Help needed truing ways on mill

[egaParticipant @ega]

I have just stripped down my Warco Economy Mill/Drill and was dismayed to find the upper Y-axis ways had been poorly machined. I have uploaded several photos to an album but the situation can be appreciated from this upside down view of the sliding base:

At opposite ends of the ways it seems that the finishing cut has not covered the full length (near end on left and far end on right in photo, see album for other views). I have checked that the underside as seen in the photo is flat and micrometer measurements indicate that the bad areas are about three thou low.

I am in Rochester, Kent. Is there anyone reasonably near, please, who is willing and able to help? The base is about 195 x 240 in size and the ways appear to have been milled rather than ground so I am looking for someone with a large mill. As an alternative to a helping hand, can anyone recommend a suitable machine shop in the area?

This machine has met my modest needs for many years and I suppose I could just reassemble it, keep calm and carry on. The only other in-house possibility seem to be the doubtful course of facing this heavy lump on the lathe faceplate followed by scraping.

Incidentally, I am rather at a loss to understand how the defective result has been achieved given that the other side is flat.

[egaParticipant @ega]

[MICHAEL WILLIAMSParticipant @michaelwilliams41215]

Hi ega ,

You could very easily file the 3 thou correction .

Cast iron files very easily anyway but with the already interrupted surface shown it would be easier still .

Scrape to final accuracy ideally but it is not always actually nescessary – your filed finish can be adequate and certainly better than what you have now .

You would need use of a big enough surface plate for checking but no other elaborate equipment .

There are systematic methods for correcting and checking surfaces .

You could just file and scrape away and hope for the best but there are better ways .

A simple method is to file and scrape small registers at each of the four corners giving you the same reference height in each place . Just file and scrape then until registers merge with rest of surface .

Lots more if you want to hear about this .

Regards ,

Michael Williams .

Edited By MICHAEL WILLIAMS on 21/02/2014 23:36:04

[I.M. OUTAHEREParticipant @i-m-outahere]

If it is machining to your expectations then leave it alone .

There is no point in worrying about a machine that is machining flat and square as you could end up adding to the problem if you don't know exactly what you are doing .

Ian

[Ian PParticipant @ianp]

If your lathe will accommodate that on the faceplate then go for it. You could lash up a toolpost grinder with a power drill or a Dremel type tool. Even if you rotated the lathe by hand, with lots of patience you would be able to grind the surface yourself. (Check that your cross slide travel will cover the diagonal of the job)

Your biggest problem is getting aligned in the first place! The finished face needs to be parallel with the opposite surface but that assumes the dovetails were correctly machined originally.

Obviously only remove the absolute minimum so you dont run out of gib adjustment.

Ian P

[roy entwistleParticipant @royentwistle24699]

I would fully agree with slotdriller Leave it alone

Roy

[RobboParticipant @robbo]

Am I right in thinking that it is the areas where machining marks can be seen that are low? If so I would suggest that it once all looked like that, and you have smoothed the rest with use.

If this is case there is little point in bringing the smooth areas down to the level of the "marked" areas, they will continue to wear down in time.

If it is sitting flat and true then leave it alone.

[Ian PParticipant @ianp]

Gray

Whilst I agree with the 'If its not broken don't fix it' technology, I doubt that things will improve when wear takes place.

The table at the moment is sort of diagonally supported with 0.003" gaps at opposite corners. Since wear will take material from the both surfaces in contact, the underside of the table will effectively become a twisted plane.

If the surface did become concave by being done on the lathe it would be far less error than the 3 thou mentioned but at least it would be symmetrical, then also much easier to correct by scraping/filing.

I am sure if this was your machine you would not use it as it is? To carry out work to the standard you set (to which I can only aspire!) needs a machine to be rigid and repeatable, gaps in dovetail ways is the last thing needed.

Your description of scraped surfaces reads like a reason for not scraping, but I am sure that is not what you meant to say.

Ian P

Edited By Ian Phillips on 22/02/2014 10:09:27

[MICHAEL WILLIAMSParticipant @michaelwilliams41215]

Since the slide block is already off the machine it would surely make sense to find out for sure what errors of flatness and parallelism it actually has . You can decide then objectively what to do .

First test is on surface plate and see if you can get thin shims in under anywhere .

Second test is bluing .

Thickness could be comparator but probably micrometer is enough .

Decide then whether errors are minor and can be ignored or are gross and need to be corrected .

I’ve mentioned hand filing/scraping but there are other non machining methods of correction possible .

If , for instance , you have one relatively big low area but with rest of surface ok you could try bonded shims .

