| Electronics > Projects, Designs, and Technical Stuff |
| (ultra)acoustic in circuit component level crack detector? |
| (1/12) > >> |
| coppercone2:
So you can get fractures in components that lead to failure, sometimes they are slow to propagate. Can you possibly make something like ultrasonic tweezers to get a profile of a part after its soldered down so you can basically use the tweezers are a VNA type device but that works with ultrasonic frequencies to see a mechanical change in a part? Or would the substrate give too much noise? i thought it might be as simple as getting some metal tweezers that are somewhat thick, mounting them to a compliant grip to isolate your hands (silicone rubber), and mount two of those cheap ultrasonic devices to it some how (the little cans with the mesh meant to be transducers that also happen to receive). The idea is that you would finish a PCB then measure the various SMD components like capacitors and then remeasure them as a part of inspection. Perhaps like after a burn in period to eliminate initial drift factors (but this is where the crack might form so I am not sure). Or you might need to cut the tweezers in half carefully with a jewelers saw and then bond them through a vibration dampener (so you measure the component), Maybe two different tools. It depends on the impedance of the component compared to the impedance of the junction its kind of like setting a VNA impedance I think? In lieu of complex dsp or pressure settings in measurements perhaps you can record the oscilloscope readout in a video form when a 'squeeze' test is performed to get rid of contact pressure related problems (since you get a a contact pressure sweep this way). I was hoping a crack might show up as a very visible anomaly. Maybe you can just bond a inductor thats going to be vibrated with a magnetic field then a real transducer. If you use a shielded inductor and shielded cable (maybe mu metal) it might reduce crosstalk through free space and limit it to magnetics. |
| cdev:
Try it! |
| coppercone2:
I don't want to waste my time if there is some kind of crack detector on the market already. I have serious doubts. And I don't know how to do the mechanical coupling well or good transducers to use. I know piezo is hard. And I don; thave a impedance analyzer. |
| cdev:
People invent huge things and don't get a penny from them all the time. It's in many cases better to sit on your ideas until you have some capital. Or they will just get stolen from you. |
| Kleinstein:
I am afraid the solder and board can have quite some effect on the ultrasonic properties. So chances are one would only see major defects and not those small initial ones. With ceramic caps, which would be my main target ther piezo electric properties of the capacitors make things more complicate, as there would be additional damping, that depends on the history of the part. The transducers of choice would be likely a piezo type. However those small part would need high frequencies - more like several MHz for 1206 size and more if smaller. So it is no am easy task. |
| Navigation |
| Message Index |
| Next page |