What's new
What's new

Home Made EDM

blake in spokane

Hot Rolled
Joined
Jan 11, 2008
Location
spokane
Has anybody made thre own tap,drill discintagrator. I have the books from Flemming & THE HOME SHOP MACHINIST. What are the pros & cons? Which one do I build or are there others?
 
I built one from Robert Langois'(sp?) book. It worked ok although I never did more than burn a few holes, spent about $500 on parts. in the end you would be better off buying one of the older 10-20 amp machines off ebay. You end up with a much better power supply as well as a tank that can fill and empty easily. You will end up with much better parts if you can submerge them in the proper EDM fluids.
 
Last edited:
Blake, I am in the process of building one. I've not put but about $20 into it, because I had most of the parts lying around. Actually, most of the money I've spent on it has been to pretty it up -- I could have made it q&d without spending a cent. But again, that's because I had parts on hand -- various digital logic chips, stepper motor, transformers, box, etc., etc.

Of course, I don't yet know if it will actually work! :( Or let me say it differently -- I have tested it enough to be confident that the mechanism will work, but I do not know if it will have enough power to be worth while. I may wind up redesigning if it doesn't.

I used ideas from a couple of designs that I found online, modified to suit the parts I had on hand. The simplest one is here: http://pico-systems.com/edm.html. The other is here: http://www.cnczone.com/forums/showthread.php?t=26271 (look at post # 12; you may have to join the forum to see the pics).

When I get mine finished, I'll be glad to let you know whether/how it works ...
 
EDM sinker and wire

I have make a EDM sinker and also EDM wire and I be start at my second EDM wire machine. But the inportant on both proces are a good power supply where you can set the erosion freqentie of the material.
 
Made the Fleming unit for under 100 bucks and it works like a charm.
If you go to the Yahoo website, they have the circut boards and parts already made up, which saves time and money (group buy)
its called
EDMhomebuilders

Rich
 
I'm going to go for a bit more juice.

I'm building my servo head based on Ben's schematic, but I'm gonna give it more juice. I'm also gonna build a little bit better slide than a drawer slide, if I don't come up with a pre made ball screw slide. I'd like to get one of the NSK units, but the $$$ is CRAZY!!!

Oh well, I might just have to build my own setup with a gib, and dovetail like a mill, or lathe slide. I'm thinking about an aluminum dovetail, with teflon glued to steel for the adjustable gib, a teflon liner for the other side, and moly/graphite dry lube. Do they make bearing bronze flat bar stock, and would that work for a gib material with a steel backing?? That may be another option. I dunno, this has to move pretty fast, and be fairly low friction, while having VERY LITTLE play. I also have a design in mind that works like roller coaster wheels on the tubular track. Some 1/2" drill rod rails with 8 ball bearings for wheels. In this case, the car stays still, and the track slides. The car has an anti backlash nut that stays still, and the leadscrew is retained by tapered roller bearings on either end of the track. Hopefully I can find a super fine ballscrew assembly that's somewhere in between 18 (doesn't matter if it's metric, I think they make 1mm pitch which is IDEAL) and 28 TPI, and between 6" and 1' long. One end of the track has the gear motor that turns the leadscrew, and the other has the electrode holder mounted to it.

Awake, the question you asked in the email, yes you use an electrolytic cap for the main discharge cap. Ben's book suggests photo flash caps. I think caps from a computer supply, or something else might work better. The finish caps should be mylar, or polypropylene. You may want a VERY small cap for the finish, I'm going to try a 10uf. I'll try some other cap types, and sizes too when I do mine. Just remember, the higher the voltage rating, the better. I'm gonna try a MUCH bigger coarse burn cap on mine, probably about 400 or 500 mfd with a 400VDC rating. I'll try a bunch of different sizes and see what happens. This is an area you can experiment with. :D
 
I'm also gonna build a little bit better slide than a drawer slide, if I don't come up with a pre made ball screw slide. I'd like to get one of the NSK units, but the $$$ is CRAZY!!!

Oh well, I might just have to build my own setup with a gib, and dovetail like a mill, or lathe slide. I'm thinking about an aluminum dovetail, with teflon glued to steel for the adjustable gib, a teflon liner for the other side, and moly/graphite dry lube. Do they make bearing bronze flat bar stock, and would that work for a gib material with a steel backing?? That may be another option. I dunno, this has to move pretty fast, and be fairly low friction, while having VERY LITTLE play. I also have a design in mind that works like roller coaster wheels on the tubular track. Some 1/2" drill rod rails with 8 ball bearings for wheels. In this case, the car stays still, and the track slides. The car has an anti backlash nut that stays still, and the leadscrew is retained by tapered roller bearings on either end of the track. Hopefully I can find a super fine ballscrew assembly that's somewhere in between 18 (doesn't matter if it's metric, I think they make 1mm pitch which is IDEAL) and 28 TPI, and between 6" and 1' long. One end of the track has the gear motor that turns the leadscrew, and the other has the electrode holder mounted to it.

