What's new
What's new

Hardinge turret alignment question

npinson

Aluminum
Joined
Apr 4, 2010
Location
Indiana
Hello

I have a Hardinge T51 Conquest with the 12 station turret and livemilling with C. Fanuc 18T control. We just purchased this machine about month and half ago, and machine has been working well, except for something strange that keeps happening.

When I first got the machine, I checked for turret alignment, "radially", first and it was right around .001 T.I.R. Again that was radial. Axially, it was pretty much nothing, maybe .0002-.0003 along the length of the flats horizontal with the spindle. Spindle itself was fine as well. I took a light cut on 2" OD aluminum and had taper under .0005 in 8in of cut.

But for whatever ever reason, the turret shifts radially out, and not by a little. I first noticed it when I tried center drilling, and it was creating a "tit" in the part. Running it manually, you could even see it bending the center drill as it was trying to find center. But the machine has not been crashed. I check the axial runout out again. Now it's around .009 total runout. :toetap:

So I play around with it, I did a rezero procedure, by holding down zero return button in jog mode, and selecting a tool station, and pressing turret index. It acted kind of funny, it rotated at normal speed but went past it's designated tool station that was called, then suddenly slowed down, did another revolution, and then indexed to the intended turret station.

Somehow, at the moment, that fixed it. Checked runout again, again back to .001-.0013 ish runout. So I'm confused now. The only thing that I can possible think of is that the curvic coupling halves are not fully seating, and kicking the turret out of position. I do have another T51 Conquest, little bit newer, and I had to realign the turret when I first got it. But every station, I think I have 4 boring bar tooling blocks in there at all times, always indicate the same, doesn't deviate. But I have noticed that the indexing time is much quicker on the newer Conquest. Not the rotation speed, but when the turret called station is reached, the hydraulic retract to seat the curvic halves is very fast. On the live milling Conquest, it's noticeably slower.

So before I drag this out too far, I put a dial indicator on the turret body housing, indexed to tool 1 and zero'ed my indicator.

These are the values that I got indexing through all 12 stations. After I found all the values, I randomly selected stations, and the indicator readings stayed the same. I attached pictures so that you can understand what I did as well....

T1= 0
T2= .0004
T3= 0.0002
T4= -0.0004
T5= -0.0002
T6= -0.0002
T7= -0.0007
T8= -0.0014
T9= -0.0019
T10= -0.0021
T11= -0.0018
T12= -0.0010


So. These values are all after the turret is seated on the coupling. In my mind, they should all be the same, or at least withing a few tenths from each other. T8-T12 is where it is pretty far out. I don't know if it's possible that the limit switch that senses when the turrent is clamped is set wrong, (ie. telling the hydraulics to stop before it's finished). Or maybe a timing belt in there is worn. However, the turret never seems to rotate as it re seats itself. It swings around, and goes straight in, not twisting at all that I can see. I did have the turret numbers get jumbled on me once about 2 weeks after I got it. The turret re zero procedure fixed that. Maybe some parts in the gearbox are worn in there, but I'm confused how the machine can be in good alignment one minute, the next it's wayyy off. I can get my indicator reading exact up and down for sure, but side to side, the worst reading I had was .01" both ways, so .02" of runout. That obviously isn't gonna fly :confused:


Also, anyone know were the batteries are for these machines? In the first Conquest I bought, the batteries are attached directly to the drives in a small grey case. They look like this.

SHS749 Battery New USA Stock Fast Shipping | eBay

But this recent T51 I have, I cannot find the batteries anywhere? :confused: Both machines have 18T controls, but I think the live milling T51 has an early version 18T. Paging through the control it's much laggier. Just to clarify, one machine, (most recent), and live milling, is 1995 with 18t. The first T51 I bought is a 1999 with a 18T.



I got tired of them going bad every 4-5 months, so I bought this

1 Pcs 4 D Cells Battery 6V Clip Holder Case Box w Lead | eBay

I soldered the old connectors from an old Fanuc battery pack into a cable wired in parallel to the D cell bat pack, making sure I had 6 volts to both connectors going to the 2 drives. Now I just change 4 D cell batteries once a year just to be safe. Got tired of the little "clock" batteries they sell you for outrageous amounts. Ok rant over with that.

If someone has seen this issue, and maybe knows what's going on, I'd love to hear about it. Otherwise, I'm going to dig in and tear it down. Something isn't right.

