The dry vocal tone combined with the comedic visual, audio and timing elements raises the level humor and snickering 10 fold...from what base level? I dunno, but a 10 fold increase did happen. Take that ToT. 🔨🔨🔨
Just subscribed to your channel. As a wire and sink EDM programmer/operator I can say that your results look pretty good for such a basic machine. If I can offer some advice if I may: -We usually use copper for the sinker EDM. Can handle much higher current; graphite is good but only if it is of low enough porosity otherwise it's worse than copper. Brass is used in the EDM hole machine, probably because of its superior stiffness when drilling e.g. 20xD on 0.3mm dia electrode. -Most, if not all commercial machines work with a EDM table, that is, a set of parameters. Most basic ones would be Time ON, time OFF, peak current and what we call GAP, measured in Volts. Most EDM machines achieve stable feed by monitoring the gap and maintaining it stable via the feed rate, while all else is held more or less constant; this yelds a much stabler arc. If you need any ideas or want to know how we do something in specific here in the pro world of EDM just let me know!
After lots of parts, if you look in the bottom of the tank you will find lots of black powder. Look at it in a microscope and they are tiny balls, like welding splatter. You are taking out tiny chunks, so the balls are tiny, some like smoke particle small. The wear thing, For each amp setting there is an ideal on time that will produce about 2% wear. Miss it and you get more. There is also a max amount of amps an electrode can handle. 25 amps per sqr inch is a safe number for most graphites, coppers are less, depending on what else is in there. Largely it has to do with the melting temps, but also has to do with resistance heating in the electrode. It also has to do with the shape of the wave form in your machine... A flushing hole through the middle of the electrode also helps with wear, by doing a better job of flushing out stuff from the burn. Submerged burns also tend to flush better and give you better control, but open air burns sure are cool :) As a note high wear burns do tend to be faster then the low wear ones. copper/tungsten works well as an electrode, graphite tends to work really well, Aluminum doesn't work that great, but if you happen to have a bunch...
Thanks, that's useful information. I get black powder in the bottom of the tank after just one part! I need to do a lot more comparative testing, so far the only thing I can see that helped with the electrode wear was changing the polarity when cutting steel. Using your value of 25A / sq inch, I should have been running the small square electrode at 600ma, I had it set to 2.5A, though the actual current is not very constant, I would need to do more averaging on the waveform to get a better idea. The pitting on the electrode and workpiece does look quite coarse to me, so maybe I was running it at too high a current.
@@AndysMachines I'll dig through some of the manuals and burn charts I've got. Poco has some nice stuff, but only about their own graphite: www.edmtechman.com/
@@stevebromley4230 i had zero knowledge about EDM ,i watch Andys 2 videos and Steve's excellent link to Poco graphite , now i'm a expert ( just kidding ) , seems its ''D.C. spark erosion of your tool mainly caused by inadequate flushing . What a wonderful resource UA-cam is , truly amazing , liked and subbed .
Love the EDM, starting to build one myself currently and have been watching videos because I was a simple build done on the cheap since its only for a couple uses then will probably gather dust awhile. Your setup is very close to what I am looking for, THANKS for sharing! I have never ran an EDM but been around them alot and diagnosed/repaired them when I was previously a CNC Field Service Engineer for Custom made Sarix EDM Drilling machines at a couple plants in Ohio (Field Service Technician in my opinion, why companies have to add Engineer to titles that are not such....). One thing I would suggest if I may is: Try using Copper, Copper/Tungsten or Graphite rather than brass. If using graphite make sure you have the proper cutting tools (coated with diamond I think) as it is very abrasive and will eat up the end mills that are made for metals.
I was hoping the electrodes would last longer, but I'm probably not using the best combination of materials, and I don't think I've got the machine completely dialled in yet. I've got some pure copper and some graphite to try next.
@@AndysMachines I've had the chance to see a similar demonstration on a commercial machine. A square hole just a bit bigger in aluminium and a copper electrode. The electrode blackened a bit but it wasn't corroded away like the brass one you showed. So I think you're right, it's just not the ideal material. I wish you success on your futur experimentations !
@@AndysMachines I think it's just the nature of a capacitor spark machine. The initial current is super high and fries the electrode along with the workpiece. The pulse machine in Ben Fleming's book is much more gentle on the electrode, but also much less energy efficient since most of the power burns up in the current limiting resistors. Basically it uses such high voltage and resistance that the variable spark gap resistance becomes negligibly small and thus the overall circuit current is determined by the well defined supply voltage and power resistors. There are probably better ways to do current limiting, but I haven't thought of any or found anyone else talking about it. I doubt anything based on current sensing and transistor switching would be fast enough.
@@dekutree64 This is actually a pulsed machine. The capacitor is switched by an IGBT which in theory cuts off the arc current before too much erosion of the electrode occurs. I think I can get much better results with some fine tuning, but even as is I can make things with this machine that I couldn't make any other way.
@@AndysMachines As I understand it, electrode wear happens primarily during initial spark formation, so the IGBT will probably only affect the size of craters in the workpiece. But you certainly can do good work with it even if you go through electrodes quickly.
Great project! Did you measured the actual peak current through the transistor? In theory, with low ESR caps it can reach hundreds of amps (assuming spark has zero resistance ), but your system is doing well with single 30A(120A peak) transistor.
Unfortunately I don't have an accurate way to measure the very short high current pulses, but yes I imagine it's in the many tens or even a hundred or so amps.
I use graphite electrodes with good results in oil. Easy to pattern, come in large diameters for welding and cutting and good longivity. Should work in water as well but haven't tried.
I would really like to make this controller, can you share the plans? As well I'm interested in making your plastic injection machine, you've done a fantastic job on it's design.
I didn't really ever completely finish this machine, so didn't produce any plans. I was going to add CNC control to the table, but I left it half-automatic half-manual. I might revisit it one day when I need some more EMD'd parts, but it's been overtaken by other projects (usual story!)
If you don't mind I would like to get your opinion. I have been thinking about building an EDM machine to make circuit boards "PCB's"... One thought is to use a N-channel mosfet instead of a igbt, I don't know if that would be better or not. My second thought would be to use a tungsten electrode from a TIG welder, seems to me it would be more durable/last longer. Please let me know what you think, and thanks.
Yes, you can use a mosfet instead, I considered doing this. You can even use several mosfets in parallel to increase the current capacity. I don't see why this shouldn't work, you can EDM copper with a copper electrode, but you might want to experiment with the polarity to see which gives less electrode wear. I haven't tried tungsten, but copper tungsten is used for EDM electrodes.
@@AndysMachines Thanks for your help and quick response. My thought was that TIG electrodes handle very high temperatures and pretty high current as well when welding, and they last quite a long time... I don't know for sure though.