You might look on Ebay for some of the Bimba & Robohand slides. I picked up a nice Robohand for $20 all it needs is a ballscrew to drive it.
 
I finally completed my home-built EDM/tap disintegrator. I played around with the design quite a bit, and re-built every part of the electronics several times (a couple of times because the magic smoke got out! :(). I wound up using a re-wired microwave transformer; so far it seems to have plenty of oomph to do the job. For the power resistors, I am using 500 watt halogen bulbs (69 cents at Harbor Freight!) -- I have a bank of three, which I can switch in in parallel. Each one adds about 2 amps to the output. For the capacitor bank, I have a 1uF permanently in the circuit, plus 120, 220, and 470 uF that I can switch in in any combination. I also have one more switch, which I intend to devote to a 47 uF or similar size whenever I locate one. (The capacitor values were determined by what I had on hand! :)

So far I have tested the unit with a 1/8" diameter brass electrode (some brazing rod that I had on hand) running at 2 and 4 amps and a 1/4" diameter brass electrode (a cut-off 1/4" brass machine screw) running at 4 and 6 amps. I tried various amounts of capacitance in the circuit. With a little tweaking of the speed and the upper/lower voltage pots, I can get a very nice, nearly continuous sizzling sound, and it seems to burn plenty fast enough for my needs -- I burned a 1/8" hole about 1/4" deep in about 5 minutes.

I still need to work on refining the mounting system -- so far it is held in an endmill holder in my mill/drill, but the spindle has too much play in it for this to be very accurate. I also still need to rig up the pump to circulate the dielectric -- I don't know whether/how much it may improve performance, but hopefully it will help to keep the liquid a little clearer so that I can see what is happening! What I have been using for dielectric may be an awful choice: paint thinner. Well, I had plenty on hand, and it does seem to work ...

I will post the schematic and other design details in a few days ...
 
Hey did you still want to try the audio amp transformer?

I just got lazy and didn't disect the amp yet. It's still sitting on the floor, don't really have a use for it. Anyways what voltage do you get out of your transformer??? I'm gonna get started on the electrical part of mine as soon as I come up with a slide/leadscrew assmebly. Kerosene is supposedly a good dielectric, mineral spirits probably isn't as good.
 
Junkyard, I will gladly take the transformer if you want to send it. It would be good to have a backup for my re-wound MOT, though so far it seems to be going strong -- it gets warm, but not hot. But I haven't tried running it for 30 minutes or more ... even if the MOT lasts, it would be nice to have something to experiment with, especially for higher voltages.

I put two windings on the MOT, one that gives me about 14VAC to run the control circuitry, stepper motors, fan (and maybe also the pump, if it has enough amps to spare)and one that gives me about 30VAC for the spark circuit. Yes, that is on the low side; I was hoping for more. If I were to do it over, I would wind it with 16 gauge wire instead of 14 gauge. In fact, I was expecting that I would either have to re-wind it, or at the least that I would have to tie the two windings together to get a higher voltage, and shoehorn in another transformer to run the control circuitry. But I decided to go ahead and try it as is, and have been pleasantly surprised at how well it has worked. The unexpected benefit is that I have amps out the wazoo not only on the spark circuit, but also on the control circuit -- which is why I am thinking/hoping I may be able to run the pump motor off of it along with everything else. Also, the box I'm using (recycled from some long-forgotten project) is already tight; I'm just as glad not to be trying to fit in another transformer.

How much power does your servo/stepper/whatever have to drive your ram up and down? That might make a difference on what sort of design you choose. I am using a fairly small stepper motor -- probably came out of a dot matrix printer, though it has been so long ago that I don't remember for sure. Fortunately, my ram seems to have very low friction; the screw turns very easily up or down. I sure wouldn't want to have any more friction (or any less motor).
 
My motor is a small gear motor.

It makes plenty of torque, I can't stop it with 12v going to it grabbing the shaft between my fingers. Follow the link, and you can see the motor. I took the noise filter off the back thinking it would interfere with the operation of the servo controller. Anyways, it's not exactly what the book called for, but I think it should work OK.

I've kinda had a bunch of stuff going on, trying to help ma with paperwork for her new car, babysitting the puppy, and helping her car shop before that. It was a big PITA, so I haven't had time to rip apart the receiver. That and my bench is a MESS, and don't have room for it.