The point mentioned about scraped surfaces having many riding points surrounded by oil sinks is certainly true but it is actually true of all machined and hand worked surfaces . Merit of scraping is that you can achieve high standards of flatness and at the same time control the ratio of rubbing area versus oil retention area to suit specific purposes .

If you want to machine the surfaces then get them surface ground professionally .

Final note – keep metal removal minimal in any case . Any large removal could alter the lateral alignment of the slides .

Michael Williams .

[egaParticipant @ega]

Gentlemen:

Many thanks indeed to all of you for your very interesting and constructive posts. The variety of approach has been quite striking and I am going to have to mull it all over before deciding what to do.

Robbo: in answer to your question, I don't think I can improve much on the paragraph immediately under the photo in my original post; the areas where the *coarse* machining marks are visible are indeed the low areas and the remainder of each way shows the marks of *fine* machining. As I said, I am puzzled by all this; if the casting had warped I would have expected the other side not to be flat.

[Roderick JenkinsParticipant @roderickjenkins93242]

I'm probably confused but does that "upper surface" actually touch anything? Does it not bear on the dovetail and gib strips.

Just a thought

Rod

[BazyleParticipant @bazyle]

Posted by ega on 21/02/2014 21:21:40:

Incidentally, I am rather at a loss to understand how the defective result has been achieved given that the other side is flat.

It is quite wide so not done at a single setting. The Operator slaps it down and does one side then spins it round and does the other side. All the castings are done at the same setting and this one was a bit thin. So is he going to reset the machine and redo both sides reducing his throughput and pay for the hour by 20%?

Remember if you do get it ground you ought to do the mating surfaces, then ground on ground is 'too good' and doesn't retain oil. That's why on good machines in the old days they would scrape a characteristic pattern ( not a full scrape in ) which I think has a name I can't think of. Now some of the far eastern stuff has a guy scratch this pattern with a broken file to make out it is quality.

[John Stevenson 1Participant @johnstevenson1]

Posted by Bazyle on 22/02/2014 13:37:42:

then ground on ground is 'too good' and doesn't retain oil.

Damn, Now got to go out into the shop and hand scape 23,743 ball races in…………………………

Plus 9,373 linear rails.

 

Sigh……Another flat battery in the car park of life

Edited By John Stevenson on 22/02/2014 13:53:08

[BazyleParticipant @bazyle]

mmmn but ball races are not sliding contact though if you disagree with my assertions you are probably right.

Frosting or flaking are the words I was after. See picture in post 14 in this thread for a proper job. The thick – thin line pattern being the signature that is sometimes immitated.

Edited By Bazyle on 22/02/2014 17:11:55

[Gary WoodingParticipant @garywooding25363]

If, when placed on its male dovetail, without the gib strip in place, it…

1. Doesn't rock when each corner is pressed down, and

2. The critical surfaces are relatively horizontal in both directions,

then leave it alone, otherwise machine it where appropriate.

[WALLACEParticipant @wallace]

I’ve often – well, once or twice – wondered why after a surfce is ground, it isn’t rotated through 90 degress and ground again at the same setting to (hopefully ) give a sort of cross effect that would retain oil.

Or am I talking coblers and scraping is the way to go ??

W.

Edited By WALLACE on 22/02/2014 17:42:49

[MICHAEL WILLIAMSParticipant @michaelwilliams41215]

Hi Wallace ,

Not cobblers at all I assure you .

When doing real engineering it is quite common practice to specify not only simple dimensions and tolerances on drawings but also the required surface roughness and the surface finish pattern .

The surface finish pattern is sometimes called the lay and defines the fine pattern to be made by the tooling .

Among the many variations of lay possible one is generated by two passes at right angles – just as you say .

There are many reasons to specify a particular lay but a common one is to ensure a difference in texture on two parts working against each other .

Another common one is to specify a lay on a ground shaft that gives favourable running characteristics on a plain bearing .

Another one again is in the raceways of high specification ball bearings for aerospace use .

Regards ,

Michael Williams .

[jonathan heppelParticipant @jonathanheppel43280]

Look up Rich King. There's quite a lively scraping scene in the US. I get the feeling that about the only scraping left in the UK is done by the very few reconditioners remaining, eg ZMT the Hardinge guys and the Bridgeport and S&B specialists whose names I forget- oh and D,S&G, but I think they would all regard yours as a scraping job. A surface ground piece would normally need hand finishing anyway. In the States they prefer to scrape all surfaces, but a specialised slideway grinder can leave a rough enough finish and it's quite common just to scrape the mating slide.

