Rebirth of a Denford MIRAC CNC Lathe

[awemawson]

They advertise their part #12786 which doesn't say it is an NUC and they advertise #14366 which IS an NUC. It was #12786 complete with monitor I was after at $799 but wanted a spec first as none is shown. For that sort of money I want to know what I am buying !

I see they have changed the registration form now to accept 'none' as to which item you own - too late, they lost a sale and I saved a lot of money !

Still had no reply to my email which is rude to my way of thinking

[awemawson]

So at long last I've managed to install a second serial port into the Compac DC7600 SFF PC that will be embedded into the MIRAC. 

I traced the correct Compac part  PA716A which despite appearances is NOT just an external socket and a ribbon cable, but has active components, and plugs into what appears to be a socket on the motherboard carrying a subset of the PCI bus.

I had previously tried PCI express serial cards but could not get them to map to the correct address and interrupt lines for the MIRAC software to use, however the  PA716A  plugged straight in, appeared in the BIOS at boot and was already on the correct address and interrupt 

So this mornings job was to lash in the new DC7600 SFF PC to prove things work before hacking the old Viglen out. Glad to say all went well and I've successfully turned  parts this morning just as before.

I now need to check airflow direction to ensure that the PC fan and the cabinet enclosure fan aren't fighting each other, then I can plan how to mount up the DC7600 SFF

[awemawson]

A quick test with a smouldering paper proved that the original cabinet fan, and the airflow through the PC will tend to pressurise the cabinet. This is correct design as it tends to keep stuff out of the cabinet unless its been through the filters.

However there seems to be no designed exit for the air. Apart from a cable duct passing into the machining space - no exit 

. . . Need to think about this and possibly adding one 

[awemawson]

Time for a bit of an update:

The replacement PC is now mounted in the enclosure with suitable metal work to suspend it in place and present a fair face to the outside world by way of an escutcheon plate.

I've made and fitted a replacement for the headstock end panel keeping fingers out of belts 

I've made a replacement for the lubrication pump access door

All await painting, but that will be done when the entire machine enclosure can be dismounted and sand blasted

The machine is working well and I've today been experimenting with the G76 threading cycle. The machine obeys the Fanuc dialect of G code and the G76 command (which occupies two lines of code) is one of the more complex ones with loads of parameters to tweak.

Net result: today I've made some of the most expensive and time consuming M6 brass hex bolts ever 

Please excuse the swarf on the pictures of the bolt 

The sliding front cover / door is obviously not the original, and whoever started making this one made a reasonable job of it but never finished, so there is no provision for a lower slider. I've various bits on order to make not only the slider arrangement but also to insert a better sized window (in 6 mm polycarbonate)

(I've a video uploading to Youtube that I'll link in when it's finished)

[tom osselton]

 This is coming along nicely!  I wish I could say the same about mine!

[awemawson]

Here is the promised video:

Making the most expensive brass M6 hex bolt ever

It was as you probably realise, a learning exercise particularly with the G76 Fanuc G code, which is a bit complicated. Note that I use the parting tool at the end to chamfer the head of the bolt before final parting off.

You'll notice that there are significant pauses before the actual G76 lines are processed - I'm assuming that the controller is doing a fair amount of calculation - remember that the controller is running under DOS although it's got a pretty fast CPU to use compared to what it would have originally been equipped with.

[seadog]

I see the need for a pneumatic chuck with bar feed 

[tom osselton]

It seems to do a nice job of it!

[awemawson]

Thanks Tom 

Seadog, a pneumatic chuck was an available option when these were made - sadly I don't have one

[awemawson]

Late in the afternoon a courier delivered a 500 mm square by 6 mm poly carbonate sheet to make the new window.

I hate cutting things like this as you only have one chance - anyway after supper and a can of Old Speckled Hen I took my courage in both hands, cleaned off the band saw table and had at it.

Didn't go too badly actually. I then finessed it on the disk sander as I'd been a bit cautious cutting the radii, so I took it 'back to the line.

Tomorrow I must face the task of cutting a similar shape in the door, but slightly over size to allow for the rubber window seal. I suspect a loosish fit is what to aim for, as fitting those rubber seals can be entertaining 

. . but that's tomorrow and I have a batch of pigs going off so that may delay things . . . .

(ps window looks opaque as protective film is still on it)

[awemawson]

No good putting it off any more - time to cut out the metal 'hole' for the window.

The window rubber takes up a significant space, so the opening needed to be larger than the polycarbonate by 9 mm all round. So that I could 'draw round'  the already cut window, I turned up an 18 mm diameter 'washer' from a bit of tufnol I had lying around, and put a 4 mm hole in the middle to take a Sharpie pen.

Overall the rubber is about 25 mm wide, so I moved the window position slightly so as not to foul the  upper slider channel.

I cut the straight bits using a 1 mm cutting disk in my angle grinder, and cut the radiused corners using my jig saw. Seems to have come out OK but the real proof is when the glazing is fitted and held in by the rubber. Once the rubber and glazing are in place there is a 'filler strip' that holds the rubber tight - hopefully I've left enough allowance to get the glazing in, but not so much to let it fall out again 

[awemawson]

Putting the cover temporarily on the machine the opening 'looks right' but I'm going to have to sort out the brush seal - probably replace it with one with longer bristles mounted further back to give more flexibility.

I don't want to mount the rubber and glazing until the cover is painted, as I suspect once they are fitted they'd be the devil of a job to get off again without damaging them.