Finally a DIY machine that is worth being shown on youtube. Great work, especially the electronics design! The bad erosion rate with electrode eating was electrode positive? Exactly the same what I just experienced on an old AGIE that I put back into action, manual says, electrode positive, but erosion rate is horrible and electrode gets eaten in no time. Changed electrode to negative and it works like charm. So at least brass needs electrode negative....
Yes, it was electrode positive, from what I'd read that was the most common set up. But with a brass electrode and steel workpiece, electrode negative worked much better. This may be because brass is not the best electrode material, I still have a lot of experimenting to do, I think I'll get better results with copper or graphite.
Wonderfull built and wonderfull result ! May i ask what's the purpose of R1 (and what value you use) ? As there's a big capacitor after, i guess it's not to limit current (or maybe to prevent short circuits damage ?)
R1 is to limit the current in case of a short circuit. I actually used a heating element for this resistor as it needs to be high wattage, it's resistance is only a few ohms, I chose it to limit the short circuit current to 4 amps so as not to overload the power supply.
You will need at least the height from the bottom of the coolant tank to the highest part of any hose when in use (usually higher than the nozzle height). Some of the smaller aquarium pumps seem like they will have plenty of flow from the specs, but they only have 0.5 meter max head which won't even reach the table, so it's important to check both the flow and head height and maybe choose one slightly bigger than you think you need (you won't get the rated flow and height both at the same time).
I have no expertise in this field, but I was wondering if a phosphor bronze, tungsten or Inconel electrode might not last somewhat longer or whether they would require higher voltage to overcome a possible higher resistance.
Also if you have XY slide as your electrode becomes worn you can do little circles that will bring them into tolerance even though your electrode is worn
Could you replace the electrode with a wire instead? Obviously you'd need a way of replacing the wire continuously but would the spark generator electronic work unchanged?
Yes, you could use the same type of spark generator for wire EDM. Usually the polarity is reversed as with a wire you don't care so much about electrode wear since new wire is being fed continuously.
Hi Andy, Very interesting project ! Did you build any circuit to protect the mosfet when there is a short circuit or do you simply rely on the stepper backing up fast enough ?
Yes, it measures the voltage across the voltage divider constantly (which is the inverse of the current flowing) and only applies pulses when it's above a set limit. Resistor R1 limits the maximum current through the mosfet and protects its during a short.
From what I understand you used water, wouldn't it be better to use some dielectric fluid ??? If I'm not mistaken commercial EDM machines use a type of kerosene or it is an extremely fluid oil (not very viscous) And for the electrode it seems to me that copper also wears out less and is more efficient, have you tried these two options I mentioned ???
Distilled /deionised water is a dielectric fluid, but you are right, commercially kerosine and light oils are used. I used water just because it was readily available. Yes, I've tried various electrodes including copper and graphite.
@@AndysMachines and how were the results with copper ?? because commercially I see that they use more copper electrodes, about graphite I was curious about the results because I have never seen an EDM machine using graphite as an electrode !!!
I haven't really done enough testing to see a clear winner. Brass is a good all rounder in terms of availability/ease of machining/wear. Graphite is very messy to machine. I didn't notice a big difference between copper and brass.
Part 1 of this showed up in my suggested videos today. I have been looking into doing a wire EDM. No real use for it but it looks like a nice toy. You give a very good explanation of the circuit and it is a pretty simple design. I may have to give this a try. What part number are you using for the transistor? I have never seen the IGBT type before.
it's an IRG4PC40 though it may not be the best choice, it's just what I had to hand. I used an IGBT as they can handle more voltage and current than most MOSFETS, though I'm actually more familiar with MOSFETS and I'm sure they would work as well but you might need 2 or more in parallel.
@@AndysMachines That is quite a part and pretty cheap. I am going to have to look around and see what I have for power supplies. I can see a quick experiment with an arduino.
a recommendation on luer lock tips, you can get a much more accurate (and higher pressure) to position water jet using standard needle tips. These can be 3d printed easily. they also work well for your cnc machine.
Yes, I actually machined a smaller nozzle and threaded it onto the end of the loc-line hose (that's the red part on the end of the normal orange nozzle). It gives better flushing with higher pressure and also uses less water, without it there was too much flow for the filter.
@@AndysMachines are you sharing your gcode or controller code? I'm struggling with the filtering right now... I planned on doing some controller work after I have better filtering. perhaps we an collaborate.
@@JesseSchoch Hi Jesse, I took a look at your circuit, it's basically identical to mine except that you're using two switching circuits. I'm relying on the spark capacitor to do most of the filtering (no high side switching) I'm also averaging the last 4-8 readings from the voltage divider, but that has only a minimal effect. Unfortunately I haven't been able to spend any more time on my EDM since I made this video, I haven't started on the gcode control yet (I will keep it very simple anyway, only a few commands).
@@AndysMachines do you recall how much power you are averaging? My supply has a power average output and it measures around 20-30W. I can push it harder but cooling the mosfets has been an issue. I need to figure out the efficiency of my system to see how much of that is going into the cut. the gate drivers are pulling 4W of power from a separate supply!
@@JesseSchoch I didn't measure the average power, but I think it's closer to 100W. My power supply can provide 240W. The IGBT and spark capacitor do get hot after a while.
It only resonates when doing rapid moves, when cutting it moves very slowly and there is no real vibration. The column of the machine is quite heavy and soaks up any vibrations there are.
I am loving it and I will definatly build something similar. It is an amazing inspiration. Although I will not go for making a injection molding mashine to make my own buttons. I have to draw the line somewhere. Thanks.
Before I had the injection moulder I used to make buttons by turning a 'top hat' shape on a lathe and drilling a hole in the panel. The flange keeps the button in place and it just rests on a tactile switch behind the panel. The problem with round ones is they can spin in the hole so anything you engrave on it won't stay upright.
60 volts at 4 amps, looks good. I think my power supply may be 5 Amps at 60 volts. I found a pressure washer at a garage sale. I have the drill edm head no motor yet. Now to find that transistor and use an arduino for my pulses. What rate would recommend for the pulses?
I think I said in the video what pulse rates I was using, it's variable over something like 20-170μs pulse length and the same range gap in between. I mostly used a short pulse with a longer pause.
Andy this is awesome! DIY EDM is already very cool, but I love the use of your previously made machines. By this point your workshop must be an appreciable % of a Von Neumann universal constructor...
Nice. But why did you opt for sink, and not wire machine. Is the wire so much harder to make, or there's some other reason. Because the wire seems, at least to me, much more universal.
Yes, a wire machine is more versatile and not really that much harder to build, just more parts to it. I may even upgrade this to a wire eventually. But there are also things you can do with a sink EDM that you can't do with a wire and that's what I needed at the time I built this.
