What's new
What's new

lathe thermal growth mitagation solutions?

You don't run G92. That's the only time I would ever do that, so yes, curious.

p.s. programming is programming, doesn't matter if it's fanuc or a numeripath, it's all telling the slides where to go.

OK, there is another app for T0 that is a total $$#! for me on this:

Anytime I go to touch off a tool, I am in JOG or MPG mode, but there is an old offset still active.
I can index the turret by hand to my tool and go down to touch it off, but even tho I have T6 touching the part, my CRT still ready T808 on the screen.

My Hardinge's would have read T600 when I was at this point as I would have canceled my offset while going home to begin with, and the Hardinge machines follow the turret on the CRT. The CRT always tells me what turret location I am on! So, a cpl of time now I have went into MDI to dump the offset so's I can git a good read on machine location to enter a new offset, and - per my example, it says T808, so I totally forgot that I indexed the turret since then, and that it doesn't follow it - and so I key in T800 in MDI, and the turret goes to index. Crashing out the turret and slipping the turret index motor to turret connection, resulting in having to dial that back in aggin....

I understand that I am totally spoiled by Hardinge re: their ladders and HMI's. Their Twin Turn is not 100% perfect, but I can't think of anything that I would want to edit as far as running their T51 machines. But - Hardinge doesn't have a twin turret machine, nor have they in several years, so ... I am learning to deal with less intuitive options...


--------------------

I am Ox and I approve this here post!
 
I can index the turret by hand to my tool and go down to touch it off, but even tho I have T6 touching the part, my CRT still reads T808 on the screen.
Wa ! I'd probably shoot that machine ! You've got it in position 6 but it says you're in 8 ?

Well that explains your reason for wanting to universally dump offsets, but I have to say I would not like that control even one little bit !
 
Only issue would be lifting them after no?
if I cannot lift by hand it’s a no go for me. No lifting equipment.

Very open to options there if there is something I’m missing

Currently I use x20 5 gal buckets.
For bigger jobs I’ll get a liquid tote (cubic meter) and leave that in the truck and toss the chips by the bucket load in as I go.

Downside of a garage shop.

I love the “work from home” thing, but it defiantly has its drawbacks. (Space, amps, distractions)
the 30 gal (wut is that in meattrick 900 liters?) cans aren't so bad, unless yer creating a bunch of small steel chips, then just fill it half way.

in a pinch you could get a cheap truck manual crane fer lifting the heavy stuff, eventually yer gonna need something to lift heavy thangs anyway... forklift etc. gantry crane, knew a guy that had a HUGE single column crane for his home shop... thing could lift the truck that brought the pallets of material... but all he had was an italian bridgport clone, and dinky manual lathe

Distractions... have a way of distracting lol, I've found the hardest part of self employment is getting going in the mornings... purdy easy to just say F it and go back to bed... but then I'm broke for 2 weeks and remember I have Shit to do.
 
  • Haha
Reactions: Ox
OK, there is another app for T0 that is a total $$#! for me on this:

Anytime I go to touch off a tool, I am in JOG or MPG mode, but there is an old offset still active.
I can index the turret by hand to my tool and go down to touch it off, but even tho I have T6 touching the part, my CRT still ready T808 on the screen.

My Hardinge's would have read T600 when I was at this point as I would have canceled my offset while going home to begin with, and the Hardinge machines follow the turret on the CRT. The CRT always tells me what turret location I am on! So, a cpl of time now I have went into MDI to dump the offset so's I can git a good read on machine location to enter a new offset, and - per my example, it says T808, so I totally forgot that I indexed the turret since then, and that it doesn't follow it - and so I key in T800 in MDI, and the turret goes to index. Crashing out the turret and slipping the turret index motor to turret connection, resulting in having to dial that back in aggin....

I understand that I am totally spoiled by Hardinge re: their ladders and HMI's. Their Twin Turn is not 100% perfect, but I can't think of anything that I would want to edit as far as running their T51 machines. But - Hardinge doesn't have a twin turret machine, nor have they in several years, so ... I am learning to deal with less intuitive options...


--------------------

I am Ox and I approve this here post!

I don't know what your Hardinge's do, but every Fanuc lathe I've ever used works this way.

As a result I never touch the turret controls on the operator panel at all, and use MDI exclusively to rotate the turret.
 
I don't know what your Hardinge's do, but every Fanuc lathe I've ever used works this way.
You're kidding ! And you guys put up with this ? Holy smokes.

I never touch off tools so wouldn't have come across this anyhow but still ... no. Bendix, accramatic, westinghouse, allen-bradley, actrion, all know and display what position the turret is really in no matter how you put it there. Eek.
 
You're kidding ! And you guys put up with this ? Holy smokes.

I never touch off tools so wouldn't have come across this anyhow but still ... no. Bendix, accramatic, westinghouse, allen-bradley, actrion, all know and display what position the turret is really in no matter how you put it there. Eek.

It's not a control thing. It's the ladder behind it.
Hardinge obviously runs their own machines in-house, and thus - they make it run in as user friendly a way as possible.

