Titivating a Wire EDM Machine

[awemawson]

A bits and pieces day today.

First I cleaned up and etch primered the 'drawing arm' that allows a replica of the part in the CNC memory to be drawn in pencil. Was going to 'satin black' it but my 'spare' can proved to be a dud  More on order.

Then I replaced the 'tray' that surrounds the work area, using ordinary sanitary caulking to seal it.

Then I made a replacement cover sleeve for the work arm - the original had corroded and looks as though rats have been nibbling it

[awemawson]

I don't like sheet metal work at the best of times, and stainless steel is the pits ! My guillotine doesn't like cutting it, and it's quite difficult to get sharp bends in the folder, despite this only being 1 mm and supposedly annealed. Turned out ok in the end.

Then I set to making the longer holder for the wire pulley - this was needed not only as the original was corroded, but also as the pulley needed shifting to the right a bit to give adequate 'wrap' of the wire around the tungsten carbide contact. It should just have been a matter of cutting the 10 x 30 mm stainless 'flat stock' to size, and drilling and tapping a few holes. However the stock was so 'un-square' I ended up milling it on all sides to square it up.

The position of the pulley is rather arbitary based on the 'that looks about right' school of design - time will tell if it is !

Now I'm puzzling how the wire is supposed to be routed when the arm has the sliding seal that attaches the 'curtains' fitting round it. Time to search the web for picture   

[awemawson]

So at last the slab of white Delrin arrived and I could make up the mounting for the tungsten carbide contact. What lovely stuff delrin is to machine - such a contrast to the stainless steel I've been working on !

Having roughly cut the slab to size on the band saw, and milled to final size I needed to spot through the existing machinable ceramic insulator to pick up the mounting hole centres. Then it was just a case of drilling and countersinking to finish that item.

Next on the list was the stubby axle that mounts the contact. Turned down from an old stainless pump shaft, axially tapped at one end 5 mm to mount it and 4 mm to take the screw that both holds it on and also takes the electrical connection. This was followed by a couple a thick washers to complete the assembly.

[awemawson]

In that last picture I've stretched a length of the edm wire round the pulley to show the way the new contact just rubs the wire.

So now I need to re-mount the arm temporarily so that I can take some measurements for the sliding seal arrangement.

[awemawson]

Having re-installed the arm, complete with the stainless steel cover that I made, it was immediately obvious that the cover wasn't long enough to make a seal for the full travel of the machine. I just slavishly copied what was there when I got it. Bad move as it's apparent now that this machine has been at some stage re-worked using bits from slightly different ones. Ah well - just bend up another one 

The black paint arrived today, so the drawing arm got another coat of paint

[awemawson]

Tomorrows job is to fabricate the seal assembly, so when at last I get my hands on some usable bellows the machine can be put back in service

[awemawson]

Well after a bit of delay as other things 'got in the way' at last the seal assembly is made and installed.

Carved from solid 20 mm square stainless steel bar it's somewhat more substantial than the original but functionally equivalent. It's main job is to tether the free ends of the bellows so that 'the curtains stay closed' as the table moves about the projecting arm.

[awemawson]

I had had to guess how much to make the seal material compress to be a decent sliding fit and yet not 'grab' on the arm cover. Having made it with 3 mm deep slots to take the seals I was erring on the cautious side, as I could easily relax the seal by sinking them deeper, but to compress them more without packing that would work loose would be a pain. As expected it was too tight and I had to sink the slots a further 3 mm to stop any tendency for them to 'roll' as it moved.

Here it is in place:

[awemawson]

Now I have failed to source suitable bellows material, so have bitten the bullet and commissioned some to be made. There goes my pocket money for many a day 

At least they will be professionally made from the right material. Downside is that they are on a six week delivery. As I had one of the original bellows that was 'sort of ok' I decided to go ahead and re-assemble the machine using that one bellows - after all I've had it for several years with neither one working !

So first of all I had to re-thread the power cables and flushing water pipe through the arm, then having replaced the arm cover the seal and bellows etc could go on

[awemawson]

So the machine is back together and apparently working. I did a dry run on paper of a spline that I cut several years ago and it seems ok. However there are still a few little niggling things that I need to sort before this project is put to bed.

Firstly, when I tried the flushing water having got things back together, the machine made a big puddle of water  Now this is coming from under the main body of the machine, and there is very little plumbing in this bit - most is in the main water tank. It will be something trivial but I'll have to take off a few panels to see what. Basically, pressurised de-ionised water is fed to the two 'flow control valves' below the plotting table, and from there is routed to the upper and lower flushing nozzles. Water then drains out of the big casting that is the work support via a flexy pipe back to the tank.

