Hand scraping

[Gordon AParticipant @gordona]

This is my first post on any forum so please forgive an lack of etiquette.

I have tried finishing flat surfaces by scraping with a modicum of success, and understand the principle of the high spots picking up the engineers blue from the reference plate.

However, researching “scraping in” of bearings on the interweb, I came across a site suggesting that after blueing a shaft and turning it in the bearing, the blue will be forced into the low spots, leaving the clear areas to be scraped.

Both methods sound logical, but I am having difficulty accepting that both seemingly inverse theories are correct. Please can anyone help clear my fog of confusion.

Gordon.

[Gordon AParticipant @gordona]

[Ramon WilsonParticipant @ramonwilson3]

Hi Gordon,

I’m not long  a member of this forum myself so know how you feel. Hope you get as much out of it as I have had so far. The interest of others and the help when asked for is very encouraging.

 

My experience of scraping is limited though I have had to carry it out on occasion both at work and at home. My understanding has always been that the surface that has been blued is scraped – the blue being ‘polished’ off by the mating surface leaving the high spots standing bright indicating where to be scraped.  The jobs I have done have worked well using that method but that is not to say it is the correct way. Knowing me I probably have it rrrs about face but it has been successful. The blue should be put on very very thinly – just a smear and well spread out – just a thin transparent layer but you probably already know that so apologies if so.

 

My friend John, a traditional time served fitter and turner on big diesel engines, is an absolute master at filing and scraping and I will ask him for you in the morning.

 

Regards Ramon

 

[KWILParticipant @kwil]

If you blue the subject  surface, the test flat or shaft will leave high spots [bright] to take off. If you blue the flat or the shaft lightly, you will leave blue on the opposing high spots, that is how I have always viewed it.

[Gordon WParticipant @gordonw]

I did flat & bearing scraping, long ago as an apprentice. Just about all I can remember is that the marking must be very thinly applied, this shows the high spots, but use common sense, sometimes, on a bearing, the highs will be polished. Stealing this thread, – I’ve just found 2 of my old scrapers, both HSS, one is a half round, hollow section, working area about 3″ long, other is triangular point about 1″ long. Question is how are they sharpened?

[Tony MartyrParticipant @tonymartyr14488]

After spending years in hand scraping marine gear box bearings I
have read a question in this forum that I feel fully qualified to
answer!

First the dolly used to bed bearings was never the
running shaft but a jig shaft whose diameter was made up of the shaft
diameter plus the running clearance. If you use the running shaft your
finished bearing, if trued bedded will be undersize

Second
only a very very thin coating of blue should be used and these will
mark the high-points grey/blue. If you have a generous coating the blue
will fill into the low-spots and the evidence is useless.

If
you are scraping in a two bearing shaft the task is straightforward;
our apprentices had to machine and scrap in a three in-line bearing set.

All
our bearing scraper, even for bearings up to 36″ diameter were made by
ourselves from steel made from the outer race of scrap roller bearings;
a flat blade curved upwards and slightly tapered with a cutting face
each side, easy to keep really sharp. We never used commercially made
triangular scrapers which tended to tip and dig-in when you were tired.

Scraping
imparts a shallow grooved surface to the bearing metal and we always
used an action that was at 45 degrees to the halving joint, the final
surface was always done so that a cross-hatched pattern was made that
encouraged longitudinal oil movement and counteracted grooving.

I
have had the chance to inspect some of the gearbox bearings that I
aligned after over 100,000 hours and the marking of the scraping was
still evident with polishing of the high-spots between the
cross-hatching only seen within plus/minus 30 degrees of the load-line
– of course years of clear oil helped!

Tony M

[Ramon WilsonParticipant @ramonwilson3]

Hi Gordon, well I think you will agree you aren’t going to get much better first hand knowledge than Tony’s.

 

However, as promised I contacted my friend and went over this afternoon as coincidentally he has the crankshaft out of his launch engine to in order to redo the bearings. I had a very pleasant afternoon being shown the correct (and totally opposite to the way I’ve done it for myself) way to blue the high spots and scrape a bearing and flat surfaces. My only experience of this at work was when I worked at a jobbing shop and was asked to do the slides and gibs on a wheelhead dresser we had made. Just had to work it out myself – typically as I said totally rrrs about face – but it did do the job’

 

So to confirm his teaching – flat or bearing – the blue should always be applied to the part not being scraped except fot the very final ‘high spotting’. He showed me that this final op can be done by allowing methylated spirit to ‘flash off’ over the part. It leaves a very microscopic ‘greyish’ layer that shows the very finest high points as bright spots when brought into contact with the mating part or a ‘spotting block’.

