Benchtest.Com - Workshop - DoAll D624-8 Surface Grinder

DoAll D624-8 Surface Grinder Continued
Preface
I've been trying to decide how best to document this surface grinder project. I don't have the time to write daily and there's really not enough new happening each day to make for interesting reading, so I think that I'll try to update weekly.

Week of March 9 to15, 2008
I was bitten by a stupid mistake. The worst part of it is that I knew better. To the casual observer, metal is a solid, tough material that doesn't bend much. To the metalworker, even an aspiring one such as myself, it is anything but that. It grows and shrinks depending on the temperature and any piece with a length longer than a few times its width will sag if suspended by its ends. I don't know where my brain was when I decided that using wedges would be a good way to support and level the table so I could periodically check the ways with my box level. What I forgot to consider with using many supports(wedges) is that it is easy to put more pressure on one wedge than the others. What happened to get me thinking about this is that I bumped one of the wedges and had to set them up again. Once I did and I went to take a print, I found that I had a low spot in the center where it had previously been flat. Darn. In the time it took to say "WTF???" I realized what I had done and that I already knew that there was a better and easier way to support a long length of metal - like the grinder table - to minimize sag. This easier method is to use the "Airy Points" formula to find where to support the table to minimize sag.


Well,I thought it was a good idea to use the wedges, but it introduced a bow in the table.	I was getting pretty close to being completed with this one way, but now I have to rescrape the dip out of the center.

There's a formula, penned by Sir George Biddell Airy (1801-1892) the English astronomer and mathematician, which states something to the effect that if an object of constant thickness and width is supported by two points separated by 5/9 of the object's length, it will incur the least amount of droop or sagging.

With the 47" table, 47 X5 / 9 = 26.1. This means the distance between the 2 supports will be26.1". This will put our supports approximately 10.45 inches from each table end. With this setup, I can now minimize the possible sag. All that's left is to do is to scrape the ends to match the dip I scraped into the center of the ways when I inadvertently raised the height of the center with the wedges. Double darn. This not only hurts my pride, but also means that I have to take more metal off of the ways than the bare minimum I was hoping for. Since the relationship between the flat and inverted V ways is fixed, I will also need to scrape more material off of the V ways to keep the table from tilting in the fore to aft (Z axis). Note: The Z axis is the axis that moves parallel to the spindle axis and the spindle on this grinder is horizontal, so the Z is the horizontal and X is the vertical axis. This is the opposite from a vertical mill.

I had no sooner set the supports up under the grinder's table on my work bench when I happened to glance at the surface grinder. Wouldn't you know that the two Z axis V ways that support the saddle appeared to be positioned using the Airy formula. It certainly made sense that the designers would have wanted no sag in the table, but I needed to check it out myself. I grabbed a tape measure to check. The two V ways were as close to 10.45" from the ends as I could measure with my tape measure.The correct position for supporting the table was staring me in the face and I still didn't think of using it.

Week of March 15, 2008
This great revelation happened on Sunday the 9th. It's now Saturday the 15th. I've been scraping every afternoon this week for a few hours and I'm still not back to where I was in terms of groupings of high spots on last Sunday. I only had to lower the ends by a few ten-thousandths,but to do this I had to go back to using a "normal" spotting technique. This is where the straight edge is blued and layed on the ways so that the blue is transferred to the ways' high spots. I find that this method of spotting works much better for getting the ways flat. Once the ways are showing color over the whole surface, I switch to bluing the ways and using the straightedge to rub off the color on the high points, which lets me see the individual silver high spots easier on the blue background.

Scraping is a rather slow process, especially when it is being done by me. I have a bad habit of not taking deep enough cuts and have to remind myself to cut deeply when I have a lot of stock to remove. A lot of stock can be a half a thousandth of an inch if you are only removing a hundred-thousandth of an inch per cycle. Since I need to work on lowering the ends of the table's ways and my straight edge is shorter than the surface I am scraping, it is necessary to print and scrape the surface in shifts. As I said in the first installment, I print and scrape the center, then the left end, then the right. Because I need to lower the ends by a couple ten-thousandths, I'll use a shortcut to help me to remove metal more quickly. That shortcut is using a file to assist me in removing stock.


Using the normal spotting technique of blue on the straight edge and transferring it to the way, I continue to mark and scrape the bow out of the way. As can be seen by the even distribution of metal colored areas, the way is pretty flat. Now it's time to scrape for a better surface quality.

