I met Benjamin Fleming at a maker fair a few years ago, and he was using kerosene (also used as lamp oil) and I think that he was using a high frequency spark stabilization circuit from a welder. That would act very similar to a pulse circuit, yet be fairly simple, with a bit less control. Ben mentioned how when he started there was NO information available without NDA's, so he had to reverse engineer what he could from talking to operators, and what he could glean from sales literature. I talked to my father (a retired aerospace CNC programmer) and he mentioned that he had once made a sinking edm electrode with 0.010 inch features in graphite that was electroplated with copper before it was to be used. I hope some of this helps.
I have a cheap Chinese TIG welder with an adjustable current setting. I’m not sure what the voltage is but suspect that it’s around 40v max. But that’s just a guess.
Hi ! don't know anything about building a spark eroda , however, have spend many years operating them. Three rules of EDM is: Flushing , Flushing , Flushing - the burnt metal has to be removed. The Most common flushing method is known as auto flushing. Here, the electrode approaches the workpiece and sparks, then retracts a couple of millimeters and approaches again. This continues until the desired depth is achieved. The speed, on time, off time can all be controlled by the machine For your big electrode the power looks good for roughing. Way to much power for those thin blade electrodes.
Wonderful video! I especially appreciate your openness and honesty in sharing experiments that did not work out, and the persistence in the face of initial unwanted results that this shows. Seeing the process of persistent experimentation is very educational in itself!
I played around with edm technology about 9 years ago. I made a hack saw with a blade made of a strip of brass sheet metal. I was able to cut aluminum in a bucket of rain water. The blade kept getting stuck so I put an air core inductor coil inline with the circuit. This causes the spark to happen when metal contact opens rather than when metal contact touches. Inductive sparking is different than regular sparking. Anyways the coil I used was a coil from a cut open microwave oven transformer. I can't remember if it was the primary thick wire coil or the secondary thin wire coil. Microwave oven transformers are easy to cut open with an angle grinder. The voltage used was about 30 volts AC. The coil gets hot and needs a cooling fan. Hope this helps.
Hello Ralf! Nice to see you still busy in the work shop. I built an EDM machine from Robert Langlois plans around 1995. It works well but the relaxation oscillator makes for slow work. What is needed is a forced oscillator to produce a high frequency for faster work. I don't have the expertise to design such a circuit, and Robert, whom I knew personally, was not interested in developing such a device suitable for hobby construction. For slotting work such as you wish to do, a wire EDM machine would work very well. Robert Langlois described and built such a machine. For slotting applications a very simple table drive is all that's required. Rotary indexing can be done by hand, perhaps using a digital indexer for simplicity. The table drive would embody the electrode drive you have on your die sinking machine. I found that a very light mineral oil worked well with copper electrodes. Normally I use distilled water with brass electrodes. For lengthy work I filter the water with a 1 micron water filter. Regards from Canada, Wolfgang
A 555 timer can be used, Then use pairs of mosfets (one to charge the cap, and the other to discharge it). You can user a 74123 to create dead time between pulses. You can also set up multiple pairs of mosfets and Caps, since cap charging time takes longer than discharge time. The easiest what do this is with a microcontroller.
For graphite, did you use pencil lead or something else? And what exactly does your power supply do? Is it a square wave? What’s the duty cycle, voltage, and current?
I wish I could help. I met Ben decades ago at a CNC convention in Michigan. I bought his book and built the unit and never finished it. I do remember Ben helping others online over the years. Have you tried to get his input on your needs? Provide him a link to your videos too. I don't know how active he is these days but that's where I'd start. It looks like your using a stepper motor rather than the faster ordinary electric motor Ben mentioned in his book. Maybe your book is newer than mine and he made some changes to the design? I got the first edition. Yours emits the wrong sound compared to those producing a stable burn. Yours sounds more like a EDM tap burner I used to run. Those were designed to be fast but they burned up electrodes fast and produced a crude hole that tapered bigger at the bottom of the cut. I do think a 90 degree electrode holder would work better to cut those long slots. Lay your part on it's side with the 90 degree tool coming in from one end. The flushing will need to be plumbed for flow through your part laying on its side. No matter how you go about it I think your going to go through many electrodes. At least they will be easy to make out of sheet material.
Hi, nice to hear your experiences with Ben Fleming, he is still active in the field and I ordered his PCB for the pulse EDM a few weeks ago, unfortunately it didn't arrived till now. I use a servo motor as Ben Fleming suggests. You are right I also think that I have a problem with too much arcs instead of sparks wich I can clearly hear in the changing sound. With my new setup and better electrode material as shown in my newer video it is much better but still not optimal. I thought abut the 90 degrees arranges electrode but first it is hard to align and second it woun't be easy to secure the right cutting depth. Thanks for your kind comment Ralf
The issue I see is that you have multiple points of contact. If you switched over to one plate instead of multiple plates I believe you have better results. Better would be to flow dielectric through the plate so the erosion material is removed. You can do this by taking two thin sheets and routing out some channels and soldering them together and use a pump to flow the dielectric fluid.
