Center drilling for cylindrical grinding: how deep?

I was doing a nearly final pass on a steel part, about 30mm (1 1/4") diameter and 300mm (12") long on a cylindrical grinder. This was between two dead centers with moly grease, 80 grit white AO wheel, taking off a couple of microns (about 0.0001"). Plenty of coolant, no sparks visible, steady low hum from the hydraulics and the hiss of the wheel. Then all of a sudden, 2/3 of the way down the part, starts periodic pulses of sparks, timed with the revolutions. I can hear it too. I hit the rapid retract, put an indicator on the part, and sure enough, it's not running true any more, 6 or 8 microns (about 0.0003") of runout, that was NOT there before. So I pulled off the part, cleaned the centers and the center drillings, fresh grease and it ran true again. Fortunately the important thing for this part was straight, round, parallel but NOT the diameter, so I was able to grind it down further.

But obviously something happened to one of the center drilling or dead center that threw it out of round. Can someone tell me what? Did some grit or swarf get in there? Or did it gaul internally? Or ??? And how can I avoid this in the future? In this case the center drillings were bigger than usual, is that part of the problem or the cause?

Here is a chart of typical center drill sizes for lathe work (click image to enlarge):



Is this also the right thing for cylindrical grinder, where the load is less? Or is smaller better to avoid problems like the one I describe above? I'm slowly learning how to use a cylindrical grinder, and am pretty sure that the experienced old hands here can tell me exactly what I did wrong and how to keep this from happening again.

You might try mari grease but what you used is OK.
MTL-NBU30 Assembly and Dead Center Grease MariTool
I like to feel tightness and re grease at last .001/.002 or so. Heat can make part and centers swell a bit and ring out the lube.

It may be that the part cooled with the lighter finishing pass and shrunk it length, allowing what had been an adequate fit to the centers to become loose.

To check this, put an indicator on the end of the part, then push it lightly on axis, each way. You should not see any appreciable movement (over a few microns), if you do then adjust your centers.

Edit: If using a true center drill, the taper may not be perfectly conical and in worse case can have two small steps from stopping and removing the drill too quickly. If doing critical work, it might be worth using a center lap to prepare the hole to a better degree of accuracy.

Straightening tip: I like to micrometer the part then come in on my hand wheel dial numbers to hand feel the parked wheel to the part. This so finding or adjusting to feel the needed difference to make part straight. Yes out of round need be considered. This is often faster than using a indicator and going end to end..
Say part is .0015 big at tail end pre-grind, so needing .00075 take from tail end dial. Most grinder dials can read very close. Yes with plenty of take-stock one might take a test grind and adjust from there.

Some materials and wheel's break down and so may require thought of where to in-feed, perhaps the wheel might break down .000000? with a long part and in feed at one end or the other. perhaps the in-feed might be at part center and and travel right or left a few times.

Don't know if it would help this problem, but I like radius centers. They're harder to find but take care of misalignment.

Conrad Hoffman said:

Don't know if it would help this problem, but I like radius centers. They're harder to find but take care of misalignment.

Click to expand...

Bell centers, mc mastercarr carries them

Milland said:

It may be that the part cooled with the lighter finishing pass and shrunk it length, allowing what had been an adequate fit to the centers to become loose.

To check this, put an indicator on the end of the part, then push it lightly on axis, each way. You should not see any appreciable movement (over a few microns), if you do then adjust your centers.

Edit: If using a true center drill, the taper may not be perfectly conical and in worse case can have two small steps from stopping and removing the drill too quickly. If doing critical work, it might be worth using a center lap to prepare the hole to a better degree of accuracy.

Click to expand...

The tailstock is spring loaded on a center grinder for that reasson isn`t it

Peter

Peter from Holland said:

The tailstock is spring loaded on a center grinder for that reasson isn`t it

Peter

Click to expand...

Are all? That implies some potential for slop at the sliding interface.

Delw said:

Bell centers, mc mastercarr carries them

Click to expand...

A bell center drill (combined drill and countersink) protects the end face, but isn't usually radiused. What I like are type R so the contact area is a ring.

