one possibility why the fire suppression is there is the machine might of ran straight cutting oil instead of water soluble coolant. When you use straight oil it can create an oil mist in the enclosure that can catch fire easily if your chips are too hot or if you create sparks. Magnesium and other metal fires are extremely difficult to put out and would be completely mitigated just by running coolant. If oil was used I would recommend you to do your best to clean out all the the coolant lines and get any remnants of oil out of the machine before running coolant. You might still run into some issues with seals going bad Ive always heard its not a good idea to switch a machine from oil to coolant or vise versa but getting a drum of oil for that thing would be very expensive.
Having done a couple of these conversions from some factory provided control to Linux CNC, Fluidnc, Mach3, etc. Think long and hard about it. By the way, you will not save any money. Doing one of these conversions creates what is known in the control systems world as an orphan. Or a control system that you have no possible outside support when things go wrong. Chewed wires are simple to repair though time consuming. I bought a Southwest Industries ProtoTrak bed mill that had all of the external cables cut, presumambly for the copper. SWI sold me all of the required cables complete with factory done connectors for about $900. Once it was put back together, being able to talk with someone who works on these things every day saved me so much time. My interest was in making chips, not learning how to adapt a new control.
I have worked with both systems, the Linux CNC and the Mach, and have implemented the MESA motion cards. These are awesome. I don't know what your level of Linux or development is. As someone with experience with all this, I suggest you return to work quickly without pulling out your hair. I would use a Centroid or dare I say a MACH3 system. The learning curve for the MESA cards is steep. However, if you are willing to endure the learning curve, the MESA for the price point is the most flexible platform for DIY.
If the Fadal control works, use it as is. The XT series machines were low end linear way weldments any way. Ive done several Lcnc retrofits and unless its a really hi end machine, its a questionable waste of time. The machine has dc motors and resolvers, which means you either swap encoders or you need a resolver card and an encoder card. The Fadal control is easy to use and rock solid. There is little to no advantage to any retro on that machine, the dc motors have limited speed, the amps are old slow technology and the machine itself is not designed for speed. Dont even consider Mach , Centroid or Acorn. Tons of Fadal parts around.
For most people, the only reason to retrofit is if the original controller doesn't have enough memory for something you absolutely need to have fully in memory in the controller, and can't really drip feed (such as Renishaw macros for probing). If you understand the macros, you can probably reduce some of them down to be smaller and simpler, at the cost of some functionality, and if you're insane, you can reverse engineer the firmware of the original controller to remove the memory limitation, and be able to add more memory to the machine than it ever supported. I may know someone that took the insane option and reverse engineered the firmware - not of a Fadal, but of a Mitsubishi M3 controller. I kind of really wanted to do a LCNC conversion, but someone convinced me it would be fun to hack the original controller to work instead. Then I got a manual lathe that I was going to LCNC, even bought some servos and mesa cards, and 3 weeks later, someone GAVE me a real CNC lathe that no one else could fix, which is now fixed, and running with it's original Seicos controller.
On the retrofit front I always rip out the old electronics. Chinese 23bit servo kits for something like this cost you sub 2k for all the motors including shipping. If the mechanics are ok it turns it into a new machine. (Mind as you said its better to get something more high end for it to make sense to burn the time and money on it, say a old Okuma/Mazak fixed gantry machine, or in general any of the larger machines).
Hi, I am not sure if you have looked into the calmotion upgrades but I would seriously consider that. I know that they now make a modern controller/pendant that is expensive but may be worth it (for the FX version). The downside is that they charge you for every little feature whereas linuxcnc is a headache up front but has all of those features built in. Calmotion has less expensive upgrades like lcd screen, drip feed, etc.
I have 3 Fadal machines that i use in production. I would not replace the 88HS control. It will take you much longer to get it to a useable state and you will lose some of the cool features it has. I also own Okuma, Doosan (Fanuc) and have played with Mazaks and help teach CNC at our local tech collage on the weekends (Haas) From my experience once you learn the quarks of the 88HS you will actually like it. I prefer to prototype working on my Fadals than my DNMs.
As people have said its likely not worth the time to retrofit over just getting replacement parts. Stuff I retrofit is generally much higher end machines or much larger machines and retrofits I do are full electronics replacement (everything electrical except the spindle motor and stuff like end-switches gets ripped out, which if the mechanics are good ends you with a basically new machine). China/taiwanese servo 23bit absolute encoder kits are just so cheap that it generally make no sense to keep the old motors for the total price of 1,5-3k all of those cost shipped to your door (3k is for full set of 5,5kW motor kits). Especially so if the old component is expensive on the second hand market as generally selling the old electronics covers most if not all the cost of the new components so you are only paying with your time.
