A question for the Juenke owners...

[m12vlp]

I have a question for the Juenke owners.

With a projectile sitting on it how much deviation do you get when you blow on the projectile?

Yeah the question sounds weird but I've been making a digital one and get a lot of variation when I blow on the projectile which makes sense given that it would have a cooling effect. I'm curious to know if the Juenke has the same sort of variation.

[Curtley78]

I was at a mates place the other week and he has a similar contraption for his arrows.

It couldn't be too hard to knock one up and fit a digital gauge to it.

I have a question for the Juenke owners.

With a projectile sitting on it how much deviation do you get when you blow on the projectile?

Yeah the question sounds weird but I've been making a digital one and get a lot of variation when I blow on the projectile which makes sense given that it would have a cooling effect. I'm curious to know if the Juenke has the same sort of variation.

[macca]

I will check on the weekend if you haven't got an answer before. I know direct breeze and temp changes cause variations.
Cheers

[m12vlp]

Thanks. I'm intrigued to see how it behaves.

[aaronraad]

What are you using instead of a taut band with the digital version?

[m12vlp]

What are you using instead of a taut band with the digital version?

I'm guessing what you mean by taut band is how to I hold down and rotate the projectile???

At the moment I'm taking measurements at static positions.

The consistency of the eddy currents measurement techniques used by these devices is dependent on the consistency of movement when the object is in motion. I have never had one of the original ones to play with but the physics involved dictates that varying the speed that the projectiles are rotated should vary the magnitude of the signal returned. By developing it around the idea of static position with mine, I'm eliminating that as a factor. That and I haven't figured out how to rotate it properly yet because I'm better with the electronics than the mechanics

The great mystery for me is really about the sensitivity of the originals. I know that the originals are sensitive to drift due to temperature because I see the same effect with my experiments and it matches what the physics says it should do.

I know that the original gives a combined result for "out of roundness", jacket thickness variation and inner jacket defects because there's not enough electronics in there to detect the different aspects of the signal response. I'm hoping I can separate out the 2 ( out of round and the rest ) because that would be pretty cool. Certainly I can see it on the oscilloscope during some of my tests.

[aaronraad]

I assume you've read the Bullet Doctor set-up and see how the calibration is performed.

Notice the calibration is done without rotation until the final check and then the variance kicks in.

In terms of the taut band I was referring to your output signal and what you are measuring it with.

I'd say if you are blowing on the projectile you are disturbing the eddy current field.

I'm not sure what sensitivity your sensor measures and is able to output, but watch the micro-epsilon video http://www.youtube.com/watch?v=K_xm5S1a_5c&feature=share&list=PLFE548F0658477E21 and you can see the impact of simple heat transfer let alone blowing(adding a force) to the object. I assume you've seen the price for a Micro-Epsilon eddyNCDT3300 with the digital readout and that's before you have the jigging set-up with the bearing and spin drive.

If your sensor is flat, consider the measuring range and trying to measure curved surfaces of relatively small radii. These flat sensors as a single unit are really made to measure a relatively flat surface parallel to the sensor face. You really need an array of sensors to measure curved items http://www.laserlinc.com/benchlinc.html.

I'd also be surprised if you are able to detect voids inside the lead core. Maybe on the surface of the interface of the lead core and the jacket but not internally. For that you need to step up to X-Ray analysis really. Unless you know of some other sensor that will do the job please share???

Keep me updated on your progress as the subject really interests me.

[m12vlp]

You raise some good points.

I did read the bullet doctor instructions about setup.
I don't believe blowing on it is causing deformation because I get very little variation from touching the projectile (unless I really lean on it).

Eddy current signal response was tricky to get my head around because at the simplest level there is both an amplitude and a phase difference depending on the nature of the physical variations present in the part under test. I've got about a forests worth of printed NDT articles and patents that I've pored through over more than 12 months and it wasn't until I got the sensor design right that it all started to come together in my head. I must have tried about a dozen different sensor configurations and associated circuits before I hit on the right one.

My sensor construction is quite variable being completely diy but my circuit is built to handle that with various calibration and adjustment points. A have characterised the signal at numerous points in the signal processing and can adjust the gain to ridiculous levels if I want but the underlying signal needs to be clean. I haven't finalised all that though. I have PCB's in the post on their way to me right now and they will be able to handle the signal better than my prototype so I'm hoping to reduce the amplitude of the signal in order to minimise the current induced into the projectile which in turn will reduce the temperature rise in the projectile.