Regards

NPinson

IMG_5318.jpg

IMG_5319.jpg
 

Attachments

  • IMG_5320.jpg
    IMG_5320.jpg
    88.9 KB · Views: 826
Also, in System, SV Mon, A Axis does have a monitor, and noticed there are 5 Alarm slots. ALarm 2 and 3 show some ones (1), but I haven't had an turret alarms at all so I can't say for sure if that's supposed to be that way.




IMG_5320.jpg
 
But for whatever ever reason, the turret shifts radially out, and not by a little. I first noticed it when I tried center drilling, and it was creating a "tit" in the part. Running it manually, you could even see it bending the center drill as it was trying to find center. But the machine has not been crashed. I check the axial runout out again. Now it's around .009 total runout. :toetap:

I would guess the curvic coupling needs replaced. I replaced my original one many years ago. The replacement unit seems to be lasting much better for whatever reason.

Check by indicating your boring bar hole, index away and back, recheck. Index the other way and back - recheck.


It acted kind of funny, it rotated at normal speed but went past it's designated tool station that was called, then suddenly slowed down, did another revolution, and then indexed to the intended turret station.

Say what? :skep:


Somehow, at the moment, that fixed it.

That didn't "fix" anything.


The only thing that I can possible think of is that the curvic coupling halves are not fully seating, and kicking the turret out of position.

The flat part of the couplings are bottoming out on each other instead of the tapers meshing first.


So. These values are all after the turret is seated on the coupling. In my mind, they should all be the same, or at least withing a few tenths from each other. T8-T12 is where it is pretty far out. I don't know if it's possible that the limit switch that senses when the turrent is clamped is set wrong, (ie. telling the hydraulics to stop before it's finished).

The hydraulics NEVER stop pulling. They are either lifting or pulling down.
I wouldn't worry about those values necessarily, but if anything - they could be showing grit in the bottom of the halves bottoming out.

I can get my indicator reading exact up and down for sure, but side to side, the worst reading I had was .01" both ways, so .02" of runout. That obviously isn't gonna fly :confused:

"Exact" in X?
Out in Y? (radially)


I did have the turret numbers get jumbled on me once about 2 weeks after I got it. The turret re zero procedure fixed that.

I hope that you don't have my ghosts cousin? :confused:

http://www.practicalmachinist.com/vb/cnc-machining/ghost-machine-again-320725/





Maybe some parts in the gearbox are worn in there

The worm is of a design that gets thicker (tooth) as you go up the length of it. So as you get lash - instead of adjusting the worm deeper like normal, you need to adjust the shaft deeper in the case to engage thicker teeth. IM/HO it doesn't work well (in this app) as you need to make a substantial move to take up the lash, and the bottom of the shaft wants to hit the inside of the casting/turret housing. You could possibly pull it out and shorten the shaft up some?

But if it's dropping in fine all over - no worries there - at least not until you put a bunch of heavy attachments on one side. ???
To check for lash in the gearbox, try your "Turret Home" proceedure aggin, but hit RESET while revolving. This should leave your Top Plate up. (There is a keep relay that you can edit for that too, but I would have to search all over for that...)
Then you can Gee and Haw the top plate around and see how much lash you have.

However - from my experience - if there is lash in there (push hard) it may not be all that much in the worm, but likely in the 2 dowel pins that are between where the gearbox turns the main shaft and the Top plate. Those need reamed back out to "clean up" and then - instead of puting pins in there the same size - I finally beat pins in there that were maybe .005 oversize! I think I fixed that problem now...


Also, anyone know were the batteries are for these machines? In the first Conquest I bought, the batteries are attached directly to the drives in a small grey case. They look like this.

SHS749 Battery New USA Stock Fast Shipping | eBay

But this recent T51 I have, I cannot find the batteries anywhere? :confused: Both machines have 18T controls, but I think the live milling T51 has an early version 18T. Paging through the control it's much laggier. Just to clarify, one machine, (most recent), and live milling, is 1995 with 18t. The first T51 I bought is a 1999 with a 18T.

The batts you are asking about are not for the control. They keep the encoders alive with the power off. (poor mans "Absolute Encoders") The older machines doo not have these, and thus - you need to HOME the machine after power cycling it. Your '99 you don't have to HOME the slides eh? Personally - I don't see this as an issue at all - other than having to replace the pesky X axis limit switch all too regularly these days. (I don't know what's up with that?)


May want to start reading here.

The 2 dowel pins that I mention are at 10:00 and 4:00 from C/L in this pic.

DSCN0319_zpsfb866508.jpg



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

I am Ox and I approve this h'yah post!
 
Thanks for the reply.

