Running Taig at 6,700rpm all the time

[saturnnights]

I cut tiny, itsy-bitsy parts with 1/16" end mills (sometimes 3/32") and since I need a nice finish (in brass), I run my Taig at 6,700rpm for the entire process. These parts typically take an hour to cut. Lately, some have been as long as two hours. I know that these bearings are rated at 10,000rpm or more, but in a realistic, practical world, what sort of life expectancy do you all see with your bearings?
The reason I'm asking is that I have two Taigs and the first one's bearings died at only something like 50-75 hours. That is probably a freak incident and Taig was nice enough to replace the bearings for me. But they did tell me that I should expect at least a couple thousand hours at that speed.
So, for those of you who run at 6,700rpm or higher, what has been your experience? So far, this was my only bad bearing experience, but I'm just sitting here drinking my morning coffee and wondering...

Mark

[900steve]

i can't reply with first hand experience, but here are a couple of interesting links:

http://www.timken.com/products/beari.../calculate.asp

http://www.bearings.machinedesign.co...ech6_60-1.aspx

i'll be interested to see what you all have to say since i'm a prospective Taig CNC mill owner.

-steve

[NC Cams]

Bearing life ESTIMATES (AKA calculations) are based upon several assumptions.

These include perfect radial loading. Add axial loads and life calc changes.

These include NO grease contamination. Reality is often far from that.

These include NO abuse (IE: spindle crashes, radial/axial brinnells,etc). In reality, stuff happens.

When trying to determine "WHY?", one needs to figure out what caused the bearings to fail. Bearing service engineers do this all the time at the factory level. At least I did as that is where new engineers got their start - doing failure analysis.

Short term, infant mortality, bearings failures are/were usually traced to installation and/or machining deficiencies. (IE: bearings brinnelled due to hammering at installation, improper press fits and/or housing/shaft geometry deficiencies, etc - these are often called "off conditions" for political correctness, in reality these are FUBAR's)

Assuming the bearings were properly installed in the first place into properly sized and/or prepared housings(don't count on it), the most common cause of bearing failures is caused thereafter by grease contamination due to dirt or coolant contamination.

The next most common cause is gease life failures. HIgh speeds really take their toll on greases. Thus, a general purpose grease which is OK for most work will NOT be the optimum choice for high speed use.

High speeds raises operating temp. High temp fries/oxydizes grease. ENough said. THus, continuous high speed use may necessitate the cleaning and regreasing of bearings more often. Don't just pump them full of grease either, too much grease causes heating problems due to excessive viscous drag induced overheating. If relube, do so with proper amount and type of grease, expecially in your case.

For example: Chevron SRI is an excellent GP grease. However, for high speed use, a synthetic high speed machine tool grease from Kluber might be a better choice. COntact a lube engineer at a grease distrutor for a recommendation.

Keep in mind that bearings are RPM rated. The rating is LOWER for greased bearings than it is for bearings run with recirculation oil. Open bearings have higher speed rating than sealed ones (as in rubbing seal). Reason: friction of seal causes heat which affects grease life.

Many machine sellers say that lots of stuff SHOULD do this or that but the realities of the design, maintenance and actual useage environment REALLY determine what WILL happen.

Finally, continuous high speed bearing use necessitates some precautions. You should NEVER, EVER fire up the machine and let it wail.

It would behoove you to start it up and let it run at a slow speed until it develops some operating temp, THEN ramp it up to speedin steps. When you DON"T have to run it, slow it down at least. Simply letting it run uses up the available amount of revs that are in the part.

Also, you want to apply the load slowly and NOT max it out in a all in fashion. For infinite life, you only run at 10% or so of the rated load. IF you run near or at 100% rated load, you'll NEVER see rated life - won't even get close.

If you check out load life calc procedures for ANY bearing, you'll see that theoretical life is a proportional funtion of speed and a disproportional function of load. Double the speed, halve the life.

Change the load and the life drops of disproportionally FASTER the HIGHER the load... If you look up the bearing P/N's you have and do some math, you'll see what I mean.

Keep in mind that the above is a general explanation. I don't know what bearings are in play or if they are the right wrong or indifferent ones for the application. I surely can't tell if any other of the critical areas that affect bearing life were infringed or violated.

In the mean time, I hope this helps.

[cartertool]

They will last a long time. Things that kill bearing life are solvents trickling into the bearings (like WD40), taking very heavy cuts, crashes.
I used the first CNC mill headstock for years without problems, at the highest speeds.

That said, bearings usually work until they fail...

[saturnnights]

They will last a long time. Things that kill bearing life are solvents trickling into the bearings (like WD40), taking very heavy cuts, crashes.
I used the first CNC mill headstock for years without problems, at the highest speeds.

That said, bearings usually work until they fail...

Hi Nick

Well, if I have any future problems, I now know where to buy new bearings

Mark

[NC Cams]

Ahhh, yes, they (bearings) will last until they fail

BUT

Why do they fail prematurely????

That is the question that REALLY seems to stump even machine tool builders some times.....

By the way, the "adjustment factors" listed in the following link

http://www.bearings.machinedesign.co...ech6_60-1.aspx

are all fractions LESS than 0.9999999999999999 due to the fact that they all REDUCE the rated life calculation denoted by Ina found on same link