Hi, I tried to use just one plate many times but the result was not better. Dielectric flow channels in 0.7mm thin graphite sheets would be a nice challenge and as I have stability problems with plates thinner than 0.6mm I think it will not be possible. But you are right flushing through the electrode is nice and if possible I use it. Thanks for your suggestions and best wishes Ralf
@@myengines2443 Graphite never lasts very long. I think the only way to cut those long deep slots is with a wire edm. You could use a disk and rotate the disk to provide even wear but the diameter of the disk would shrink from errosion. You could also try cutting horizontally with a longer plate instead of plunge cutting vertically, such that the copper plate length is the total cut length perhaps you get enough life out of the electrode for one slot (depending on depth of cut).You could probably get two cuts if you use both sides (plunge horizontally for the first cut, than reverse Y-axis to cut the 2nd slot on the opposite side) The only other way I can think of to cut faster is with a band saw (you would need to cut the blade and weld reweld it for the internal slotting which probably is not practical. Unless you can cut through the entire cylinder, which I suspect is not an option. Another option might be to cut radially using a small slitting saw, or on a lathe with boring bar with a groving tool.
Hi, I tried all mechanical ways to cut the slots you mentioned and a few more. It worked but was a hell of work as you can see in my older videos. With my new setup and graphite electrodes I get good results and soon I will begin to make my new heat exchangers. You can see my progress in my new video. A wire EDM is to expensive/complicated and it would take a too long time to complete. Using a horizontal electrode it would be difficult to burn to exact depth or I need several gangways to just cut one slot. This would take too long as I have to cut hundrets of slots. Thanks for your suggestions and with best wishes Ralf
My uneducated guess would be to try tungsten wire, maybe bent to do a slot. BTW, most EDM I have seen has used an x,y,z moving wire to cut. What is the advantage of using a plane? To my admittedly ignorant way of looking at it, it seems like sparks/arcs might come off the surface of the plane parallel to the cut surface which would continue eating away the workpiece and tool? Well, maybe not if the electric is super well controlled and the cutting edge is consistently closer to the bottom than the plane is to the sides but that seems tricky to achieve. Anyway, thanks for sharing your results.
Hi, the advantage of a sinker EDM is its simplicity. You just need z control (up and down) and no cutting wire management which makes it MUCH easier. Wire EDM in the home workshop is a hobby for itself my sinker EDM is a project of a few weeks. Yes, the electric control is very important so I decided to buld also a true pulse EDM next. Still waiting for the PCB... Thanks for your kind suggestions and with best wishes Ralf
Hi, if you are interested into the electronics the book of Ben Fleming describes eveything in detail. It is even understandable and buildable by the electronic beginner. If you order it in the right online shop it isn't expensive. Greetings Ralf
This is incredible! If you can get it working it would be a massive feat. there's not many ways to easily machine deep thin slits in aluminium in the home shop :)
Hi Joseph, now it works fine as you can see in my newer video, even in 1.4571 stainless steel which is much more demanding than aluminium... Greetings Ralf
As I am an electronic newbie I am not sure. As far as I know no optocopplers or something similar are used but there is a short detection for retraction. Actually I have a problem with the retraction when a short occurs and I blow up a resistor (its an impessive explosion...) Unfortunally I havent any time in the moment but as soon as I can I will repair the circuit and post a video. It would be very nice if you could specify your suggestion. Many thanks and with best wishes Ralf
Lamp oil is the one product that comes closest to Professional Dielectric Fluid for Sinker EDM machines.. for Wire EDM, Deionized water works best.. but you have to treat that with Filter Resin as the conductivity rises when Eroding..That is why i have a resin filter Cartridge in my Wire EDM machine.. (Brother HS-100)
What if you used a zener diode to prevent voltage dropping below a certain threshold. Also how do you know if the circuit isn't simply shorting across debris or the tool/workpiece. Maybe you should be using a voltmeter and ammeter during your tests.
seems pulsing is the way to go. the voltage is only present before the arc forms. need it to break down the dielectric. once its conductive, the voltage drops dramatically. measure a neon sign when its running... they can be 20-30V... despite the transformer saying 15kv. because its limited to 30mA. it will only spit 15kv out when its unloaded, or has over 500k resistance... (eir) any lower resistance, higher load, and it will remain at 30mA but sag the voltage accordingly. anyway. when the arc is formed, its composed of ions in the dielectric, and the electrode materials. polarity dictating the majority of the transfer from negative to positive, much like welding. dcen dcep AC.... as you maintain the arc, the electrode starts breaking down. your "points" start to round off, arc stability becomes erratic on DC, and the material doesnt want to come OUT of the arc, instead transferring to the other electrode. the pulse, if you have say, a variable capacitor (ahem... not so easy? but then again, ive dunked an old moving plate under oil and run it at 10kv...) you can charge to a specific voltage for the breakdown, but only store enough energy to dislocate those metallic ions from the workpiece... then cease the current, leaving said atomised particles floating in mid-solution where they can be flushed away. i n this regard, allowing the capacitor to discharge on its own terms isnt the way to go, rather, once it reaches a certain level its cut off entirely, conduction ceasing abruptly, ideally at max flow. which, with some sampling shouldnt be too hard to discriminate. theres a brief rise as conduction commences, with a peak at full conduction at which point the standard RC rules of a discharging capacitor apply. whilst theres an arc, the ions/particles tend to REMAIN in the arc. its supplying heat, energy, and its got those dielectric and magnetic effects... youre burning, oxidising, and destroying the dielectric rather than the electrode/workpiece. liken it to blowing dust with two air guns. one lifts the dust with short rapid puffs, directed where you need, whilst the other blows each lifted cloud of dust clear... whereas if you blast both you just make a mess. you want a corona discharge, high tension, but as soon as theres current applied it reaches a certain value then is cut-off. think of the lightning rod... you need the surfaces to produce those spikes, which they cant do once theyre actually conducting. tungsten copper electrodes? another aspect to look at is the AC welding of aluminium, the cleaning action of reverse polarity. you just want to get the surface layer suspended, free... then push it away. its not quite the same with TIG as the main goal is the metal stay in place and simply melt, you just have to bust the oxides off. anyway. if anything. you are basically trying to rip that charged surface layer off the plate of a capacitor, then blow it away whilst suspended. the electrode and workpiece IS a capacitor, after all. not heat anything. simply tear a layer of charge and associated particles off.