I am only ref: their machines built in Elmira, not the commodity ones built in Taiwan.
I don't know anything about those.


-----------------

I am Ox and I approve this here post!
 
I don't know what your Hardinge's do, but every Fanuc lathe I've ever used works this way.

As a result I never touch the turret controls on the operator panel at all, and use MDI exclusively to rotate the turret.

I had a Tornos that we recently retired that had to doo every stinkin' thing in MDI. I hated it!
Everything was 2 screens away and ....

Then go look at an 8x Yamazaki. What's it called??? ... Hyper sumpthin' I think? Quadrex! Hyperquadrex! I think that';s it.

Now go look at the HMI on that bad boy!
It is 2-1/2 to 3' tall, with a CRT (screen of some sort) that you can read from 30' away!
More buttons than Carter ... OK, well that's an exaggeration. Carter had quite a few of those little pills...
But there is a button for everything on that thing!
You don't need to back up 3 pages and enter it in MDI!

I bet that's a friendly machine to run!
But I have full intentions of skirting any machines with GPS locators in them and their service reputation precedes them.


---------------------

Think Snow Eh!
Ox
 
playing with a new to me lathe. ran a batch of parts and slooowly comped thermal growth over several hours. approx 0.0022" very slow and consistant growth. shop probobly went up 2-4 deg through the run as well.
then took a 10-15 min break and it cut the next part 0.0016 smaller than the last. She shrunk fast!!!!! this might be a real problem kicking the door open once in a while. +20 inside and its -25C outsidein winter ... anywhooooo.

now its got me wondering what i can do to mitigate the majority of the growth in the first place?
my simple ideas:
add a pond pump and radiaror to take some heat out of the coolant. place the radiator atop the mistaway to blow the heat out?
add a few fans to the enclosure to pump shop air through to reduce the heat buildup

any other simple ideas?

i dont have much extra power i can throuw at it, pretty power limited unfortunatly
This post COMPERATIVE MEASURENT posted by me several years ago is exactly describing such case and the simple method to cope with the problem. The method is simple, works always and immediately, no worm up times or radical environmental control needed. The program was written for OKUMA, but can easily be adapted to any control.

Stefan
Cogito Ergo Sum
 
Can it be reliably mapped out?
So we run grinders.
They like continuous cycles. They do not like lunch breaks, 10 minute break time or the bad overnight cold startup.
So we have basic "Hope it works" comp for shutdown and this best guess is in the part routing. (lunch comp or fix is different from overnight...duh)
For us we know it it goes one way or the other. Comp and then remove all that comp as things get going.

I walk into a cold machine.. what is needed? A warm machine but I have been gone for lunch ... what is needed.
One needs to know how she acts. They are are different.
SPC charts and logging size control is helpful.
A machinist often "gets this" in your head but notes help other guys running the same.
 
Last edited:
You're kidding ! And you guys put up with this ? Holy smokes.

I never touch off tools so wouldn't have come across this anyhow but still ... no. Bendix, accramatic, westinghouse, allen-bradley, actrion, all know and display what position the turret is really in no matter how you put it there. Eek.
This is one argument we don't need to have - I have used some much better lathe controls than Fanuc.
 
You're kidding ! And you guys put up with this ? Holy smokes.

I never touch off tools so wouldn't have come across this anyhow but still ... no. Bendix, accramatic, westinghouse, allen-bradley, actrion, all know and display what position the turret is really in no matter how you put it there. Eek.

OK, so how doo you set tool offsets [in a lathe]?

I git that these days you may have a touch probe, but that's not traditional.


---------------

Think Snow Eh!
Ox
 
I did not read all the comments btw. But you should first check the preload on the X-axis ballscrew. A ballscrew will get warmer if used. The preload should comp for that. If it's to loos the axis will drift. You definitely run a warmup program when the machine did sit for a few hours. I'm running a warmupprogram between 30 and 60min depends on the tolerance area i'm working in. If the preload is oke and you did warmup the machine and still there is that amounts of error you could solve it by reducing the rapid speed of the x axis in the parameters. I do that for my z-axis. I did make my own thermal comp program for the z axis. But this program in only accurate between 20 and 40 degrees. Now when my program starts and te temperature of the z-axis is above 38 degrees it will decrease the rapid speed of the z axis with 10%. Normally when its running for full days the temperature will settle at 37 degrees when the rapid is at 70 a 80%.
 
I'm refraining on commenting about g28,g30,g52,g53... each is sorta machine dependent, and each has its use, being sending the turret/spindle somewhere safe for a tool change, is in my opinion should be part of every program... as well as a basic safety line at the beginning of each tool... Though I don't think using it as a way to limit X axis travel is enough to make a significant change in temp, as the Z axis will do far more actual work then the x axis... to say nothing of the spindle. Without an ail cooler, is all pretty much spitting into the wind.

Even if you rapid all the way back to machine home fore every tool change, an average that would be what 8 tool changes? compared to how many roughing/pecking/threading cycles each with a rapid move in Z?
 