 

His blue, ‘Micrometer’ brand in a tin was a lot thinner and spreadable than my ‘Hi-spot’ in a tube. That is almost wax like and probably the reason I’ve always found it difficult to get it to transfer. The thing that made such impact was how light the ‘touch’ was required between parts for the blue to transfer.

 

I asked him about sharpening, triangular is hollow ground on the face of a wheel and then stoned, flat is held vertically, flat face to the direction of travel and then rubbed back and forth over an oil stone, rolling from side to side for the relevant curvature. I was told that there are no set guidlines for this, other than an obtuse angle is required, the degree depending on the material and to some extent the operators preference.

 

I’m glad you raised this question as I have learnt a lot from it too, particularly the actual bluing up process. Nothing like a good practical demo.

 

I took several pics so here are a couple of the bearing having just been blued, not much as you can see and the light transfer of it onto the bearing. The last is for interest – it’s his engine built from scratch by a good friend of his who also served his time with him at the same place. Hope this is of some further use, Tony has explained it so well there is little further to add.

 

Regards – Ramon

 

 

 

 

 

Edited By Ramon Wilson on 24/02/2010 21:22:11

[Les Jones 1Participant @lesjones1]

Hi Tony and Ramon,

                                        I think your last posts combined would make a good article in ME or MEW. I have made a point of copying them into a word document for future reference.

Thanks for the very well presented information. Also thanks to Gordon for asking the question in the first place. It made me remember my father re doing the big end bearings of his Morris 8 when I was about 10 years old. (about 56 years ago.)

Les.

[Tony MartyrParticipant @tonymartyr14488]

Hi Les and Ramon

Ramon’s photographs remind me of an important point that I missed mentioning.

You have the be very careful about not distorting the bearing shell in a vice by pinching it across the diameter (note how it is correctly clamped in the photo) on the same theme many shells tend to ‘spring out’ when taken out of their caps so we always did any scraping with the shell installed – this is vital otherwise you can end up with too little clearance at the halving joint.

The key point, which Ramon’s message has supported, is that you use a very very thin film of marking blue that is evenly applied to a clean shaft and then almost wiped off again with your fingers (the best non-fluffy material) otherwise you get a complete covering of blue that is useless.

This subject would be a good theme for a ‘You Tube’ video as the amount removed and the direction of the scraper action and the preparation of the dummy and the problems of correcting alignment etc etc are best learnt by watching and then doing – which was what 5 year craft apprenticeships provided in the ‘good old days.

Tony

[Nigel McBurney 1Participant @nigelmcburney1]

Hi      Interesting info from Tony Martyr,about scraping a bearing to a  mandrel so that the correct clearance is obtained,never seen that before in print. A lot of fitters made their scrapers from flat and triangular files.

[James BParticipant @jamesb]

I have been interested in the technique of hand scraping for a while – excellent information from Ramon and Tony – thank you…!

 

Where this applies to a shaft in a bearing, there is one area I have not been sure of the correct method, is scraping parallel areas, for instance, the slideways of a machine – the instance in question would be the saddle of a milling machine. Getting one side true, using a gauge or master is clear in principle (if a little different in practice..!) but how would one go about ensuring the second slideway is parallel to the first – is this a case of measurement, or should you have a gauge to cover both sides? Or would you use the mating slideway to get a working fit?

 

Also, on a similar direction, if you wanted to scrape a surface larger than your available reference surface, is there a preferred method? Overlapping the scraping area? Or should this be avoided?

 

Thanks,

 

James

[Ramon WilsonParticipant @ramonwilson3]

James,

I expect Tony to have much more to comment on this than I but even though to date I have carried this out in a somewhat dsylexic manner the results were successful.

 

Sometime back I made a new, longer table for my mill and scraped in the old and new slideways. With nothing else available I used a straight edge ground from gauge plate (soft state). The initial side of the saddle dovetails were done and squareness checked using a DTI off the inner machined but unused face. The other sides parallelism was checked by micrometer across two ground rollers held in the vees of the dovetails.

 

Once done the sadlle was refitted, the gib put in and a slide fit checked for any tightness in places over the travel. This was then repeated for the new table. Despite my application of blue to the wrong surface(!) it worked and I’m more than happy with it’s accuracy and alignment.

 

I know that to scrape a flat suface you need three surfaces and from the way my friend John used my small 10 x 15 plate to scrape his much larger one then I guess it can be ‘overlapped’ but perhaps Tony or others can take over from here.