Thanks for the replies!

Buck - I'll look for some of that grease. But as you said, the grease
I used was OK. I was taking off almost nothing, total of perhaps 10
microns, less than 0.001". The part was not even a bit warm, and when
I took it off the centers there was plenty of grease there.

Milland - the tailstock is spring loaded, and the part was not in any
way loose. It's not shrinkage. No axial play as you describe. Center hole
had been lapped with a 150 grit lapping point. But I think it might
still have some annular steps inside because it's too big (diameter
and depth). Note also that the spring loaded tailstock is a perfectly
lapped finish in the barrel. There is zero play even with a micron
indicator and significant side pressure. Absolutely zero. Remarkable
Swiss construction!

Buck - I was always infeeding on the left (workhead) side because
after grinding a cylinder section I needed to swivel the table to
grind a taper which has a precise transition on the left. I was
keeping the left side of the wheel pristine for that. So I was being
attentive to only break down the right corner of the wheel as I fed
in. But don't think this explains why suddenly the part shifted on
the center. Total amount I had removed with a freshly dressed wheel
was a few microns, a couple of tenths imperial.

Conrad - I'm honestly not sure if ball centers are a good idea for
cylindrical grinding. If I use those, then the dead center is
resting on a ring of contact rather than a cone. This may force out
the grease from that contact ring or make it easier for grit/swarf/coolant to enter
alongside. It's not mentioned in any of the books or articles I have
read about cylindrical grinding.

Peter - Yes, you are right, this is not part expansion or shrinkage! The tailstock spring is exactly for that.

Experts - My original question remains: what is the best size for center drills? Could an oversize center-drill mark be responsible for what I saw? Is smaller better than bigger here?

Cheers,
Bruce

Hello Ballen:

I have conducted a garage survey of center drill holes. The 3 inch diameter Ford tractor axle uses a 1/2 inch "C" dimension hole with two 5/8 inch diameter face drive pin holes.
The 2" diameter Lamina die guide pin uses a 3/8 inch "C" dimension center hole with a single 1/4 inch diameter face drive pin hole.
The 1.5 inch diameter Lamina die guide pin uses a 1/4 inch "C" dimension center hole.

The #4 Morse taper drills are using 3/16 inch diameter "C" dimension holes.

All of these bars have holes larger than the recommended size in your chart . The Lamina pins are used with preloaded ball bushings and have very small tolerances for out of round error and taper error.

One observation: None of these center ground bars was driven with a lathe dog. There were all face driven either with a pin or a tang drive.

************************************************************************************************

Have you eliminated the lathe dog as a cause for the bar lifting off of the center?

Have you adjusted the tailstock spring compression force to insure that the force applied to the bar by the grinding wheel and the lathe dog will not lift the bar off of the headstock or tailstock centers?

The center hole bearing surface should have a minimum contact pressure to maintain a thin oil film. If the pressure is too low the tailstock center bearing will float on a thick oil film and the bar center line will shift with grinding wheel force . This would cause a unintended taper to be ground on the part. If it is too high the oil film will be lost. The larger than recommended center holes will permit higher grinding forces to be applied to the part without loosing the oil film.

Hi Robert,

Robert R said:

All of these bars have holes larger than the recommended size in your chart . The Lamina pins are used with preloaded ball bushings and have very small tolerances for out of round error and taper error.

Click to expand...

Thanks for this. I have also been looking at a lot of center holes recently.

One observation: None of these center ground bars was driven with a lathe dog. There were all face driven either with a pin or a tang drive.

Click to expand...

My part was also face driven with a pin. It's a test bar, and I wanted to go right to the end without flipping it end-for-end.

(When I first started cylindrical grinding, I screwed up a few parts because the drive was not good and applied forces in the wrong direction. So I'm very attentive to that now.)

Have you eliminated the lathe dog as a cause for the bar lifting off of the center?

Click to expand...

It's a good question and the answer is yes. There is no lathe dog, and no misdirected drive forces. That's not the issue.

Have you adjusted the tailstock spring compression force to insure that the force applied to the bar by the grinding wheel and the lathe dog will not lift the bar off of the headstock or tailstock centers?