Would adding some inductance in series help stabilize the arc? This is what they used to do with arc lights. Also, how do you control the feed? Is it a constant rate, or is there a control loop that is driven by the average current through the electrode?
Hi Ralf, I have been running a hansvedt spark edm which I think is an RC type given its advanced age. I have obtained results similar to you. what might make it work better, but I have not tried, are BeCu electrodes or a higher voltage spark from the RC circuit with a shorter RC time constant. Also, your use of multiple blade electrodes seems to me to impose a burden on the RC circuit similar to having the RC tuning incorrect. If you watch your video you will see that the spark is hopping amongst the blades in contrast to what happens with a concentrated discharge within a more electrically limited discharge region (intense local heating and higher ion density). Have you tried to cut with a single blade? And another thought, as the electrode drops down into the slot, the sides of the electrode can continue to conduct laterally. Have you tried to varnish or paint the sides of your blade electrode so it is conductive only on its intended business end? These are just some ideas that I have not tried while also thinking about the pulsed edm. Regards, Mark
I'd be careful with beryllium; IIRC high toxicity. Agree on trying a single blade to start, before attempting the additional complexity of multiple blades. Perhaps the second step would be widely spaced blades. Possibly rotating disks for better dielectric flow? Though that way lies a wire EDM.
Hi, I made best experiences with electrode positive and didnt reversed it for graphite on aluminium. Do you think this will give better results? I havent much experience with aluminium because stainless steel is more important for me. Greetings Ralf
Hi, Could you provide minimum requirements of resistor and capacitor specs, and the DC source ratings for voltage and amperage required to generate sparks
I think he will need to appropiately filter water and to use a deionizer too. I was thinking in this interesting process as one of my best friends works in this area here in Spain but we agree that is very difficult to diy a very accurate system, even he uses it occasionally in criticall operations with an industrial grade machine due to the time consuming task an high costs.
Hi, yes good flushing is important, with kerosene (lamb oil) I also got good results but now I use diesel which is even better. You can see it in my new video. Greetings Ralf
Ralf, Ive been watching your videos for a few years now. I've always wanted to know what your final plans for the stirling engine will be? Is it for backup power? Just for show?
Hi Martin, thanks for your longstanding loyality :-) At first Stirling engines are my passion and I want to develop a relatively simple useful machine with reasonable power. Unfortunatly this is no easy task for a single hobbyist and not even for a company. You have much more setbacks than successes. I try many variations of the Stirling engine and I also often improve my workshop as you see in this video. This takes a lot of time and I only post a video when I think I have something new to show. My final plan is to develop a relatively simple RELIABLE (thats the main challenge I think after more then 10 years of Stirling research and development) Stirling engine for generating power for ourselves (we have much wood, I can produce biogas, the sun is shining...) and maybe for others. It is facinating for me to 'produce' mechanical energy just from heat. Greetings Ralf
Bravo........run copper wire.........flow....u are trying to drill with a flat.....if round maybe rotate electrode.......also diesel........clever.....what volts...amps.....cheers.....
Fascinating! I'm doing some EDM experiments now and this could be useful. For the electrolyte, have you heard of conductive particles? They are said to improve surface finish by making the sparks smaller, but I thought it might stabilise the sparks.
Hi, this sounds very interesting but I have never heared of conductive particles in the dielectic. If you know something more about it please let me know... Ralf
I would have thought that conductive particles are (1) what you get when the fluid is used, (2) prone to providing unwanted 'chain' paths to increase the erosion distance, decreasing accuracy.
You need to control you on/off times more accurately Nice machine just needs a bit more work Interested in your thermal displacement engines. I once restore a large water pumping engine.....it was at least 100 years old,,,,Stirling type using wood fire Maybe 1 hp 2 hp I will post video soon and share with you
Could you just do simple wire EDM? That way you could worry less about it being a flat sheet. I don't know how you have to cut so I'm not sure if that'd work for your application.
Yes you are right a wire EDM is very nice but much more complex to build. When I would decide to build a wire EDM I would not have any time for my main obsession Stirling engines. Thanks and greetings
@@myengines2443 You can buy an old wire edm for very little swap out the motors to run on Mach3 and clean up the flushing system. That is what I am currently doing and building this generator for the spark.
Hi, yes I tried just one blade, it works without difference. I cant control current or frequency as it is a simple R/C machine. My new experiments with diesel as the dielectric and graphite show very promising results which I will show in an update video soon. Thanks for yout suggestions and with best greetings Ralf
@@myengines2443 You can control discharge energy by the size of the caps and inductors in series with the caps. The Inductors will provide current impediance to limit the current. You need to use a recovery diode so the energy stored in the inductor does feed back into the mosfet (possibiliy damaging the mosfet.
