Shop OBEX P1 Docs P2 Docs Learn Events
Eddy Cuurent Analyzer (Metal tester) — Parallax Forums

Eddy Cuurent Analyzer (Metal tester)

CenlasoftCenlasoft Posts: 265
edited 2009-09-16 22:29 in Propeller 1
Hello,
I have finished a project called "Eddy Current Analyzer" using the Professional Prop Board. The circuit works great on the breadboard and now its time for me to make it on a PCB. I used ExpressPCB (attached file) and if anyone uses this program, I have a few questions.
1. The trace from C1 to the sip connector maybe too long for this is the pin that excites the RLC circuit. I cannot see anyother way to route it. Any suggestion would be appreaciated.
2. I want to basically plug in this to the prop board for now until I can put the prop and the circuit together on a board. I am new at this pcb design and would appreciate any help.
3. I was also thing about using the Schmartboard to plug in this circuit.

Thanks,
Cenlasoft (Curtis)

Post Edited (Cenlasoft) : 8/29/2009 3:00:11 PM GMT
«1

Comments

  • Bobb FwedBobb Fwed Posts: 1,119
    edited 2009-08-27 17:21
    Attached is a few changes.
    You should get in the habit of not making 90-degree corners in the traces. 45-corners creates less EMI.
    Also, unless you need all those ground connections, a single ground connection and a ground plane (the green plane I added) is effective for power transfer and as an EMI shield between traces and a little from outside interference. I am not quite sure what all you are trying to do with this PCB (an electrical schematic would be nice -- ExpressPCB also offers ExpressSCH which integrates nicely). If you aren't contraned by size, this whole setup could easily be re-arranged to get C1 close to L1 and the SIP connection.
    Is U1 an EEPROM?
    Also, it seems intimidating, but the QFP propeller isn't that difficult to solder to a PCB by hand, just use fine solder and a pointy soldering tip.
    You should also always use some type of snap (0.025- or 0.005-inches). It makes the traces straighter and easier to add parts later. You can turn it off for non-standard spaced items, but generally 0.025-inches is optimal spacing.
    You have the border of the board just as the default, yet you all your components in a small portion of the board. You should determine your maximum size (or create a size that limits the wasted space --- to save money when getting the PCBs made).

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    April, 2008: when I discovered the answers to all my micro-computational-botherations!

    Post Edited (Bobb Fwed) : 8/27/2009 5:30:28 PM GMT
  • RaymanRayman Posts: 14,849
    edited 2009-08-27 17:48
    I would run all the traces on the top layer and leave the bottom layer as just the ground plane, if possible...

    If this is some kind of RF circuit, you might want things closer together.

    Also, if you want a lot of boards, it looks like you could fit two copies of that layout on the board, in order to get twice as many...

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    My Prop Info&Apps: ·http://www.rayslogic.com/propeller/propeller.htm
  • LeonLeon Posts: 7,620
    edited 2009-08-27 18:03
    90 degree corners aren't relevant to EMI, but look untidy and can cause etching problems.

    Leon

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    Amateur radio callsign: G1HSM
    Suzuki SV1000S motorcycle
  • Bobb FwedBobb Fwed Posts: 1,119
    edited 2009-08-27 18:18
    Leon said...
    90 degree corners aren't relevant to EMI, but look untidy and can cause etching problems.
    Leon

    ...I got that info from some other thread on this forum...oh well, good practice for trace problems.
    I think Express PCB (on their website) says they have issues with traces smaller than 0.012".

    Well, attached is a [noparse][[/noparse]slightly] rearranged board, re-sized and traces moved to top layer.

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    April, 2008: when I discovered the answers to all my micro-computational-botherations!
  • CenlasoftCenlasoft Posts: 265
    edited 2009-08-27 18:19
    Thanks everyone,

    I am new to PCB design and I appreciate the suggestions. Let me give a little background on this project. It is a analyzer to recover hallmarks (stamps or engravings) on metal artifacts. I am an archaeologist and I love to build my own instruments. This circuit allows me to scan an object line by line with a 1mm diameter coil (probe). As I pass over a discontinuity the voltage changes and I then plot each line to form a 2-D picture. I have attached some samples. My results are crude and I have to plot the data in using VB 6.0. Most of it is not automated yet. One image thats attached is of a silver coupon that has been engrave by a silversmith, the other is the same coupon that has the hallmark polished off, the third is my crude recovery of the hallmark. I will try your suggestions and will post the whole project when I finish it. Thanks, Curtis
  • RaymanRayman Posts: 14,849
    edited 2009-08-27 18:41
    Latest version looks good. But, for ExpressPCB (assuming you want to save money), you should make your board be 3.8"x2.5" (the default) to get the great $51 miniboard service... You could probably fit 2 or maybe even 4 of these circuits on one board and then cut them up with a jigsaw or something...