@@AndysMachines Thank you for the clarification. I'd prefer to make the wire variety, as regardless of the method used, laser, CNC router, CNC lathe etc., I almost exclusively make small quantities of parts for prototyping. And wire EDM is attractive to me because of it's precision. Of course nowadays there's also a laser/waterjet combination which has similar possibilities, but the prices of such systems are for what I do astronomical. Even the generator like the one from BaxEdm is too expensive for my needs, but trying to put together something like you did doesn't seem such a problem to me, and I believe I'm capable of creating a mechanical part, too, CNC and wire+fluid management. But I suppose it's best to try to assemble the genertor, and probably also test it in sink configuration anyway. I may get back with some additional questiuons if I do try this, if you'll be so kind to answer them. Thanks again.
Well done! Dealing with micro compters and now, microcontrollers, the automation feels like familiar. However, as my use for EDM is small I could think of manual pumping water as well as manual feeding the electrode down. Looking at part 1 will be the next step. By the way, creating 60 volt and the current needed calls for some knowledge...
I've been enjoying your videos for quite a while Andy! I've been breaking my head over this one (no electrical engineer), do you interpolate the 0-5V reading to an averaged arc amperage? I understand that 1) 0V:= shortcircuit = steady state just under 4A I'm guessing depending on R1 after a capacitor discharge spike of I=60V/(measured shortcircuit resistance), and that 2) 5V:=60V cap := 0A arc current. Are the spike and 0A the two points you interpolate between? Also, would it not be enough to use the potential divider to measure the voltage across the cap and advance the electrode to reach a target capacitor voltage? Really looking forward to how this build progresses!
I've tried to keep the current reading/gap adjustment as simple as possible, I'm feeding the voltage from the potential divider straight into an A2D pin on a microcontroller and taking a reading every 2ms or so. The averaging effect of the resistor and capacitor is not that great and the voltage read is still very spikey, the yellow trace on the screen is directly related to the voltage read, there is no extra filtering. Every time the stepper motor steps I compare the last A2D reading with the value of the current limit, if the current is below this reading (ie. the voltage is above the set limit) then the motor steps, if the current is above the limit then the motor pauses until the next step. I could do something much more complicated with a target gap voltage and PID algorithms, but this simple way seems to work fine.
@@AndysMachines Thanks for getting back so quick! I understand everything you mention besides the direct (inverse) relation between the 0-5V reading and the arc current (the magnitude of the yellow trace)
@@deebo7276 Not an engineer either, but I think I can answer your question. The capacitor stores power from the power supply. When the burn mode is activated, the power supply begins charging the capacitor. As the electrode isn't close enough to the work, the spark current is zero, and the voltage quickly rises to the power supply voltage. As the electrode comes into range of the work, and with the IGBT pulsing, sparks begin to occur. These sparks pull power from the capacitor's storage, thus causing its voltage to fall. The stronger the sparks are or the more frequently they occur, the lower the voltage on the capacitor. When the spark current is high, the voltage on the capacitor is low as it is still being recharged by the power supply. Thus one can see the inverse relationship; low voltage on the capacitor implies high spark current, high voltage on the capacitor indicates low spark current.
@@nuclearusa16120 Yes, that's exactly it. The resistor is before the capacitor so as the current increases there is more voltage across the resistor and less across the capacitor.
Thanks! I did try copper, but can't really say yet how it compares to brass. I also got some graphite rod to try, but the stuff is extremely messy to machine and the dust is not good for machine tools, so this is waiting until I have a real need for it.
Is there any chance of plans and code? Everything looks and works great, but there´s no information to it, aside to the Voltage-divider for detecting capacitor-current... In YT, we either find systems that work really just partially, and are full of informations, or systems that work greatly, but without informations...
This is still actually a work in progress, I shared as much information as I really could on the build. Mechanically and controller-wise there's still a lot to do before it will be 'finished'.
exelente pero el electrodo mejor de cobre, yo use arduino para controlar el gap con un divisor de tension y para el ciclo de lavado y como dielectrico mejor usar kerosene o gasoil sds
Do the sparks affect the microcontroller and stepper motor driver performance? Will losing steps occur ? Or after some times, will z axis not go back to home or original position (or zero reading)? Is the pcb is normal breadboard type?
The microcontroller, no. But the cheap stepper drive I used did occasionally get affected and take extra steps. This wasn't a problem as I was only moving in one axis but If I ever get this machine converted to 3-axis I will use better drives that are more immune to noise.
I'm sure you could convert a 3D printer into an EDM machine, but it would be quite a big conversion and not easily reversible or suitable for use as both. The Z axis would really need much finer resolution, so would probably need a different leadscrew and gear reduction. (I have an Ender 3 myself).
I think the biggeset challenge would be dealing with all the electrical noise induced by the process, a 3D printer just isn't designed with such a harsh environment in mind. All the long wires would make perfect antennas, triggering random resets of the controller.
Because I didn't have any at the time. Apparently there is a special type of graphite for this and I now have a rod of it, but it's messy stuff to machine and not good to get in machine slideways so I haven't got around to using it yet.
That's cool ! . For people who are not expert with electronics it could be great if you can provide a diagram about your edm circuit. I don't know if it can be possible. By the way... your project looks great !
Try to get ahold of some EDM graphite(it's literally just called that). Extremely dense compared to what you'd think graphite would be, machines like air-cutting(plus some chipping), and lasts a lot longer. Same deal with EDM copper, but far more expensive. You can also in theory try using graphite gouging rods. Way softer, but stupidly cheap in... rod form. Would need to turn the copper coating off.
Yes, I do have some graphite rod, sold as an electrode (I believe for EDM), but I haven't tried it yet as it's very messy stuff to machine. I haven't heard of 'EDM' copper, but I also have a pure copper rod.
Hello really nice video and experiments ! you can control the current via the resistor, take a 0.1 ohms and a current amplifier, with protections, and feed it on the microcontroller, it will be ok,. but to improve results a microcontroller should be used to keep constant I * time value. and the arcing time is not constant so you must detect the start of the arc and then cound and trig off. Good experiments !
Yes, you could. You'd need a metal-cutting laser of course. Another more simple way would be to use a resist mask and etch them chemically. I just wanted to see what the EDM could do.
Hi Andy, love the project. Do you have any plans to try graphite electrodes? I'm wondering if something like an electric motor brush could be a more durable electrode. Plus it would be very easy to shape.
Yes, I do have a graphite rod to use as an electrode that I haven't tried yet, it was quite expensive. I hadn't thought of motor brushes, that might be a good source of material to experiment with.
Obviously de ionised water is the safest di electric but it looks like you are running into contamination problems with the high tool wear. I understand that there has been some reserch into using glycerine as a safer dielectric than dielectric oil.