The glazing rubber obviously has a slot for the 'glass', a slot for the metal frame, but it also has a strange shape moulded in to take an infill trim that is not just decorative but serves to tighten the grip of the rubber on the other components. It uses a special tool to insert in - fortunately this is the same rubber profile that I used on my Traub lathe so I already had the tool 

[awemawson]

While I was doing all this, the postman kindly brought me a block of Delrin that I will be fashioning into a lower slider to take the weight of the cover.

First it needed slicing length ways on the band saw,  then cleaning up on the mill and a suitably placed channel milling for it to sit comfortably on the lower slide bar.

Undecided as yet whether to make it a short minimalist bit so that the door is effectively on three points (two up at the top)  or to keep it full length

[awemawson]

I ended up fitting two short guide blocks, one at either end on the lower sliding rail.

Today's job: Make a little bracket to stop the sliding door lifting off it's rails. Simple in concept but very fiddly to decide how to form it and measure up for the making. Not only has it to function - it has to be possible to fit it !

Cut out a bit of 2 mm plate and hot bent it as 2 mm has a tendency to crack with tight bends. The bending went OK except I made one bend in the wrong place 

Never mind - get it cherry hot, flatten it out and put the bend in the right place - plonker !!!

Once bent roughly to shape I fitted it to the door off the lathe and tweaked things to fit - it works actually quite well 

[awemawson]

So having made the lift stop all we need now is a handle on the door and I can get on painting it 

Two spacers of Delrin cut from a 25 mm black rod, a 25 mm x 4 mm x 310 mm  round edged bar and some M6 bolts cut to length and we have a handle. Pop a couple of holes in the door and bingo there we are 

Now I've had problems with the Simonize satin black spray paint I've been using chipping off unless the metal has been sand blasted. OK I can sand blast the handle, but the door it too big for my cabinet blaster and the weather isn't conducive to using my big pot blaster. So I'm experimenting with a (new to me) paint intended for vehicle chassis work. Reputedly sticks to anything, doesn't need a primer, and minimal surface preparation.

Handle duly sprayed and awaiting final verdict when it's cured

[awemawson]

Initial tests of the paint sample are very promising  After 90 minutes drying it passes the 'can I scratch it with my thumb nail test' which the Simonize failed miserably. Makers advise 24 hours for full cure.

So I've cleaned up the door and sprayed the inside with it's first coat. I'll give it another coat later today and leave it over night before turning it over and doing the outside.

[AdeV]

You do realise of course that the shiny new door is going to make the rest of the machine look a bit shabby.... and in need of a paint job of it's own....? 

Full strip-down to follow? 

[awemawson]

Yes Ade, the cream coloured cover assembly would seem to lift off after about a million M4 button head screws have been removed. But I'm waiting for a bit of fine weather so that I can remove it, take it outside and grit blast it with my Hodge Clemco.

Meanwhile i'm trying to track down some very tough paint in that cream colour

[hermetic]

Hi Andrew, that glass rubber bead is very reminiscent of the original Austin mini screen seal, I have one of those bead fitting tools too!. We used to put the screen into the bead, put a loop of string into the gap where the metal would be,  place the whole lot over the hole, inserting the metal at one side and then pull the string inside the car, which lipped the rubber seal over the metal! Easy, especially with a bit of liquid soap or swarfega to lube the rubber. The inner lip that fits to the metal is far softer than the bit that holds the screen, and when the loop method fails, as it sometimes does, you can usually lip the rest of the rubber over with your fingers.
Phil.

[awemawson]

Hi Phil,

I just watched a YouTube video where they did just that 

I'm thinking of stretching the rubber around the window, cutting it exactly and supergluing it into a circle, so it will hold on and doing precisely as you describe 

[vtsteam]

Great project, Andrew, sorry I missed it earlier. I really admire your lathe. Perfect size -- to me at least. 

Until recently I used DOS based TurboCNC on a couple of machines I have -- a homemade hot wire foam cutter, and a gantry style router. So I'm enjoying the fact that you've restored a DOS CNC system on your newer computer.

Right now at work I'm attempting to bring back into operation a 1998 vintage MTI dicing saw purchased by the CEO at auction, reportedly for $250  . Multiple issues so far, but weeding through them, Andrew style! 

[awemawson]

Hi Steve, glad to have you along for the ride 

I'm re-learning things about DOS6.2 DOS6.22 WIN98 and XT that I forgot years back 

[awemawson]

OK I admit - I'm impatient 

This paint is pretty good stuff and dries quickly to a very tough finish. I've managed to give the door two coats both inside and outside, and also the coolant pump access door and the sliding door lift stop.

Mind you, at over £12 per 400 mL aerosol - and so far I've used well over one and a half - cheap it ain't 

However I'm very impressed with the finish and no doubt given a full 24 hours cure it's hardness can only improve (I hope !)

[vintageandclassicrepairs]

Hi Andrew,
There was a link to this company who will mix any colour machine paint you need on the Harrison Lathe forum recently
https://www.ebay.co.uk/itm/1-Litre-Harrison-Lathe-Beige-Engine-Enamel-Paint-High-Temp-/263107968126

Craftmaster Paints Ltd
Adam Brown
Acadamy Works
Norman Way Ind Est
Over
Cambridgeshire
CB24 5QE
United Kingdom

Enjoying the progress reports on the lathe
John

[awemawson]

Thanks John, I've fired off a message to them