Few silly things I said in there after you pointed them out. When I realigned the 99 T51, yes it's obvious that the hydraulics are always pushing, or pulling. The mesh on the curvic only works that way. I turned my air off to to the machine so the "top plate" it's a turret hardinge, would relieve of hydraulic pressure when air line pressure dropped. Oops


And yes, 99' T51 machine has absolute encoders.

95' T51 machine I have to zero in the morning. However, not the C axis. When I first got it, I assumed I had to. Used to run some Okuma & Howa Act 3's for my father that had C axis, and you had to zero the C. Chuck didn't rotate however, odd.

At one point when I was sweeping the bore of the boring bar block, when the needle was in the "Y" quadrant, I just took both my hands and held onto two opposing boring bar blocks, and tried to twist the turret by hand. I got the indicator to move about .0007 by just doing that. So yea, I'm in agreement that the coupling is just worn.

But, based on that, the coupling, is it designed to be dual contact. Like similar to BIG-Plus tooling?

If not, what would be wrong with relieving the flats, but not the angle where is meshes, on at least one side of the coupling? This might be moronic statement, but is that an option? Worst case, you would still have to replace both halves of the coupling I assume?

As far as batteries go, what backs up the parameters on these machines then? Because the 99' has only had servo drive batteries replaced in the 3 years I've had it.
 
Also OX,
Whats odd with these machines are the hours. I got the 99' from a facility where all the did was hard turning gears. I watched a few of it's brothers making some parts, and they usually ran 2-3 tools so I imagine over the course they did't have a enormous amount of tool changes. Like say a C axis lathe. The 99' has the big bore option, but with out a tailstock. Spindle bearings, ballscrews still sound excellent. Cosmetically, it doesn't look bad, but not nearly as nice as the 95' I just got. But the 99' has 15,3xx hours. The 95' has 4,8xx hours on it right now. Both just run great. But the large hour difference, it seems odd that that coupling would be worn in already at under 5000 hours. Now I don't know if the original owners replaced the coupling at some point on the 99'.

One thing that irks me a little is the 95' is missing the eccentric bushing for the tailstock. I called Hardinge and they want $1800 for that, but worse, a 6 month lead time. I'm considering clamping on to a Mt5 reamer with an S20 collet, and opening that hole up and just reducing down to a MT4 live center. I'm I crazy for thinking that? I would have to find a way to control the hydraulic forward on the tailstock in some way. Thought about using some form of hydraulic ram forcing against the tailstock body, then carefully reducing pressure on the ram to allow the tailstock body to advance, slowly. I got a sweetheart deal on the 95', so I don't mind spending the bucks to do it right. But the cheapskate in keeps me thinking too. I did get in contact with the company that had the 95'. It's an aerospace shop up North, and they have 2 more. I asked them about the eccentric bushing. They said they have it :bawling: I bought it from a rigger. They did mention they are going to be selling the last 2 T51's with all the specific tooling with them in the next year. So I'm trying to hold on and just snatch up the other 2 machines and the tooling. My machine came with 5 radial live holders, and only 1 Z axis live holder. I'm keeping my eyes open for more on eBay, but if you happen to know of anyone that has more Z axis holders, I'm interested. Even partial trade on X axis tools if interested. I have ran them all now so far, and the bearings sound great in all of them as well.

Regards,
Npinson
 
At one point when I was sweeping the bore of the boring bar block, when the needle was in the "Y" quadrant, I just took both my hands and held onto two opposing boring bar blocks, and tried to twist the turret by hand. I got the indicator to move about .0007 by just doing that. So yea, I'm in agreement that the coupling is just worn.

:confused:

There should be too much hydro pressure on it for you to move that by hand [I would like to think]. Is your "System Pressure" back on the hydro pump in the green zone?
If not - that could 'splain why you think the top plate seats slower?


But, based on that, the coupling, is it designed to be dual contact. Like similar to BIG-Plus tooling?

No - there should be daylight between the crest/roots.


If not, what would be wrong with relieving the flats, but not the angle where is meshes, on at least one side of the coupling? This might be moronic statement, but is that an option?

You could try that. I don't see why it wouldn't work.


As far as batteries go, what backs up the parameters on these machines then? Because the 99' has only had servo drive batteries replaced in the 3 years I've had it.

K'ant say for sure. I know that I replaced one on my 18i once (in the last 15 years), but I'm not sure if I have had to replace the batt on the 18 ever (1997) and I'm sure that I never changed one on my 16 - and I've had that for over 10 years. With that said tho - the 18 is powered up (if not running) most all the time. The 16 - not as much. I'd say that others can help with that better than I can - but I'm not sure why to bother changing them if you're not getting an alarm - unless you were puting in storage.



