Advice on Bridgeport converted to CNC for sale

[Charlieman22]

Gents,
Newcomer tying to get up to speed on CNC BP or BP clone conversions.
Went to look at a Bridgeport today - not yet listed for sale.
Surprised to find it was converted to CNC already.

He is the original owner.
Showed me the receipt from 1973(ish) for $6500 all in including vice.
It's a job shop and he is retiring.
Very little use now.

Set up uses a 486 computer (that is not a typo) and an ancient cable (R something) from the computer to the machine.
The computer is on last legs - perhaps a better description might be last knees at this point.
He ran what ever program was last used. X/Y/&Z axis all hummed and spun and it moved back to .005 of the zeroed out origin point (to my surprise!), at least once.
Pics of the conversion below.

Questions:
- If I want CNC - would it be a better value to just convert a machine with new system myself?
- Is there a scenario worth considering where perhaps some but not all of the components on this machine should be replaced/upgraded? Why the 486 computer?
- On a newer clone, that has powered x/y/z axis - is that feed system good enough to convert to CNC - or do you need to put new CNC specific servos on to operate properly?

Few pics below of the Bridgeport. The last image is of a clone I'm looking at that has all three axis powered.
Thanks for taking the time to read.
Anything that could help get me educated much appreciated.
-CM

 

[projectnut]

I can't speak for this particular machine, but it is of the same vintage as the Bridgeport I purchased many years ago. Mine has a 1 1/2 hp spindle motor with the variable speed drive. It had the same power feed on the table and came from the factory with a set of collets by 1/16" and a vise.

Mine came from a local high school and looked to be in about the same condition. When I disassembled it to move it to the shop I found the original paperwork in the column. The school had paid a little over $1,400.00 for the machine and the accessories. The price of $6,500.00 seems a bit excessive when a more expensive base machine only cost $1,400.00 from the factory.

I would guess the conversion was made many years later, possibly around the early to mid 1980's. That's the era of the 486 computers. By the mid 1990's things changed to the early Pentium platform.

 

[wrat]

I would guess the conversion was made many years later, possibly around the early to mid 1980's. That's the era of the 486 computers. By the mid 1990's things changed to the early Pentium platform.

I believe you're off by about a decade.
In the mid 80s (1985) a 386 rig was a $13,000 computer. The 486 was touted, when found, by the early-90s. Heh, heh... those were the tragic days - if you brought a computer home, it went obsolete as you unboxed it.
Regardless the ancient chipset, I'd be leery of a BP that's already converted by a working shop. These things are often beat TO DEATH by their own NC drivers without their owners knowing it; often without deceleration blocks in the coding but snapped corner to corner. And then being left to run day-in and day-out.
Remember, the whole idea of NC is to make the machine move longer, faster, and more often than a hand crank can do it.
So it's possible - not always the case, natch - but possible this thing has had 20 years of cranking for every year of NC service.
[yes, yes, i know there's always that one guy that never heard of such a thing....]

The wear is where you don't see it: in the ways and bearings. Pernicious wear in the form of bowing and not scoring.
It might "run just fine" but if they're making parts that have to be good to .030 and you're wanting to do setup and make cuts to .005 or less then caveat emptor.

Now with all that doom and gloom out of the way, this thing might be a heck of a find. Cast aluminum pulley housings? That's great! Still runs? That's very cool. I'd buy it for the right price.

Another point, if this is a 486-based system, this hearkens from an era where the actual programming was usually done elsewhere. In the office. Then the code was loaded into the machine at the computer (NC) as opposed to fully programmed in the shop at the machine (CNC). It takes next to nothing computer-wise to simply move the machine to XYZ so for decades, these were just data buckets loaded from something else - everything from punch tape to a Cat5 cable (DNC).

Finally, "last legs" is often simply neglect. These boards, yes, even a lowly 486, will run past your lifetime if care is taken up front. The trouble is often one of hostile environment. Oil and smoke vapor condense on the board that's not housed in a sealed cabinet producing random and frustrating errors. A PC of that (or almost any) era is doomed if it's housed in a production shop environment - unless it's of a hardened type.

