Jafo milll repair

[.RC.]

I was watching some videos on youtube of Jafo Jarocin mills like what I have and noticed they have a rapid travel feature. Mine has a rapid button, but all it ever did was engage the feed motor and move the axis in whatever feed speed was selected at the time. In the videos pushing the rapid button made the axis rapid, then went back to normal feed speeds when released.

So removing a cover and peering into the gearbox revealed an electrical wire heading into the innards. I suspected an electromagnetic chuck, and found where the wire exited from the feed box. I also found a blown fuse, bypassing it made the 24V light work and 24V to this mystery wire when the rapid button was pushed.

I would not normally had done anything about it, but on this mill for some reason the table points up to the moon, so I wanted to fix that anyway. And the knee gearbox was full of crud which needed cleaning out. It is of Russian communist era design, so heavy castings, simple enough design but a lots of hours to get things apart as have to pull things off to get other things off to get the gearbox out.

The removed feed gearbox looks like this

361594

To remove it the z axis screw has to be removed, along with other things due to it's L shape. Upon removal an electromagnetic chuck was revealed. The gearbox splits into two and after much swearing at russian/polish designers of the era the open circuit clutch coil was removed. It was epoxied in so I cooked it in my heat treatment oven until the epoxy decomposed and the coil revealed. The clutch is a Stromag ER5 24V 37W. I was intending to try to make the coil myself, but I have no coil winding machine and a shop in Rockhampton has done these before so they did the coil for me.

All I had to do was clean up the epoxy a bit as they made it a bit too proud. Testing on 24V reveals it works.

the old coil

361595

361596

repaired

361597

In the process of putting the gearbox back together.

Which leads me onto finding out why the table points into the air.

With the knee stripped but still on the machine, putting a square onto the knee butting up against the column reveals the problem is not there.

So are there is some minor wear on the knee ways I will scrape them in again.

I have started with one flat way and here is a pic half done.

361598

It only took half a dozen passes to get this far and another four and it was done.

Now working on the little guiding ways, then make the other flat way parallel to the finished one. I have had to grind up a skinny steel straight edge to do the guiding ways. Slideway grinder was interesting to work, there is a 0.0002" bow I could not grind out. The grinding wheel would grind on the ends and just dust the middle which an indicator said was a 0.0002" rise on the ends, but I could not grind it out. Just more things to learn to do I guess.

[.RC.]

I got the gearbox back together and it has been cleaned and wrapped in plastic. Incidentally that plastic sheeting you put under concrete before you pour it makes good material to wrap things up in.

It is an interesting design. The bronze crossed helical gear you see here spins all the time and it is connected to the lubrication pump and the electromagnetic clutch. So when you press the rapid button and the clutch engages, power is fed through this gear to the axis's being moved.

The worm wheel you see next to it is connected to the gears that selects the various different feed rates. Now the bronze worm wheel has an overrunning clutch of some description affixed on the inside of it. So when the rapid is engaged the worm wheel keeps going at the same speed but the overrunning clutch lets the shaft spin at rapid speed.

361616

Here is the mating gearbox that contains the gears that select the various feed rates.

361617

Joined together again

361618

Onto scraping now.

To spot the small vertical shears that guide the saddle I reground my steel straight edge that was made out of 75X20mm flat bar. I had milled the top of it and warped it a bit. So I slideway ground it but could not get a 2 tenths bow out of it. So I scraped it flat. Although steel is not fun to scrape.

361619

Really the guide way is not worn badly, only a 0.02mm difference between the ends and the middle. Pretty quick to bring back to flat.

Getting there

361620 361621

Another pass after this run and we were done.

361622

[jhovel]

You make all sound easy.... I still haven't got my Cash mill back together.... maybe I should invite you to come for a visit and help me.... at least help me stay motivated :)

[.RC.]

I have a few stock prods here Joe, they are great motivators :D

Here is the finished skinny guide way. It is parallel with it's mate.

I measured it with a micrometer and gauge blocks as a scraped surface is not actually flat flat and a micrometer on it's own will not give a good ready as at some point you might be in a scraping hole and another point on a high spot.

361640 361641

[.RC.]

Did not get as much achieved today as I had hoped.

All the top slides are done. If you were doing it with a kingway type device you would do it the following way.

Scrape the flat horizontal way flat, then scrape the outside vertical guide way flat. Then assemble your kingway type device as per the photo and put it on those two ways adding a bit of oil to them to make it slide easier.

