they are good NiCads that in this case I soldered together to make an external battery pack.

I am using an old Pentium 1MHz computer that I cut down to place in the control cabinet. When I went to turn it on, it would not boot. I removed every drive until I only had the hard drive connected, I realize then that all the drives were connected as slaves without a master, hence no boot. That took a few hours to sort out.

The limit switches for my mill are simple micro switches from ebay. They work very well and are simple.

I also installed 2 5" 12VDC fans that draw air through a furnace filter to keep the cabinet pressurized with clean air. A vent on each side was added to permit air flow from the middle and top of the cabinet.

A sliding door was cut in the cabinet door with my plazma cutter, notice that paint does not even get burned with this tool. It is very fast as well. The door will be infront of the computer and allow me access to the USB port for loading files and to also push the on button for the computer. The window is made from 3/16" Polycarbonate sheet.

I did all of the testing for the back board in the basement before it was mounted in the shop. I am so glad I did as there were some problems that I will cover latter in the next post.

Ark1

TonyK.

[Ark1]

There are times that I shake my head and ask why am I doing any of this and until now things were progressing smoothly. The Sound Logic Break Out Board manual gives the instruction to power up the board and test the related pins for voltage. I followed the instructions and all was well.

Mach3 requires an XML file to configure the ports and pins to the Break Out Board and this is where the trouble started. I could not get Mach3 to communicate with the Break Out Board. The manual suggests that the most common cause is the 25 pin cable not being a straight through cable. I rang the cable out pin for pin and found it to be a straight through cable. The Mach3 manual then states that before proceeding this must be resolved. No @$!*&.

I embarked on ringing out 3 more cables, running tests with XP and got no where. I emailed screen shots to both James Cullens and Bob Campbell of sound logic. I also emailed pictures of the Break Out Board and the LED's Lit.

First reply was that the Mode jumper was on the wrong set of pins. The manual shows this to be in a latter step that I had not gotten to. No big deal, just shift the jumper. Still no communication between Mach3 and the BOB. Both James Cullens and Bob Campbell were emailing me at 10 PM trying to help me get through this problem. Fantasic service, I wonder if these guys sleep at all. I tried every suggestion they gave me and still did not get the #4 LED to light. I had the opposite, LED's 1,2,&3 were lit, the step and direction LED's were Lit ( Run Away ) I was going no where. Finally I ran a LAN cable from my router to my PC mounted on the back board, dumped Mach3 and reloaded from the InterNet. Installed the XML file provided and success! 16 hours of trouble shooting and guessing were over. Some how a file was corrupted that I had loaded on to a flash drive to transfer Mach3 between my home PC and the Mill PC. At least this was resolved in the warmth of my basement and not my cold shop. I now could get Mach3 to reset, and manually turn on the spindel relay and coolant pump relay.

Before installing the back board I ran a 30 AMP 240 Volt supply from my main house panel to a small 8 Circuit load centre beside the control cabinet.

My Son helped me carry the back board from the basement and install it into the cabinet. The connection process of terminating the wires from the bulk head connectors was started.

The spindel seal in contact was routed through a manual switch for Auto ( under Mach3 control) or manual mode if I want to manually use the mill.

I now started on following the Gecko Drive 320X installation manual that requires the motors not to be connected. Since I used Amphenol connectors for the motors I did not connect the X, Y or Z axis motors. The AMT 102 encoders were connected to the Gecko Drives. Testing by hand for error and fault worked correctly and the Z axis motor was plugged in. The Servo held and the Jog buttons in Mach3 allowed the motor to turn in each direction. I next connected the X axis and found that it turned 1/4 turn and then went into fault. According to the Gecko Drive manual it suggests that reversing the encoder imput channels or the armature leads should correct this problem. I powered down the drives with the reset button in Mach3 and reversed A and B channel input leads at the X drive. I then powered up the drives in Mach3 where a flash over occured and the control relay board that energizes the power supply also burnt up.

At this point I was totally lost as to what had happened. No fuses blew or circuit breakers tripped. The encoder obtains it's 5 volt supply from the Break Out Board and has nothing to do with the Power supply. I removed the relay PC board and repaired the PC board traces on the back of the board. This relay energizes the 240 volt supply for the power supply primary coils.

I then reset Mach3 and observed a flash over at the power supply. Again no fuses blew or circuit breakers tripped. A voltage check did not show out put at the 77 volt DC power supply terminals. The Antek power supply has dual secondary windings and power supplies. The X and Y axis are on one secondary the Z soley on the other secondary. At the time of failure the Z and X motors were connected and dithering. The power supply was removed and I found both secondary input traces on the power supply regulator boards melted. I repaird the traces by jumpering and proceeded to bench test with a clip cord. Energizing each primary winding 1 at a time, the load so excessive that the lights in my shop dimmed. I disconnected the circuit from the supply and removed both power supply regulators to find that the AC input of both rectifier bridges shorted. I suspect that the rectifier bridges were mis-labeled and were not 1000 volt PRV rectifier bridges but 100 volt bridges.

I sent an email to John Angro of Anteck and after 4 days sent another email to the sales director who indicated that someone may contact me shortly. Antek sent me 1 powersupply PC board and bridge, I replied that I still needed one bridge to repair my power supply and was willing to purchase the parts, Antek did not respond to my email. I sourced out a 600 Volt Peak Reverse Bridge and made repairs. A month was lost waiting for parts to make repairs.

Since I saw a flash over on the X Gecko Drive I also contacted Marcus Freimanis of Gecko drive, absolutly great service and help from him, he responded to every email I sent him in less than 24 horus. I purchased a spare drive and mailed the X drive back for evaluation. The X drive tested out okay and it has been kept as a spare.

This kept the project from advancing but allowed me to tidy up a few things that I had put on the back burner with the repair of the machine. That I will cover in the post next.

Ark1

TonyK.

[diyengineer]

I READ IT ALL!! AWESOME WORK! It looks amazing!!!!

That really sucks about Antek not getting back to you at all. I cant believe they would not even send you a email back. I hope i will not have that problem.

Keep up the good work! I can't wait to see a video of it run! They just don't make them like they use to!

[Mega Graham]

Great job on the rebulid, good to meet you and hear about the project. We sold the Tungaloy milling inserts for this machine.

MeGa Industries Inc.
5109 Harvester Road, Unit 3A
Burlington, Ontario L7L 5Y9
Tel:905.631.6342
Fax.905.631.6341
Toll Free 1.800.665.6342
graham@megaindustries.com

Mega Industries Inc.

[RotarySMP]

This is coming along nicely. Good work.

Have you contacted Machine Tools CZ? They deal in TOS machines and might be able to help you find the manuals.

[Ark1]

Graham thanks for the comment on the mill conversion.

Just to let you know the facing mill with the inserts you supplied me with worked great. I have a video that I will be posting with the Tungaloy Facing Mill.

Many thanks to your great service.


Ark1

TonyK.