Scraping - lessons learnt (the hard way) and other pearls of wisdom

[mm289]

So for those of us putting our recent training to good (or bad) use, and for anyone else brave/mad/foolish enough to while away hours of otherwise useful time scraping lumps of metal, I thought it might be an idea to start a "lessons learnt" thread that we could all contribute to.

The idea being that we will all learn stuff (usually from mistakes) we have made and if we share it here there is a slim chance the rest of us will read it and not make the same mistake, or learn the same lesson 

So feel free to contribute, doesn't have to be a lesson learnt, could be a top tip or trick you have seen on the web etc. I will start the ball rolling....

After nearly a week of preparation I finally got to do some scraping on the Centec mill today. Plan was to do the ways on the column and then work my way upwards.

Anyway, got the column on the bench, spent a good half day measuring and prevaricating before eventually deciding to take the plunge and hit it with the Biax.

Worked out how much I needed to take off (about 4 thou) and started hogging metal. Checking with a straight edge that is only about 40% the length of the way, and then every so often using the prism Pete lent me which is pretty much the full length of the way but very awkward to use.

Got the way reasonably flat on the vertical (end to end) axis (albeit the column is on its side in the pictures  ) but was only showing on the outside of the way when printing. Figured the inside was worn so started knocking of the blue on the outside where it was printing. (Pic 1)

After a while,,,,,,,,,,,,, checked my measurements with a bit of ground stock I am using as a long straight edge and couldn't understand why I had a side to side error on the bar/way. Put my surface gauge on and it showed a 2 - 3 thou error at each end of the way, but with the OUTSIDE lower than the inside. (pics 2 - 5)

I was perplexed. Checked and rechecked, got a cup of tea, came back, checked again - couldn't understand why the bit that was printing blue was LOW.

Because it is only a narrow way and I have limited tools I was struggling to figure it out. I checked my dial gauge set up (pic 6) and used some slip gauges on the way to measure again and still got an error.

Whilst playing with a slip gauge I decided to stand it on end (so it was the same width as the way) and do a rock test. B**ger me, but it rocked slightly.

After another half hour of mucking about I figured out that where I had been scraping the way I hadn't got in underneath the dovetail and had created a ridge there. So as I put my straight edge on it was sitting on this ridge and the outside edge of the way, and the blue was only showing on the outside edge. Of course , I was then scraping the outside edge further and just creating a bigger ramp 

I know at the course it was mentioned about using a hacksaw blade to cut a recess in the end of the dovetail - is this why? (must have been asleep for that bit of explanation  )

Anyway, that's what I did and then spent another two hours hogging of a load of metal from the inside of the way to try and get it flat again 

Slowly getting there as seen in last pic, can just about get a print in the middle area now, but what a faff  8 hours later and I have one way half scraped - and that is with a machine scraper

Lesson learnt: don't assume the blue'd part of the machine is high, it could be because the straight edge is pivotting on something.

Must be time for beer 

Cheers,

Paul.

[Graham Stabler]

Yep, I think that's the reason for the hacksawing. Good that you diagnosed it in the end!

[mm289]

Just as another update to this.....

I had another go at the ways tonight as I still wasn't happy with the flatness and the ridge on the inside of the dovetail. I ended up using a hacksaw blade to take a much deeper cut, and played around with the Biax on a very short throw to try and cut off the lip.

Was working but taking a while and not easy so I dug out one of Pete's narrow dovetail hand scrapers. Blacked up the inside of the dovetail with a marker pen and by using a DTI worked out where the ridge was. Then used the scraper a bit like a snow plough, so turned it so it was running parallel to the way with the blade sitting on the ridge. the marker pen helped to make sure I had it landed just right, then ran long scrapes all the way along the way/inside of the dovetail rather than short little scrapes.

Lifted the ridge of nicely and quickly with no scratching and gave a much better result than using the Biax!

Was then able to blue up and tidy up and now have a much better bearing pattern with the bearing points reaching a lot further into underneath the dovetail than they did before.

Didn't take anymore pics but if its not obvious from the description above let me know and I will "model" for some 

Cheers,

Paul.

[PekkaNF]

Very good you got it sorted.

On few pictures you are indicating from the scraped surface. Fine sometimes and sometimes needle does so much weeble wooble it's hard to understand what's going on. Because human brain tends to even out and sort of average - which is fine when centering round objects - but not very ideal when trying to gage surface.

I have learned that when high spots are killed and and the straight edge feels stable, best way to indicate is to "bridge" 20-100 mm length of the surface and measure it. It shows general slope/dip pretty nicely at early phase of the work.

Not sure if this monoloque makes sense?
http://madmodder.net/index.php?topic=11870.msg140313#msg140313

I was thinking of ordering a 25 mm gage block, but remembered I had small comparator table, about 50 mm dia disc, that top surface is really flat and parallel to bottom surface. It has a camfer - I can zero indicator to one place, move indicator exactly over another point, slide the "scraper's block" under the indicator stylys while observing the reading and see from the indicator large scale slope of the surface or record representative reading.

I found it frustrating to better than 0,01 mm indication without scraper's block and pretty reliable with one.

Pekka

[mm289]

Hi Pekka, that's exactly how I am measuring flatness from side to side  I put a slip gauge block under the indicator then (0.5") is the same as the depth of the way then swing it over to the other way and do the same. Have now got both ways parallel (flat) to within 0.02mm which is good enough for me.

I also used a slip gauge when I was testing for the ridge and flatness across the way.

I am just sourcing an precision level and will be using that as well to check things and will post the results up on the main thread - if they look good 

Cheers,

Paul.

[PekkaNF]

Strange. Is't it?

There is observation/problem. 

Then is the thinking process.....

and many ways to solve the same problem

I'm using the feeler gauge specially on round/abrassive parts to average reading and to protect the indicator stylys.

Pekka