I haven't noticed any problems after the water has been in use for a while, I don't think any of the particles are soluble and don't seem to change the conductivity, they all settle out eventually leaving clear water, the filter catches most of them.
This video you made is very successful. I also think of doing this from CNC DRİLL EDM. I need an electronic card and digital display. I'm thinking of running it with G code. Where can I buy electronic card and digital display?.
If you want to build something similar, really all the information you need is in the videos. I haven't produced any plans as you probably wouldn't want to build it exactly the same way as I did and it would be extremely hard to locate all the exact same components I used. The basic circuit and theory of operation is explained in the videos. I didn't build it with open sourcing in mind, I built it for my own application. I'm not going to release schematics and code etc. for the UI/control system as 1. It's not finished yet and 2. It's quite a lot more complicated than it needs to be and specifically tailored to my application, so probably nobody else would want exactly the same design. I don't see what more information I can really provide, but if anything's not clear in the video explanation I'm happy to answer questions.
@@AndysMachines i wonder what parts u use to make arc & control it (normal analog way ) .....that el part drawing is nice ,but confusing when u dont know edm ,which transformers ,resistors, caps to use to get nice spark :)
@@MrCcfly You really only need a power supply that's at least 40V at several amps. The capacitor and resistor just need to be able to handle the power you are putting through them or they will get hot. Camera flash capacitors are good (low ESR) and high wattage resistors (you can connect many in parallel).
If memory serves me right your electrode should be positive a little bit more on time use your off time to control the flushing you should see the electrode becoming a little silver as replating should be occurring otherwise you need more on time. I would stay away from brass straight copper is best otherwise you have two different things happening the zinc is coming out much sooner undermining the copper and avoid mixed graphite like copper infiltrated graphite again you have two different things going on graphite should be negative electrode never been a big fan of graphite it's nasty leaves horrible filth copper is one of the best items to use as an electrode that's 100% recyclable. If you want better round hole you can rotate the electrode slowly and you can even very slowly feed a copper infiltrated tungsten block by the electrode to true it up that way you can control the diameter very tightly before you begin burning by doing this and continuing to rotate the electrode in the part you can make round and hold tenths of thousands of an inch on your diameter if you go to that trouble. Instead of water you might wish to just use some diesel fuel at your dielectric you will smell the smell of acetylene that is what comes off is flammable it may flash into little flames here and there if you're near the surface unless the oil gets hot it should not catch on fire but I always keep a fire extinguisher around anyway the water will make hydrogen gas which you have no warning only an explosion.
Thanks for the advice, there's a lot to learn regarding all the different electrode types and variables. I did get some pure copper and graphite to experiment with. Yes, graphite is messy stuff, I don't like machining it for that reason.
@@AndysMachines I have many years of experience from work at one of our national laboratories where we cut everything but die steel. I wrote a rough book on non-conventional machining non-conventional material company that I was working with I believe sold that information to sodik. Mostly on time off time settings different materials including plutonium that was done with wire machine much was wire and a good amount was die sinking. We were using wire machines to make parts that you would normally make on a mill only much more efficiently. Two of the most advanced sodick wire machines we were not able to use them as the software was quirky mostly used fanuc w series machines that no problem because of illegal semiconductors there are no American machines being manufactured back in 95 so they were imported with many names just it went against an agreement to not bring their machines to our market. It ended up a good thing in a way it opened the door for new technology like a soft wired power supply this is big change as most EDM supplies are what they are you don't like the pulse type you must buy a new supply this one allowed many things that were not available then you could put your parameters on a floppy and share them. Fanuc supplied many companies with controls right up until this agreement was no more than no one could get a controlled or a power supply so then that's just when they begin to make their own trying to remember the machine that we had purchased and there was no dealer in Arizona we got it very inexpensively everyone was afraid of this machine I took one of my more difficult programs and went spent a week with it looking under its skirt we took delivery of this machine and a week later they sell many with dealers in place
Aww, more sciency bits, please, and less of the embarrassing dance music. We're here for the Open University vibe. Please be assured: we LOVE the sciency bits. Don't apologise for them. Grow a beard; wear a corduroy jacket, a loud tie and some flared trousers. Let's party like it's 1974. Yay, science. 😁
The dry vocal tone combined with the comedic visual, audio and timing elements raises the level humor and snickering 10 fold...from what base level? I dunno, but a 10 fold increase did happen. Take that ToT. 🔨🔨🔨
Just subscribed to your channel. As a wire and sink EDM programmer/operator I can say that your results look pretty good for such a basic machine. If I can offer some advice if I may:
-We usually use copper for the sinker EDM. Can handle much higher current; graphite is good but only if it is of low enough porosity otherwise it's worse than copper. Brass is used in the EDM hole machine, probably because of its superior stiffness when drilling e.g. 20xD on 0.3mm dia electrode.
-Most, if not all commercial machines work with a EDM table, that is, a set of parameters. Most basic ones would be Time ON, time OFF, peak current and what we call GAP, measured in Volts. Most EDM machines achieve stable feed by monitoring the gap and maintaining it stable via the feed rate, while all else is held more or less constant; this yelds a much stabler arc.
If you need any ideas or want to know how we do something in specific here in the pro world of EDM just let me know!
The Lord sent me here to keep me humble. Thank you for doing that so well.
After lots of parts, if you look in the bottom of the tank you will find lots of black powder. Look at it in a microscope and they are tiny balls, like welding splatter. You are taking out tiny chunks, so the balls are tiny, some like smoke particle small.
The wear thing, For each amp setting there is an ideal on time that will produce about 2% wear. Miss it and you get more. There is also a max amount of amps an electrode can handle. 25 amps per sqr inch is a safe number for most graphites, coppers are less, depending on what else is in there. Largely it has to do with the melting temps, but also has to do with resistance heating in the electrode. It also has to do with the shape of the wave form in your machine... A flushing hole through the middle of the electrode also helps with wear, by doing a better job of flushing out stuff from the burn. Submerged burns also tend to flush better and give you better control, but open air burns sure are cool :) As a note high wear burns do tend to be faster then the low wear ones. copper/tungsten works well as an electrode, graphite tends to work really well, Aluminum doesn't work that great, but if you happen to have a bunch...
Thanks, that's useful information. I get black powder in the bottom of the tank after just one part!
I need to do a lot more comparative testing, so far the only thing I can see that helped with the electrode wear was changing the polarity when cutting steel. Using your value of 25A / sq inch, I should have been running the small square electrode at 600ma, I had it set to 2.5A, though the actual current is not very constant, I would need to do more averaging on the waveform to get a better idea. The pitting on the electrode and workpiece does look quite coarse to me, so maybe I was running it at too high a current.
