Mounting a rare earth magnet in a thin plate

Problem statement:
Mount a small (nominally 3.2 x 1.6mm) NdFeB magnet in an aluminum model airplane spinner backplate (for a tachometer).
The thing needs to work with a model airplane engine, so it's going to be a high vibration environment.
The easy to get magnets (which is what I have) are nickel plated, so they're slippery as hell. They're also not machined to super-tight tolerances, and NdFeB is brittle as hell (it's optimized for magnetic hardness, not physical strength!).
The best I can think of is to drill the hole to a slip fit and then epoxy. I have visions of actually achieving a press fit, then watching magnet after magnet crumble into clinging dust trying to press them in.
Suggestions?
--

Tim Wescott
Wescott Design Services
  Click to see the full signature.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

In the alarm industry rare earth magnets are often used for problem or difficult installation situations. Perhaps a magnet like this with a screw mounting hole might be of some use to you.
http://grisk.com/recessed/mighty_mag.html
I would think even using epoxy and a light interference fit with the resin squeezing up through the hole, or the use of a screw would work fairly well. To be fair, we do not typically use these in a high vibration environment in the alarm industry, but magnetic reed switches and magnets are often used on machinery where abrasive grit will work into most other types of switches and destroy them. I would expect there is some vibration in those environments.
Anyway, the rare earth magnets pictured on that page might give you other options.
Drop me a line if you like.
alarm(underscore)wizard dfkgjs-to-confuse-email-harvesters-asljdprhrtph (at)hotmail(dot)com
I have found when using a raw bare metal rare earth magnet they will not hold up to more than the lightest interference fit. In practice we often get them already glued into a plastic sleeve which will take a rather firm interference fit, and is how they are typically installed such as the magnets on this page.
http://grisk.com/recessed/20rs.html
I would point out that a rare earth magnet may have to wide of a magnetic field for optimum detection by your sensor. As it spins past it will be detected for a wider range of travel. Depending on the sensing speed of your sensor, its reset time, and the distance from the axis of rotation you may wind up with a continuous or nearly continuous activation.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload


I would point out that a reed switch is far to slow to be an adequate sensor for your application. I just used them as a reference item in the uses of which I am accustomed.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

Yup. I use them all the time for Hall-Sensor magnets on my machines. Drilling slightly over-size and potting them completely works sometimes. For more rigorous environments, consider sandwiching them between two plates, each drilled half-way, and with the magnet potted and the plates epoxied together.
LLoyd
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
On Thursday, May 1, 2014 12:12:52 PM UTC-4, Tim Wescott wrote:

Maybe an over size hole, epoxy and then a cover plate/tab that's held with a screw.
What's your detector? Could you use an LED and photo diode?
George H.

Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
On Thu, 1 May 2014 10:15:48 -0700 (PDT), snipped-for-privacy@gmail.com wrote:

Works good with a reflective spot on the back of the spinner and a photo-tach. Or like a lot of the model plane guys do, a shiny tip on the prop, and a phototach reading out at the tip instrad of at the hub - better chance of getting accuracy at speed.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
On Thursday, May 1, 2014 5:35:18 PM UTC-4, Clare wrote:
<snip>

Maybe some of the nice retro reflective tape that 3M makes? George H.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
On Thu, 01 May 2014 10:15:48 -0700, ggherold wrote:

It's in a very uncontrolled environment both for dirt and for light (it's gotta work in full sun, in just about every conceivable orientation), and there's no room to shoot light through an aperture -- it would have to be reflective.
--

Tim Wescott
Wescott Design Services
  Click to see the full signature.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
Tim Wescott wrote:

No aperture needed for a photodetector sensor. Go find an old VCR and rip out one of the IR spindle sensors and its reflector. The sensor is the size of a TO-92 case transistor. You can simply blacken the rear of the spinner and polish 2 spots on it to act as the reflectors. Mill a recess a couple thousands into the back of the spinner for the sensor to tuck into and it should be fine. something like this..
http://www.marktechopto.com/img/products/photo-reflector.jpg
--
Steve W.

Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
On Thursday, May 1, 2014 11:20:44 PM UTC-4, Tim Wescott wrote:

<snip>

Hi Tim, I've never done light detection in full sunlight, so I may be full of shit. But there is some nice 3M retro reflective tape that Phil H. (of SED) turned me on to. As long as you have enough dynamic range in the detector (to take care of full sun) cna't you you then AC couple the signal? (White leds are blinding these days.) (Will it kill you if you get a few false counts as full sun blinks through the spinner onto the detector. That's going to be a very small solid angle) I've no answer for dirt and grease.
George H.

Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
On Fri, 02 May 2014 08:53:21 -0700, ggherold wrote:

I am starting to think in terms of opto -- but usually the decision tree for industrial equipment for optical vs. magnetic starts with "will there be dirt on the surfaces?" -- if the answer is "yes", then the following box says "use magnetic sensing".
The spinner comes with big lightening holes, much larger than the magnet I'm thinking of using. I'm considering making up some high strength epoxy putty with some chopped carbon fiber strand and microballoons, and holding the magnet in with that. It's kind of a cro-magnon solution, but it may work.
Now, if only someone made rare-earth magnets in the form of little screws, I'd be all set!
--

Tim Wescott
Wescott Design Services
  Click to see the full signature.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
On Fri, 02 May 2014 12:53:09 -0500, Tim Wescott

A sort of left hand turn off the topic but why not just use an optical tach for setting static speeds and (perhaps) and acustic tach to calculate the increase in rpm when flying?
One example: https://play.google.com/store/apps/details?id=com.javiery.rpmgauge
--
Cheers,

John B.
  Click to see the full signature.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
On 5/1/2014 12:12 PM, Tim Wescott wrote:

I use silicone, it's more resistant to vibration that epoxy.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
On Thursday, May 1, 2014 12:12:52 PM UTC-4, Tim Wescott wrote:

Maka a steel button slightly larger in dia than your magnet. Press the steel button into a hole and let the magnet field hold the magnet to the button.
While you are doing this make another steel button so that it weighs as much as the first steel button plus the magnet. Press fit this button in a hole to balance the weight of magnet, etc.
Dan
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
On 5/1/2014 11:12 AM, Tim Wescott wrote:

How big is the spinner, Tim?
Is there room to mount a magnet with a center hole - AND a short steel cross bar (arms) inside the spinner?
Getting the magnet located at the center of the shaft would reduce the effects of vibration (impact impulses) and G loads on the magnet itself.
That's going to double the pulse rate, but it would fer-sure be easier to balance!
IF the polarity will allow it, anyway...
What RPM are you talking about?
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
On Thu, 01 May 2014 11:12:52 -0500, Tim Wescott

Easiest way is slip fit and epoxy. Best way is 2 peices, with relief machined in both halves, with the magnet sandwiched between, held in the hole with epoxy or silicone or polyurethane glue,
Or better yet, use radially mounted magnet rod, in hole drilled across the diameter of the spinner, but not all the way through, so the magnet is slung to the outside by cent force, and can NOT get out.
The problem with either solution is balanace - the magnet is heavier than the aluminum removed to put it in.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
On Thu, 01 May 2014 17:31:27 -0400, snipped-for-privacy@snyder.on.ca wrote:

+1

Adding mass to a rotating crank can be iffy, adding considerable vibration and shorten crank life. (Ask Ford with its cut-down-V-8 V-6 diesel. A CA fleet truck dealer told me in 2007 that only two of the hundreds they sold had NOT come back for major engine repairs within a year. I ended up with the Tundra.) Balancing with an equal weight of steel rod 180-degrees out would double that mass and any problem associated with it.
LJ's 2-cents: Optical seems the best route. If it's a dirty enviro, shield it, wot?
--
If government were a product,
selling it would be illegal.
  Click to see the full signature.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

Still... why not a vibrational pickup? Then there's _nothing_ to get dirty, no imbalancing of the engine, and no 'attachments'. Just mount the pickup _near_ the motor mounts and be done with it.
They've got them in teesy sizes that will handle up to a couple of hundred KHz, can consume NO power (some active ones will pull a couple of mA.)
Lloyd
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
On Fri, 02 May 2014 10:38:13 -0500, "Lloyd E. Sponenburgh"

Vibe tach? Nevahoiduvit. Are they available?
--
If government were a product,
selling it would be illegal.
  Click to see the full signature.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

All day long. They're almost a commodity item. Anything with a reliably repeatable cyclic vibration is a candidate for the method.
They even sell sub-$20 elapsed-time meters for gasoline engines that are only activated when vibrated by the running motor.
Lloyd
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

Polytechforum.com is a website by engineers for engineers. It is not affiliated with any of manufacturers or vendors discussed here. All logos and trade names are the property of their respective owners.