@@myengines2443 I have seen the guy from de the baxedm using a series resistor made with 500W lamps reflectors (tungsten core) to limit the current, i didn't saw that part in your setup, maybe it's possible to add it? I think that you have excessive current/voltage, sound very sparky and not so much as frying bacon
Hi Anton, I don't know which burn you mean (M?) but now it takes about an hour to burn 3 slots 1x8mm trough 11mm thick 1.4571 Stainless steel. I dont know the average power consuption of the EDM but it cant be much, the machine has a 2A fuse on the 230V net so I think it is in the region of maximal 100-200 Watts. (2x 24V 2A transformers are for the power line). Greetings Ralf
@@myengines2443 wow thats actually incredible that it takes only 100-200 watts, I thought its much much more because of high voltage needed. I was afraid that you pay for such high tolerances with high electricity bill.
@@KyrychenkoAnton You don't need high voltage. lower voltages will work just fine and also allow more precision cuts because the electrode will be closer and the spark gap much smaller.
Hi, I tried tungsten copper as electrode material as you can see in my newer video. It worked quite well but the thinnest sheets distorted badly. You mean TIG welding tungsten wire? Maybe I will give it a try. Greetings Ralf
@@myengines2443 There is a wire used in Mig Welding; Its a sacrificial wire; sold on spools. (its Cheep.) There are two types of wire used for mig welding, and they are gas and gasses welding wire spools. The one that you would want to use is for gas welding. Using this wire you can focus the electricity onto one spot at a time. Focussing the burn to a single spot. Opposed to larger surface areas. Using larger surfaces as an electrode, the ark will would be more difficult to control; causing heat to warp the electrode. As the electricity takes the path of less resistance. With Mig wire that ark will occur in the one controlled location. And when the Amps begin to climb feed it more wire to maintain a constant burn. The Amps will increase as the tip of the wire begins to deteriorate.
Hi Carl, plz just write your problems in the comments so all can benefit and/or help. I am just an electronic newbie but maybe others can help better. Greetings Ralf
Hi, yes electrochemical machining seems to be very interesting but all I read about it is very complicated and not suitable for the home workshop. But if you know something how to do it easy please let me know. Thanks for you kind comment Ralf
Hi, I do not know much about the ECM technology isn't it to complex for DIY? If not do you know a good description or presentation of a DIY ECM Machine, it looks very interesting. Thanks for your help and with best greetings Ralf
@@myengines2443 Youre capable of building your own 3d printed electrodes and drop on the z, ecm is very slow and produces a lot of waste. sludge and hydrogen.
Hi Ravi, you can buy it directly at Ben Fleming (and his PCB's also) www.homebuiltedmmachines.com Or with a bit of search you will find a seller in your region. Greetings Ralf
using plates is a bad idea. As the plates move down a very very small error in x or y will cause plates to short to work piece. Consider doing the slots with a wire EDM for better results.
Yes you are right a wire EDM would be very nice but this will be very much work so I would not have any time for my original hobby developing Stirling engines. With my new pulse EDM even the thinner sheets work well. Just look at my newer videos.
I tried it but it tends to distord under wear. Otherwise it is very nice and resistant. Look at my newer videos to see my results with it. Thaks for your suggestion. Greetings
There is a problem with pulse EDM is the machine must be big and consumes a lot of energy Because it uses MOSFETs to regulate voltage, may not suitable for home shop. I am using diesel with satisfactory result. Kerosene is even better than EDM oil but too dangerous not worth using.
Hi, I plan to build a pulse EDM also with a book of Ben Fleming now and ordered his PCB. It is a bit bigger and the transformers are bigger and more expensive but not to much. I also use diesel now (ARAL ultimate diesel smells less!) but before I used kerosene (lamb oil) which worked also very good. Greetings Ralf
@@myengines2443 Read papers of Mr. Wojciech Mysinski if you want to know how a commercial machine works. They are fee on the net. Your job is difficult even on a commercial machine.
I met Benjamin Fleming at a maker fair a few years ago, and he was using kerosene (also used as lamp oil) and I think that he was using a high frequency spark stabilization circuit from a welder. That would act very similar to a pulse circuit, yet be fairly simple, with a bit less control. Ben mentioned how when he started there was NO information available without NDA's, so he had to reverse engineer what he could from talking to operators, and what he could glean from sales literature. I talked to my father (a retired aerospace CNC programmer) and he mentioned that he had once made a sinking edm electrode with 0.010 inch features in graphite that was electroplated with copper before it was to be used. I hope some of this helps.
Hi,
thanks for your thoughts and suggestions.
Greetings
Ralf
I have a cheap Chinese TIG welder with an adjustable current setting. I’m not sure what the voltage is but suspect that it’s around 40v max. But that’s just a guess.
Hi !
don't know anything about building a spark eroda , however, have spend many years operating them. Three rules of EDM is: Flushing , Flushing , Flushing - the burnt metal has to be removed. The Most common flushing method is known as auto flushing. Here, the electrode approaches the workpiece and sparks, then retracts a couple of millimeters and approaches again. This continues until the desired depth is achieved. The speed, on time, off time can all be controlled by the machine
For your big electrode the power looks good for roughing. Way to much power for those thin blade electrodes.
Wonderful video! I especially appreciate your openness and honesty in sharing experiments that did not work out, and the persistence in the face of initial unwanted results that this shows. Seeing the process of persistent experimentation is very educational in itself!
Thanks for your kind comment!
