diff --git a/electronics/readme.md b/electronics/readme.md index 2eeec65d864c8bbb1d0e8b78727c78fba4ddc942..02581b5ff53fb477ff1ec258776b14b373e47d8d 100644 --- a/electronics/readme.md +++ b/electronics/readme.md @@ -168,15 +168,34 @@ It's nice to see that the voltage divider output of the spark capacitors gives a Here's what the sparking looks like: + +  -#### next steps: +And here's the hole I made… The cut is remarkably asymmetric; the tool (resistor lead) had som surface damage but remained largely in tact while a significant amount of the the workpiece (bullet connector) was worn away: + +<img src="images/edm_testing7.png" width="350px"><img src="images/edm_testing8.png" width="350px"> + +I also did a test with a brass GIK and another bullet connector (this was before I figured out the right polarity so the bullet connector took way to damage than the GIK part): + +<img src="images/edm_testing2.png" width="350px"><img src="images/edm_testing3.png" width="350px"> + + + +#### Next Steps: - figure out how to increase frequency + - I think this will likely involve an auxillary charge pump + - Try a smaller capacitor… this should get initially charged much faster and as long as it has enough charge to switch the MOSFET, it should be no different than a big hefty capacitor. + - according to the datasheet, they're theoretically capable of getting 20kHz PWM using a [15nC transistor](https://www.mouser.com/ProductDetail/Infineon-Technologies/BSC252N10NSFGATMA1?qs=K00xGehIljsuZG6JAgT75A%3D%3D) (4x less gate charge than mine)… but that's 4x, not 40x so I'm a little confused still. This transistor would probably work for me given that I likely over-spec'd the one I originally picked. + - or... I could switch to P-channel MOSFET's and save having to do all of this charge pumping. + - this one is pretty good: [P-channel 150V, 300W, 55nC](https://www.digikey.com/product-detail/en/ixys/IXTA36P15P/IXTA36P15P-ND/1995361) or [100V, 60nC, 50W](https://www.digikey.com/product-detail/en/rohm-semiconductor/RSJ250P10TL/RSJ250P10TLCT-ND/3443680) + - these can be driven with a relatively [simple level shifter circuit](https://www.infineon.com/dgdl/Infineon+-+Application+Note+-+Power+MOSFETs+-+OptiMOS+P-channel.pdf?fileId=db3a304341e0aed001420380cc13101b) - try higher power (test 2x 50V supply) - put it in the case for >50V - start working on control: - - sample VC (with ADC) in realtime and see how clear a signal I get (at 500Hz, I'd expect it to be pretty good with ~2Msps) + - sample VC (with ADC) in realtime and see how clear a signal I get (at 500Hz, I'd expect it to be pretty good with ~2Msps)… Zoomed out on the scope, I can see the pulses quite clearly even at only 100Ksps. - start with just having this give an output to tell me to turn the adjustment screw or not - - make a stepper motor board and connect the two together - - or… just use the tabletop instron... \ No newline at end of file + - make a stepper motor board and connect the two together (maybe with an opto-isolated serial connection?) + - or… just use the tabletop instron… or better yet, the Desktop EDM! +- pretty soon I'm also going to have to start worrying about a constant source or clean DI water. For these little tests I can just squirt some in a beaker but that won't work for long given how hazy the water gets after just a few minutes of constant burning. \ No newline at end of file diff --git a/electronics/video/rapid_burn.gif b/electronics/video/rapid_burn.gif new file mode 100644 index 0000000000000000000000000000000000000000..11b169e71fb1cf2a0adf7a2fb20e586602e17c73 Binary files /dev/null and b/electronics/video/rapid_burn.gif differ