Some of my testing involved involved layers of copper tape stuck together on a flat surface to create varying thicknesses. I was able to characterise signal response variations for both liftoff and metal thickness which agreed with what I saw on video's in this series (http://www.youtube.com/watch?v=1YUSn___VxQ) and was my first real breakthrough in sensor design. Once I had that I applied it to the projectile and was able to observe similar variations again as I saw on the flat surface.

I couldn't tell you exactly what part of the projectile is being measured but I know by the theory that it is at least the diameter of the probe so some of the measured area will run up the sides of the projectile.

I will make no claims of being able to measure voids in the lead core (and I know you aren't accusing me of making such claims ). As lead has no electromagnetic properties the sensor cannot measure it it or even detect it's presence. Most of the hype around the original machines comes from a fundamental misunderstandings about what it actually does. The Juenke and my device measure material conductivity and the distance from sensor to the part. Material conductivity is affected by flaws and variation in the material such as thickness variations and cracks. Projectiles really don't have cracks in them but they definitely have thickness variations. I believe temperature affects conductivity and shape so influences both aspects of the measurement.

I would love to have and array of sensors and be able to just insert a projectile in them however the sensitivity for liftoff is so great that any variation in positioning the projectile will drown out the ability to detect the actual variations in the projectile itself. That's why the 4 BB's works. It allows for consistent positioning.

I have to run for work

[macca]

Hi, ran the test. Blew on it for all my lungs worth. Tried it from all angles and it had no effect on the reading. Tried it with a hand held fan and it did move it one deviation point.
Thought about it and repeated test with a thermometer gauge . The fan lowered the temperature as well so that may be the reason for the movement.

But normal breathing or hardblowing doesn't bother it. Tried it on twenty odd to be sure.

Hope it helps.
Cheers

[m12vlp]

Wow thanks for doing that. It's a very interesting result and gives me plenty to think about.

[m12vlp]

Just an update of my general progress.
http://youtu.be/NiOQghdUBic
http://youtu.be/NiOQghdUBic

While I have a functioning prototype I want it to be more sensitive and more accurate than it is. I have another design revision that I need to build which will hopefully be cheaper and more accurate.

[m12vlp]

I've been lax and have not been following this up.

I'm about 5 revisions on from my last post and have a number of working units including the prototype (which is 2 revisions old) used in the video here:
https://www.youtube.com/watch?v=V8MLwajWtRY

[aaronraad]

Do I spie with my little eye something beginning with an Arduino board?

Have you notice any differences in placing the shank of the projectile on the bearings as opposed to the boat-tail and/or ogive?

[m12vlp]

Do I spie with my little eye something beginning with an Arduino board?

If you're asking if the micro is an atmel atmega328p the answer is yes.

Have you notice any differences in placing the shank of the projectile on the bearings as opposed to the boat-tail and/or ogive?

The rear bearings need to be on the main bearing surface of the projectile because otherwise the boattail wants to slide down into the bearings and away from the backstop. The ogive is different because sliding down into it pushes back against the backstop. The pre-production versions i'm playing with at the moment have an adjustable backstop so i've been able to play with the positioning.

I get quite significant variations depending on the position of the projectile on the bearings. I'm in the early stages of working out which correlates to better accuracy on the target.

[aaronraad]

Getting a bearing right on the pressure ring near where the base meets the shank would be my target for the rear.

For the front bearing the preference would be right on the meplat but otherwise approximately the point on the ogive where the diameter is -0.001 of the shank diameter or the nominal calibre.

Projectile manufacturing methods differ but that base/shank junction the last contact point out of the barrel and break point of air flow for the remaining journey.

Ogives on modern match grade target projectiles are very good quality and not likely to show up any particular faults. The area around the meplat though can be compromised depending on the manufacturing process, but because of it's size and area doesn't typically have the same impact as the base/shank junction. Errors at the meplat will still leave you with a match grade projectile, just not the best possible result. Very difficult to measure but sometimes so bad it can be visualised, think of it as meplat tipping gone wrong!