What Price Machine Rigidity?

Machinists know that there is no substitute for beef when it comes to the rigidity of their machine tools. That made me wonder whether we could verify and quantify this notion in some way. How much beef does it take to achieve a level of rigidity?

I frequently recommend to G-Wizard users that they think of rigidity in terms of horsepower. Horsepower from the spindle is what pushes against the rigidity of their machine, and barring harmonics (chatter) excited by certain vulnerable frequencies (which are a factor of both rigidity and damping), rigidity may be the limiting factor for many machines, especially small ones.

What I was looking for, then, was the relationship between horsepower and the weight (beef) of the machine. So, I fired up Google and started cranking horsepower and weight specifications into a spreadsheet for various commonly seen CNC machines. The results are striking:

The relationship between horsepower and weight holds up pretty well: about 600-1000lbs per HP on average…

I was pretty surprised at how well the numbers trend. There are some outliers–the knee mill and an older Cincinnati, for example, are way up there at over 1000 lbs per HP. In the knee mill’s case I know it doesn’t take advantage of all that weight because my IH mill is just as rigid as a brand new Bridgeport knee mill based on some testing I’ve done. My mill, BTW, has at least 150lbs of extra weight in the form of an epoxy/granite fill, and it provides a noticeable improvement in rigidity. I conclude that 600lbs per horsepower is the lower end of “professional grade” rigidity, 1000lbs per horsepower is the upper end for well-made beefy machines, and 800lbs is perhaps a happy medium.

I had plotted a horizontal mill in the data out of curiousity, but it was a major outlier.

I will speculate that the weight requirement declines at higher spindle speeds because the mass that contributes to rigidity starts to be a greater percentage of the machine’s overall mass for those machines, and hence that is perhaps a truer number. After all, the weight of the motor sitting atop the spindle isn’t contributing much to rigidity!

For pros, you might consider plotting this information when trying to choose between several different new machines to purchase for your shop. Some machines have a reputation of being better for aluminum because they’re too light weight for more demanding jobs. You should be able to see some of that from this kind of data.

You can get a sense of the performance of some very light hoobyist machines too. Consider:

- A Sieg X3 (Grizzly G0619) weighs 418lbs with a 1HP spindle = 418 lbs/HP. Not bad, but you might derate the 1HP.

- A Sieg X2 (Grizzly G8689) weighs 149lbs with 3/4 HP = 199 lbs/HP. That’s getting iffy. I would definitely derate to perhaps 3/8 or even 1/4HP if rigidity matters at all.

- A Grizzly G0720 (like the G0704 Hoss is modding?) weighs 641lbs with 2HP = 320 lbs/HP. I would again, derate this mill slightly. To get to rigidity similar to the professional machines would mean derating to 1 HP, for example.

There are even tinier mills out there. Taig says their mills weigh 65lbs. Forget horsepower, lets measure in watts for a machine like this. If 600lbs/HP is right, and 1 HP = 735W, then 1lb = 0.8W. About 52W for the Taig. Any more will start to compromise the machine’s rigidity.

I should note that after numerous conversations, the average hobby machinist doesn’t like to push their machine too hard anyway, so they may instinctively be avoiding the edge of the rigidity envelope. Still, it’s useful to know when a cut should be watched or reconsidered. Therefore, I will introduce a feature in G-Wizard before too long that is aimed at setting realistic horsepower limits for smaller machines so they perform better within their envelope of reasonable rigidity. It isn’t that you couldn’t take more aggressive cuts, it’s just that when you do, you should be aware they are not conservative, they will challenge the machine’s rigidity, and as a result, you may start to have problems with accuracy, chatter, and tool wear (or breakage in a worst case). To provide this kind of input, I have a little more mathematically sophisticated way of modeling the over-simplified version I give here. This is some research I’ve been doing for quite a while, but I wanted to show a little preview.

 

Related posts:

1. Machine Tuning as Preventative Maintenance
2. Carbide vs HSS and Chatter for Small Mills
3. Deep Pockets

4 Comments

• 
  Robert 2011/03/26 at 2:53 am

  That’s a great way of looking at the limits of a machine. I started out with a Sherline a long time ago with no CNC experience. G-Wizard with that feature would have saved me a lot of experimentation.

  Robert

  Reply
• 
  Robert 2011/03/26 at 2:53 am

  That’s a great way of looking at the limits of a machine. I started out with a Sherline a long time ago with no CNC experience. G-Wizard with that feature would have saved me a lot of experimentation.

  Robert

  Reply
• Making G-Wizard Better for Beginners and Small Machines « CNCCookbook 2011/04/01 at 3:12 pm

  [...] second feature we’ve added for smaller machines is Rigidity Compensation. We recently published some of the results from a long-running study we’ve been making into finding good proxies for machine rigidity. A [...]

  Reply
• Making G-Wizard Better for Beginners and Small Machines « CNCCookbook 2011/04/01 at 3:12 pm

  [...] second feature we’ve added for smaller machines is Rigidity Compensation. We recently published some of the results from a long-running study we’ve been making into finding good proxies for machine rigidity. A [...]

  Reply