I'm refraining on commenting about g28,g30,g52,g53... each is sorta machine dependent, and each has its use, being sending the turret/spindle somewhere safe for a tool change, is in my opinion should be part of every program... as well as a basic safety line at the beginning of each tool... Though I don't think using it as a way to limit X axis travel is enough to make a significant change in temp, as the Z axis will do far more actual work then the x axis... to say nothing of the spindle. Without an ail cooler, is all pretty much spitting into the wind.

Even if you rapid all the way back to machine home fore every tool change, an average that would be what 8 tool changes? compared to how many roughing/pecking/threading cycles each with a rapid move in Z?


If he's trying to hold tenths in X, who cares about Z?



------------------------

Think Snow Eh!
Ox
 
If he's trying to hold tenths in X, who cares about Z?



------------------------

Think Snow Eh!
Ox
you hear that whooshing sound? no? the x axis going to be doing nearly the same amount of work whether you reduce rapid distances or not, a few inches under 0 load isn't going to change how hot it gets, the Z on the other hand does most of the work, and since they are in the same "room" as it were i.e. the cabinet they are sealed up in, they will live at more or less the same temps.

Plus, I'd be willing to bet a sizeable amount of money that the ball screws by themselves don't create a lot of heat, since they are pretty close to zero friction even under heavy load, the motors/servo's yes, but again they all live in the same cabinet
 
Nevermind that on a lathe you can't always just rapid away in Z for clearance, different tools have drastically different clearance requirements, so you end up moving in X just for clearance, unless you want to spend the time to "optimize" every tool so it has absolute minimum clearance, sure, thats a thing, but for a job shop, its not exactly wise, and really, not going to gain F all in temp control, which is the issue at hand remember
 
EKrol,
30 to 60 minutes of non-productive warm up time?
On a 8 hour shift? That would hurt.
I've werked in shops were this was considered "normal"

IF < big if... you are trying to deal with tight tolerances, then it kind of makes sense, but really a 10 min warm up and back off the critical tools a safe amount and its not such a big deal
 
you hear that whooshing sound? no? the x axis going to be doing nearly the same amount of work whether you reduce rapid distances or not, a few inches under 0 load isn't going to change how hot it gets, the Z on the other hand does most of the work, and since they are in the same "room" as it were i.e. the cabinet they are sealed up in, they will live at more or less the same temps.

Plus, I'd be willing to bet a sizeable amount of money that the ball screws by themselves don't create a lot of heat, since they are pretty close to zero friction even under heavy load, the motors/servo's yes, but again they all live in the same cabinet

"As I see it"

A)
Sure, they're all in the same enclosure, but screw heat and ambient heat Shirley are not the same.
Heat from the Z will amortize with the whole of the machine. And it will slowly change temps all together.

B)
Shirley the screw must warm up by it'self. If in fact the motor is a major influencer of the screw temp, then it setting idle over break wouldn't have any effect, as the motor is still holding position all that time. And you can Shirley see changes in X size long before the whole machine comes up - be it coolant, castings, and whatnot. A 10 minute warm-up cycle isn't warming up the coolant and castings, just the screws.

C)
I would not think that there are many times that you would need to clear in X to index unless you either pull the part out part way through the cycle? As long as you pull back far enough for your longest tool to clear, you should be golden. If this is not the case, I would like to see an example as I can't picture it. I am talking single turret here, which I think he has.


-----------------------

That's the way I see it. Saturday, March 16, 2024.
Goodnight.
 
"As I see it"

A)
Sure, they're all in the same enclosure, but screw heat and ambient heat Shirley are not the same.
Heat from the Z will amortize with the whole of the machine. And it will slowly change temps all together.

B)
Shirley the screw must warm up by it'self. If in fact the motor is a major influencer of the screw temp, then it setting idle over break wouldn't have any effect, as the motor is still holding position all that time. And you can Shirley see changes in X size long before the whole machine comes up - be it coolant, castings, and whatnot. A 10 minute warm-up cycle isn't warming up the coolant and castings, just the screws.

C)
I would not think that there are many times that you would need to clear in X to index unless you either pull the part out part way through the cycle? As long as you pull back far enough for your longest tool to clear, you should be golden. If this is not the case, I would like to see an example as I can't picture it. I am talking single turret here, which I think he has.


-----------------------

That's the way I see it. Saturday, March 16, 2024.
Goodnight.
thermal dynamics, an often quoted seldom understood theory (folks like to use it while disparaging electric vehicles) heat travels in 3 different ways, radiation, convection, conduction.

ball screw heat is largely from either convection, with some, probably very little through conduction, i.e. from the surrounding environment, or dirrect from the servo, but 2 things, the serve is divorced via belt drive, and even direct mount usually has a rubber cushion between which would insulate them from most of the heat... maybe a tiny bit of heat vie radiation, but it would be minimal...

Now the ways to create heat, being friction, fission, or induction... the servos and motor create heat via induction, mostly the spindle motor... there is very little friction heat on machines, some but, its minimal, Fission... would be a very bad day.
 








 
Back
Top