Proximity sensor capable of sensing flat spots on shaft?

[leeboy4130]

Hi, my question is if a proximity sensor would be capable of sensing flat spots ground into a rotating shaft. For example, if the proximity sensor (rated at 8mm max air gap) were mounted so that the face of the shaft were 7mm away, would a flat spot ground to a depth of 6mm be enough to switch the sensor (meaning this low spot would be out of range to the sensor)? For reference, the shaft is about 4-5/8" in diameter, made of 4140 steel, and is rotating at 2100 RPM. With 4 flat spots (for balancing) that would be a pulse frequency of 140Hz. I was looking at inductive senors but with it being 4140 I believe my air gap would be too small after the correction factor.

This would by the cleanest way for me to sense the RPM of the shaft, otherwise I would need to sense the attached gearbelt via another setup.

Thanks,
Lee

 

[1gibson]

Some reason you can't use an optical pickup and paint the flat spots?

 

[leeboy4130]

There is a constant mist of synthetic oil that is sprayed in this area. That might gunk up the optical sensor, do you agree? Also, my boss is somewhat leery of optical sensors in general. I was hoping to find an inductive or hall effect type sensor for the job, just not sure if a flat spot would be enough of a gap change to switch the sensor. The shaft I would be grinding on is high priced so I can't easily do a test run. When I call sensor manufactures I get someone reading off the same information I already have (standard target size, frequency, etc.).

 

[Compositepro]

Magnetic speed sensors are typically used with a gear (commonly 60 teeth) to generate a pulse train. You can get split gears that clamp onto a shaft. Red Lion has a variety products for your application, as do other companies.

 

[GregLocock]

Use a pair of magnets in a collar, please.

Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376

http://eng-tips.com/market.cfm?

 

[MintJulep]

It might be possible.

It will probably work perfectly when adjusted perfectly before you ship.

It will be a maintenance nightmare and unreliable in the real world.

If shaft speed is important, then it's important enough to deserve a dedicated system to measure.

 

[dhengr]

Leeboy:
Why not set a laser beam perpendicular to the shaft and just outside the radius to the flat spot on the shaft, with the source on one side of the shaft and the target on the other. You would get an indication every time the flat spot matched the laser beam line. And, if you set the laser warm enough you would ultimately trim the shaft to that length too.

 

[tygerdawg]

The oil mist & associated gunk may be problematic, but look into something like Keyence's capacitive analog-output displacement sensors. Looks like a prox, but gives a precise analog signal across a very tight range. If this shows promise, then you'll have to deal with
[ul]
[li]the typically short sensing range[/li]
[li]whether your control system can handle the frequency, response time, and signal hysteresis[/li]
[li]sufficiently discriminate the displacement change[/li]
[li]electrical interference issues[/li]
[li]and probably a few other things I haven't remembered[/li]
[/ul]

These things only work well on ferrous metals, though.

The resolution is typically less, but another option may be ultrasonic sensors (try Banner Engineering, SICK, others). I got a very nice little ultrasonic analog-output sensor from Baumer (Swiss, I think) that was designed for sensing the liquid level inside test tubes. I was surprised at the resolution (+/-0.1mm if I recall).

TygerDawg
Blue Technik LLC
Virtuoso Robotics Engineering

http://www.bluetechnik.com

 

[IRstuff]

Your approach to use the rated performance is misguided, since the requirement is simply a threshold above which the manufacturer can ship the unit.

However, there are sensors that can detect that level of change:

http://www.omron-ap.com/product_info/E2C-EDA/index.asp

supposedly can resolve down to 1um.

TTFN
faq731-376

 

[dgallup]

Wheel speed sensors are used all the time but you will get a lot cleaner signal with geometry where the gap varies more rapidly than 4 flats will. You also want to keep the gap down as much as possible within reason. Take a look at a typical wheel speed sensor for automotive ABS for instance, these work in all kinds of rain, snow, mud and are quite reliable.

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

The Help for this program was created in Windows Help format, which depends on a feature that isn't included in this version of Windows.