I played around with edm technology about 9 years ago. I made a hack saw with a blade made of a strip of brass sheet metal. I was able to cut aluminum in a bucket of rain water. The blade kept getting stuck so I put an air core inductor coil inline with the circuit. This causes the spark to happen when metal contact opens rather than when metal contact touches. Inductive sparking is different than regular sparking. Anyways the coil I used was a coil from a cut open microwave oven transformer. I can't remember if it was the primary thick wire coil or the secondary thin wire coil. Microwave oven transformers are easy to cut open with an angle grinder. The voltage used was about 30 volts AC. The coil gets hot and needs a cooling fan. Hope this helps.
Thank you for sharing your results - very thought provoking. Not an area that I've tried but now I'm tempted :-)
We had a commercial EDM machine where I worked in the 1980’s. It used graphite electrodes under kerosene.
wow... great job done. I wish you luck !!
Hello Ralf!
Nice to see you still busy in the work shop.
I built an EDM machine from Robert Langlois plans around 1995. It works well but the relaxation oscillator makes for slow work. What is needed is a forced oscillator to produce a high frequency for faster work. I don't have the expertise to design such a circuit, and Robert, whom I knew personally, was not interested in developing such a device suitable for hobby construction.
For slotting work such as you wish to do, a wire EDM machine would work very well. Robert Langlois described and built such a machine. For slotting applications a very simple table drive is all that's required. Rotary indexing can be done by hand, perhaps using a digital indexer for simplicity.
The table drive would embody the electrode drive you have on your die sinking machine.
I found that a very light mineral oil worked well with copper electrodes. Normally I use distilled water with brass electrodes. For lengthy work I filter the water with a 1 micron water filter.
Regards from Canada, Wolfgang
A 555 timer can be used, Then use pairs of mosfets (one to charge the cap, and the other to discharge it). You can user a 74123 to create dead time between pulses. You can also set up multiple pairs of mosfets and Caps, since cap charging time takes longer than discharge time. The easiest what do this is with a microcontroller.
For graphite, did you use pencil lead or something else? And what exactly does your power supply do? Is it a square wave? What’s the duty cycle, voltage, and current?
I wish I could help. I met Ben decades ago at a CNC convention in Michigan. I bought his book and built the unit and never finished it. I do remember Ben helping others online over the years. Have you tried to get his input on your needs? Provide him a link to your videos too. I don't know how active he is these days but that's where I'd start.
It looks like your using a stepper motor rather than the faster ordinary electric motor Ben mentioned in his book. Maybe your book is newer than mine and he made some changes to the design? I got the first edition.
Yours emits the wrong sound compared to those producing a stable burn. Yours sounds more like a EDM tap burner I used to run. Those were designed to be fast but they burned up electrodes fast and produced a crude hole that tapered bigger at the bottom of the cut.
I do think a 90 degree electrode holder would work better to cut those long slots. Lay your part on it's side with the 90 degree tool coming in from one end. The flushing will need to be plumbed for flow through your part laying on its side.
No matter how you go about it I think your going to go through many electrodes. At least they will be easy to make out of sheet material.
Hi,
nice to hear your experiences with Ben Fleming, he is still active in the field and I ordered his PCB for the pulse EDM a few weeks ago, unfortunately it didn't arrived till now.
I use a servo motor as Ben Fleming suggests.
You are right I also think that I have a problem with too much arcs instead of sparks wich I can clearly hear in the changing sound.
With my new setup and better electrode material as shown in my newer video it is much better but still not optimal.
I thought abut the 90 degrees arranges electrode but first it is hard to align and second it woun't be easy to secure the right cutting depth.
Thanks for your kind comment
Ralf
The issue I see is that you have multiple points of contact. If you switched over to one plate instead of multiple plates I believe you have better results. Better would be to flow dielectric through the plate so the erosion material is removed. You can do this by taking two thin sheets and routing out some channels and soldering them together and use a pump to flow the dielectric fluid.
Hi,
I tried to use just one plate many times but the result was not better.
Dielectric flow channels in 0.7mm thin graphite sheets would be a nice challenge and as I have stability problems with plates thinner than 0.6mm I think it will not be possible.
But you are right flushing through the electrode is nice and if possible I use it.
Thanks for your suggestions and best wishes
Ralf
@@myengines2443 Graphite never lasts very long. I think the only way to cut those long deep slots is with a wire edm. You could use a disk and rotate the disk to provide even wear but the diameter of the disk would shrink from errosion. You could also try cutting horizontally with a longer plate instead of plunge cutting vertically, such that the copper plate length is the total cut length perhaps you get enough life out of the electrode for one slot (depending on depth of cut).You could probably get two cuts if you use both sides (plunge horizontally for the first cut, than reverse Y-axis to cut the 2nd slot on the opposite side)
The only other way I can think of to cut faster is with a band saw (you would need to cut the blade and weld reweld it for the internal slotting which probably is not practical. Unless you can cut through the entire cylinder, which I suspect is not an option.
Another option might be to cut radially using a small slitting saw, or on a lathe with boring bar with a groving tool.
Hi,
I tried all mechanical ways to cut the slots you mentioned and a few more.
It worked but was a hell of work as you can see in my older videos.
With my new setup and graphite electrodes I get good results and soon I will begin to make my new heat exchangers.
You can see my progress in my new video.
A wire EDM is to expensive/complicated and it would take a too long time to complete.
Using a horizontal electrode it would be difficult to burn to exact depth or I need several gangways to just cut one slot.
This would take too long as I have to cut hundrets of slots.
