Measuring Low Speed with Cellphone 1

[BUGGAR]

I have to design some shock bumpers to stop an 8000 pound X-Ray machine that has to be repositioned frequently to accommodate odd-shaped parts. We will move it on air bearings.
But I will need some means of knowing how fast this thing may move. Before I come up with a fancy engineered solution for measuring how fast this thing is moving, can I just put a cell phone on top of it with the right app and get the speed directly? I’m tentatively looking at a maximum speed of 3 inches per second. Acceleration doesn't matter. I tried asking some young techno people but they didn’t really understand physics so they couldn’t help. The shock bumper reduces the g’s to stop this thing so I can figure that but I need to know the speed based on actual tests.

I’m designing the shock bumpers using auto shock absorbers. There’s a race shop down the street with a shock dyno and a huge supply of used but still good racing shocks. I’ll submit my calculations for this shortly for everyone's critique. I think I can use a shock out of a SCCA Class D Modified, whatever that is.

Thanks all.

 

[Strong]

Google: measure speed with iphone
Example:
Turn Your iPhone Into A Speed Radar
Pasted from <

http://www.iclarified.com/14342/turn-your-iphone-into-a-speed-radar>

Google: measure speed with android phone
Example:
SpeedView: GPS Speedometer
Pasted from <

https://play.google.com/store/apps/details?id=com.codesector.speedview.free&hl=en>

For ultrasound method
Google: ultrasonic distance and speed sensor
Good methods using low cost Arduino computer

The project can be difficult if the load on air bearing has rotation and is free to move in many/any horizontal direction as compared to straight-line motion without rotation. You could check out robotics (Battle Bots) for ready-made components and computer code.
Walt

 

[racookpe1978]

I'd measure distance moved, measure time traveled.

Unless you get some sort of alarm rigged up to tell the "pushers" that it is too fast (that it is approaching a limit, BEFORE they exceed that limit), they still will go too fast by mishandling.

 

[IFRs]

I don't think phones and non-military GPS are accurate enough for small distances. I don't think a police radar gun is accurate at those low speeds either. I'd use the distance / time method.

 

[3DDave]

I think it won't be so easy. I'd add a swivel wheel with an incremental encoder to measure the rate. If the operators are paying attention, two swivel wheels and some pots would give velocity vectors for two points so that the rotation rate and linear rate could be determined. The case to worry about is a direct reversal of a swivel where it will give really odd info that is hard to integrate over the short time of reversal, but as long as the movers know it will be OK.

This is certainly in arduino territory and, unlike the phone, can also automatically apply braking, for example by cutting the air flow to the bearings.

If I was to look for a phone solution it would include visual flow analysis to pick up details on the ceiling, but I don't know that anyone would make an alarm for going too fast. It would also need a distance measurement to calibrate the flow speed.

 

[BUGGAR]

Thank you everybody. The guys at the race shop were telling me about how they race against each other indirectly on select streets with their cell phone results going to a "secret" website. If you want to know, the local hotshoe does 85 mph on the offramp from 8 East to 805 North. It's apparently a long sweeping rising curve, single wide lane with no traffic interference, and located where radar won't work. I always wondered who was hitting the guardrail there all the time.
But the consensus is that cell phones won't work for slow speeds.
I'll rig up a Mythbusters grid and see how fast they can push it. There are OSHA rules on how hard an employee is allowed to push, so a little F = ma will get me started.
I've got the shock calculations done; the car shocks work quite well, at least theoretically. I'll send these when I get into the right format.

 

[GregLocock]

Phone GPS won't work indoors. Camera and ruler is cheap and easy for a one off but hard to automate.

Cheers

Greg Locock


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

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

 

[BUGGAR]

I'm going to try to attach the preliminary analyses here.

 

[3DDave]

Buggar,

Is the kinetic energy calc right? I recall a G/Gc as being part of using lbf weight calc. I usually use slugs to be sure.

 

[GregLocock]

I think you've got a units muddle somewhere, and frankly an energy calc is not going to be an easy approach. I suggest a stepwise model in the time domain. Let's see, initial acceleration is about 1/8 g, say 50 in/s/s, so the event time is going to be of the order of 3/50 second, and so the distance travelled is of the order of 9/50 inches. As it turns out the event is almost a constant deceleration, so you can use the old v^2=2.a.s



Cheers

Greg Locock


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

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

 

[BUGGAR]

Thank you again for the input. Yes, I erred on the kinetic energy of the moving X-Ray and it should be closer to 8 Ft Lb. My gut feeling is that my velocity is too low (I haven’t done the Mythbusters grid and camera routine yet).

But looking at it another way, again using energy, the machine will be moved by one man and has a maximum distance to move before hitting a wall of six feet. OSHA limits the amount a person can push an object to 50 pounds. So a man pushing the machine with 50 pounds of force for a distance of six feet imparts an energy of 300 Ft Lb. This would be an extreme case but still within the realm.