Secondly, when the machine was delivered years ago, one of the two sliding guards that move around as the work table moves was missing. As they are supposed to be bolted together to form a rectangle with a large square hole in the middle, half of one isn't much use! So I never fitted them - I did however order up a sheet of 2 mm aluminium to make one. The machine body has a 'kerb' of 8 mm x 10 mm aluminium bar that catches on the lip of the covers and limits their motion. One had apparently been forcibly ripped off as it's remains were still attached - I've never made the covers as I didn't have the bar !!!! Bar stock on order 

Thirdly, the upper wire guide assembly racks up and down like the Z axis of a mill, the idea being to minimise wire length and water flushing distance for thinner work pieces. To this end the cooling and flushing pipes and power leads have sufficient free length to allow the downwards movement, but they flap about and get tangled in the polycarbonate anti-splash shields - so they need some form of re-routing or maybe just tidying up

Fourthly, having just re-threaded the wire I am even more convinced that the 'Vee Ditch' wire guides are the way to go, although it will need a bit of ingenuity to fit them and construct suitable flushing nozzles I'm sure it will be worth it!

[awemawson]

Well the list is getting ticked off one by one!

The leaking water WAS trivial I'm glad to say. Just a Jubilee clip that responded to a few more turns to tighten it  It is one of two each side of the 'Conductivity Cell' with electrodes used to monitor the conductivity of the water.

Looking at the rust stains below this assembly, it has probably been dripping for quite some time 

There is one more thing though to add to the 'to do' list on this machine: Re-learn how to program it The basic moving around is simple G code, but there are pages and pages of 'cutting conditions' that need referencing in code (and understanding !) It's a couple of years since I last did it so hopefully it'll all come back so long as the appropriate grey cells haven't died in the mean time 

[DMIOM]

........Thirdly, the upper wire guide assembly racks up and down like the Z axis of a mill, the idea being to minimise wire length and water flushing distance for thinner work pieces. To this end the cooling and flushing pipes and power leads have sufficient free length to allow the downwards movement, but they flap about and get tangled in the polycarbonate anti-splash shields - so they need some form of re-routing or maybe just tidying up.....

Just watch how you re-route / tidy them! linky

Dave

[Henning]

........Thirdly, the upper wire guide assembly racks up and down like the Z axis of a mill, the idea being to minimise wire length and water flushing distance for thinner work pieces. To this end the cooling and flushing pipes and power leads have sufficient free length to allow the downwards movement, but they flap about and get tangled in the polycarbonate anti-splash shields - so they need some form of re-routing or maybe just tidying up.....

Just watch how you re-route / tidy them! linky

Dave

Considering how Mr. Mawson seems able to make a tidy job of even the worst of basket case machines, I'd hazard a guess that if he'd been in charge the satelites would not only be in the right orbit, but they would have self repair system and also a system to get them into the right orbit should they not behave themselves 

Not many comments, and I guess it's because other people too are so impressed and at the same time feel so inadequate?

It's a stirling job again Mr. Mawson and i thoroughly enjoy following along, even if i have troubles understanding everything you do!   

[Spurry]

Not many comments, and I guess it's because other people too are so impressed and at the same time feel so inadequate?
It's a stirling job again Mr. Mawson and i thoroughly enjoy following along, even if i have troubles understanding everything you do!   

Ain't that the truth!
Pete

[awemawson]

Well thankee kindly chaps

"It's a stirling job again Mr. Mawson and i thoroughly enjoy following along, even if i have troubles understanding everything you do!    "

Now you are assuming I understand what I'm doing - don't be too trusting 

[awemawson]

So after a brief delay when I had to rescue a dropped screwdriver from the machines innards  (Must make that sliding cover !) I made a start on tidying up the wiring to the rear of the machine where it fouls the polycarbonate splash covers.

At some time a retrospective mod has been added to ventilate the 'biscuit tin' the the U & V servo mechanisms sit in. This box is suspended on the bottom of the 'Z ram' which is manually driven up and down. It comprises a wimpy under powered little fan remotely mounted in a box, and both the feed and return air is piped though 1/2" bore flexible piping that totals 2.2 metres in length. As you can imagine the air flow through this arrangement is microscopic though presumably it must have helped an over heating problem. The box that the fan is in, and it's adaptor on the back of 'the biscuit tin' is well engineered, but it's design is dreadful ! Also power to the fan (24v DC) is by two single core wires dangling all over the place and lead into the fan box through one of the air pipe holes - very crude. The side of the fan that you cannot see in the pictures butts up against a 'funnel' machined from some engineering plastic reducing it's orifice to 1/2" !!

The result of all this is that the 1/2" flexible pipes and fan feed wires were a mess - some re-engineering called for.

I removed the 'fan box' dismantled it, and put a cable gland in it's end plate suitable for a more conventionally covered cable to power it

[awemawson]

Now this description is slightly out of order, as to re-wire the fan the rear machine panel had to come off - but this had had to come off anyway, as I wanted to lose some of the drive cable back into the chamber. Although it would cheerfully slide back into it's gland I had no idea what it might foul, as I'd not been in there before. As it turned out there was room to pull the cable into the chamber, and rationalise the number of cable glands - blanking off two unused ones.