Fix the stock control / wiring. not worth the time to do a retrofit on that machine. You will not be able to sell it easily when you decide to upgrade to a better machine
one possibility why the fire suppression is there is the machine might of ran straight cutting oil instead of water soluble coolant. When you use straight oil it can create an oil mist in the enclosure that can catch fire easily if your chips are too hot or if you create sparks. Magnesium and other metal fires are extremely difficult to put out and would be completely mitigated just by running coolant. If oil was used I would recommend you to do your best to clean out all the the coolant lines and get any remnants of oil out of the machine before running coolant. You might still run into some issues with seals going bad Ive always heard its not a good idea to switch a machine from oil to coolant or vise versa but getting a drum of oil for that thing would be very expensive.
Having done a couple of these conversions from some factory provided control to Linux CNC, Fluidnc, Mach3, etc. Think long and hard about it. By the way, you will not save any money.
Doing one of these conversions creates what is known in the control systems world as an orphan. Or a control system that you have no possible outside support when things go wrong.
Chewed wires are simple to repair though time consuming.
I bought a Southwest Industries ProtoTrak bed mill that had all of the external cables cut, presumambly for the copper. SWI sold me all of the required cables complete with factory done connectors for about $900. Once it was put back together, being able to talk with someone who works on these things every day saved me so much time. My interest was in making chips, not learning how to adapt a new control.
I have worked with both systems, the Linux CNC and the Mach, and have implemented the MESA motion cards. These are awesome. I don't know what your level of Linux or development is. As someone with experience with all this, I suggest you return to work quickly without pulling out your hair. I would use a Centroid or dare I say a MACH3 system. The learning curve for the MESA cards is steep. However, if you are willing to endure the learning curve, the MESA for the price point is the most flexible platform for DIY.
If the Fadal control works, use it as is. The XT series machines were low end linear way weldments any way. Ive done several Lcnc retrofits and unless its a really hi end machine, its a questionable waste of time. The machine has dc motors and resolvers, which means you either swap encoders or you need a resolver card and an encoder card. The Fadal control is easy to use and rock solid. There is little to no advantage to any retro on that machine, the dc motors have limited speed, the amps are old slow technology and the machine itself is not designed for speed. Dont even consider Mach , Centroid or Acorn. Tons of Fadal parts around.
For most people, the only reason to retrofit is if the original controller doesn't have enough memory for something you absolutely need to have fully in memory in the controller, and can't really drip feed (such as Renishaw macros for probing). If you understand the macros, you can probably reduce some of them down to be smaller and simpler, at the cost of some functionality, and if you're insane, you can reverse engineer the firmware of the original controller to remove the memory limitation, and be able to add more memory to the machine than it ever supported. I may know someone that took the insane option and reverse engineered the firmware - not of a Fadal, but of a Mitsubishi M3 controller. I kind of really wanted to do a LCNC conversion, but someone convinced me it would be fun to hack the original controller to work instead. Then I got a manual lathe that I was going to LCNC, even bought some servos and mesa cards, and 3 weeks later, someone GAVE me a real CNC lathe that no one else could fix, which is now fixed, and running with it's original Seicos controller.
On the retrofit front I always rip out the old electronics. Chinese 23bit servo kits for something like this cost you sub 2k for all the motors including shipping.
If the mechanics are ok it turns it into a new machine. (Mind as you said its better to get something more high end for it to make sense to burn the time and money on it, say a old Okuma/Mazak fixed gantry machine, or in general any of the larger machines).
Cool Stang brother.
Hi, I am not sure if you have looked into the calmotion upgrades but I would seriously consider that. I know that they now make a modern controller/pendant that is expensive but may be worth it (for the FX version). The downside is that they charge you for every little feature whereas linuxcnc is a headache up front but has all of those features built in. Calmotion has less expensive upgrades like lcd screen, drip feed, etc.
I have 3 Fadal machines that i use in production. I would not replace the 88HS control.
It will take you much longer to get it to a useable state and you will lose some of the cool features it has. I also own Okuma, Doosan (Fanuc) and have played with Mazaks and help teach CNC at our local tech collage on the weekends (Haas)
From my experience once you learn the quarks of the 88HS you will actually like it. I prefer to prototype working on my Fadals than my DNMs.
Consider tu use one Acorn Cnc from Centroid. Their products are amazing. Greetings from switzerland
As people have said its likely not worth the time to retrofit over just getting replacement parts.
Stuff I retrofit is generally much higher end machines or much larger machines and retrofits I do are full electronics replacement (everything electrical except the spindle motor and stuff like end-switches gets ripped out, which if the mechanics are good ends you with a basically new machine).
China/taiwanese servo 23bit absolute encoder kits are just so cheap that it generally make no sense to keep the old motors for the total price of 1,5-3k all of those cost shipped to your door (3k is for full set of 5,5kW motor kits). Especially so if the old component is expensive on the second hand market as generally selling the old electronics covers most if not all the cost of the new components so you are only paying with your time.
Cool!
Fix the stock control / wiring. not worth the time to do a retrofit on that machine. You will not be able to sell it easily when you decide to upgrade to a better machine
Why do people, men especially, collect so much crap over the course of their lives....?
All my "crap" got destroyed 3 years ago in a fire, so I guess I don't have an answer for you