Hardinge Lathe Accuslide CNC Lathe Control Retrofit

slodat · Jul 27, 2020

Hardinge Accuslide CNC Lathe Control Retrofit

Some trading with a buddy resulted in this guy coming home the other day..

Before we loaded it on the trailer:

Hardinge lathe with Upgrade Technologies Accuslide CNC conversion with Fagor control. I will be installing new servos and Centroid control. Very similar to my mill conversion. Should make a really nice CNC lathe for my shop.

Got the lathe unloaded off the trailer and roughly in its resting place. It will stay on the cribbing until the controls are installed. Overall it’s in excellent condition. Was quite dirty. Still some cleaning left. Starting to sort out the existing control circuits. Most of the new control components are ordered. The existing Accuslide CNC conversion mechanical components look to be in great condition. I’ll have to source or make a few way cover parts that are missing. I have a 3 jaw chuck on the way.

I sorted out how to move the Accuslide away from the spindle. This will allow me full use of the X axis travel.

The lathe will be quite similar to my mill build: Centroid Acorn control, DMM AC servos and Hitachi SJ-P1 spindle VFD running closed loop vector control.

slodat · Jul 29, 2020

Re: Hardinge Accuslide CNC Lathe Control Retrofit

Parts are starting to arrive. This is the control panel layout. Test piece is made out of 1/8 mdf on the laser to check dimensions, layout, etc. I think this is what I’ll be going with.

2oolhound · Jul 30, 2020

Show off!

I know very little about them but I have touched one. I bid on it in auction but was beat by over 10 X my bid. Pretty nice lathe, I'll be following along here.

slodat · Jul 31, 2020

Touchscreen arrived. Got a great deal on Amazon Warehouse deals. Almost half price.

MDF mockup looks good.

The Hitachi VFD is able to be mounted through the back of the cabinet, placing the cooling fans and heat sink outside. I’ll stand the cabinet off the side of the machine about 3”. Chips shouldn’t be a problem because it’s under the chip tray.

matt_i · Jul 31, 2020

This looks like another great CNC machine taking shape!

Is this machine already ballscrew-ready without replacing Acme?

That is a clever mounting for the VFD cooling...I have to take a closer look at that...have an old machine I'm working on where that could benefit

How do you "tool" it, just a gang of cutters attached to the cross-slide with space between them for actual use? Or are you going to turret-post it...thinking like Dorian's QTCP post or ?

slodat · Aug 1, 2020

Re: Hardinge Accuslide CNC Lathe Control Retrofit

It has ball screws already. I’m replacing an existing Fagor CNC control. Tools mount on the gang block. I’m not sure about how that all works yet. I have some learning to do.

DMM AC servos arrived today. I did a lot of measuring prior to ordering. I knew they were close. They are! The original servos had 3/8” shafts. I found a 1/4 coupler of the same style. Worked great with the DMM’s 1/4” shaft. I had to turn down the heads of the socket head cap screws so they would go in to the bracket straight.

The X axis bracket is going to have to have about 0.4” milled off so the coupler will engage. I’m thinking I can modify the existing part. Not sure about extending the existing threaded mounting holes further in.. I may have to make a new bracket.

Line scratched showing where motor would mount up:

The Z axis coupler has a much smaller gap. I think it will need to be closed up, too.

slodat · Aug 6, 2020

Re: Hardinge Accuslide CNC Lathe Control Retrofit

Got the part modeled.

Original motor mount:

All I have on hand is some 3.5” round stock. This is definitely the most complicated (feature rich?) part I’ve made. I’m sure the first will be a test part.

I ordered a 3 x 3.5” x 12” bar from Xometry. Hopefully I’ll only use a 3” chunk of it for the final part and have plenty left for the Z motor mount.

edit: forgot the mounting holes earlier.

slodat · Aug 6, 2020

Re: Hardinge Accuslide CNC Lathe Control Retrofit

Starting to get the components laid out. I’d like to get everything in the existing two enclosures.