Then I put 'cable wrap' around the power and water feeds to make them into a cable form, put the rear panel back and secured the cable form and air pipes with 'P clips'. I've reduced the length of the air pipes and exposed cable lengths to a bare minimum, so that as the Z axis goes up and down the resulting loop of cable is as small as possible (to avoid hitting the splash guards)

It's not ideal - I did consider installing an 'Energy Chain' arrangement but that would not be easy - I'll see how it works for now.

...hey ho ... onwards and upwards .....

[awemawson]

This afternoon I cracked on with the sliding cover that is missing. Did I say I don't like sheet metalwork  Big wobbly bits that slice your fingers 

So I dug out the 1 x 2 metre sheet of 2mm aluminium that I bought perhaps two years ago for this job - those sucker things intended for glass work a treat handling sheets single handed. I had hoped that the cover would come out of a width of the sheet but no such luck - it has 8 mm upstands (or rather downstands in the attitude that it is used) and is 1 metre wide, so short by about  11 mm !

I cut the sheet approximately to size to give me an off cut that I could use to check bending allowance for this particular material on my particular bender. Turns out that each bend needs 5.5 mm ie the sheet needs to be 1 metre plus 11 mm for the pair of bends on the side, and the bend point needs to be 7.5 mm from the edge of the sheet for the upstand to be the correct height.

I made up a test piece aiming for 150 mm with an upstand at each end, and it worked out ok.

Then I did some test cuts on my corner notcher to get the edges to meet when folded.

Then I closed my eyes and went for it bending the three required upstands - when I opened them again it seems to have miraculously worked 

[awemawson]

Continues

[awemawson]

So now to cut out the rectangle in the middle. Jig saw, angle grinder panel saw or nibbler ?

Jig saws tend to make sheets leap up and down, angle grinders clog on aluminium, panel saw is a pain so the nibbler drew the short straw. Now I have to say this is not my favourite tool - whenever I use it it gives me grief and less than perfect results. To get anything like a straight line it needs to run down a guide, and clamping the guide generally prevents the working punch enter the cut. I suppose really I need a much wider guide that can be clamped well back from the cutting face.

I got there in the end with only a little wobbly bit at one end that wont show as its where the original cover clamps on to it

[awemawson]

Now conveniently, marking out the fixing holes is just a case of spotting through from the original cover. I did this, and was about to drill holes suitable for a Hank Bush as the sheet is only 2 mm thick. Then it dawned on me - nothing can project below the sheet, as it slides over the 8 mm high 10 mm wide 'kerb' that surrounds the hole in the machine. Anything sticking down below the under surface will foul on this kerb 

Curiously where the two halves of the cover join, there are four holes on each side. The outer two are counter sunk for apparently a 4 mm screw, and the inner two are drilled but not counter sunk.

So the quandary is how to bolt the two halves together without anything sticking out below. Theoretically I can tap the cover that I have made M4 and fix using M4 x 4 mm screws (do they make them so short?) but I have a feeling that M4 in only 2 mm of soft aluminium is going to have very little strength. Not sure that I have a choice unless you have any suggestions?

Going to sleep on it before I start tapping !

[AdeV]

Could you drill holes in the right place on the kerb, so that captive nuts welded under the bottom guard sheet drop into the holes, thus laying flat, when the sheet is in the right place? The upper sheet then slides over the top, and 2x screws go into the captive nuts.

I presume the "kerb" moves with the machine/guard sheets?

[RussellT]

Countersunk bolts from below?

Russell

[awemawson]

Ade, no the 'kerb' is fixed to the body of the machine and catches the bent down edges of the covers as the table moves them around. It's a bit like a skirt round the table but as the rectangular hole in the pair of covers is bigger than the table support it doesn't move until that table motion passes a certain point. Hard to describe so I hope you get it.

Russell, yes I've been thinking along those lines, but the original cover is already countersunk from above on four of its eight holes so they must have originally screwed in from above. All very odd !

I suppose that there's not much to lose by trying tapping at 4 mm . I've found some M4 pan head x 4 mm screws on ebay - I'd prefer button head to keep the profile low but I'll probably go with them and see what happens.

Thanks to you both for the interest and suggestions

[RussellT]

the original cover is already countersunk from above on four of its eight holes so they must have originally screwed in from above. All very odd !

It does seem odd - especially that only half the holes are countersunk.  It makes you wonder whether that's a mod - and perhaps someone discovered that it wouldn't work with a protruding screw - hence the missing guard?

If you're concerned about the threads in the aluminium perhaps you could loctite some studs into them so that they wouldn't wear - of course your plan always allows for the fallback position of contersinking from below.

Russell