    Also, you might want to add some mounting holes so you can secure the board to something...

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    My Prop Info&Apps: ·http://www.rayslogic.com/propeller/propeller.htm
  • JRetSapDoogJRetSapDoog Posts: 954
    edited 2009-08-28 14:06
    Interesting project, Cenlasoft. Thanks for posting your question. Due to your post I installed ExpressPCB and played around with it.

    And thanks to Bobb for your embellishments, including the ground plane, as well as your helpful comments for us beginners.

    I noticed that my Prop System Module (PSM) from Rayman is almost exactly (if not exactly) 3.8"x2.5". Wonder if this PCB service was used.

    Does anyone have suggestions for inexpensive PCB software if one wanted to do production but didn't want to be limited to one company?

    Any other relevant tips or interesting experiences with PCB design and fabrication? Well, that might be in other scattered threads, here.

    I'm curious if the ExpressPCB program (and similar) work well for parts with small geometries, such as the upcoming Prop II (not BGA stuff).
  • Bobb FwedBobb Fwed Posts: 1,119
    edited 2009-08-28 16:32
    As for the "production" aspect of using ExpressPCB, they do help you out a bit. Once you order some boards from them, you can pay an extra $60 to get a Gerber version of the file. This isn't ideal, but ExpressPCB has great customer service, fast PCB turn around, and ease of use (start to finish -- Free ExpressSCH to doorstep) is amazing. I too am looking to get boards assembled, and need a Gerber version of my boards. I also need another quality program to go to the next step. Some type of integration into a program like ExpressSCH would be fantastic as well!

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    April, 2008: when I discovered the answers to all my micro-computational-botherations!
  • TreeLabTreeLab Posts: 138
    edited 2009-08-28 17:37
    This is a very cool application. is the probe measuring how subsurface defects affect the inductance?
    Cheers!
    Paul Rowntree
  • CenlasoftCenlasoft Posts: 265
    edited 2009-08-28 18:20
    Hi Treelab,

    Yes, I can change the frequency and value of the coil (probe) or capacitance and go deeper into the metal surface. My goal is to make an Inductive (eddy current) based CAT scan using the prop and an RLC circuit. These results results would not have been possible if not for the members of the forum who help me and especially Beau Schwabe. Parallax is a great company and their products are awsome. I will post the entire project soon.

    Cenlasoft (Curtis)
  • CenlasoftCenlasoft Posts: 265
    edited 2009-08-28 18:27
    Hello,

    I forgot to mention Hanno Sander and Viewport. His help and Viewport allowed me to see what was happening in the circuit and also save the data fromthe plots.

    Thanks,

    Cenlasoft (Curtis)
  • TreeLabTreeLab Posts: 138
    edited 2009-08-28 19:16
    Yes, this is the best forum that I have ever been involved with. And Hanno is a great asset.

    Could your device be used to identify stamped ID numbers (guns, engine blocks....) that have been ground off by persons of ill-repute?

    Cheers!
    Paul Rowntree
  • CenlasoftCenlasoft Posts: 265
    edited 2009-08-28 19:40
    Hi,

    That's exactly what I am attempting now. The FBI came to investigate another archaelogical matter and saw the instrument and was interested. I will be doing some test next week.

    Thanks,

    Curtis
  • Beau SchwabeBeau Schwabe Posts: 6,568
    edited 2009-08-28 19:53
    Cenlasoft,

    Wow, you got the FBI's attention? cool!

    BTW) thanks for the plug, I can't wait to see your project report!

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    Beau Schwabe

    IC Layout Engineer
    Parallax, Inc.
  • LawsonLawson Posts: 870
    edited 2009-08-29 05:31
    This is really cool. Have you thought of using write heads from old hard drives? Salvaging the voice coil drive for the heads would also give you a really fast scanning system. A co-worker of mine is a bit of a recycler, several parts of the Lidar we're building came from old hard drives [noparse]:)[/noparse]

    For my own PCB work I've moved up to TinyCAD for schematic capture. Not the best, but works well enough. To layout a board I use FreePCB. I find FreePCB to be an excellent manual layout program. Well thought out UI and enough documentation to learn how to use it. (though not Parallax level documentation) Oh yea, both pieces of software are free.