Thanks for your suggestions and with best wishes
Ralf
My uneducated guess would be to try tungsten wire, maybe bent to do a slot. BTW, most EDM I have seen has used an x,y,z moving wire to cut. What is the advantage of using a plane? To my admittedly ignorant way of looking at it, it seems like sparks/arcs might come off the surface of the plane parallel to the cut surface which would continue eating away the workpiece and tool? Well, maybe not if the electric is super well controlled and the cutting edge is consistently closer to the bottom than the plane is to the sides but that seems tricky to achieve. Anyway, thanks for sharing your results.
Hi,
the advantage of a sinker EDM is its simplicity.
You just need z control (up and down) and no cutting wire management which makes it MUCH easier.
Wire EDM in the home workshop is a hobby for itself my sinker EDM is a project of a few weeks.
Yes, the electric control is very important so I decided to buld also a true pulse EDM next.
Still waiting for the PCB...
Thanks for your kind suggestions and with best wishes
Ralf
nice work, it would be great to make a video on electronic parts and circuit and wiring
Hi,
if you are interested into the electronics the book of Ben Fleming describes eveything in detail.
It is even understandable and buildable by the electronic beginner.
If you order it in the right online shop it isn't expensive.
Greetings
Ralf
This is incredible! If you can get it working it would be a massive feat. there's not many ways to easily machine deep thin slits in aluminium in the home shop :)
Hi Joseph,
now it works fine as you can see in my newer video, even in 1.4571 stainless steel which is much more demanding than aluminium...
Greetings
Ralf
If the slits run all the way through.. use wire edm...
Do you have a galvanical isolated DC-Voltage circuit in parallel to your spark circuit to detect shorts and reduce downwards feed rate?
As I am an electronic newbie I am not sure.
As far as I know no optocopplers or something similar are used but there is a short detection for retraction.
Actually I have a problem with the retraction when a short occurs and I blow up a resistor (its an impessive explosion...)
Unfortunally I havent any time in the moment but as soon as I can I will repair the circuit and post a video.
It would be very nice if you could specify your suggestion.
Many thanks and with best wishes
Ralf
Lamp oil is the one product that comes closest to Professional Dielectric Fluid for Sinker EDM machines.. for Wire EDM, Deionized water works best.. but you have to treat that with Filter Resin as the conductivity rises when Eroding..That is why i have a resin filter Cartridge in my Wire EDM machine.. (Brother HS-100)
What if you used a zener diode to prevent voltage dropping below a certain threshold. Also how do you know if the circuit isn't simply shorting across debris or the tool/workpiece. Maybe you should be using a voltmeter and ammeter during your tests.
Você é incrivel! Grande abraço meu amigo!! Congratulations of Brasil!🙏👏👏👏👏😊😊😊😊
Many thanks
seems pulsing is the way to go.
the voltage is only present before the arc forms. need it to break down the dielectric.
once its conductive, the voltage drops dramatically. measure a neon sign when its running... they can be 20-30V... despite the transformer saying 15kv.
because its limited to 30mA. it will only spit 15kv out when its unloaded, or has over 500k resistance... (eir) any lower resistance, higher load, and it will remain at 30mA but sag the voltage accordingly.
anyway. when the arc is formed, its composed of ions in the dielectric, and the electrode materials. polarity dictating the majority of the transfer from negative to positive, much like welding. dcen dcep AC....
as you maintain the arc, the electrode starts breaking down. your "points" start to round off, arc stability becomes erratic on DC, and the material doesnt want to come OUT of the arc, instead transferring to the other electrode.
the pulse, if you have say, a variable capacitor (ahem... not so easy? but then again, ive dunked an old moving plate under oil and run it at 10kv...) you can charge to a specific voltage for the breakdown, but only store enough energy to dislocate those metallic ions from the workpiece... then cease the current, leaving said atomised particles floating in mid-solution where they can be flushed away. i
n this regard, allowing the capacitor to discharge on its own terms isnt the way to go, rather, once it reaches a certain level its cut off entirely, conduction ceasing abruptly, ideally at max flow. which, with some sampling shouldnt be too hard to discriminate. theres a brief rise as conduction commences, with a peak at full conduction at which point the standard RC rules of a discharging capacitor apply.
whilst theres an arc, the ions/particles tend to REMAIN in the arc. its supplying heat, energy, and its got those dielectric and magnetic effects... youre burning, oxidising, and destroying the dielectric rather than the electrode/workpiece.
liken it to blowing dust with two air guns. one lifts the dust with short rapid puffs, directed where you need, whilst the other blows each lifted cloud of dust clear... whereas if you blast both you just make a mess.
you want a corona discharge, high tension, but as soon as theres current applied it reaches a certain value then is cut-off. think of the lightning rod... you need the surfaces to produce those spikes, which they cant do once theyre actually conducting.
tungsten copper electrodes?
another aspect to look at is the AC welding of aluminium, the cleaning action of reverse polarity.
you just want to get the surface layer suspended, free... then push it away. its not quite the same with TIG as the main goal is the metal stay in place and simply melt, you just have to bust the oxides off. anyway.
if anything. you are basically trying to rip that charged surface layer off the plate of a capacitor, then blow it away whilst suspended. the electrode and workpiece IS a capacitor, after all. not heat anything. simply tear a layer of charge and associated particles off.
Inspiring effort, thanks
Thanks!
The big machines use a speed rate, and peck and drill method, not a plunge cut. Also the nozzle, it helps to have fluid pushing the material out.
Would adding some inductance in series help stabilize the arc? This is what they used to do with arc lights. Also, how do you control the feed? Is it a constant rate, or is there a control loop that is driven by the average current through the electrode?