@@AndysMachines I'll dig through some of the manuals and burn charts I've got. Poco has some nice stuff, but only about their own graphite: www.edmtechman.com/
@@stevebromley4230 i had zero knowledge about EDM ,i watch Andys 2 videos and Steve's excellent link to Poco graphite , now i'm a expert ( just kidding ) , seems its ''D.C. spark erosion of your tool mainly caused by inadequate flushing . What a wonderful resource UA-cam is , truly amazing , liked and subbed .
So what would be the ideal voltage? And frequency? How would you figure out the sweat spot in terms of energy used and amount of material removed?
Love the EDM, starting to build one myself currently and have been watching videos because I was a simple build done on the cheap since its only for a couple uses then will probably gather dust awhile. Your setup is very close to what I am looking for, THANKS for sharing!
I have never ran an EDM but been around them alot and diagnosed/repaired them when I was previously a CNC Field Service Engineer for Custom made Sarix EDM Drilling machines at a couple plants in Ohio (Field Service Technician in my opinion, why companies have to add Engineer to titles that are not such....).
One thing I would suggest if I may is: Try using Copper, Copper/Tungsten or Graphite rather than brass. If using graphite make sure you have the proper cutting tools (coated with diamond I think) as it is very abrasive and will eat up the end mills that are made for metals.
OMG so excited to see him build a syntheciser and make EDM with it :)
The kind of neighbors I wish if I had!
That panel is a work of art
Looking great, didn't know the electrodes wore out so fast though. Self made buttons are some overkill dedication, well done
I was hoping the electrodes would last longer, but I'm probably not using the best combination of materials, and I don't think I've got the machine completely dialled in yet. I've got some pure copper and some graphite to try next.
@@AndysMachines I've had the chance to see a similar demonstration on a commercial machine. A square hole just a bit bigger in aluminium and a copper electrode.
The electrode blackened a bit but it wasn't corroded away like the brass one you showed.
So I think you're right, it's just not the ideal material.
I wish you success on your futur experimentations !
@@AndysMachines I think it's just the nature of a capacitor spark machine. The initial current is super high and fries the electrode along with the workpiece. The pulse machine in Ben Fleming's book is much more gentle on the electrode, but also much less energy efficient since most of the power burns up in the current limiting resistors. Basically it uses such high voltage and resistance that the variable spark gap resistance becomes negligibly small and thus the overall circuit current is determined by the well defined supply voltage and power resistors. There are probably better ways to do current limiting, but I haven't thought of any or found anyone else talking about it. I doubt anything based on current sensing and transistor switching would be fast enough.
@@dekutree64 This is actually a pulsed machine. The capacitor is switched by an IGBT which in theory cuts off the arc current before too much erosion of the electrode occurs. I think I can get much better results with some fine tuning, but even as is I can make things with this machine that I couldn't make any other way.
@@AndysMachines As I understand it, electrode wear happens primarily during initial spark formation, so the IGBT will probably only affect the size of craters in the workpiece. But you certainly can do good work with it even if you go through electrodes quickly.
I could see this setup being used for “broaching” hardened steel, 50+ RC, for special dies.
Ich musste so Lachen als er das Kabel aus dem Crimpstecker gezogen hat :D
Great project! Did you measured the actual peak current through the transistor? In theory, with low ESR caps it can reach hundreds of amps (assuming spark has zero resistance ), but your system is doing well with single 30A(120A peak) transistor.
Unfortunately I don't have an accurate way to measure the very short high current pulses, but yes I imagine it's in the many tens or even a hundred or so amps.
Love your X axis design. ;-)
6:00 I was expecting this 🤭
👍
You can use an aquarium or fish tank pump filter its more effective.
Just an idea
I use graphite electrodes with good results in oil.
Easy to pattern, come in large diameters for welding and cutting and good longivity.
Should work in water as well but haven't tried.
I would really like to make this controller, can you share the plans?
As well I'm interested in making your plastic injection machine, you've done a fantastic job on it's design.
I didn't really ever completely finish this machine, so didn't produce any plans. I was going to add CNC control to the table, but I left it half-automatic half-manual. I might revisit it one day when I need some more EMD'd parts, but it's been overtaken by other projects (usual story!)
Excellent design and build, thanks for sharing.
If you don't mind I would like to get your opinion. I have been thinking about building an EDM machine to make circuit boards "PCB's"... One thought is to use a N-channel mosfet instead of a igbt, I don't know if that would be better or not. My second thought would be to use a tungsten electrode from a TIG welder, seems to me it would be more durable/last longer. Please let me know what you think, and thanks.
Yes, you can use a mosfet instead, I considered doing this. You can even use several mosfets in parallel to increase the current capacity.
I don't see why this shouldn't work, you can EDM copper with a copper electrode, but you might want to experiment with the polarity to see which gives less electrode wear. I haven't tried tungsten, but copper tungsten is used for EDM electrodes.
@@AndysMachines Thanks for your help and quick response. My thought was that TIG electrodes handle very high temperatures and pretty high current as well when welding, and they last quite a long time... I don't know for sure though.
Of course mine wouldn't be doing what yours does. mine would basically be drawing lines, so no shaped electrodes, just a point.
Finally a DIY machine that is worth being shown on youtube.
Great work, especially the electronics design!
The bad erosion rate with electrode eating was electrode positive?
Exactly the same what I just experienced on an old AGIE that I put back into action, manual says, electrode positive, but erosion rate is horrible and electrode gets eaten in no time. Changed electrode to negative and it works like charm. So at least brass needs electrode negative....
Yes, it was electrode positive, from what I'd read that was the most common set up. But with a brass electrode and steel workpiece, electrode negative worked much better. This may be because brass is not the best electrode material, I still have a lot of experimenting to do, I think I'll get better results with copper or graphite.
Wonderfull built and wonderfull result !
May i ask what's the purpose of R1 (and what value you use) ? As there's a big capacitor after, i guess it's not to limit current (or maybe to prevent short circuits damage ?)
R1 is to limit the current in case of a short circuit. I actually used a heating element for this resistor as it needs to be high wattage, it's resistance is only a few ohms, I chose it to limit the short circuit current to 4 amps so as not to overload the power supply.
@@AndysMachines Thx ! I think i'll try this buit to :)
This pump seems perfect for other liquid cooling applications.
Yes, I've used several of these as coolant pumps for mills, lathes etc. it's a 1200l/h version, smaller ones don't have enough head height.
@@AndysMachines What head height is the sort of minimum? I currently have a little bit of a to much coolant so i sort of drown in it on my mill.
You will need at least the height from the bottom of the coolant tank to the highest part of any hose when in use (usually higher than the nozzle height). Some of the smaller aquarium pumps seem like they will have plenty of flow from the specs, but they only have 0.5 meter max head which won't even reach the table, so it's important to check both the flow and head height and maybe choose one slightly bigger than you think you need (you won't get the rated flow and height both at the same time).
