Spindle Run-out Inspection Rig

[MarcodC]

Hi Everyone,

I urgently would like to get some of your suggestions regarding several methods that one can use in order to inspect shafts/spindles that my company is receiving from our suppliers.

I'm aware of supporting a shaft in a suitable V-Block and using a dial indicator, but there are several conditions that could be present, e.g. Ovality, Taper, Bending, curved or Bowed, Eccentricity and Concentricity.

I have to design a suitable Inspection Rig for our quality department to inspect these parameters. Any of your suggestions or recommendations would be greatly appreciated.

Hope to be hearing from you soon.
Thank you.
Best Regards,
Marco.
Johannesburg.
South Africa.

 

[MikeHalloran]

- Sketch out a design for each fixture/ setup you would need to inspect each identified parameter to the specified tolerance, for each and every shaft or spindle drawing that you have ever issued or used.
- Then search for common elements among those tools, and buy the minimum complement you would need to make every one of the required setups.

;---

The assignment to design a 'rig' suggests that you inspect a lot of shafts, which means you could and probably should be looking at a fair amount of automation. I can provide one example of what's possible beyond v-blocks.

Forty years ago, I worked in a car axle plant. We had one really cool tool called a 'Rondi-Gage'. You would place an axle shaft in a pair of v-rests, and then press a button to start a cycle, which would:
- Engage a pair of center points to pick up the drilled centers in the ends of the shaft.
- A measuring head would descend on the bearing diameter, and a pair of pins would engage the flanged end, and spin the shaft slowly for exactly two revolutions.
- Sometime during the cycle, a circular chart maybe 300mm in diameter would also start rotating, and a pen would trace out a greatly exaggerated representation of the bearing diameter's profile at the plane of measurement, with resolution down in the millionths of an inch. The whole cycle took less than a minute.

[ If an operator had defeated the shaft grinders' "sparkout" setting (in order to process 8 hours worth of axles in 6 hours and create idle time for playing checkers and socializing), there would be a noticeable bump in the roughly circular trace on the chart. The bump would not be detectable with a dial indicator, but would result in shortened life of the bearing pressed over the shaft diameter. ]

That's what was possible four decades ago. You can probably do a lot more now, but what you _have_ to do depends on what's on your drawings. The level of automation you can afford depends on how many units you make and inspect.





Mike Halloran
Pembroke Pines, FL, USA

 

[MarcodC]

Hi Mike,

Thank you for getting back to me.
I've heard of an automated device very similar to the one you described at the car axle plant. Unfortunately, I have to find a simpler layout for our inspection guys to use.

We do inspect a fair bit of shafts, but not enough to grant state of the art inspection equipment. We actually only need to check the run-out of these spindles. Is there a specific correct way to check run-out? What i'm trying to say...is using a dial indicator suitable, whereby the shaft is mounted on V-Blocks? Our spindles also have several stepped down diameters. Should each diameter face be inspected for run-out, or should the shaft be mounted at the bearing journal surfaces, and checked for run-out at the other end of the shaft?

Best Regards,
Marco da Costa.

 

[unclesyd]

The Cadillac of the shaft measuring devices. We had 2 very similar devices that we sold as scrap when our research group dissolved. The salvage people were real proud that they got $25.00 for some junk.

http://www.marposs.com/library/m67_english.pdf

We used Bench Centers to check our shafts. The metrology group added a few more fixtures to facilitate measurement of more features of the shaft. If you look at the Marposs gage you can the possibility of additional instruments.

http://www.subtool.com/tp/9205.shtml

Ia am trying to find my literature on the making of a ring gauge to check camber of a shaft. This is a very simple tube of a certain size based on your tolerances that will act as a go or no-go gage.

 

[dgallagher]

As well as static tir, it is best to monitor dynamic spindle runout. A very good system is offered by Lion Precision. Here is a link:

http://www.lionprecision.com/targa/index.html

http://www.lionprecision.com/targa/targaarticles.html

Dave G.

http://www.ReliabilityDirectStore.com

 

[MarcodC]

Gents,

Thank you for all the input. I'll definately have a read through all the articles. I'll get back to you regarding my thoughts and opinions.

Speak to guys soon.
Best Regards,
Marco.

 

[GageDr]

Marco,

I am currently working on a project like this for a customer that machines shafts, pinions, and side gears. If you are only inspecting one or two items per part, it is probably more cost effective to locate the part between centers and use dial or electronic indicators.

If you are looking for multiple checks per part, with the use of a multi I/O system (metronics gage-check) and analog or digital probes, you can check multiple parameters with common probes.

For instance, you can have two opposing probes that check a diameter, and use one of those probes to check the run-out. At the same time those two probes can be compared to two more probes on another diameter to check the concentricity of the two diameters.

http://www.triadgages.com

 

[GageDr]

Marco,

I apologize for a possible miss statement I may have made in the previous post. I said, "...it is probably more cost effective to locate the part between centers...". Locating between centers may not be an option. How you locate the part depends on the datums specified on the drawing.

If you have a call-out on your drawing for run-out, it should specify to which datums and what the tolerance should be. If the GD&T specifies run-out to the centerline of the shaft, then centers are the proper way to go. If it specifies run-out to another diameter then the standard would be to locate that diameter(s) in VBlock(s).