 

Regards – Ramon

[Tony MartyrParticipant @tonymartyr14488]

Hi James

Of course the danger of using a flat ‘master’
that is significantly smaller than the ‘child’ is that you may not pick
up macro distortions like bend across the full width – which is why the
3 plates method works as long as the plates are all the same size. With
just two plates one (A) could be crowned and the other (B) hollow and
you could get a good blue marking on both but when you tried both on
(C) you would find how they differed.

The 3 plates system
used to be an apprentice test task and it was inspected with a piece of
plate that had a one inch square hole in it with which the contact
points per square inch were counted at any point of the surface of
every plate. It took me years of practice before I could produce a
‘feather’ pattern with a scraper on flat cast iron or steel halving
joints but when you got it right it looked wonderful and I have always
wondered if the process has been automated since I saw a Maag gear
grinder a few years ago on which the scraper pattern was too good to be
true.

Flat scrapers have a slightly curved front edge and
are best used at an angle of about 30 degrees from forward rather than
straight ahead. You sharpen the edge by holding the handle vertically
above the stone and rocking the cutting edge to and fro in line with
the blade – ruins your stones and I have a collection of grooved ones.

Interestingly
when I worked on a big Swiss turbine site I learnt their technique
which is to hold the scraper much more steeply and use a draw stroke as
the cutting stroke, their scrapers were held at the end of a 4 foot
long tube that rested on your shoulder and had a weight at the far end
– this allowed one to develop a sort of bouncing motion that was
transferred to the cutting edge and produced a ‘snowflake’ pattern that
you may see on the beds of German and Swiss machine tools.

Once
again a very light covering of blue was used on the master surface
which, in the case of turbine and marine gears was a very large cast
iron surface plate that was suspended face down from a crane. Once the
contact points were close enough and there were no hollows on both
bottom half and casing the two were put together and the joint was
inspected by proving with a 1.5 thou feller gauge – You only called the
foreman to carry out this inspection if you were very sure it was right!

Finally
the surface produced is not truly flat, as in mirror-like or flat as in
gauge blocks that can be wrung together, but they are truly planar and
the surface is excellent for retention of lubricant between surface as
required in bearings and machine tool sliding surfaces.

Tony M 

[Sub MandrelParticipant @submandrel]

I too would tip my hat to Tony and Ramon,

 

What an informative set of postings!

 

Neil

[James BParticipant @jamesb]

Hi Ramon, Tony,

 

Thanks again for your comments – very helpful – I get involved in the restoration of older machines, and my next, I think, I will be looking to try my hand scraping..

 

Please excuse my ignorance, but the three plates system – is this just 3 seperate master surfaces to cross check the results on the child surface, or is there more to this?

 

Thanks,

 

James

[Tony MartyrParticipant @tonymartyr14488]

James

You have to produce 3 to get one!

It is an averaging process that only works with 3 (or more)

If you choose on as an arbitrary reference (say A) you blue it and try it to B& C in turn in several orientations because you might have one with a hollow or a twist etc.

If you then scape the best to A (say B) you then use B as the parent of C. Then you use C as the parent of A – by this time you might have a good bedding in any combination but more likely you have to continue with the round robin until the errors are all averaged out and you have three surface plates (keep one and sell the other two)

Two plates won’t work because the parent imposed its distortions on the child – as in real life!?

Tony

[John McNamaraParticipant @johnmcnamara74883]

Hi All

Scraping Mild steel.

Over the weekend I decided to try making a small test piece using an offcut of 350 x 75 x 50 hot rolled channel. I milled all the faces and the bottom of the legs first, then got to work scraping it to size. Mild steel is not easy to scrape and I wanted to see how long it would take, in preparation for a bigger project.

So I set to work scraping the 75mm face using Prussian blue on the surface plate to mark the piece.

Some observations:

It is very difficult to scrape in two crossed over directions as you can with cast Iron. The scraping tool digs in. Scoring of the work is a major problem The small 20mm wide curved nose tool used was ground then stoned to a polish. I ended up scraping in fine lines.

After every scrape you have to remove all burrs before cleaning the scraped part and applying the part to the surface plate (having re-spread a thin even transparent coat of bluing), this is the same with cast Iron except the burs are firmly attached… In the end I stoned them off as a file tended to leave them sitting a little proud making false markings .

The first step was to grind the thin edges of the channel parallel to the top. I did this on successively finer sheets of abrasive paper on a flat plate by applying more pressure on one corner you can steer it down and make the top parallel.

I was not able to get a good finish (scratches) however was able to get a reasonable spotting pattern. In the end I decided to finish off by grinding the high spots down with Mylar backed abrasive strip 18mm wide about 600 grade wrapped over the tip of my finger. Using this on the bluing high spots worked quite well.

As the process preceded I was able to get the face flat to within about .0005" high point to low point on the indicator.

