There seems to be a perverse tendency in the EDM construction articles to concentrate on the construction and skimp on the application. The EDM How-To Book is a welcome exception to this failing. A chapter gives all the steps in setting up for your first burn, describing not only What, but Why. It then goes into alternate and advanced types of electrode and flushing, stressing the importance of the latter. Other authors should take note.

Here we are ready for the first trial burn. The tank is clamped to the mill table. The ram is held in the mill quill and lowest gearing selected to prevent the quill rotating (my joke of a mill has no anti-rotation clamp). Power has been removed from the mill, and the shop fire extinguisher is within easy reach. First the electrolyte delivery system needs to be tested. The tank is filled to a depth that covers the rather large base of the suds-pump and power applied.

I had no choice for electrolyte. Not even the local refinery people had heard of EDM oil. I did find a supplier two states away who would supply a 20 kg bag of "resin" that could be added to water and used as the electrolyte. I did not even bother to ask the price. So household kerosene was used. The pump was started and soon the discharge nozzle was gushing kero. So was two or five other places in the plumbing! Soon there was kero everywhere. I ended up applying PTFE "plumbers' tape" to all screwed joints to cure the worst of this. There is still a tendency for the flexi-hose joints to leak kero. I now think arranging the delivery hose so all the little segments are over the tank opening (just like in Ben's book) is a much better idea.

Here's the test piece. The last bit of that miserable 4-40 tap has been screwed into a piece of scrap aluminium plate (a failed Sugden Special conrod) and cut off with a Dremel cut-off disk (have you ever tried to intentionally break a tap? It's not easy). The suggestions are to use an electrode that will burn away the tap core, allowing the cutting edges to be picked out of the hole with a scriber tip. Make sense to me. Here we go...

Pritty pictures, aren't they? The test piece is clamped to a piece of hefty angle screwed to the work holding plate. A length of 1/16" OD brass tube has been clamped vertically to the insulated ram block and visually aligned on the broken tap. This must be done "dry". Every schoolboy—and probably a few schoolgirls as well—knows that light gets bent when you look at things under water, so alignment has to be done before the work is submerged. You will also need to lower the level to inspect progress. The quick-fill and empty feature is suddenly starting to sound better and better. Anyway, with the generator clamps applied, the pot is used to slowly lower the ram until sparking starts. The book says "adjust for the sound of frying bacon". This was quite easy and very little adjustment was needed—none, in fact! The glow of sparks is visible under the surface and if you stop the pump, black gunk is seen rising at an alarming rate.

Here is the result of about one minute of EDM'ing with the brass tube and only the "fine cut" capacitor engaged. We see I was not precisely centered and the electrode is too small to reach the flutes in the 4-40 tap, although we are close. Note how the tube has made negative donut cut. Not so obvious is the erosion rate of the brass tube. This was about 4:1 more than the cut depth. In fact, the misalignment resulted in a sliver of un-eroded brass that passed down one tap flute neatly cutting a key-way through the aluminium threads about 0.01" wide. Neat, but bad for the real job still to be done.

The brass electrode was replaced with a piece of 0.076" diameter copper (from a heavy duty electrical cable) and very carefully centered on the tap from the other side. This time, it worked perfectly. The core was eroded away and the remains of the tap, looking like little thread-chasers, were easily picked out. The Cirrus crankcase was then carefully mounted and squared up. This required more tiresome emptying of the tank. With the copper electrode equally carefully squared and aligned, the tank was filled and the bacon fried. Result: perfect! The tap is gone and the 4-40 thread is completely undamaged, even when viewed under magnification. This photo shows the cleaned out hole and the fragments removed from the trial burn. The "actual" job just produced little granules that were cleaned out with a new 4-40 tap (very carefully, I might add ).

For grins, a piece of square brass tube was used on the test item to make a blind, square hole. Measurement shows the hole to be 0.016" larger than the uneroded electrode, equating to the 0.008" kerf that seems to be standard for most EDM's. Close examination of the electrode surface shows the side facing the stream coming from the hose nozzle is clean, while the other three sides are very pitted. This indicates to me that it is desirable to either flush from two directions, or better, through the electrode—just like the book says. Hmmmm...

Back: Plumbing

EDM How-To Index

Next: Conclusions

  Model Engine News Home

This page designed to look best when using anything but IE!
Please submit all questions and comments to [email protected]