Basic drop testing software needed

[BretMan]

I'm a novice regarding accelerometers and am making an effort to put together a simple indoor, single axis drop testing apparatus to do some tests. I think I have the hardware parts identified which will be compatible. I still need the data acquisition software to give me the readings of the Gs at impact. I don't think they'll be much higher than 200g. The only software I've found is at least several hundred $ and provides much more than I need now.

Does anyone know of any data acquisition software than can serve my needs? I just need it to provide the Gs at impact - really basic. It can be old or used or ugly but I do need it to be reliable. If it's free that would be BIG help too as the whole expense of this is for an R&D experiment.

I'm running Windows XP Pro (SP3). Thanks!

Bret

Miami,FL USA

 

[IRstuff]

Doesn't that pretty much depend on the sensor? Don't most sensor come with software that will log data into a file? At which point, you scarf up the data into Excel and scale it appropriately?

http://www.analog.com/en/mems/high-...s/evaluation-boardstools/resources/index.html

TTFN

FAQ731-376

 

[BretMan]

Thanks for the tip. That would be great if it did come with the accelerometer. I checked the link and what I think was the associated pdf for it and it say "Evaluation ...". It also mentions that it comes with connecting cables to the PC. I'm not sure if the supporting software is specific to the particular sensor, nor if, say it does come with that sensor, that's a sensor I can use for my application.

It doesn't seem to be free/included but I will ask them.

Bret

Miami,FL USA

 

[analogkid2digitalman]

Do you have a digital storage oscilloscope already on-hand that you can download captured data from? Then like IRstuff suggested, import the file and analyze in Excel.

-AK2DM

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
"It's the questions that drive us"
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

 

[BretMan]

Ok, update: I've sent out questions to various accelerometer manufactures and am getting feedback where it looks like they don't come with the software after all. This isn't good news, yet I must do these tests. Might there be used drop testing equip. somewhere by any chance? If so is there a supplier?

The lowest cost new set-up so far includes a universal accelerometer, a connecting cable and signal conditioner. I mentioned to them the idea/option of replacing a PC w/software with an LCD-display that would have a "hold" feature so I can note the readings, then reset it. I asked what the compatibility specs need to be so I can hunt for such a display.

Please tell me more about your oscilloscope idea. I don't have one. Are there used ones available? How would that be connected and then to the PC w/ Excel?

Bret

Miami,FL USA

 

[IRstuff]

They don't come with ANY software, or don't come with the software you asked for?

If you buy the evaluation board from Analog Devices, you get the software that captures data.

TTFN

FAQ731-376

 

[MacGyverS2000]

What are you dropping?

Dan - Owner

http://www.Hi-TecDesigns.com

 

[BretMan]

What adds to the confusion, to me, is that some manufacturers seem to supply only certain components and not everything I need. So if I don't get a complete "kit" for my needs, there will be a good chance I'll buy an incompatible remaining part and have to look yet again.

This would be a snap to do if I had unlimited funds and could just buy a turn-key drop test system. That would be several thousand dollars I can't risk at this point. So I'm trying to put something together that will do for now.

Analog Devices does have the evaluation board which includes the software but that also adds to the cost and it's too much.

I want to keep it under $500. The lowest I have so far with known compatible components for sensor,cable, and signal cond. is $900. That can be connected to a LCD display I need to still find. Eveb so, I still feel I'm being up-sold to more than I need.

I will be dropping helmets down at a single axis to the floor. I understand that helmets have about a 10-15 millisecond impact event time. However, the sensor will be mounted to a test head inside the helmets. I don't know what the event time is for that and I believe that's an important detail for choosing the accelerometer.

I've also found a couple like this at much lower cost:

http://www.futureelectronics.com/en...eters/Pages/3351632-SMB172-0273101224001.aspx

Can I get by with that and if so what would be the remaining compatible components?

Bret

Miami,FL USA

 

[IRstuff]

You're funny...

You bitch about the cost of an eval board, but now you want to look at a bare component?

TTFN

FAQ731-376

 

[MacGyverS2000]

I thought you wanted a package deal with datalogging software, etc.? So why look at bare devices?

You gotta pay to play, so if you want it all in one package, it's going to cost you... no shortcuts.

Dan - Owner

http://www.Hi-TecDesigns.com

 

[BretMan]

My experience is if you buy a package deal that should be less expensive - but I don't know and don't have a God complex assuming supreme knowledge about this ... If that's how it is then there's no other option.