Next I used a clean new sheet of 1200 wet and dry paper on a smaller surface plate (Not my good plate) and using this paper first overall then back to spotting with blue got the surface down to .0002" moving the piece under an indicator set on a stand placed on the surface plate.

At this point most of the scrape marks have disappeared being ground away by the second stage.

Then trouble struck!

The same process was applied to the sides. A lot easier having developed a process doing the top. however the top face started to move a curve in the middle .004" high appeared as the work progressed……. No doubt residual stresses. I should have done all three sides at once.

It is fixed now well almost. a thou to go. hopefully one night this week.

Cast Iron is a lot easier to work with but this confirms the steel bearing rail supports for the Epoxy mill can be hand scraped if necessary.

Cheers

John McNamara

 

 

 

Edited By John McNamara on 18/06/2012 16:09:27

[David LittlewoodParticipant @davidlittlewood51847]

John,

As your experience proves, if you want to finish MS to very precise dimensions, it is a good idea to anneal it first to remove the locked-in stresses. Bright drawn mild steel, as the name suggests, is drawn cold, which leaves a lot of internal stresses; if you machine this, especially if the machining is asymmetric, then the piece will distort. Whether this is disastrous or not depends on how much it distorts, but of course also on how accurate you need the piece to be.

Tubal Cain's book in the Workshop Practice series contains details on how to stress-relieve:

**LINK**

David

[maurice bennieParticipant @mauricebennie99556]

Hi Ramon, I have three black squares 4ins with a white circle in the middle ,inside that a horizontal white bar where the photos should be .Can anyone explain please?

thanks Maurice.

[The Merry MillerParticipant @themerrymiller]

I have the same problem as Maurice (three black squares I mean!!)

Len. P.

[John McNamaraParticipant @johnmcnamara74883]

Hi

The three blanks are caused by the links the Ramon used no longer existing.

Or maybe the server has changed and they need to be re-pointed to the correct path.

Cheers

John

[Ramon WilsonParticipant @ramonwilson3]

' Morning Guys, I noticed this late last night and too late to do anyting about it.

I don't know exactly what has caused this link to be broken as the images are still on my Picasa web album and as far as I'm aware still accessible. Though I have changed computers all my files were transferred and this has not happened on the other site I use -HMEM – as far as I know

However this has also happened on the Aircraft General Discussion thread on here.

Because there is no ability to edit a post on this site beyond the short time allowed after posting nothing can be done, I'm afraid, to ressurect the images within the text. I can however put the images up in an album and will do so later.

I have a horrible feeling that this is going to occur elswhere on my other posts but that will have to wait until it manifests itself as time is rather short at the mo

Thanks for pointing it out though

Regards – Ramon

[John McNamaraParticipant @johnmcnamara74883]

Hi All

I stumbled across this video of scraping a bevelled straight edge.
For material he cut a section from an old surface plate then milled it. A good idea…. I have seen rusted or badly worn ones go for scrap value. Have to keep my eye out.

Interestingly he does a lot of scraping by dragging with a hand made hook type scraper; he also uses the conventional push type scraper that also appears to have a brazed on tip. And……. here is another use for old carbide tips!

The narrow width hook type scraper appears to remove metal rather well, it would be good for the early passes. I am not sure it would be as good as the round type for spotting the later passes.

The 2 part video does not have a voice over just scratching noises, and in the middle a funny face? a bit strange but the technique he uses gets the job done and he has obtained a good result.

**LINK**

Regards
John

[IanTParticipant @iant]

Good video John – only problem is that I seem to be watching as much YouTube this week as TV (although the former was certainly much more educational!).

This guy seems to be using the exact technique that Tony mentions earlier in this thread (with reference to a Swiss Turbine) and both the "pulling" action and the long shafts (supported on the shoulder) also seem to match his description od Swiss practice. My scrapers are the more traditional 'file-like' ones but I will be tempted to try these Swiss ones, as I have trouble with my fingers these days if I subject them to too much "stress" (probably linked to my arthritis). I think these may be easier on them (possibly spreading the load?). Very sore thumb today just from using an 8mm tap (& holder) yesterday.

Regards,

IanT

[VersabossParticipant @versaboss]

Hi all,

in my humble attempts at scraping (2 mills, a shaper, a small lathe up to now) I always used that pulling method with a long scraper resting on the shoulder. I think I have ten times more control over the direction and length of the stroke and can use more pressure on the scraper if necessary. Although I have to say my scrapers seldom cut so nicely as demonstrated in that video!

What I don't have is a small beveled master for scraping vee ways; I'm glad having seen it can be produced with the means I have…

Regards, Hans R.

[Author]

Posts