Spindle drive and AC components:

Control and servo drives in the control panel enclosure:

American Locomotive · Aug 7, 2020

I'd really be weary of the lovejoy-style couplings. Under high loads you'll get lost some lost motion unless your control is using direct positional feedback from linear scales. You could mitigate it somewhat using a very high durometer spider I suppose.

Have you looked into bellows (or even helical) couplings? You can find them pretty inexpensively on Amazon, and they have far more torsional rigidity.

But, for what your trying to do, the lost motion of the jaw coupling's spider may not be an issue.

slodat · Aug 7, 2020

Re: Hardinge Accuslide CNC Lathe Control Retrofit

Coupling is what came on the machine. The GS couplings are zero backlash made for the application from what I’ve read on their website.

American Locomotive · Aug 7, 2020

Ah, zero backlash is not the same thing as zero lost motion.

Backlash basically refers to the distance the motor (or handwheel) can physically move when switching direction before the table/tool/slide/whatever responds. Lost motion on the other hand includes things like deflection and flex in components (the "spring windup" I like to call it).

So while the GS coupling may be 0 backlash, it almost certainly has quite a bit of lost motion under load due to the spider.

The machines I used to work on all used bellows or helical couplings, but our tolerances were down in the +/- 0.0003 range usually. Not sure if a standard coupling would be an issue for you. Just something to consider though.

slodat · Aug 7, 2020

The original maker of the retrofit had specs in the tenths, so I'm thinking these couplers are fine. If not I'll sort it out when I get to it. I'm not concerned about the couplers at this point. There are a lot of these conversions in production shops making parts.

slodat · Aug 10, 2020

Re: Hardinge Accuslide CNC Lathe Control Retrofit

Made each end of the motor mount in 1/2 material to check dimensions. Went well and I caught a few mistake that would have scrapped the bigger material had I not found them.

Other end:

slodat · Aug 12, 2020

Re: Hardinge Accuslide CNC Lathe Control Retrofit

The first servo motor mount is coming along. What I started with 3x3x3.5:

Close to fitting correctly. Need to shave a couple faces.

slodat · Oct 13, 2020

Finally finished a few adjustments on the servo mount. All went together nicely. I’m new to this cnc machining. Overall, quite happy with it. Bonus is, it will do the job! On to the next.

What I started with:

Control electronics starting to come together.

cnc-me · Oct 14, 2020

Is that the Acorn board? I got a small Lathe that needs one. Right now it's running Zeus, a DOS controller that runs on a PC. Zeus is pretty good, but it's really a mill control made to act like a Lathe control.

Firebrick43 · Oct 14, 2020

American Locomotive said:
Ah, zero backlash is not the same thing as zero lost motion.

Backlash basically refers to the distance the motor (or handwheel) can physically move when switching direction before the table/tool/slide/whatever responds. Lost motion on the other hand includes things like deflection and flex in components (the "spring windup" I like to call it).

So while the GS coupling may be 0 backlash, it almost certainly has quite a bit of lost motion under load due to the spider.

The machines I used to work on all used bellows or helical couplings, but our tolerances were down in the +/- 0.0003 range usually. Not sure if a standard coupling would be an issue for you. Just something to consider though.

I would tend to agree with you on a standard 2 wing lovejoy. It’s not actually the polyurethane spider it’s the shape and material of the coupler. It about 10 times less stiff than an equivalent bellows style coupler.

However the similar(but different geometry) 3 wing European stainless steel curved jaw couplers with the no back lash spiders have the same stiffness as as the bellows type couplers. And in the German Cnc machines I work on use them on the axis servos. Some of the old Burkhard Weber’s made in the early 90’s still used bellows but the new waldrich colburg, stama’s, shaundt’s, and junkers all used the curved jaw couplers.

https://www.wmh-trans.co.uk/image.php?filename=292-003-041.jpg&stockcode=292-003-041&category=0023000300070006&width=1000&height=1000

slodat · Oct 15, 2020

cnc-me said:
Is that the Acorn board? I got a small Lathe that needs one. Right now it's running Zeus, a DOS controller that runs on a PC. Zeus is pretty good, but it's really a mill control made to act like a Lathe control.