    Lawson

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    Lunch cures all problems! have you had lunch?

    Post Edited (Lawson) : 8/29/2009 5:38:57 AM GMT
  • ErNaErNa Posts: 1,752
    edited 2009-08-29 10:54
    Fantastic! That's the way, I like it. And also the idea of using a read head (of an old hard disk or even floppy drive) to sense signals. Read/write heads may also be used for excitation and sensing. If you scan only very small areas, what about using a CD-drive head, it allows for scanning one direction and changing the distance to follow "hilly" probes!

    And: you have to be curious and ask: "what's behind the scene?". Finding such an interesting application under "ExpressPCB"

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    cmapspublic3.ihmc.us:80/servlet/SBReadResourceServlet?rid=1181572927203_421963583_5511&partName=htmltext
    Hello Rest Of The World
    Hello Debris
    Install a propeller and blow them away wink.gif

    Post Edited (ErNa) : 8/29/2009 10:59:18 AM GMT
  • ElectricAyeElectricAye Posts: 4,561
    edited 2009-08-29 13:25
    Curtis,

    this is a very cool project. And for the sake of documenting it for future reference, you might consider clicking on the pencil-shaped edit button at the top right corner of your forum screen and edit your title for this post, and include several words to describe what this project is all about. It's too good a project to be lost to history as a generic expressPCB topic!

    cool stuff,
    Mark
    smile.gif
  • CenlasoftCenlasoft Posts: 265
    edited 2009-08-29 14:56
    Thanks everyone for the compliments. I never thought about using a hard drive head, great idea. This weekend, i was thinking about making an XY type apparatus with servos or stepper motors to scan the objects. I have several stepper motors and servos that I have not used yet. I have a hard drive also. Thanks for the suggestions. Also thanks for suggesting that I change the topic name. I didn't know I could do that. I also want to add another feature to my metal analyzer, I want to measure the phase angle as the probe comes close to the object. In eddy current analysis, an rlc circuit and an oscilloscope set to XY can do this. I want to use Viewport (which has an XY function) to view the angle. The only problem is that I don't know how I would measure the phase angle. The phase angle can tell us the ID and conductivity of the metal tested and also the thickness of a coating on it.

    Thanks again, I am working on posting the schematics and data for everyone to see. Curtis
  • Nick MuellerNick Mueller Posts: 815
    edited 2009-08-29 15:31
    Damned interesting! I always thought, that these stampings can only be recovered by etching.
    Does your process work on any metall?

    Photos and schematics are *really* welcome (at least by me).



    Nick

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    Never use force, just go for a bigger hammer!

    The DIY Digital-Readout for mills, lathes etc.:
    YADRO
  • CenlasoftCenlasoft Posts: 265
    edited 2009-08-29 16:49
    Nick,

    I have tried copper, aluminum, steel, silver, gold, and lead with great success. I am working this weekend to finalize the pictures and circuits.

    Curtis
  • Nick MuellerNick Mueller Posts: 815
    edited 2009-08-29 17:21
    Oh, and what's about the size of the sample you showed? A few mm for the hallmarks?

    Really great work! Fascinating! I'd like to build my own, but I have no use for it. smile.gif


    Nick

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    Never use force, just go for a bigger hammer!

    The DIY Digital-Readout for mills, lathes etc.:
    YADRO
  • CenlasoftCenlasoft Posts: 265
    edited 2009-08-29 18:17
    Hi Nick,

    The area that includes the hallmarks is about 7.8 mm squared. Thanks for your interest. Today I am trying to make a very small coil. Its hard to wind. I want a 2mm diameter for the coil. This will give me about 1 mm resolution. I will also make many different coils. The value I want is 40 uH and I will use 36 g magnet wire. I will also try the harddrive head method. I wish I could have a method to wind these coils automatically. Thanks,

    Curtis
  • Nick MuellerNick Mueller Posts: 815
    edited 2009-08-29 18:58
    If you tell me what diameter of the wire (in mm, not AWG!), the core and how many windings, I'll wind you a set. I can turn a core out of POM or PA if you want. I need a drawing of the core (in mm please, I hate imperial units).
    I can also measure the exact inductivity.