Hi Ralf, I have been running a hansvedt spark edm which I think is an RC type given its advanced age. I have obtained results similar to you. what might make it work better, but I have not tried, are BeCu electrodes or a higher voltage spark from the RC circuit with a shorter RC time constant. Also, your use of multiple blade electrodes seems to me to impose a burden on the RC circuit similar to having the RC tuning incorrect. If you watch your video you will see that the spark is hopping amongst the blades in contrast to what happens with a concentrated discharge within a more electrically limited discharge region (intense local heating and higher ion density). Have you tried to cut with a single blade? And another thought, as the electrode drops down into the slot, the sides of the electrode can continue to conduct laterally. Have you tried to varnish or paint the sides of your blade electrode so it is conductive only on its intended business end? These are just some ideas that I have not tried while also thinking about the pulsed edm. Regards, Mark
Hi Mark,
thanks for your thought-provoking impulses I will try them out and show the results as soon as I have good outcomes.
Greetings
Ralf
I'd be careful with beryllium; IIRC high toxicity.
Agree on trying a single blade to start, before attempting the additional complexity of multiple blades. Perhaps the second step would be widely spaced blades.
Possibly rotating disks for better dielectric flow? Though that way lies a wire EDM.
for aluminium with graphite have you reverse polarity ?
Hi,
I made best experiences with electrode positive and didnt reversed it for graphite on aluminium.
Do you think this will give better results?
I havent much experience with aluminium because stainless steel is more important for me.
Greetings
Ralf
@@myengines2443 You want the electrode to be negative to avoid anode corrosion (ie if you try electrolysis, the anode corrodes considerably fast.
@@guytech7310 I've heard that aluminum and steal require opposite polarization. th-cam.com/video/AQCpueMkZ6o/w-d-xo.html
Did you drill flush holes in the electrodes?
Hi, Could you provide minimum requirements of resistor and capacitor specs, and the DC source ratings for voltage and amperage required to generate sparks
I think he will need to appropiately filter water and to use a deionizer too. I was thinking in this interesting process as one of my best friends works in this area here in Spain but we agree that is very difficult to diy a very accurate system, even he uses it occasionally in criticall operations with an industrial grade machine due to the time consuming task an high costs.
I think that better flushing directly into the electrode gap would help a bit. Kerosene is supposed to burn a bit better iirc.
Hi,
yes good flushing is important, with kerosene (lamb oil) I also got good results but now I use diesel which is even better.
You can see it in my new video.
Greetings
Ralf
Ralf, Ive been watching your videos for a few years now. I've always wanted to know what your final plans for the stirling engine will be? Is it for backup power? Just for show?
Hi Martin,
thanks for your longstanding loyality :-)
At first Stirling engines are my passion and I want to develop a relatively simple useful machine with reasonable power.
Unfortunatly this is no easy task for a single hobbyist and not even for a company.
You have much more setbacks than successes.
I try many variations of the Stirling engine and I also often improve my workshop as you see in this video.
This takes a lot of time and I only post a video when I think I have something new to show.
My final plan is to develop a relatively simple RELIABLE (thats the main challenge I think after more then 10 years of Stirling research and development) Stirling engine for generating power for ourselves (we have much wood, I can produce biogas, the sun is shining...) and maybe for others.
It is facinating for me to 'produce' mechanical energy just from heat.
Greetings
Ralf
change to lower voltage. i have built similar diy edm but mine cylinder plus needle solenoid valve for movement.
Very informative
Bravo........run copper wire.........flow....u are trying to drill with a flat.....if round maybe rotate electrode.......also diesel........clever.....what volts...amps.....cheers.....
Sir I want a edm control panal
Fascinating! I'm doing some EDM experiments now and this could be useful. For the electrolyte, have you heard of conductive particles? They are said to improve surface finish by making the sparks smaller, but I thought it might stabilise the sparks.
Hi,
this sounds very interesting but I have never heared of conductive particles in the dielectic.
If you know something more about it please let me know...
Ralf
I would have thought that conductive particles are (1) what you get when the fluid is used, (2) prone to providing unwanted 'chain' paths to increase the erosion distance, decreasing accuracy.
You need to control you on/off times more accurately
Nice machine just needs a bit more work
Interested in your thermal displacement engines.
I once restore a large water pumping engine.....it was at least 100 years old,,,,Stirling type using wood fire
Maybe 1 hp 2 hp I will post video soon and share with you
This sounds very nice, I would love to see your engine.
Greetings
Facinating!
Could you just do simple wire EDM? That way you could worry less about it being a flat sheet. I don't know how you have to cut so I'm not sure if that'd work for your application.
Yes you are right a wire EDM is very nice but much more complex to build.
When I would decide to build a wire EDM I would not have any time for my main obsession Stirling engines.
Thanks and greetings
@@myengines2443 You can buy an old wire edm for very little swap out the motors to run on Mach3 and clean up the flushing system. That is what I am currently doing and building this generator for the spark.
have you tried using only one thin blade,or less current or high frequency pulses
Hi,
yes I tried just one blade, it works without difference.
I cant control current or frequency as it is a simple R/C machine.
My new experiments with diesel as the dielectric and graphite show very promising results which I will show in an update video soon.
Thanks for yout suggestions and with best greetings
Ralf
@@myengines2443 You can control discharge energy by the size of the caps and inductors in series with the caps. The Inductors will provide current impediance to limit the current. You need to use a recovery diode so the energy stored in the inductor does feed back into the mosfet (possibiliy damaging the mosfet.