Yes, this is Centroid Acorn. I highly recommend it. This is my third machine I’ve used Acorn on. Very happy with it.

slodat · Oct 15, 2020

It’s always a challenge fitting and mounting all of the components, especially when using the existing enclosures. Came up with a second tier on the HMI rear door. This is where the original electronics were. Turned out well. Everything is secure and the terminals and I/O status LED’s on the lower board are still accessible and visible. Lower layer is gigabit Ethernet switch and 1616 Ethernet I/O expansion. Upper layer is Acorn control board and 8 output interposing relay board.

manwithtools · Oct 15, 2020

Nice work! There is a bit of an art to making everything fit in the available area, when it's less than optimal to begin with.

American Locomotive · Oct 15, 2020

Firebrick43 said:
I would tend to agree with you on a standard 2 wing lovejoy. It’s not actually the polyurethane spider it’s the shape and material of the coupler. It about 10 times less stiff than an equivalent bellows style coupler.

However the similar(but different geometry) 3 wing European stainless steel curved jaw couplers with the no back lash spiders have the same stiffness as as the bellows type couplers. And in the German Cnc machines I work on use them on the axis servos. Some of the old Burkhard Weber’s made in the early 90’s still used bellows but the new waldrich colburg, stama’s, shaundt’s, and junkers all used the curved jaw couplers.

https://www.wmh-trans.co.uk/image.php?filename=292-003-041.jpg&stockcode=292-003-041&category=0023000300070006&width=1000&height=1000

Interesting! Most of the German machines I used to work on (Index, Traub) and Japanese machines (Tsugami) all used bellows couplings, and these were new machines (2010-2019 Vintage). Some of the Index's had belt-drives to their axis mechanisms, but they compensated by having scales.

Firebrick43 · Oct 18, 2020

American Locomotive said:
Interesting! Most of the German machines I used to work on (Index, Traub) and Japanese machines (Tsugami) all used bellows couplings, and these were new machines (2010-2019 Vintage). Some of the Index's had belt-drives to their axis mechanisms, but they compensated by having scales.

The more brands I have worked on the more I have come to realize how “set in their ways” some machine builders are. Sometime good, sometimes bad, sometimes just odd and different. Bellows couplers used to be universal across Japanese, German, and American but I would say they are more the exception now.

Even your comments about compensating with scales is interesting to me as all the machines I have worked on used scales. And as a result my previous employer ordered a gaggle of 8800 hmc mazaks and a dozen of their intergrex e670h. When a group of us attended the Mazak maintenance training the trainers were completely baffled that are machines had scales and we were baffled why they would ask why as every machine we had came with scales.

Well Mazak direct drives and oil cools their ballscrews so even a machine with a 20’ long ball screw should maintain a 20 micron accuracy over its length.

And shuaundt has hydrostatic screws which are just strange.

American Locomotive · Oct 18, 2020

In our case, the Index multi-spindles didn't use scales, but all the screws were directly connected with bellows couplings.

Their 3-turret lathes on the other hand use a complicated "scissors mechanism" (as they call it) to move the turrets in X and Y. It allows for the machine to be extremely compact compared to just about everything else on the market, but I believe the mechanism sort of inherently requires scales since it relies on a lot of things working together and there's probably a lot of lost motion that can stack up.

In general, I think the scales (depending on where they are mounted) allow you to maintain accuracy easier under high cutting loads, since the machine can basically compensate for its own floppy-ness .

cnc-me · Oct 19, 2020

slodat said:
It’s always a challenge fitting and mounting all of the components, especially when using the existing enclosures. Came up with a second tier on the HMI rear door. This is where the original electronics were. Turned out well. Everything is secure and the terminals and I/O status LED’s on the lower board are still accessible and visible. Lower layer is gigabit Ethernet switch and 1616 Ethernet I/O expansion. Upper layer is Acorn control board and 8 output interposing relay board.

Know what you mean, seems like you always need more room than you have.
Here's one I did a while back it's for a small AXYZ CNC Router.

View media item 107367

slodat · Oct 21, 2020

cnc-me said:
Know what you mean, seems like you always need more room than you have.
Here's one I did a while back it's for a small AXYZ CNC Router.