I’m trying to keep this graphic because that is how I design marine fenders, and I don’t know any better. And the shock dyno prints out a nice curve that I scan and insert into autocad, convert to a force-distance chart and read the energy and reaction values directly. And I like pictures.

Next step is to go back to the race shop and have him chart some more shocks. He also has a chassis dyno going all the time and feeling upwards of 800 horsepower thirty feet away makes for a great field trip.

 

[3DDave]

For those without AutoCAD, there's a nice utility called Enguage for converting images of charts into numeric values.

http://digitizer.sourceforge.net/

 

[zeusfaber]

Keep at the back of your mind that, while the OHSA would prefer you not to ask anybody to push more than 50 Lb, that figure is not a hard limit on what ill-advised people can actually do.

A.

 

[3DDave]

In thinking of it more, if two mice with traditional mouse balls were used, they could be gutted and their incremental encoders monitored by a microcontroller; this instead of a swivel wheel and pot. Otherwise a similar device could be built.

 

[Ron]

I'm not sure there's anything special about a shock absorber for SCCA Class D (such as DS). Essentially any shock absorber can be used. Given that, considering your mass, you might want to consider something a bit more robust such as an off-road shock absorber. Adjusting for stop length might be critical. If you "bottom out" the shock you might blow its seal.

Greg....you're the car expert...does this make sense?

 

[BUGGAR]

This what's so nice about adapting a car shock - there's so much out there to chose from and I can do actual testing to find the right one. I'm finding I can absorb the energy easy but I have to watch the reactive force. A flatter shock curve is better and I should find one without the low speed knee. However, I do ships' fenders and I could use some knowledge here if you would, Greg. I'm shooting for as constant deceleration as I can get. But maybe the fall off at the knee is good because it will be taken up by any return spring I use?

I was conservatively stingy with my 4 inch stroke and will increase this.

Another guy is trying to design a sort of mechanized fence to sweep away anyone between the moving X-Ray and the wall. I gotta get this onto YouTube, it's getting hilarious.

 

[Compositepro]

I do not see the point in measuring the speed. It will be whatever it is pushed to be. You already did the calculation based on a 50 lb force OSHA standard. Obviously this is not a "worst case", which is what you have to design for. Measuring will not help to decide what the worst case could be. Once you decide what the worst case is, the calculation is simple. Is the worst case so large that you cannot easily select a damper with a safety factor of 10? How about using a powered load mover so the speed is limited and there is no stress on anyone?

Your problem is identical to moving a load with a pallet jack. With a pallet jack one way of stopping the load is to lower it. You can stop your load by releasing the air to the air bearing. Have your bumper do this automatically if it touches anything.

https://www.appletonmfg.com/cart-mover.php

 

[GregLocock]

Assuming the idea is to bring the thing gently to a halt at some location a spring and a bit of friction will be very helpful. Shock absorbers tend to run a fair bit of friction, 2 lbf or so from memory.

Cheers

Greg Locock


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

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

 

[gruntguru]

A few thoughts.
- How level is the floor? A small slope will impart a lot of KE to a 4 ton vehicle.
- Make sure you select a gas shock - the pressure will auto-extend it after each use. (or add a spring as suggested by Greg)
- Is the bumper mechanism attached to the wall or the vehicle? How do you ensure the correct location on the wall/vehicle is impacted? Is there another bumper for the return journey?

3DDave. I love the mouse encoder idea.

je suis charlie

 

[BUGGAR]

Floor level and flatness is critical with air slides so the floor had a levelling compound and vinyl tiles (replaceable) were installed. The bumper shocks will be mounted onto the X-Ray base. I’m analyzing the shocks for directionality and angled impact the same as we do for ships berthing against dock fenders. That’s the advantage of my marine design background. I will probable use return springs, light enough so they don’t enter the energy equations.

For Greg: I have shock reactions, what impact factors would be recommended? Shocks are typically rated up to 10 inches per second – is that the maximum expected shock speed in a vehicle or just a “rating” speed to calibrate shocks. I’m going to the race shop today for dyno curves. Will cold testing be different? I noticed previously that he cycled the shocks several times before recording. Our machine hopefully will never exercise its shocks, how will this affect the rate? We’ll cover the exposed shafts like the off-roaders do so they don’t freeze up. Thanks for your help.

We’ll make these up from telescoping tubing flanked by two or more car shocks and a light return spring inside the tubes. The owner likes the concept (they are low cost) and wants to build some more that are portable with aluminum tubes and frames and place them around the shop when he moves other equipment. There is a containment curb around the areas and I’ll size them to keep from blowing his curb out.

I like the mouse encoder idea also. I've got a spare mouse that I'm taking out to the garage to play with right now.