361649

Slide it to either end and take note of the reading. Is it way way more then you expect like I found. Put some weights on the other end and see what it does.

361650

Low and behold you find out your work-piece is not stable.

Make it stable and continue on.

So I made it stable and scraped the mating horizontal way flat and parallel.

Then to check the parallelism with the inside vertical guiding way I put an indicator on as shown

361651

That checked and done I turned my attention to the dovetail way.

A mild initial scrape and check squareness. Really you want the nose pointing up something like 0.02mm over 500mm.

361652

[kwijibo99]

Thanks for this thread, it's very interesting and includes some handy tips, I particularly like the one about using gauge blocks between the micrometer anvils when measuring scraped surfaces.

When you were scraping your steel straight edge did you grind your blade differently or stick with the same angles used for CI?

Keep up the great work.
Cheers,
Greg.

[.RC.]

I stuck with the same angle as cast iron for scraping the straight edge.

With all the scraping work on the knee done I moved onto the gib. Taking the first cut yielded it was made of steel not cast iron.

I did scrape it fairly well flat but did not do as good a job had it been cast iron. Why they made it out of steel is beyond me.

361742

Today I got the innards of the knee put back in and I have trialed the repaired rapid clutch and it works wonders. The feeds are very fast.

[.RC.]

One thing when checking out older second hand machinery is, has parts of it become contaminated through coolant getting into feed gearboxes like aprons on lathes or in this case into the knee of the mill.

I think this mill in the past has has significant amounts of coolant get into the knee. While nearly all the gears and shafts and bearings are pristine, the knee up/down gears have suffered badly, even though are right up the top. it is still perfectly serviceable but it is disappointing to see things like this happen. I have seen it happen on several machine tools I have bought now.

361745

The lube system is working :)

361746

[PDW]

I think this mill in the past has has significant amounts of coolant get into the knee. While nearly all the gears and shafts and bearings are pristine, the knee up/down gears have suffered badly, even though are right up the top. it is still perfectly serviceable but it is disappointing to see things like this happen. I have seen it happen on several machine tools I have bought now.

Yeah - had exactly the same issue when I tore down the Vicky mill. Full of nasty brown slime and fine swarf. Under all of that the ways etc were pretty good. I did have to make a few new parts.

Funny thing about bevel gears in a drive system that slide on bushings to engage/disengage dog clutches. If they don't slide because some 'fitter' didn't fit them, the power feed doesn't work either. 10 minutes work polishing things is all it took....

PDW

[.RC.]

Have got the lower saddle scraped onto the knee, except for the gib.

361788

[.RC.]

Been dabbling a little bit with it.

Had to wait on an opportunity to get some oil seals. Which I now have got. I have pulled apart the little right angle gearbox that transfers the feed from the knee to the saddle. It also incorporates the saddle cross slide nut.

Inside the box is just a set of spiral bevel gears. It should contain oil, but it had all leaked out, but the bearings and gears were all good. It has two oil seals on the horizontal shaft but they had a wear groove in them. it was not deep and I had new seals so I ground out the groove on the OD grinder. Did not take much. A little trick someone told me awhile back with oil seals if you have a groove or a slightly smaller shaft to seal is to remove the spring on the seal lip, undo it, shorten it a bit with some side cutters, them screw it back together and put it back on the seal, that makes it put a bit extra pressure on the lip. I took about about 2mm off the spring on a 35mm ID seal.

Here is the grinding going on. I have a magnetic chuck in the workhead, which has a 3 jaw chuck stuck to it, which holds the workpiece.

361845

same setup showing the groove being ground out

361846

For those that have never undone a lip seal spring, here is what one end looks like. Of course as you know, you cut the other end off to shorten.

361847

Of course as luck would have it my phase converter died just as I went to start grinding. Turns out a very large ceramic resistor snapped one leg off the circuit board. Is fixed now and I got the grinding done.

[.RC.]

BTW I am writing this in two forums if anyone sees the same thing in another forum.

Here for all the good people and elsewhere for the benefit of the unwashed :D :D

Moving along now



I have got the feed transfer gearbox back together and installed. It incorporates one of the underside saddle retainers (or what ever is the proper name), which I ground to close up the clearance to about 0.0005"

361883

I also scraped the saddle gib and guess what? It was steel as well.