After I've scraped the center 36 inches, I now have about 5½ inches on each end of the ways that didn't get scraped in the first pass. I blue up the straight edge and set it on the ways with about 2 inches overhanging the end I plan to work on. Because the ends are high, the print shows high spots near the end and doesn't show any more high spots (very dark color) until about the center of the ways. I only scrape the end high spots. I don't touch the marks near the center. It does make sense not to scrape the center of the straight edge until the whole of the side being worked on shows high spots, but I always have the temptation to work on all high spots I see. After I scrape the high spots on the end, I grab my 15inch mill file with its handle removed. This file is slightly bowed lengthwise. I turn the convex side down and I lay it flat on the ways - well actually it isn't flat, it rocks a little. With downward pressure on the center of the file, I make a few cuts over the areas I have scraped. If you choose a file that has a nice bow to it, you will be able to remove metal from only the areas that have been scraped without cutting the surrounding areas. Since the teeth of the file act like multiple scrapers, a couple passes of the file cuts much more metal than a couple passes of the scraper. By only using the file on the area I have scraped (the high spots), the chance of removing too much metal or metal from the wrong area is lessened.

I now have high spots over the entire center 36 inches of the ways. I have the left 5½ inches scraped very close to level with the center 36 and the right end is still a tenth or so high. Tomorrow I'll return to working on the right side..


These shots are the prints taken after changing from spotting with blue on the straight edge to reverse spotting with blue on the ways.

I was outbid on a nice looking 90° inverted V scraping template on Ebay tonight. It was rather short at 20", but would have been nice for doing lathe saddles. I really liked that it was hollow and ribbed to keep the weight down and the rigidity up. I did win a matched pair of 36"cast iron straightedges a couple weeks back that are in pretty nice shape. They will need a bit of scraping to clean up a couple nicks and burrs, but when they're finished I will have a nice large set of straight edges for scraping that will also serve as parallels for machine setup. I have found that generally I prefer to spot from a granite template. This is mostly because of the way that granite hold sand transfers the Prussian blue, but also because there is less chance of a fleck of swarf doing damage to granite than to cast iron. However, the down-side of granite is the weight. The granite masters I own and have seen weigh more than a cast iron master of the same length.


Lost Auction: A nicely made V template.	The flat way is done and I'm starting the inverted V.

Week of March 23, 2008
I was able to spend a little more time scraping this week and have finally gotten the flat way of the table finished to about 15 points per square inch. Before I get started on scraping the inverted V ways of the table, I want to scrape one of the cast iron straight edges I purchased. I have 2 of these 36" two-sided straightedges. Both have one scraped and one ground side. The ground faces are pretty flat, but have lots of nicks and scratches that I have taken a stone to. The scraped surfaces also have nicks and scratches, but these sides will need to be rescraped.The reason I purchased the cast iron straight edges in addition to the granite one is that the outboard side of the inverted V doesn't have much room between it and the lip that surrounds the bottom of the table. Because of this lip, the granite straight edge won't fit in the small area. The 3' cast iron straight edges are shaped like an "I" beam as seen from the end view. The width of the finished surface (scraped or ground) is about 1½" and the column of the"I" is only an inch. The thinner "I" section will allow it to fit past the lip on the table so I can spot the outboard side of the way. I am looking forward to scraping the straight edge. Not only is it something different to scrape, but I am looking forward to checking the parallelism of the two faces of the straight edge with a couple of my Metrology tools. Namely the K&E 71-2022 autocollimator and my shop made copy of the Rahn Planekator with its Mahr Supramess test indicator.



My shop made copy of the Rahn Planekator allow me to read the distance between my Starrett "A" surface plate and straight edge. It shows the granite straight edge and surface plate are still pretty close to "dead flat." On the Supramess scale, each division is 0.5 µm(0.000019").

I originally made the Planekator copy to help me lap a surface plate I purchased that was in pretty poor shape. This 18X24" Metro plate surface plate was one of the few items I've purchased on Ebay that turned out to be a poor deal. However it did start me on learning how to resurface a surface plate. The plate was badly pitted. Some acidic substance like Coke/Pepsi had probably been spilled on it, so I needed to lower the whole surface by a few thousandths. I purchased a12" square, serrated lapping plate and scraped it as flat as I could get it. I then started with 500 mesh diamond powder and began re-working the surface. The shop made Planekator helped me to keep the surface flat as I removed material. I also made a copy of the Rahn Repeat-O-Meter to help me eliminate localized high or low spots. At this time the plate is back to about "B" quality specifications and I was using 3000 mesh diamond powder.to smooth the surface. B quality specs for a 18"X24" plate are not more than 0.0003" between the highest and lowest points on the surface (overall flatness, measured with the Planekator) and not more than 0.00011" between any 2 adjacent points (repeatability, measured with a repeat meter). This project has been set aside until I get the grinder done.

Week of April 6, 2008
Next on the agenda is to divide my time between scraping the inverted V ways of the grinder table using the ground cast iron straight edge as a template and to continue scraping the other 3'cast iron straight edge which will be used as a template when I get to the point of finish scraping the table's V ways. I guess I'd better get back to work. These things won't scrape themselves.

Grinder 1