I have no expertise in this field, but I was wondering if a phosphor bronze, tungsten or Inconel electrode might not last somewhat longer or whether they would require higher voltage to overcome a possible higher resistance.
EDM with EDM. Nice.
Also if you have XY slide as your electrode becomes worn you can do little circles that will bring them into tolerance even though your electrode is worn
Yes, I believe that's called orbiting? And is a function built in to many commercial EDM machines.
Could you replace the electrode with a wire instead? Obviously you'd need a way of replacing the wire continuously but would the spark generator electronic work unchanged?
Yes, you could use the same type of spark generator for wire EDM. Usually the polarity is reversed as with a wire you don't care so much about electrode wear since new wire is being fed continuously.
I love that's subscribe button 😂😂😂
Hi Andy,
Very interesting project ! Did you build any circuit to protect the mosfet when there is a short circuit or do you simply rely on the stepper backing up fast enough ?
Yes, it measures the voltage across the voltage divider constantly (which is the inverse of the current flowing) and only applies pulses when it's above a set limit. Resistor R1 limits the maximum current through the mosfet and protects its during a short.
From what I understand you used water, wouldn't it be better to use some dielectric fluid ??? If I'm not mistaken commercial EDM machines use a type of kerosene or it is an extremely fluid oil (not very viscous)
And for the electrode it seems to me that copper also wears out less and is more efficient, have you tried these two options I mentioned ???
Distilled /deionised water is a dielectric fluid, but you are right, commercially kerosine and light oils are used. I used water just because it was readily available. Yes, I've tried various electrodes including copper and graphite.
@@AndysMachines and how were the results with copper ?? because commercially I see that they use more copper electrodes, about graphite I was curious about the results because I have never seen an EDM machine using graphite as an electrode !!!
I haven't really done enough testing to see a clear winner. Brass is a good all rounder in terms of availability/ease of machining/wear. Graphite is very messy to machine. I didn't notice a big difference between copper and brass.
Wow, your project got useful so fast!
Part 1 of this showed up in my suggested videos today. I have been looking into doing a wire EDM. No real use for it but it looks like a nice toy. You give a very good explanation of the circuit and it is a pretty simple design. I may have to give this a try. What part number are you using for the transistor? I have never seen the IGBT type before.
it's an IRG4PC40 though it may not be the best choice, it's just what I had to hand. I used an IGBT as they can handle more voltage and current than most MOSFETS, though I'm actually more familiar with MOSFETS and I'm sure they would work as well but you might need 2 or more in parallel.
@@AndysMachines That is quite a part and pretty cheap. I am going to have to look around and see what I have for power supplies. I can see a quick experiment with an arduino.
a recommendation on luer lock tips, you can get a much more accurate (and higher pressure) to position water jet using standard needle tips. These can be 3d printed easily. they also work well for your cnc machine.
Yes, I actually machined a smaller nozzle and threaded it onto the end of the loc-line hose (that's the red part on the end of the normal orange nozzle). It gives better flushing with higher pressure and also uses less water, without it there was too much flow for the filter.
@@AndysMachines are you sharing your gcode or controller code? I'm struggling with the filtering right now... I planned on doing some controller work after I have better filtering. perhaps we an collaborate.
@@JesseSchoch Hi Jesse, I took a look at your circuit, it's basically identical to mine except that you're using two switching circuits. I'm relying on the spark capacitor to do most of the filtering (no high side switching) I'm also averaging the last 4-8 readings from the voltage divider, but that has only a minimal effect. Unfortunately I haven't been able to spend any more time on my EDM since I made this video, I haven't started on the gcode control yet (I will keep it very simple anyway, only a few commands).
@@AndysMachines do you recall how much power you are averaging? My supply has a power average output and it measures around 20-30W. I can push it harder but cooling the mosfets has been an issue. I need to figure out the efficiency of my system to see how much of that is going into the cut. the gate drivers are pulling 4W of power from a separate supply!
@@JesseSchoch I didn't measure the average power, but I think it's closer to 100W. My power supply can provide 240W. The IGBT and spark capacitor do get hot after a while.
amazing job
Loved both attemps (pt1 & pt2)
I wonder if taking the vibration out of the timing puller might not improve finish a wee skosh.
It only resonates when doing rapid moves, when cutting it moves very slowly and there is no real vibration. The column of the machine is quite heavy and soaks up any vibrations there are.
I am loving it and I will definatly build something similar. It is an amazing inspiration. Although I will not go for making a injection molding mashine to make my own buttons. I have to draw the line somewhere. Thanks.
Before I had the injection moulder I used to make buttons by turning a 'top hat' shape on a lathe and drilling a hole in the panel. The flange keeps the button in place and it just rests on a tactile switch behind the panel. The problem with round ones is they can spin in the hole so anything you engrave on it won't stay upright.
60 volts at 4 amps, looks good.
I think my power supply may be 5 Amps at 60 volts.
I found a pressure washer at a garage sale.
I have the drill edm head no motor yet. Now to find that transistor and use an arduino for my pulses. What rate would recommend for the pulses?
I think I said in the video what pulse rates I was using, it's variable over something like 20-170μs pulse length and the same range gap in between. I mostly used a short pulse with a longer pause.
Great job
Congratulations you did a great job
I didn't understand what kind of electronics you used
They are pic microcontrollers?
This actually uses Atmega 328 microcontrollers (the same as used in arduinos), though I have used PICs in many of my other projects.
Andy this is awesome! DIY EDM is already very cool, but I love the use of your previously made machines.
By this point your workshop must be an appreciable % of a Von Neumann universal constructor...
Thanks Alex!
Yes, some of my machines just make parts for other machines, which make parts for other machines, which...
@@AndysMachinesso glad it isn't just me building machines to fix machines that I bought to fix another machine.
Just curious -- instead of brass electrodes, could you use tungsten? Would it last longer than brass?
Very good video, i enjoyed it and plan to use some of the ideas when i build my EDM machine.
So big thx for the inspiration :)
Great Video , Nice work
should have glued the X axis button on the plastic container so that when you push the X axis "button" you push the plastic container along. ;)
Now that's an idea! Maybe I'd need one either end, X+ and X-.
Nice. But why did you opt for sink, and not wire machine. Is the wire so much harder to make, or there's some other reason. Because the wire seems, at least to me, much more universal.
Yes, a wire machine is more versatile and not really that much harder to build, just more parts to it. I may even upgrade this to a wire eventually. But there are also things you can do with a sink EDM that you can't do with a wire and that's what I needed at the time I built this.
@@AndysMachines Can the same generator work for wire, too, or you'd need to modify it?
Yes, it should work for wire too, though I think often the polarity is reversed for wire vs. sink EDM.