    Don't worry about the shipping (I'm in Germany). Drop me a pm.

    Don't think I can get my coilwinder back running (it had a Prop for the counter), but I'll do it on the lathe.


    Nick

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    Never use force, just go for a bigger hammer!

    The DIY Digital-Readout for mills, lathes etc.:
    YADRO
  • Beau SchwabeBeau Schwabe Posts: 6,568
    edited 2009-08-29 19:42
    Nice offer Nick!.. I was going to suggest winding on a toothpick.

    Cenlasoft, remember you can also place two inductors in series to get the required inductance for your LC circuit. This way you can off-load some of the inductor on the "sense head" making it smaller. I wouldn't make the ratio too high though between the sense head and the other series inductor 1:4 would be about the max I would go. Much more than that and your signal is harder to deal with.

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    Beau Schwabe

    IC Layout Engineer
    Parallax, Inc.
  • CenlasoftCenlasoft Posts: 265
    edited 2009-08-29 20:07
    Thanks Nick,

    I will try a few things today and tomorrow and if they don't work, I definitely take you up on your offer. I want to at least pay the shipping.

    Thanks Beau, I'll try your suggestions and post the results.

    Curtis
  • CounterRotatingPropsCounterRotatingProps Posts: 1,132
    edited 2009-08-29 20:08
    Curtis,

    how far down into the metal can you go and get reasonable results?

    Mentioning the CAT scan made me think of a very cool experiment. If you could find a univerisity physics dept. nearby that has a small superconducting lab, you could emerse the sample and coil into liquid helium. The results would be interesting at the very least - you might get deeper penetration. Although all current flows near superconductive temps behave strangely, it would be worth a few trys. And if you get results that interest the physics prof enough, with your connection to the FBI, you might even be able to get some grant funding to study it further!

    cheers
    Howard
    - a firm believer in excluding the Pauli Exclusion Principle [noparse]:)[/noparse]

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
  • CenlasoftCenlasoft Posts: 265
    edited 2009-08-29 20:16
    Thanks Howard,

    I teach at a university, I am a professor of Archaeology and Anthropology and the head of the physics dept is my friend. I did get a grant to see if it was feasable to use non-destructive testing on artifacts. We ordered instruments, supplies, and parallax products. The grant was from the National Park Service. The low temperatures are a great idea. I have been able to go below the suface to about 6mm. I am new to serious electronics so I am learning as I go. I have been· a Parallax customer for about 8 years though and have enjoyed electronics.



    Thanks

    Curtis
  • CounterRotatingPropsCounterRotatingProps Posts: 1,132
    edited 2009-08-29 21:03
    You're welcome, Curtis.

    When you get a chance to discuss the supercooling idea with your physic's dept. friend, please let us know what he thinks.· Many of us here are learning electronics too - and some I'd bet are fascinated with superconductivity - something that few get to play around with, however.

    A·lot of work has been done concerning the effects of superconductivity involving a multitude of coils at the National High Magnetic Field labs in Los Alamos, and at UF and USF (Florida).·· It might well be worth getting in contact with them.· The director is Dr. Boebinger. These folks are wonderful; I am sure he could get you all in contact with the right people there.·

    The Tallahasee facility has the worlds largest superconducting magnets. And they have a full shop dedicated just to coil winding.· They've made all kinds of coils, from very large, to sewing-bobbin sized. (That's the smallest I've seen there, but they probably could make any size.)· I mention this specifically because they have the expertise in coils used in unusual circumstances - might be a great resource.

    NHMFL: http://www.magnet.fsu.edu/

    cheers,
    - Howard in Florida

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
  • Clock LoopClock Loop Posts: 2,069
    edited 2009-08-29 22:07
    Another great use of the props power!

    Someone mentioned using hard drive heads for things a while back...
  • CenlasoftCenlasoft Posts: 265
    edited 2009-08-29 22:45
    Hey Howard,

    I hope that coil lab was here now. These tiny coils are hard to make. I'll talk to my friend about superconductivity. I am having fun though, I'll think I will try the two coils in series. If Beau is around or anyone else, how far can the two coils be apart? I was thinking about putting one near the circuit and one on a probe connected by a shielded cable. thnaks, Curtis
Sign In or Register to comment.