If the price was right (no more than maybe $3K) I wouldn't laugh. At $2k, I'd snap it up. If he wants sticker price of $6.5K I'd walk. Because just the 'conversion kit' for something like a Precision Matthews is about $1400 -- and that's after buying the mill at around $4K (plus of minus loads of extras). That's around $6K for entry level, but a brand new, precise machine.

But I'm not in Los Angeles and the market here is pretty spotty.

 

[Charlieman22]

The school had paid a little over $1,400.00 for the machine and the accessories. The price of $6,500.00 seems a bit excessive when a more expensive base machine only cost $1,400.00 from the factory.

They do sound similar. Not sure why it would be such a discrepancy. I think there was no more than $2K of additional items, tax, delivery...

Regardless the ancient chipset, I'd be leery of a BP that's already converted by a working shop. These things are often beat TO DEATH by their own NC drivers without their owners knowing it; often without deceleration blocks in the coding but snapped corner to corner. And then being left to run day-in and day-out.

This is precisely my concern.

Another point, if this is a 486-based system, this hearkens from an era where the actual programming was usually done elsewhere. In the office. Then the code was loaded into the machine at the computer (NC) as opposed to fully programmed in the shop at the machine (CNC).

This is the case.

Finally, "last legs" is often simply neglect. These boards, yes, even a lowly 486, will run past your lifetime if care is taken up front. The trouble is often one of hostile environment. Oil and smoke vapor condense on the board that's not housed in a sealed cabinet producing random and frustrating errors. A PC of that (or almost any) era is doomed if it's housed in a production shop environment - unless it's of a hardened type.

All the above. I had to help him try to start it. Managed to start it once before it froze and then no amount of un-powering would get it to do anything by make that beep sound from the board. 20 years of dust and grime.

If the price was right (no more than maybe $3K) I wouldn't laugh. At $2k, I'd snap it up. If he wants sticker price of $6.5K I'd walk. Because just the 'conversion kit' for something like a Precision Matthews is about $1400 -- and that's after buying the mill at around $4K (plus of minus loads of extras). That's around $6K for entry level, but a brand new, precise machine.

He is asking $4500 - though there is room to negotiate. But I have some trepidation about negotiating even to $3K or $3500 if its beaten to death from years of CNC. Side note - someone on another thread I had posted on regarding moving of machines noted that the control box might be replaceable with an Acorn model control box- which would allow a windows 10PC.

Still trying to understand if the powered axis I see on modern clones (like the PM machine you note) are sufficient as drivers for CNC or if you need to use what appear to me to be larger motors like what I see on many of the CNC conversions?

 

[markba633csi]

Looks like it uses beefy DC servo motors- nice hardware but old. Worn, for sure but for hobby use probably fine. Personally I would only want to pay around 2500$ max
-Mark

 

[Janderso]

I saw one of those go for $400 at a Bidcal auction. Nobody knew what to do with it. It didn't have a DRO though.

 

[macardoso]

If it were me, and the price was right, and the machine was not too worn, and I could get it home, then I'd be replacing the controls.

The actual choice of control software (Mach 4, Centroid, Linux CNC, etc.) is of little relevance to me. What I would be concerned about is the DC servos. Those are big boys and will need interface to the old amplifiers (if functional and trusty) or new amplifiers to replace them. Older machines only had tachometers on the motors and glass scales on the axes. This is different from the digital encoders expected by most modern DC servo drive control systems. The motion control board must then be able to interface with whatever drive you want to use. If the drive is analog input then the motion controller is significantly more expensive than one for controlling pulse input drives.

Alternatively you can replace the motors with modern AC servos, but I'd budget (at a minimum) $2000 to get motors, drives, and cables.

It can work, but expect to need to dive in head first, learn about the controls, and select replacements. There will be cost associated with this.

Personally, I search for these kind of projects because they are fun for me, others maybe not. You need to do your own soul searching to figure out if you want a project machine or a machine that works out of the box.

 

[Charlieman22]

Marcardoso - good run down. Drinking water from fire hydrant on this one - but that's a norm for me when jumping into something new. Goal is to not get my lips ripped off.