@@AndysMachines Thank you for the clarification. I'd prefer to make the wire variety, as regardless of the method used, laser, CNC router, CNC lathe etc., I almost exclusively make small quantities of parts for prototyping. And wire EDM is attractive to me because of it's precision. Of course nowadays there's also a laser/waterjet combination which has similar possibilities, but the prices of such systems are for what I do astronomical. Even the generator like the one from BaxEdm is too expensive for my needs, but trying to put together something like you did doesn't seem such a problem to me, and I believe I'm capable of creating a mechanical part, too, CNC and wire+fluid management. But I suppose it's best to try to assemble the genertor, and probably also test it in sink configuration anyway. I may get back with some additional questiuons if I do try this, if you'll be so kind to answer them. Thanks again.
For very thin materials like that I would use Photo etching but I do very much like the EDM machine for other applications
Yes, ECM (electrochemical machining) is another good way for thin materials you might photo etch.
Well done! Dealing with micro compters and now, microcontrollers, the automation feels like familiar. However, as my use for EDM is small I could think of manual pumping water as well as manual feeding the electrode down. Looking at part 1 will be the next step.
By the way, creating 60 volt and the current needed calls for some knowledge...
I've been enjoying your videos for quite a while Andy! I've been breaking my head over this one (no electrical engineer), do you interpolate the 0-5V reading to an averaged arc amperage? I understand that 1) 0V:= shortcircuit = steady state just under 4A I'm guessing depending on R1 after a capacitor discharge spike of I=60V/(measured shortcircuit resistance), and that 2) 5V:=60V cap := 0A arc current. Are the spike and 0A the two points you interpolate between? Also, would it not be enough to use the potential divider to measure the voltage across the cap and advance the electrode to reach a target capacitor voltage? Really looking forward to how this build progresses!
I've tried to keep the current reading/gap adjustment as simple as possible, I'm feeding the voltage from the potential divider straight into an A2D pin on a microcontroller and taking a reading every 2ms or so. The averaging effect of the resistor and capacitor is not that great and the voltage read is still very spikey, the yellow trace on the screen is directly related to the voltage read, there is no extra filtering. Every time the stepper motor steps I compare the last A2D reading with the value of the current limit, if the current is below this reading (ie. the voltage is above the set limit) then the motor steps, if the current is above the limit then the motor pauses until the next step. I could do something much more complicated with a target gap voltage and PID algorithms, but this simple way seems to work fine.
@@AndysMachines Thanks for getting back so quick! I understand everything you mention besides the direct (inverse) relation between the 0-5V reading and the arc current (the magnitude of the yellow trace)
@@deebo7276 Not an engineer either, but I think I can answer your question. The capacitor stores power from the power supply. When the burn mode is activated, the power supply begins charging the capacitor. As the electrode isn't close enough to the work, the spark current is zero, and the voltage quickly rises to the power supply voltage. As the electrode comes into range of the work, and with the IGBT pulsing, sparks begin to occur. These sparks pull power from the capacitor's storage, thus causing its voltage to fall. The stronger the sparks are or the more frequently they occur, the lower the voltage on the capacitor. When the spark current is high, the voltage on the capacitor is low as it is still being recharged by the power supply. Thus one can see the inverse relationship; low voltage on the capacitor implies high spark current, high voltage on the capacitor indicates low spark current.
@@nuclearusa16120 Yes, that's exactly it. The resistor is before the capacitor so as the current increases there is more voltage across the resistor and less across the capacitor.
Andy, have you tried using carbon or copper electrodes? Excellent video
Thanks! I did try copper, but can't really say yet how it compares to brass. I also got some graphite rod to try, but the stuff is extremely messy to machine and the dust is not good for machine tools, so this is waiting until I have a real need for it.
Is there any chance of plans and code? Everything looks and works great, but there´s no information to it, aside to the Voltage-divider for detecting capacitor-current... In YT, we either find systems that work really just partially, and are full of informations, or systems that work greatly, but without informations...
This is still actually a work in progress, I shared as much information as I really could on the build. Mechanically and controller-wise there's still a lot to do before it will be 'finished'.
HALL EFFECT current device. Totally isolated and FAST.
exelente pero el electrodo mejor de cobre, yo use arduino para controlar el gap con un divisor de tension y para el ciclo de lavado y como dielectrico mejor usar kerosene o gasoil sds
Do the sparks affect the microcontroller and stepper motor driver performance? Will losing steps occur ? Or after some times, will z axis not go back to home or original position (or zero reading)? Is the pcb is normal breadboard type?
The microcontroller, no. But the cheap stepper drive I used did occasionally get affected and take extra steps. This wasn't a problem as I was only moving in one axis but If I ever get this machine converted to 3-axis I will use better drives that are more immune to noise.
I wonder if a conversion kit could be made for an ender 3, using some sort of common rod stock as an elecrode.
I'm sure you could convert a 3D printer into an EDM machine, but it would be quite a big conversion and not easily reversible or suitable for use as both. The Z axis would really need much finer resolution, so would probably need a different leadscrew and gear reduction. (I have an Ender 3 myself).
I think the biggeset challenge would be dealing with all the electrical noise induced by the process, a 3D printer just isn't designed with such a harsh environment in mind. All the long wires would make perfect antennas, triggering random resets of the controller.
hey, is that explosion from The Last Action Hero?
It might look similar!
(Well done, I thought somebody would have spotted it by now!)
it's really necessary to use brass as electrode? Greatings and like from Sapin!
No, it's not necessary, in fact brass is not the best electrode material, it wears quite quickly but is ok for short term use.
@@AndysMachines i mean because of the damage on electrode thank you 4 all
I’ll sub just because you engraved a button
Beautiful work
Why not use graphite for the electrode?
Because I didn't have any at the time. Apparently there is a special type of graphite for this and I now have a rod of it, but it's messy stuff to machine and not good to get in machine slideways so I haven't got around to using it yet.
That's cool ! . For people who are not expert with electronics it could be great if you can provide a diagram about your edm circuit. I don't know if it can be possible. By the way... your project looks great !
Try to get ahold of some EDM graphite(it's literally just called that). Extremely dense compared to what you'd think graphite would be, machines like air-cutting(plus some chipping), and lasts a lot longer.
Same deal with EDM copper, but far more expensive.
You can also in theory try using graphite gouging rods. Way softer, but stupidly cheap in... rod form. Would need to turn the copper coating off.
Yes, I do have some graphite rod, sold as an electrode (I believe for EDM), but I haven't tried it yet as it's very messy stuff to machine. I haven't heard of 'EDM' copper, but I also have a pure copper rod.