I had always had trouble the the saddle cross feed engagement so I decided to fix that while it was in pieces. That took longer then expected. Essentially the shaft that moves a spline to engage the dog clutches was binding. The mechanism is a real silly idea and the later models did away with it.



Went to put it all together with a new seal and of course in fantastic Russian design, the seal does not even go anywhere near the housing.. It just sits there floating in air. You can just see it in the photo

361884

Here is a fix and I worked out why the Russians lost the cold war.


To fix the seal I made up a for want of a better word seal extension. If I knew about this beforehand I would have bored the hole out bigger when I had the knee stripped right down and done an even better repair.

Here is the seal extension piece.

361885

It fits in this hole.

361886

Like this

361887

This is the bit it runs on. I have thinned out the piece with the wire clip on it. This piece butts up against a tapered roller bearing and is the retainer for the bearings. Thinning it by 4mm gave me sufficient room for the seal extension to work.


361888

Putting it back together revealed the whole assembly would not bolt up flush on the front of the knee housing. There was a 2mm gap.

Trying to find the issue finally yielded this. A manufacturing error where a small raised section that should have been machined away at the factory was hitting the knee housing first. A quick grind with a carbide burr in a die grinder sorted it out.

361889

[KBs PensNmore]

Nice work and good to see that you worked a way around the problem :2tsup:
Kryn

[.RC.]

So where were we? Ahh yes we had just reassembled the knee.

Next up was to put the upper saddle section onto the lower saddle. I need new felt wipers and have ordered some engineering felt off ebay. I also ground all the under saddle retainers and fitted them with a smidgen of clearance there.

I put the upper saddle on, which still had the table attached and checked some alignments and the front was still pointing up in the air. So the problem all along was in the saddle/table ways.

When I first got the mill I disassembled it and it has an unusual coolant drain system, in that there was a hole in the middle of the table on one side, which drained into a long slot which was between the ways on one side. Then went out a threaded hole onto some tubing into the base. Now the threaded hole had obviously been blocked for decades. The slot was full of rusty swarf and crud. The ways on that side had suffered from the rust and coolant. and measuring they were approx 0.1mm lower then the other side. There was my alignment problem. This has been fixed in two ways. Firstly I was not impressed with the skinny ways that were on that side. So I pocketed out a section and inserted a cast iron block. The photos will explain better. The cast iron block was glued in with a substance that is green and comes in a red container.

When both were down the upper saddle was put on the slideway grinder where it was ruff ground in. Which I have just completed and I will finish grind it tomorrow some time.

First up before I pocketed the clot I needed to sharpen my sole carbide endmill that I had chipped a corner. Another job for the slideway grinder :D I actually do not have a suitable diamond wheel for doing this easily on the TC grinder.

361939

What it looks like on the mill prior to removal

361940

Starting to mill the pocket

361941

One pocket milled

361942

Measuring the slot width with gauge blocks

361943

One side done and fitted

361944

Rough ground, ready for finishing later on

361945

[KBs PensNmore]

You'll have a "new" mill by the time it's assembled.:2tsup:
Thanks for the WIP updates.
Kryn

[Steamwhisperer]

I agree with Kryn.
I'm loving this thread Richard, is there anything you wont have a crack at?
I still want to live at your house:D

Phil

[Oldneweng]

Great. Amazing possibilities open up just by buying a slideway grinder. A carbide endmill. Love it.

Dean

[.RC.]

The table flat ways are ground. Not going to touch the dovetails as they are not worn by any significant margin.

Took a long time to grind the ways, probably more because of other things that are more important to do taking precedence along with I could only grind a little bit at a time due to heat from the grinding process. Steep learning curve. You start grinding in the morning and after a short time, the wheel starts expanding and I also think the spindle expands a bit as it heats up. So you have to watch it like a hawk after you start.

Also last night when I was about done it showed a 0.01mm rise at one end that I had been battling to remove. The table was slightly warmer then ambient temperature.

This morning it was flat with the grinder ways.

Ideally now would be someone with a suitably accurate planer to plane the top and hope like hell the table does not warp.


362022

[jhovel]

Richard,
why wouldn;t you take a lick off with the grinder? I assume you are just trying to clean it up? Otherwise, why not scrape it to correct any geometry relative to the dovetails?

[.RC.]

Not sure what you mean Joe? Take a lick off which bit?

[jhovel]

Not sure what you mean Joe? Take a lick off which bit?

I was referring to the top:

Ideally now would be someone with a suitably accurate planer to plane the top and hope like hell the table does not warp.