But if I can use the Bosch type accelerometer, which is $32 that would save me about $400 over the high end ones from the current lowest set up. So I'd be paying less to play. Is there a problem with that?

Bret

Miami,FL USA

 

[desrod]

Sensors give analogical signals (ex: 0-5 Volts for 0-200g) that must be converted to digital signal if you want to use it with your PC. You can use either an acquisition board/box that usually comes with software (found a nice one with multiple channels that was USB connected but $$$) or you can use the sound card on your PC and homemade soft for simple applications like yours.

Getting the signal could be done using C++ and you can then build a text simple file or even use it as a function in Excel I think. Nice project for a EIT.

Desrod

 

[BretMan]

Thanks Desrod - that's good creative problem solving at work. I have heard of using Excel via the PC sound card (also in this post) and I've heard about using BASIC Stamp modules but the programming involved was something else I personally should avoid. Are you an EIT?

Bret

Miami,FL USA

 

[VEBill]

Your Google-fu is very weak Greasshopper... ;-)

Gulf Coast Data Concepts
X250-2 Impact Sensor, $199

http://www.gcdataconcepts.com/x250-2.html

The X250-2 Impact sensor is a high-g, three axis accelerometer data recorder. Data is continuously stored onboard a 1GByte microSD card. To retrieve recorded data, plug into any USB host and the data is instantly available through Excel compatible .csv files.

3-axis +/-250g accelerometer
Internal, rechargeable Li-ion battery
8, 16, 32, 64, 128, 256 or 512Hz user selectable sample rate
Accurate (2 PPM) time stamped data using Real Time Clock (RTC)
Easily readable comma separated text data files
Data recorded to a field removable 1Gb microSD card (included)
Data transfer compatible with Windows/Linux/Mac via Universal Serial Bus (USB) interface (no special software required)


*** Double-check that 512 Hz sampling is good enough.***


There's also a nicer model, but it's $600:

http://www.gcdataconcepts.com/x250-hd.html

 

[IRstuff]

You cannot compare a bare component against a board mounted chip, with the requisite power conditioning and connectorization.

The cost of the Analog Devices evaluation board was only $186, although, it will only do 100 Hz datalogging using the provided firmware:

http://www.analog.com/en/sensors/inertial-sensors/adxl345/products/product.html

http://search.digikey.com/scripts/DkSearch/dksus.dll?site=us&lang=en&mpart=EVAL-ADXL345Z-S

http://www.analog.com/static/imported-files/eval_boards/UG065_2-9-10.pdf

TTFN

FAQ731-376

 

[desrod]

Bret,

Here's an interesting link that does what I suggest in my previous post.

http://www.zeitnitz.de/Christian/scope_en

Depending on what your are trying to do and if you already have a sensor, that might help.

PS: EIT = Engineer In Training I might be still learning but I'm too old for that.

Desrod

 

[BretMan]

Thanks for those leads!

I can spend $15 for that oscilliscope software. Now I have to figure out if it's compatible.

I think the selection process workflow should start with what the even time is of the impacts as described. Then select the appropriate sensor for that event time. I got this from a software salesman:
"Rule of thumb: The fastest event you can measure should be about 10x slower than the inverse of the highest frequency (ideally 50x if you want to split hairs, but 10x will only yield a few percentage points of error):
Example 1: 1 / (400Hz) = 2.5msec. The fastest event you should try to measure, then should be 2.5msec x 10 = 25msec
Example 2: 1 / (2,000Hz) = 0.5msec. The fastest measurable event in this case is 0.5msec x 10 = 5msec"

Then hopefully that oscilliscope would be compatible. Does that seem correct?

Bret

Miami,FL USA

 

[MacGyverS2000]

VE,

That's a bit shy of 2ms per sample... it's des's call, but 5-8 samples seems a little low to me for mapping the entire event. but I've been known to be wrong...

Dan - Owner

http://www.Hi-TecDesigns.com

 

[BretMan]

That same salesman was who told me that in the helmet tests he knows of, they have event times of about 10-15ms. In my tests, the sensor will mount on a dummy head inside the helmet, so I would assume that would increase the event time a lot due to the cushioning in the helmets.

Bret

Miami,FL USA

 

[MacGyverS2000]

That's a big assumption, and accepting the word of a salesman is almost always deadly...

Dan - Owner

http://www.Hi-TecDesigns.com