Acceleration - mathematics and equipment advice

Nyeros

Hello all,

I'm currently studying at University, and need some advice.

I am looking to measure the acceleration of a large (7 tonne) unit during a drop test (of around 10mm), and I've been told the acceleration will be less than 1g (9.81 m/s^2). However I'm not sure what the best equipment would be to measure this acceleration (very limited knowledge of accelerometers etc).

Also, I am interested in the mathematics behind this, so if anyone can point me in the right direction of the calculations required it would be much appreciated.

Thanks in advance.

Martin Hebel

Here's an accelerometer Parallax offers, with sample code to read the acceleration.
http://www.parallax.com/detail.asp?product_id=28026

A possibility to send the data wirelessly as it falls using Xbee's or other RF units.

As far as the math... what were you looking for? Anything in freefall (on earth) will fall at a rate of 1g (9.8m/s), but drag and other factors will cause it to be less.

-Martin

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
StampPlot - GUI and Plotting, and XBee Wireless Adapters
Southern Illinois University Carbondale, Electronic Systems Technologies

PAR

Nyeros said...
Hello all,

I'm currently studying at University, and need some advice.

I am looking to measure the acceleration of a large (7 tonne) unit during a drop test (of around 10mm), and I've been told the acceleration will be less than 1g (9.81 m/s^2). However I'm not sure what the best equipment would be to measure this acceleration (very limited knowledge of accelerometers etc).

Also, I am interested in the mathematics behind this, so if anyone can point me in the right direction of the calculations required it would be much appreciated.

Thanks in advance.

If you are really trying to measure acceleration over a 10 millimeter distance (vertical drop) from an initial resting state, will you really be caring about drag and other factors which may influence acceleration?

The rate would be affected (mostly by air resistance) if you were dropping 7 tons of unrolled aluminum foil, but a 7-ton cube of aluminum would not manifest easily-measured drag/etc over a 10mm distance from an initial resting state.

So, I'm not sure whether you were assigned a simple problem (ignore drag/etc), in which case Martin's observation that the rate is more or less a known constant is good enough, or whether you were given a seriously difficult problem, in which case much more would have to be known about your drop test setup and experimental requirements. Like Martin said, "What were you looking for?"

PAR

Martin Hebel

I missed the 10mm when I answered.· Depending on the release mechanism, this might not even be far enough to make forces and such from the release mechanism negligible.

-Martin
·

Capt. Quirk

It seems to me the·issue is time. Time·is going to determine·what product to use and how to transmit or store the data.

bobledoux

If the bottom of the unit is flat compression of air between the unit and floor may break the fall.

Could you physically measure the changing distance to the floor? Perhaps a piece of conductive foam between two plates, placed under the unit. As the unit falls it crushes the foam changing the resistance between the two plates.

LSB

You don't need an accelerometer, or foam. You don't need to lift or measure anything because Galileo proved everything drops at the same rate [noparse][[/noparse]edit: excluding drag, etcetera] (9.81 m/s^2). All you need is a calculator to convert m/s^2 to mm/s^2 --definately easier than converting sensor readings to mm/s^2 and a lot easier than lifting 7 tons.

The 2G acceleometers that Parallax sells are well documented and cheap. Find them (sensor, documentation and circuit diagrams) in the store-- In 2 hours you'll have a working circuit that demonstrates what Galileo proved in the 16th century. Oh, if you must drop something, drop a 7 ounce weight scaled proprtionately and measure that.

Not blaming you, but this sounds just plain silly to me.

Post Edited (LSB) : 3/29/2007 7:16:01 AM GMT

Nyeros

Thanks for your responses and advice so far.

I've got some more information which will clear up a few issues.

The drop test is a requirement of an IEC standard for transportation·of a 7 tonne transformer which will be housed in a (new)·protective container (2m high by 2.5m wide). To satisfy this standard, the unit must withstand a freefall of 10mm. The test cannot be carried out more than two times, therefore the first·drop will be used to assess the acceleration, and the second for the test (i.e. change height of drop test so that unit experiences free fall (or close to it). The height cannot be higher than 15mm to safeguard the transformer.