[.RC.]

Planed surfaces on top are probably the best since then you do not have a dead flat surface that everything you put on the table will stick to, instead the surface is wavy.

The surface of this table has been planed. A lot of Euro stuff seems to be. Even Deckels I believe.

Another way out is to flake the surface. Seen some images of CNC machining tables with heavily flaked surfaces eg: https://www.youtube.com/watch?v=tlaLyfUBfKs

Turns out the table is bowed. approx 0.05mm

[jhovel]

Whoa that video is interesting! Do you think it's a stationary blade or a at a very slight angle /out of tram?
So you could scrape it flat or grind it and then flake it - that would look nice.
On that issue: have you ever tried to push the flaker along a straightedge? I wonder if that might help us 'sporadic' scrapers to make it look 'pretty'? You could use a couple of suitably evenly spaced perforated strip with locating pins against the straightedge to get the lines to remain parallel and at the same angle. I mean guiding against the front of the Biax, not against the blade
Since you are unlikely to maintain the same forward speed, it would still have the 'hand flaked' look - just 'neater' than we are otherwise capable off freehand....

[.RC.]

Not sure Joe, I think it is a blade shaped type tool so that in one direction it makes a skinny groove and the other direction a wider groove. That would give a half moon type shape when moved in a semi circle motion.

I have finished grinding the table and I am really happy how it turned out. It just seemed to take forever though. There till nearly 11pm last night and took a good proportion of today as well as evenings through out the week. It is out around 0.01mm or so. it is 1200 long or so. Does not appear to have warped.

Unfortunately this picture does not do the real thing justice.

362037

[jhovel]

That looks fantastic, Richard!
Are you going to flake it?

[.RC.]

Yes I will flake it Joe, although a bit of practice will happen first.

Put the table on the surface plate

Results are.

dovetail flat ways sub 0.01mm parallel (0.01mm indicator really did not move from one side to the other). However they both are 0.01mm low in the centre over the 1250mm length ways. The table may have warped a tiny bit when I ground the top. It is user error, not machine error.

Table top.

total error I could find was 0.02mm and that was at the extreme ends at diagonals. Remember the grinding wheel is a cup and only 100mm diameter and the table is 320mm wide. So I had to grind in four strips. The middle two T slot sections are 0.01mm higher then the outer two. I knew that the middle was slightly higher when I finished and before I removed the table from the grinder, so I could have fixed it but since the middle will wear first, it will hopefully wear in.

Slightly better pictures here. The grinding head is adjustable for nodding tilt at any time so that is why you can easily get the cross hatch pattern

362045 362046

[jhovel]

That looks bloody fantastic Richard.
I see Phil gave you some pointers on PM. I'm sure you will do a grand job!
Have you considered using a staightedge as a guide for the flaker?

[.RC.]

Have not thought too much about it Joe.

This evening went a bit quicker then I expected.

Do not think I have gone from roughing to finishing so quickly.

The dovetail spots nicely as well and it was not touched other then a bit of scraping when I was scraping the top.

I have an overhead crane of sorts set up above the surface plate that is also under my scraping table. It is simply some heavy walled 3" pipe 4500mm long that sits up in the trusses and on it runs a cheap chinese crane. I think it is rated to 800kg.

With the upper saddle on the scraping table, I could simply blue up the bottom ways of the table and slide it into the saddle then lower and spot. I did not count but probably only had to do it ten times or so and I had really good spotting all over.

To check alignment that the table top is parallel with the bottom of the base I simply lifted the assembly onto the surface plate and measured it with a height gauge. It is good. 0.01mm or better cross and very similar lengthways.

362049


I did probably another three runs after this picture.

362050



Where the inserts I installed end and the original part begins is totally indistinguishable.

I have to flake the ways and table top and look at redoing the lubrication grooves. And cut the felt up for wipers.

Have to do the steel gib as well. Not looking forward to that.

And all I set out to do when I started this was fix the rapid feeds. This is what you do when you have no life. LOL :D:D

[YBAF]

.....This is what you do when you have no life. LOL :D:D

Heh, I think most people on the planet would love to have the spare time to be an artisan. :cool:

[KBs PensNmore]

Any chance of some better shots of your "overhead crane" please, particularly the roller set up.:2tsup:
Thanks
Kryn

[BaronJ]

Hi Richard,

That is one great big chunk of granite you have got there.