I've rewound microwave transformers for a bunch of different purposes. You can get whatever the wattage was for the microwave out of the transformer. The wattage remains constant, the voltage can be whatever you want to put on there for windings. Current depends on how fat the windings are once the voltage gets low enough you couldn't possibly max out on watts. For instance, a 5v secondary gives you 200a if you had a 1kw nuker. You'd need some FAT wire to do that without the voltage dropping from the wire resistance. Nuker transformers get warm, that's normal. You ever see how bright the arc is from one of those before you disect it?!?!?! Once you get it started, you can stretch the arc to over 3"! It doesn't have the starting capability like a neon transformer does. You can do LOTS of stuff using nuker transformers!

http://www.surpluscenter.com/item.asp?UID=2008032514572784&item=5-1158&catname=electric
 
Pictures

Here are some pictures of the controller box. I'm still working up a readable schematic.

Inside the box -- a bit of a rat's nest. Upper left corner - transformer; upper right corner - power board; lower right corner - capacitor board; lower left corner - logic board.

EDMController1.jpg


Closeup of the logic board. The two 8-pin ICs are 555 timers; one provides the pulse for the stepper motor; the other provides a pulse for a PWM control for a pump motor.

EDMController2.jpg


Front panel; from left-to-right, top row: Main power switch, speed pot for stepper motor, four switches to choose capacitance, up/down switch, speed pot for PWM control for pump motor, jack for pump motor. Second row: LED, various miscellaneous holes left over from previous projects, pin connector for stepper motor, lower gate pot, upper gate pot. On the top of the box, muffin fan is to the right, and velcro strips to attach resistor array to the left.

EDMController3.jpg


Top of box with resistor array attached. Resistors are 500 watt halogen bulbs in home-built mounts. The two red switches allow the middle and right bulbs to be switched in; each additional bulb adds around 2 amps.

EDMController4.jpg


Back of box. Power cord and fuse on the right. Receptacle on the left provides connection for the resistor array (right-hand receptacle) and for the electrode (left-hand receptacle).

More to come ...

EDMController5.jpg
 
I've got a good replacement for your lightbulbs.

Your 500w halogens might cause a BIG problem. If you leave that unattended, and the electrode shorts, you'll have the FULL OUTPUT of the transformer! The resistor is supposed to limit the output so it doesn't heat up the electrode when it shorts. Imagine a flaming pool of kerosene because the electrode shorted and got WAAAAYYY TOO HOT! If you always plan on watching the machine, that would be OK, but you effectively have a 0 ohm resistor there. Might as well just not have them, I doubt they even start to light, even if you had just 1. Light bulbs could work, but they'd need to be about 100w halogens, or a single 300w. They'd still need to be a low enough voltage to light, which would be hard to find, and expensive. Maybe with a higher voltage, 300w of 120v bulbs would work OK. Halogens also have a reduced lifespan, especially with fluctuating current like an EDM produces. I've got an adjustable resistor that would work good as a replacement. You've got too much current flow capability there. ESPECIALLY when you raise the voltage. I ordered one of these resistors for my EDM, cheaper than the other resistor options I've seen. Should work well for you too. :cheers:

http://www.allelectronics.com/cgi-bin/item/20-375/search/20_OHM_375_WATT_ADJUSTABLE_RESISTOR_.html
 
I did a good bit of testing before I went this route -- in particular, I tested it with the electrode shorted to see what the maximum amps would be. With one 500W bulb, I get 2 amps going through the circuit at a dead short. With 2 bulbs, I get 4 amps. With 3 bulbs -- you guessed it, 6 amps. And yes, the bulbs do light. It is actually kind of nice, because the bulbs give visual feedback for the amount of current flowing.

I do expect that the lifespan may be limited, but I got a bunch of these bulbs for .69, so I should be able to experiment a while before it becomes an issue.

I certainly expect that the resistor would be better ... but when I added up the cost to get the amps I was looking for, it was more than I was ready to spend.

I definitely do not intend to leave this running unattended in any case, so hopefully I can catch any problems before they get out of hand.
 
Hi Rich,
Good to see your name in print. I just joined and am trying to figure out how the forum works. I need some change gears for a Emco Compact 10 and thought this would be the best place to look. Send me your email address. Thanks
Steve Vitkovits
 
Edm

Good Lord,

You guys never seen the '66 Popular Science DIY project. 1/2 wave bridge (Single diode) two light bulbs, one to set the current one for the R portion of the RC circuit and two 40 uf caps.... one as a smoothing filter the other one for the C of the RC.

Downfeed was .... how shall we put this????? Oh hell, lets just call it interesting at best health hazad at worst.

Cyclotronguy
 
Last edited:








 
Back
Top