View media item 107367

This is my third Acorn conversion (5x8' router, mill and now the lathe). I really like it. How's it working out for you?

cnc-me · Oct 23, 2020

slodat said:
This is my third Acorn conversion (5x8' router, mill and now the lathe). I really like it. How's it working out for you?

It's not, I don't have any Acorn powered machines yet, though I have always been a fan of Centroid controls. Back in the late 90's I was converting a 5 x 10
router, and checked into Centroid, but their controls back then cost more than the machine. Went with CNC PRO by Doug Yeager, It's pretty good but has its limits. I would say that it's better than Mach 3 which did not exist back then.
CNC Pro seems to have a problem with running at higher speeds anything north of 85 FPM is going to cause problems. On mills and small routers it's pretty decent, but large routers or faster mills it's not gonna cut it. It's now open source so anyone can modify it.

slodat · Dec 3, 2020

The control panel.

slodat · Dec 6, 2020

Re: Hardinge Accuslide CNC Lathe Control Retrofit

Control panel sitting on the lathe.

slodat · May 23, 2021

Finally getting to wiring in the new control. It’s a lot in a small space.

macgee · May 23, 2021

Awesome job Slodat,

Keep it up and Looking forward to seeing it making some chips!

slodat · Sep 27, 2022

I've been chipping away at the CNC lathe project the last few evenings. It's close to being ready to power up and start checking things out. Plenty of little details to get dialed in yet. Once it's powered up, it goes pretty quickly on the remaining little stuff. The control board, i/o expansion, ethernet switch and a row of terminal strips are on the control panel rear door. The upper enclosure holds the servo drives, control computer, and front panel.

The lower enclosure houses the AC power conditioning and distribution, spindle VFD, DC power supplies, and a few relays.

It will be nice getting this machine up and running. I have a couple parts I want to run on it.

RickP · Sep 28, 2022

That looks like a LOT of electrical components -- good thing you did this kind of thing once or twice in your day job!

It really is amazing that old iron like your lathe can be brought into the 21st century, and made useful again for your type of manufacturing.

GeddyT · Sep 29, 2022

What's the capacity? I might have a job to throw at this machine some day.

slodat · Sep 29, 2022

GeddyT said:
What's the capacity? I might have a job to throw at this machine some day.

1-1/16” is the 5C collet capacity.

slodat · Oct 2, 2022

Almost done with wiring on the lathe. Waiting on some stuff to arrive to finish the encoder wiring and I'll be done.

Lower cabinet:

There are a few spots that could use a blanking plate. 3d printed this one. Worked out well.

It looks like originally there was a cover for the axis ball screw. It's long gone.

Cut this out of 12ga mild steel. It's way over kill, but the sheet was on the plasma. Worked well.

Close to be wrapped up.

I have a lot of learning to do before I'll be doing anything meaningful with the lathe. I ordered a few tools. I have a lot of tool holders I bought a while back. Need to get a small assortment of tools and sort out using the conversational programming to make a part. The encoder will allow for threading.

Hardinge Lathe Accuslide CNC Lathe Control Retrofit

ALLIANCE MEMBER

Attachments

ALLIANCE MEMBER

Attachments

Well-known member

ALLIANCE MEMBER

Attachments

Well-known member

ALLIANCE MEMBER

Attachments

ALLIANCE MEMBER

Attachments

ALLIANCE MEMBER

Attachments

Well-known member

ALLIANCE MEMBER

Well-known member

ALLIANCE MEMBER

ALLIANCE MEMBER

Attachments

ALLIANCE MEMBER

Attachments

ALLIANCE MEMBER

Attachments

Well-known member

Well-known member

ALLIANCE MEMBER

ALLIANCE MEMBER

Attachments

ALLIANCE MEMBER

Well-known member

Well-known member

Well-known member

Well-known member

Attachments

ALLIANCE MEMBER

Well-known member

ALLIANCE MEMBER

Attachments

ALLIANCE MEMBER

Attachments

ALLIANCE MEMBER

Attachments

Well-known member

ALLIANCE MEMBER

Well-known member

ALLIANCE MEMBER

ALLIANCE MEMBER

ALLIANCE MEMBER