Very nice workmanship on the scraping :)

[.RC.]

Any chance of some better shots of your "overhead crane" please, particularly the roller set up.:2tsup:
Thanks
Kryn

I shall, the rollers are pretty simple.


Hi Richard,

That is one great big chunk of granite you have got there.

Very nice workmanship on the scraping :)

Thanks. It is 1.6m X 2.5m I think, made in South Africa during the apartheid years and sold to the South African aircraft manufacturing facility (Atlas Aircraft).Then later a South African company had big ideas to set up a machine tool rebuilding facility in Brisbane which turned out to be a bad idea. They could not sell the plate and eventually I ended up with it. Was grade AA when new but from the few measurements I have taken it is only really a workshop grade now as it has some wear holes in it. No chips in it though the surface is immaculate so it would be relappable if one wanted.

[.RC.]

Pictures of the overhead crane system

362089 362090 362091 362092

[PDW]

Pictures of the overhead crane system

362089 362090 362091 362092

Cow cockies & fencing wire - cliches really are based on fact, hey?

PDW

[KBs PensNmore]

Thanks for that RC. As you said it's very simple.:2tsup:
Kryn

[.RC.]

Well that is the table completed.

362107

[KBs PensNmore]

I'll bet the brand new ones don't look like that. :2tsup:

[jhovel]

That is fantastic work Richard!
I looks just brilliant.
How did you choose the spacing of the diagonals? My sens was they would be closer together...
How hard was it to start near the edge that often?

[.RC.]

I had no idea how close to put them Joe. I agree though they are probably a bit too far apart. That is 25mm spacing's. I think 20mm would have been better. It is just to reduce the surface area a bit. Was no trouble starting near the edge.

[.RC.]

All pretty much done. Left to right alignment is pretty good. Cross ways is just within new tolerance (0.02mm/300mm)

I am happy with the results. I am no professional at this and it was a good learning experience. The bridgeport style head on it needs attention. It rattles like an old Kingswood.

362139

[Steamwhisperer]

Outstanding Richard.
I've loved every bit of this thread. Thanks. :2tsup:

Phil

[.RC.]

I finally have power again and made a not very exciting one minute video showing the now operational rapids.

It is 70MB in size for those whom are internet challenged.

https://www.dropbox.com/s/2crq0oi6ovl0rg7/Jafo%20mill%20fixed.mp4

One thing I have not mentioned about this mill is just under the table it has a keyed sleeve that rotates with the power feed. I assume the idea being you put a shaft in there and can use it to drive rotary tables and such.

I am starting to warm to this mill quite a bit now. It has great potential.

[Ryan in Melbour]

Left to right alignment is pretty good. Cross ways is just within new tolerance (0.02mm/300mm)


Could you elaborate on how you check this for a newbee please?
I would be interested in working out what my mill is like.

[Gammaboy]

Could you elaborate on how you check this for a newbee please?
I would be interested in working out what my mill is like.
Dial indicator held in the chuck, move table underneath is probably the most basic way. For running across the t-slots you would probably want to drop on a parallel to give the indicator tip something to run on.

[KBs PensNmore]

Looks real good RC, what are you going to restore now?
Kryn

[.RC.]

Could you elaborate on how you check this for a newbee please?
I would be interested in working out what my mill is like.

I think checking a mill is complicated and requires some some precision tooling like a largish square. Put a 300mm parallel on the table, dial indicator onto it and check the x and y axis. If there is major discrepancy put feeler gauges under one end and see if that zeros out most of it.


Looks real good RC, what are you going to restore now?
Kryn

Nothing for awhile. I do have to get the big rotary table working. and fit the DRO to the 10EE. Plus a heap of fencing, and other stuff.

[YBAF]

...and fit the DRO to the 10EE. Plus a heap of fencing, and other stuff.

Have yer got your-self one of these Oz made jiggers yet...

Petrol Post Driver- JACKAROO (http://horsleywholesale.com.au/contractor-petrol-post-driver-en.html)

I had a play with one the other day and they sure beat the old rammer pipe.

[.RC.]

Do not have one of those models, we have one of the more reliable original models from a different local manufacturer.

[YBAF]

Do not have one of those models, we have one of the more reliable original models from a different local manufacturer.

They did mention that the earlier manufacture had some issues with being over hardened in the hammer to picket area. Apparently fixed now.

RC, whats the name (or link) of the machine you've got?