Hello really nice video and experiments ! you can control the current via the resistor, take a 0.1 ohms and a current amplifier, with protections, and feed it on the microcontroller, it will be ok,. but to improve results a microcontroller should be used to keep constant I * time value. and the arcing time is not constant so you must detect the start of the arc and then cound and trig off. Good experiments !
Awesome machine, great video :)
This is the most useful cold fusion reactor I have ever seen 😊
Surly you would laser cut them at this thickness?
Yes, you could. You'd need a metal-cutting laser of course. Another more simple way would be to use a resist mask and etch them chemically. I just wanted to see what the EDM could do.
Hi Andy, love the project. Do you have any plans to try graphite electrodes? I'm wondering if something like an electric motor brush could be a more durable electrode. Plus it would be very easy to shape.
Yes, I do have a graphite rod to use as an electrode that I haven't tried yet, it was quite expensive. I hadn't thought of motor brushes, that might be a good source of material to experiment with.
cool
10:07 your coffee is ready
Amazing!
Obviously de ionised water is the safest di electric but it looks like you are running into contamination problems with the high tool wear. I understand that there has been some reserch into using glycerine as a safer dielectric than dielectric oil.
I haven't noticed any problems after the water has been in use for a while, I don't think any of the particles are soluble and don't seem to change the conductivity, they all settle out eventually leaving clear water, the filter catches most of them.
superbe ! congratulations!
The display that you have built is awesome what did you use to control it.
It's a 320x240 2.4" LCD display. I'm driving it with just an 8 bit microcontroller (ATmega328) using custom code I wrote in assembly language.
Congratulations you won a new subscriber
This video you made is very successful. I also think of doing this from CNC DRİLL EDM. I need an electronic card and digital display. I'm thinking of running it with G code. Where can I buy electronic card and digital display?.
Can i use Raspberry Pi as a controller perhaps?
You could, but that would be totally overkill. Plus you'd have to wait a minute or so for it to boot up when you turn it on, this powers on instantly.
Cool
Wow I wish I has access to that, I need to make a punch and die and it would be a breeze with that. Excellent machine
How about a sheet metal tack welder that won't electrocute myself and a copper coil kiln 😌
Great project Mr Andys, with Patreon filiate can i acess this project ? Draws? Codes? Tks
If you want to build something similar, really all the information you need is in the videos.
I haven't produced any plans as you probably wouldn't want to build it exactly the same way as I did and it would be extremely hard to locate all the exact same components I used.
The basic circuit and theory of operation is explained in the videos. I didn't build it with open sourcing in mind, I built it for my own application. I'm not going to release schematics and code etc. for the UI/control system as 1. It's not finished yet and 2. It's quite a lot more complicated than it needs to be and specifically tailored to my application, so probably nobody else would want exactly the same design.
I don't see what more information I can really provide, but if anything's not clear in the video explanation I'm happy to answer questions.
@@AndysMachines i wonder what parts u use to make arc & control it (normal analog way ) .....that el part drawing is nice ,but confusing when u dont know edm ,which transformers ,resistors, caps to use to get nice spark :)
@@MrCcfly You really only need a power supply that's at least 40V at several amps. The capacitor and resistor just need to be able to handle the power you are putting through them or they will get hot. Camera flash capacitors are good (low ESR) and high wattage resistors (you can connect many in parallel).
Andy you have to much time on you'r hands.. jokes aside I love your work.
No, not really! If only I could spend more time on stuff like this!
Esque tu pourras faire une vidéo pour voir comment ta fais ton profil??
If memory serves me right your electrode should be positive a little bit more on time use your off time to control the flushing you should see the electrode becoming a little silver as replating should be occurring otherwise you need more on time.
I would stay away from brass straight copper is best otherwise you have two different things happening the zinc is coming out much sooner undermining the copper and avoid mixed graphite like copper infiltrated graphite again you have two different things going on graphite should be negative electrode never been a big fan of graphite it's nasty leaves horrible filth copper is one of the best items to use as an electrode that's 100% recyclable.
If you want better round hole you can rotate the electrode slowly and you can even very slowly feed a copper infiltrated tungsten block by the electrode to true it up that way you can control the diameter very tightly before you begin burning by doing this and continuing to rotate the electrode in the part you can make round and hold tenths of thousands of an inch on your diameter if you go to that trouble.
Instead of water you might wish to just use some diesel fuel at your dielectric you will smell the smell of acetylene that is what comes off is flammable it may flash into little flames here and there if you're near the surface unless the oil gets hot it should not catch on fire but I always keep a fire extinguisher around anyway the water will make hydrogen gas which you have no warning only an explosion.
Thanks for the advice, there's a lot to learn regarding all the different electrode types and variables. I did get some pure copper and graphite to experiment with. Yes, graphite is messy stuff, I don't like machining it for that reason.
@@AndysMachines I have many years of experience from work at one of our national laboratories where we cut everything but die steel.
I wrote a rough book on non-conventional machining non-conventional material company that I was working with I believe sold that information to sodik.
Mostly on time off time settings different materials including plutonium that was done with wire machine much was wire and a good amount was die sinking.
We were using wire machines to make parts that you would normally make on a mill only much more efficiently.
Two of the most advanced sodick wire machines we were not able to use them as the software was quirky mostly used fanuc w series machines that no problem because of illegal semiconductors there are no American machines being manufactured back in 95 so they were imported with many names just it went against an agreement to not bring their machines to our market.
It ended up a good thing in a way it opened the door for new technology like a soft wired power supply this is big change as most EDM supplies are what they are you don't like the pulse type you must buy a new supply this one allowed many things that were not available then you could put your parameters on a floppy and share them.
Fanuc supplied many companies with controls right up until this agreement was no more than no one could get a controlled or a power supply so then that's just when they begin to make their own trying to remember the machine that we had purchased and there was no dealer in Arizona we got it very inexpensively everyone was afraid of this machine I took one of my more difficult programs and went spent a week with it looking under its skirt we took delivery of this machine and a week later they sell many with dealers in place
How does this channel have such low views
Sylvester Stallone can take a lot more voltage than this. He likes it when it's around 64kV. :)
Aww, more sciency bits, please, and less of the embarrassing dance music. We're here for the Open University vibe.
Please be assured: we LOVE the sciency bits. Don't apologise for them. Grow a beard; wear a corduroy jacket, a loud tie and some flared trousers. Let's party like it's 1974.
Yay, science. 😁
Got to be an easier way
use copper
Dislike - I was looking for an Electronic Dance Music machine…
Oh god please NO. That music is terrible.
Yes, everybody's taste if different, but I think you're missing the point, it's EDM music! (It's supposed to be kind of a joke)
Are these parts for making iris mechanisms for your new Terminator?
Correct! I'll have a new video out on this in hopefully a few days.
@@AndysMachines Cool! I just discovered your channel and it's all really interesting :)
Cool