Click to expand...

Yes. The force applied by the grinding wheel is downwards, and much less than the force due to gravity. But again, the question is a good one. The truth is that I don't know any systematic procedure for setting the tailstock spring force. I typically make it as small as possible consistent with keeping the part firmly in place. Is there a better way? Buck, Cash, if you are reading this, is there some rule-of-thumb that you use for tailstock spring force?

Your final remarks on contact pressure are right on the money. If the center and the drilling both had perfect 60 degree angles then I can easily estimate the contact area of the ring where they touch and the pressure (spring force/contact area). Unfortunately if they are not exactly the same angle then the contact area is less, and hard to determine.

What's a reasonable maximum pressure before the grease film will be lost? I am using Liqui Moly LM47 long-life MOS2 grease Liqui Moly 3510 LM 47 Langzeitfett + MoS2, 100 g: Amazon.de: Auto According to the data sheet this is a lithium based molybdenum disulfide grease, NLGI class 2. I think it's a typical garden-variety MOS2 grease. The data sheet says that in use the contact surfaces are embedded with a thin layer of MOS2 which will continue to act as a lubricant even if the grease film is lost.

Meanwhile I trimmed the length of my part on both ends (parallel test bar with a shortened MT4 taper end + nub on the taper end) and made new center holes. Fortunately the bar is not too hard. In the new center holes I followed the guidance of the table in my first post. Since the diameter of the part is about 30mm = 1 1/4" I used a #3 center drill with diameter C about 5mm = 3/16". (I did it on my Deckel FP2 in horizontal mode, using a centricator to find the center. So the new center marks have 4 microns of TIR, less than 0.0002". I just need to tickle the bar to make these the true centers.)

Thanks again for your thoughtful comments,

Bruce

Conrad Hoffman said:

Don't know if it would help this problem, but I like radius centers. They're harder to find but take care of misalignment.

Click to expand...

Just a +1 on this. I like those also.

Next best are the center drills with the little extra chamfer, they tend to protect the center from handling nicks. Plus they give you a definite stopping point, so all the centers will be the same depth if you're doing a batch of parts.

Riten has a chart showing the stress acting on the center for a given supported part weight and the corresponding recommended center size.
http://www.riten.com/wp-content/uploads/2014/12/hole-center-specs.pdf

The message of the chart is that a 60 degree arc of the dead center surface area is supporting 1/3 the part weight.
If a grinder is supporting a work piece that weighs 100 pounds and the grinding wheel is applying a force that is in the range of 10 pounds then the area of contact will not move significantly. The center of rotation of the part will not shift.

If the part weighs 10 pounds and the grinding wheel applies 10 pounds of force to the part the area of contact on the dead center will shift 45 degrees. There will be a corresponding shift to the part center line.

The solution is to apply a significant preload to the dead center spring. A 100 pound preload on the 10 pound part would insure that not only would the center line not shift but that the contact area on the dead center would be a 360 degree arc rather than a 60 degree arc. The operating stress on the center and the grease would not change significantly in spite of the high preload because the area of contact on the center had been increased by roughly 6 times.

Milland said:

Are all? That implies some potential for slop at the sliding interface.

Click to expand...

Most are and that shift from thermal length means hopefully following the true part axis.
Not sure how many here are using micron filtered grease but not doing that does give a "jump" as described.

I'd more suspect hole finish as I see nothing about lapping, fine finishing, polishing it.
A very tiny bit tears off or worse yet "rolls over".

No way a 60 degree hole and center fit each other. Both have manufacturing tolerances.
If this hits inside or outside life can change. (outside means needing perfect square ends)
So we have carbide ball centers, line/circle contact and high pressure per area which leads to other oh-poops if the part is not very hard.

Looking at the ceneterhole and center after the bad grind with a microscope may lead to answers.
Bob

I like using a radius type center drill, after "spotting" a recess.
If grinding after HT be sure to use a center lap to clean things up before grinding.

Did you use HSS or carbide centers?
Sounds to me your part resonates with it's rotation speed and center pressure. If it happens again try a lower speed of the part.