# Help me record nipple displacement!! :-)

#### Laura9003

Joined May 23, 2010
12
Hello--
I've got an interesting project...I need to build a simple circuit to measure...get this...nipple displacement! I am designing a sports bra and will need to test my prototype. I want it to simply record nipple position at rest, and then take another measurement every microsecond or so while my model is running on a treadmill. I'll worry about how to hook the apparatus itself up, but I need suggestions on the best simple circuit to build to record these measurements.
I do have a BSEE but I never used it so I'm out of practice and I can't remember how I would do this. I appreciate the help!

Laura

#### retched

Joined Dec 5, 2009
5,208
I could donate my time as a visual inspector and data logger. But Im guessing you dont want a human.

Oh. Do you only care how far apart from breast to breast laterally or up and down also?

If so, you could latex on a string to each nipple, and use a light spring to keep the nipples naturally were they belong during the at rest point.

As the spring changes size, the resistance or inductance changes.. I cant remember.

Using the spring as your sensor, you would know how far apart and how much closer the nipples got to each other during the test.

That would only do lateral movement. If you want to measure every direction, a camera with a dot following program could follow the two dots.. One on each nipple.

#### Laura9003

Joined May 23, 2010
12
LOL if your eyes sample and record on a microsecond basis that could work!
I actually am only interested in vertical. I have lateral...um...controlled already. I am thinking there has to be a simple off-the-shelf configuration that I could just hook into my pc and download into a software ap or something. I've just been out of the loop so long I don't know what's out there. Probably looking for a microprocessor-based solution.

#### R!f@@

Joined Apr 2, 2009
9,749
Count me in

#### retched

Joined Dec 5, 2009
5,208
Well, which way do you want to sense them? Using the reflective dot and camera method?

#### SgtWookie

Joined Jul 17, 2007
22,221
Consider using accelerometers. They make them quite tiny nowadays; some have outputs for two axis, some three.

You could calculate the displacement by the G-forces measured.

Radio Shack even stocks a 2-axis accelerometer:

The accelerometer IC itself is in the center of the relatively huge board. It's perhaps 0.150" square.

#### Laura9003

Joined May 23, 2010
12
rm, I don't know. I'll have to look into both to decide. I am looking for a relatively inexpensve option.

Sgt. Wookie, thanks for your input. I will look into accelerometers. G-forces could play very well into my marketing dialog, so it could be helpful to be collecting that data along the way.

#### Markd77

Joined Sep 7, 2009
2,806
Do you mean millisecond? Microsecond seems tricky.

#### Laura9003

Joined May 23, 2010
12
No, I meant microsecond but I didn't really think it through. Just remember working in on that scale back in school. Now that you mention it, micro isn't necessary for this application. Milli it is!
Thanks.

#### Laura9003

Joined May 23, 2010
12
Sgt Wookie-
Looked at the accelerometer...this could work. I guess all I would have to do is measure vertical acceleration? I could find some EKG Lead-type things and connect them. Didn't notice how I read output but I'm sure it will be obvious...

#### SgtWookie

Joined Jul 17, 2007
22,221
I haven't read the specs in awhile, but seems to me that they are usually digital serial data.

If you look at the specs of the base IC's for these kinds of things, you'll find that they can be used in one or more ranges; ie: +/-2g, +/-8g, or something like that. It's been awhile since I've looked.

The device you see on Radio Shack's page is "packaged for the hobbyist". It's fine for working up a test circuit on a breadboard. However, you probably want to pick up a few of the IC's themselves, and do some experiments with them. The IC's themselves are maybe $5 to$12 or so.

You might use "Acceleration Sensor IC" for search keywords on suppliers' websites.

You would need some kind of microcontroller to monitor the sensors, and to interface with your computer. I think you'll find that the data rate is in the vicinity of 100 samples/second, or 10mS. Still, I would think that should be more than enough for human-induced motion, unless they're slamming into a concrete wall.

#### Laura9003

Joined May 23, 2010
12
OOH! I'm excited. Read around on the Parallax website and reminded myself how this all works. I can do the accelerometer system for cheap and it will give me what I need. Thanks everyone for your help. I'll report back on results.
Seems like a funny topic but it's important to female athletes.

#### SgtWookie

Joined Jul 17, 2007
22,221
Here's a random MEMS IC that Digikey stocks:
http://search.digikey.com/scripts/DkSearch/dksus.dll?Detail&name=497-8548-ND
Family:Accelerometers
Axis:X, Y, Z
Acceleration Range:±2.3g, 9.2g
Sensitivity: 18mg/digit, 72mg/digit
Voltage - Supply: 2.16 V ~ 3.6 V
Output Type: Digital
Interface: I²C, SPI
Mounting Type: Surface Mount
Package / Case: 16-LGA

That's a lot of data for a few bucks in a small package.

I can't comment on whether measuring just one axis would be enough, as I have no formal training in physiology. However, if you plan ahead and include all of the data (this one covers movement in all three axis), as well as selectable sensitivity ranges up front (2.3 and 9.2g's), you will likely have your bases covered.

#### SgtWookie

Joined Jul 17, 2007
22,221
I'm not certain that you will achieve an acceptable data rate with a low-end Parallax uC. They are OK for prototyping, but keep in mind they are running a version of Basic in a tokenized interpreted mode; which slows them down a great deal.

Also, the per-unit expense would be prohibitive for a production/commercial environment.

You would likely get much better performance and enormous cost reductions out of another manufacturer's offerings.

#### Laura9003

Joined May 23, 2010
12
Good to know, to all of it. I really appreciate your input. Right now I am thinking small potatoes in the testing department, non-scientific results only, for my own information for the most part. However, you know how things evolve beyond your imagining sometimes and I certainly agree that testing for everything now puts me in the right position later when in case I decide I do need all the data. Will save time by planning well.
Either way, I have time. Still designing the prototype, and have to write the patent before I can get it built. So I can keep thinking about this part in parallel with the other parts and I'm sure I'll end up with a good system.

#### Laura9003

Joined May 23, 2010
12
As for where to measure, I believe my garment design has lateral motion worked out so I started out interested in vertical displacement only. But as I think of it as a marketing tool, it occurs to me all measurements will help.
There are actually a few people in Europe studying Breast Biomechanics and consenus says breasts during running displace 50% vertically, 25% side-to-side, and 25% in-and-out.
There's your random fact of the day.

#### retched

Joined Dec 5, 2009
5,208
Any numbers on the male parts?

#### SgtWookie

Joined Jul 17, 2007
22,221
So, you see? I wasn't all that far off.

But to keep things in perspective, you might want to also sample data from other parts of the anatomy, such as G-sensors attached near the breastbone, the spine, etc. so that you have something to relate the "nipple data" to. You would then have absolutes and relative data.

You might look at something like the dsPIC series by Microchip; they have built-in RF capability. Data polled from the accelerometers; then transmitted to a USB-compatible receiver that is connected to your workstation for mass storage.

Or if you want to keep it simple, just use one or more large serial static RAM ICs. Download the data later. Get a PIC that has I2C or SPI interface as well as USB interface.

#### SgtWookie

Joined Jul 17, 2007
22,221

#### retched

Joined Dec 5, 2009
5,208
Sorry...just wondering...