I'm not much help on the tailstock. The only CNC lathe that I have with a tailstock swings 40".
Otherwise I have all sub-spindles.


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

I am Ox and I approve this h'yah post!
 
Yea, I would have thought too much hydro pressure as well to see any movement. If my own body can get it to flex .0005 or more, what's it doing taking a roughing pass. Since I have started using it, I have not done much on center work, mostly OD tools. But I have not had any problems with size control. But I tried the same thing on the 99', and I couldn't get that turret to show any movement at all on my indicator when trying to move it by hand. Even when I did the turret alignment, I had to loosen all 12 bolts fastening the top plate to it's coupling half. Either put the machine in E-stop condition, or bleed out it's air so the hydro system turned off. Then I clamped in a 1x1" 30" long solid bar into one wedge bar station, and it was everything I had to nudge it back where it was on center again.

But both machines show system pressure in the green. Can't remember the PSI exactly, but it was toward the top end of the green area on the system pressure gauge.

One thing I may try this week sometime is try to take the hydro filter out and inspect it. I haven't had any issues with chuck clamping force. And adjusting the pressure for the tailstock, it still moves slower or faster based on what it's set at. But it's obvious that it's slower. I would have to time the difference, but it's a good 2.5-3 seconds for the turret to reseat after it's done rotating. Even thinking about that, makes me hopeful maybe something is not right with the top plate cylinder. I have not had any turret unclamped alarms yet, but makes me wonder if the cylinder can't pull the two curvic halves fully together because of an obstruction. I've never taken these turrets apart yet, much less one with the live tooling.
 
Not an actual "cylinder". It's built right into the body.

Should clamp in 1 second.

There is a by-pass on the filter. If the filter get's plugged, it just goes around it.

I have a sub collet actuator that is slow, and I haven't figgered it out yet. Maybe lazy valve?


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

I am Ox and I approve this h'yah post!
 
Well, I don't know right now. I checked everything again and 3 tool stations are reading right now. Worst I had was .0013 total and the best was .001

I had to drill 5 1/2" into steel this evening with an 1" twist drill, and it never caused any problems. Checked it periodically and no change. Tomorrow I'm going to be doing some heavy roughing in steel with a left handed OD tool, so I'm going to check right after that's done.

Checked the hydraulic pressure. All good there. T11 was one station that had excessive runout that I was talking about yesterday. I indicated it this morning and voila, somehow it was good. It was .0015, then I just backed the turret far enough away that tools would clear, and jumped to T12, then back to T11 and checked again. Same. Also to T10, then back to T11. Still the same. So I'm trying to create a situation
that caused it to be off so much again.

So that's the update at the moment. I think I have a spare solenoid that came with the 99' T51, so tomorrow I'm going to track down the solenoid that clamps, unclamps the top plate, and possibly swap them to see if the turret reseats faster. I'm also going to check that the 12 bolts are all tight on the face of the turret. If someone else before me aligned the turret and forgot to tighten everything up, it could be shifted without too much pressure from roughing etc. But that can't explain why it's off, then it's back and good again.

Is it possible that there is some debris between the mating couplings, and occasionally causing a center problem?

Regards
Npinson
 
My original coupling was way corroded.
But have had the turret apart a few times recently, and I expected to replace the couplings again back around 2010 (?) when we had it apart, and they looked good then, and still looked good last yr when we had it tout to replace hydro seals. I didn't see any signs of rust to speak of. The fact that the first one seemed to wear quicker than the second, and that the first one rusted and the second one doesn't seem to - makes me wonder if there was an "A" change to the coupling print? (material)

Yes - there could be rust in the bottom of the teeth.

I have WAY more hours on mine than either of yours.
I think that I am up to around 1.3 million cycles.



edit:

Now - it could be that _ that big O-ring seal behind the turret is worn a bit now and doesn't hold coolant anymore - yielding better results? I don't think we replaced THAT seal - so - maybe that is a good thing?

I have become a much bigger fan of "giving the coolant somewhere to get back out" more than "trying to keep it out to begin with". The fact is that it WILL get in. Try reasonably to keep it out, but make sure that if it does get in - that it can find a drain!

This is for anything - turrets, thrust blocks, even liquitite. A hole at the bottom of a droop of Liquitite is a GOOD thing if it has either end anywhere near coolant!

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

I am Ox and I approve this h'yah post!
 
This thread kinda goes all over. I've seen curvics that were just all crusty and needed to be cleaned up.

Where are you in Central IL?
 








 
Back
Top