Scraping is by no means entirely obsolete, but largely only seen on the finest machinery these days. I am not refering to the butchered "flaking" on oriental machinery which the machinery dealers use as a selling point to the unwary.

If you've never tried scraping, it can be quite hypnotic, but by heaven is tedious!

[WALLACEParticipant @wallace]

Thanks for the replies – I can see the advantge of scraping in that it’s something that can be done with simple equipment – albeit with considerable skill in manual dexterity.

But does it still have a place in industry ? I would have thought a modern all singing grinder would produce a better (whatever thar is ! ) surface finish in a lot less time and a higher degree of repeatability – but possibly not as cheaply as a man whose been doing it all his life with a scraper ground up from an old file..

W..

Edited By WALLACE on 22/02/2014 23:37:39

Edited By WALLACE on 22/02/2014 23:38:30

[jonathan heppelParticipant @jonathanheppel43280]

It's the other way round. Scraping is an extremely expensive process. Not just the man hours, but the temperature controlled factory space in which to work. For this reason, I expect it will be totally obsolete in a few years for new builds, but check out the Dixi site. (Super precision Swiss borers- they are very proud of their hand finished machinery) Rich King has been spending a lot of time in Taiwan teaching scraping classes because their premium machines are still hand finished, though their cheap stuff is farmed out to the PRC. There must be automated processes, but they don't yet have hegemony. No old files either- power scraping machines and carbide tips, though it seems the Swiss pull style still uses steel, even just carbon sometimes. There are some good you tubes and also threads on US forums.

Dixi site says 500 hours of scraping per machine! Still if you want the best……….

http://www.diximachines.com

Edited By jonathan heppel on 23/02/2014 00:29:29

Edited By jonathan heppel on 23/02/2014 00:34:20

Edited By jonathan heppel on 23/02/2014 00:40:13

[egaParticipant @ega]

I see that this thread has broadened into an interesting discussion of more general matters.

So far as my problem is concerned, there seems to be a near-consensus in favour of either leaving not too bad alone or, if machining is resorted to, having the offending ways surface ground. This brings me back to my original quest for a suitable machine shop and so far I have been unsuccessful – one disadvantage, perhaps, of living in the soft underbelly of England. I will continue my search when the world gets back to work tomorrow.

Renewed thanks to all those who contributed. The responses have been genuinely helpful.

[MICHAEL WILLIAMSParticipant @michaelwilliams41215]

Hi ega ,

(1) Don’t tell us your personal address but whereabouts do you actually live ??

There are still useful jobbing machine shops in many parts of country though sometimes they keep rather a low profile these days .

There are at least three specialist slideway grinding firms as well .

Give us a clue and may be able to point you in right direction .

(2) If it were me and I thought the job worthwhile I would do it by hand – maybe take me one weekend .

Regards ,

Michael Williams .

[egaParticipant @ega]

MICHAEL WILLIAMS: I will PM you.

[Ketan SwaliParticipant @ketanswali79440]

Posted by ega on 23/02/2014 10:18:08:

I see that this thread has broadened into an interesting discussion of more general matters.

So far as my problem is concerned, there seems to be a near-consensus in favour of either leaving not too bad alone or, if machining is resorted to, having the offending ways surface ground. This brings me back to my original quest for a suitable machine shop and so far I have been unsuccessful – one disadvantage, perhaps, of living in the soft underbelly of England. I will continue my search when the world gets back to work tomorrow.

Renewed thanks to all those who contributed. The responses have been genuinely helpful.

I would suggest you leave it alone. You should re-read Grahams posts on this thread. He has the ability to scrape, and he has a clear understanding of mating surfaces. By doing what you intend to do by taking it to a machine shop, you will probably open up a different can of worms and a never ending story.

There are so many factors to be taken into consideration before you go on this journey. If you want to understand in more detail, you can call and discuss with me. I am in and out of office tomorrow. If you can't get hold of me tomorrow, you can call me during the week.

Ketan at ARC

[John Stevenson 1Participant @johnstevenson1]

Scraping was common in days of old because bed surfaces were formed by milling or planing, neither which gave a finish accurate enough.

When grinding came into force people still expected to see flaking.

Take a look at a modern CNC which have to hold tolerances all day we would kill for and in thier case it's not about cost but do you see flaking ? short answer is no.

My Beaver CNC which is really old technology is built the same as a modern Haas in that the ways are ground and one, usually the top has a series of circles milled in that link up and also link to an oil feed.

If it stands for any length of time and you move the slides a long way you can see the pattern left on the lower slide of the circles.

In theory the surfaces shouldn't run on one another anyway, that's what slideway oil is for, it been developed to preserve the integrity of the machine slides.

[Author]

Posts