The company are planning to use a crane to hoist the container, then release the brake to drop it. I think this was the crucial piece of information that I overlooked before (and should of mentioned in my original post). I think the friction from the cable/crane etc is the reason the acceleration will be lower than 9.81 ms^2. They are also interested in the mathematics behind my calculations during both tests.

I will also be considering other options that may be better to conduct the test (instead of crane etc).

Hope this extra info has cleared up a few points. I will check the accelerometers that you have suggested.

Thanks again.

stamptrol

The test is supposed to show that the transformer could withstand the usual bumps and bruises during handling and transportation.

The documentation has to show that the acceleration was at least gravity without being reduced by the crane or other influences.

Get accelerometers (definitely more than one) for the first test and hook them up to a good analog recorder or fast digital recorder. I think you'll need to have a quick release shackle on the lifting point in order to get close to the free-fall case. Even though its only a 7 tonne load, you may need a hydraulically actuated shackle or an explosive bolt type to get quick action.

If you're lucky you won't have to bump the unit too many times!

I assume you'll do a bunch of electrical tests after the drop to show up any problems before its delivered to site?

Cheers

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Tom Sisk

http://www.siskconsult.com
·

Mike15

I have made a data logger from the stamp an accelrometer and an RTC. It then stores all the data to an external eeprom and downloads the data through the serial port.

I use this as a payload in my rocket to determin acceleration. I use high G accelrometers to withstand the high Gs caused by rocket flight.

Capt. Quirk

The duty cycle of·the Mesmic 2125 is 10 ms, then record that to an eeprom. But the transformer is going to hit the ground in 1 ms and you need precautions for the impact too. So you need to find a sensor that can measure several times in less than 1 ms and a processor that will keep up.

What I think you need is a distance sensor vrs. time stamp, that outputs a voltage.·Use a fast 8 bit ADC with a narrow reference voltage range to improve the resolution.·Save the data to the processors internal ram, maybe using a 16 variable index array. Once the index is done, put the program to sleep. (or some other method, that doesn't destroy the·data variables·collected)

Post Edited (Capt. Quirk) : 3/31/2007 8:13:52 AM GMT

Skywalker49

Mike15 said...

I have made a data logger from the stamp an accelrometer and an RTC. It then stores all the data to an external eeprom and downloads the data through the serial port.

I use this as a payload in my rocket to determin acceleration. I use high G accelrometers to withstand the high Gs caused by rocket flight.

Hi Mike .. wrong thread ?!

Have a nice day,

Ed

LSB

First, let me apologize for breaking my cardinal rule of forum posting: NEVER judge the value or legitimacy of a project; just answer the question or keep your fingers shut. My bad, thanks for the reminder.

Second, let me suggest there are other means of gathering data that may indirectly answer your need (some or none may apply). Perhaps it may be possible to measure the air displaced under the unit as it drops, or the vacuum created in a cylinder above. A lever, properly positioned, would scale the drop and allow for use of high speed photography to verify time/ distance measurements. Similarly, a 10mm pulley married to a 1mm pulley or a 100mm pulley would scale the drop by a factor of +/- 10.

I think too that the rate of fall is governed by the laws of physics, measuring impact force as “instantaneous weight at impact” might satisfy your requirement. The deformity of a block of metal of known hardness might then describe the impact force sufficiently for your needs. If its posible to engineer some void in the floor beneath then lasers or ultrasound ould be employed without crushing the equipment.

The problem is an interesting one—the answers are known, only the means is left to discover. I'd love to see video of your "final"

iamscottym

I had to work on a similar, though much smaller scale, project for another company. I don't have my mat sci book handy, so I can't cite the formulas specifically, but you basically have to work out how much the material of the object and what it's falling onto (concrete, dirt, etc) will deform. How much deformation there is determines how quickly the object must decelerate. In my case, it was steel falling onto nylon, and even that resulted in some intense g's. I forget the specs exactly, but we needed one hell of an accelerometer to measure it. It had to have a high g tolerance(several hundred g's minimum), and run at >1khz or better. Generally, this falls into a class of accelerometers called 'impact accelerometers', because they are made for this type of application. They're a few hundred bucks each.

bjhamltn

Couldn't you just measeure the elapse time of the drop and then derive what the acceleration was from that?