@@myengines2443 I have seen the guy from de the baxedm using a series resistor made with 500W lamps reflectors (tungsten core) to limit the current, i didn't saw that part in your setup, maybe it's possible to add it? I think that you have excessive current/voltage, sound very sparky and not so much as frying bacon
That is fascinating! Can you tell please how long the first M letter cut took and also how much KW's of power it takes in average?
Hi Anton,
I don't know which burn you mean (M?) but now it takes about an hour to burn 3 slots 1x8mm trough 11mm thick 1.4571 Stainless steel.
I dont know the average power consuption of the EDM but it cant be much, the machine has a 2A fuse on the 230V net so I think it is in the region of maximal 100-200 Watts.
(2x 24V 2A transformers are for the power line).
Greetings
Ralf
@@myengines2443 wow thats actually incredible that it takes only 100-200 watts, I thought its much much more because of high voltage needed. I was afraid that you pay for such high tolerances with high electricity bill.
@@KyrychenkoAnton You don't need high voltage. lower voltages will work just fine and also allow more precision cuts because the electrode will be closer and the spark gap much smaller.
Use mig wire as your electrode; you can the better focus the electrons.
Hi,
I tried tungsten copper as electrode material as you can see in my newer video.
It worked quite well but the thinnest sheets distorted badly.
You mean TIG welding tungsten wire?
Maybe I will give it a try.
Greetings
Ralf
@@myengines2443 There is a wire used in Mig Welding; Its a sacrificial wire; sold on spools. (its Cheep.) There are two types of wire used for mig welding, and they are gas and gasses welding wire spools. The one that you would want to use is for gas welding. Using this wire you can focus the electricity onto one spot at a time. Focussing the burn to a single spot. Opposed to larger surface areas.
Using larger surfaces as an electrode, the ark will would be more difficult to control; causing heat to warp the electrode. As the electricity takes the path of less resistance.
With Mig wire that ark will occur in the one controlled location. And when the Amps begin to climb feed it more wire to maintain a constant burn. The Amps will increase as the tip of the wire begins to deteriorate.
Hi, can i contact you for help completing my build, i have the PCB and parts, thanks
Hi Carl,
plz just write your problems in the comments so all can benefit and/or help.
I am just an electronic newbie but maybe others can help better.
Greetings
Ralf
If you want to cut thin steel why dont you use electrolysis
Not the steel is thin, the slots are thin.
They have a length up to 60mm.
I dont think electrolysis can do that but I am happy to be proven wrong.
awesome technology. we also show videos of this technology on our channel. it's a fantastic technology.
Maybe try electrochemical machining?
Hi,
yes electrochemical machining seems to be very interesting but all I read about it is very complicated and not suitable for the home workshop.
But if you know something how to do it easy please let me know.
Thanks for you kind comment
Ralf
Please link the ebook
There is no ebook, you have to buy the book.
Would you benefit from doing ECM over EDM for certain applications?
Hi,
I do not know much about the ECM technology isn't it to complex for DIY?
If not do you know a good description or presentation of a DIY ECM Machine, it looks very interesting.
Thanks for your help and with best greetings
Ralf
@@myengines2443 Youre capable of building your own 3d printed electrodes and drop on the z, ecm is very slow and produces a lot of waste. sludge and hydrogen.
Can I expect the power schematics
Hi,
you can buy the book of Ben Fleming.
It is a very detailed description of the EDM and a very good read.
Greetings
Ralf
Where to buy this book
Hi Ravi,
you can buy it directly at Ben Fleming (and his PCB's also)
www.homebuiltedmmachines.com
Or with a bit of search you will find a seller in your region.
Greetings
Ralf
using plates is a bad idea. As the plates move down a very very small error in x or y will cause plates to short to work piece. Consider doing the slots with a wire EDM for better results.
Yes you are right a wire EDM would be very nice but this will be very much work so I would not have any time for my original hobby developing Stirling engines.
With my new pulse EDM even the thinner sheets work well.
Just look at my newer videos.
Tungsten copper may work better...
I tried it but it tends to distord under wear.
Otherwise it is very nice and resistant.
Look at my newer videos to see my results with it.
Thaks for your suggestion.
Greetings
There is a problem with pulse EDM is the machine must be big and consumes a lot of energy Because it uses MOSFETs to regulate voltage, may not suitable for home shop. I am using diesel with satisfactory result. Kerosene is even better than EDM oil but too dangerous not worth using.
Hi,
I plan to build a pulse EDM also with a book of Ben Fleming now and ordered his PCB.
It is a bit bigger and the transformers are bigger and more expensive but not to much.
I also use diesel now (ARAL ultimate diesel smells less!) but before I used kerosene (lamb oil) which worked also very good.
Greetings
Ralf
@@myengines2443 Read papers of Mr. Wojciech Mysinski if you want to know how a commercial machine works. They are fee on the net. Your job is difficult even on a commercial machine.
@@Spark-Hole I am having a look at Wejciech papers now. Thanks for the reference
th-cam.com/video/KgG_2Bdb2VM/w-d-xo.htmlsi=Xd7TT1gAcuoZIB5F
check this out, the best homemade machine. th-cam.com/video/KgG_2Bdb2VM/w-d-xo.htmlsi=U5svjANBNDUjUkFm
固态继电器 电火花加工
what am I watching?
A home made spark eroder, a good way to burn your house down