My understanding of the components is rudimentary - but your explanation is helpful. Sorta like not being able to use a modern Amp with an old phonograph.

It can work, but expect to need to dive in head first, learn about the controls, and select replacements. There will be cost associated with this.

For sure I need to determine this.

Two questions:
1. Assuming DC motors are good (they seam to be) and their respective amplifiers are as well - what would your approach be for setup? (and are these amplifiers inside the large black box?). That is - what would you buy/swap out hardware wise (I assume it would solve for a digital rather than analog input?)
2. Are the modern servos like what I showed in last pic of my first post capable/suitable/desirable for changing to CNC or would I want to upgrade those as well, if I went with the modern clone option to get in to the game?

Thanks!

 

[macardoso]

Marcardoso - good run down. Drinking water from fire hydrant on this one - but that's a norm for me when jumping into something new. Goal is to not get my lips ripped off.

My understanding of the components is rudimentary - but your explanation is helpful. Sorta like not being able to use a modern Amp with an old phonograph.

For sure I need to determine this.

Two questions:
1. Assuming DC motors are good (they seam to be) and their respective amplifiers are as well - what would your approach be for setup? (and are these amplifiers inside the large black box?). That is - what would you buy/swap out hardware wise (I assume it would solve for a digital rather than analog input?)
2. Are the modern servos like what I showed in last pic of my first post capable/suitable/desirable for changing to CNC or would I want to upgrade those as well, if I went with the modern clone option to get in to the game?

Thanks!

Take your time, I think the consensus currently is the price is high and the machine may be worn. If it turns out to be in good shape and you negotiate price, then you might get a really nice machine.

1) I'd assume the motors to be good. Typical maintenance is brush replacement. The drives will be old and I would guess they run tachometers. This might not be what you want. It would limit you to using the Centroid Oak control board ($1820 - $2260 depending on options) or KFlop (cheaper but quite complicated and little support AFAIK - I would not recommend for a first time CNC conversion). Alternative would be to replace with a modern digital DC servo drive which accepts pulse inputs and uses an encoder on the motor. This would cost you for the drives (moderate cost), and you'd need to install encoders on the motors (cheap enough, but a bit of work). We would need to know the motor nameplate data or at a minimum the ratings on the old drives. We would also need to know the DC supply voltage into the drive. This would cost you for the 3 servo drives, plus encoders, but allows you to pick any motion controller and software (~$200 - $600).

AC servos and drives will out perform DC servos, but that might not matter to you. New motors, drives, and cables might be the route to go if the DC drives are too expensive or difficult to find. These also allow you to use any motion controller. Motors will almost guaranteed be a different frame and shaft size from what you have and will require work to interface to the existing mechanics.

2) The servo in your image is a power feed and is not suitable for motion control. Nice to have but only for manual work

I'd reserve $2k-$3k for a conversion. Maybe it can be done cheaper, but we would need to get into the specifics of what is on the machine.

Mike

 

[JimDawson]

The school had paid a little over $1,400.00 for the machine and the accessories.

The school must have gotten a special deal. Those machines were were in the $6500 range in the 70's. Around that time I bought several new clones at about $5500 each, in the mid-90's I bought another new one w/DRO for $8000

That is the oldest Centroid system I have ever seen, well past its prime, and needs to be replaced with modern controls. The motors and drives are most likely OK since the machine runs. This means you can reuse the motors, drives, and the existing power supply. I'm still using the original DC motors and drives with modern controls on my 30 year old machine. I'm going to make a wild guess here that Centroid was using Copley drives at that time. In any case they would be analog drives and thus compatible with Dynomotion, Mesa, and Galil products. And also should be compatible with a modern All-in-One DC Centroid system.

The value of the machine depends on the mechanical condition, the existing controls have no value. If the ways are not excessively worn, then it should be fine. These machines run forever if kept properly lubricated and are not abused. The fact that is has ball screws is valuable if you want a CNC machine, the hard work has already been done. The fact that it is a manual/CNC machine is very valuable in my book. Never seen a conversion that still had the original power feed attached, very cool.

At $4500, that does not seem unreasonable, but does require a personal inspection.

Since that old 486 is still running, it might have some value on eBay.