What's new
What's new

Machining and scraping a Denis Foster’s 36” Featherweight Straightedge at Tuckahoe

Paolo_MD

Stainless
Joined
Apr 6, 2013
Location
Damascus, MD
Denis Foster (dgfoster) has kindly donated a casting of one of his 36” Featherweight straightedges to the Machine Shop Museum of the Tuckahoe Steam and Gas Association for us to machine, scrape, and raffle it.

Details about the raffle can be found in this thread.

Here we will document the whole process of machining and scraping it. Be patient, since we are a voluntary organization and, right now, there are a lot of things going on at the Museum. But, our goal is to have the straightedge ready before the end of the raffle (Saturday, November 3) and we plan to start working on it this week, during the annual show.

The straightedge was delivered to my home boxed in a sturdy wooden box wrapped in cardboard, which was glued together with hot glue.

Here are a few pictures of how it looks like, right out of the box (as usual, click on any picture to display a larger version of the same).



























We have already set it up on our 1920s Rockford planer, trying to introduce as little distortion as possible.

Following an advise from Richard King, we will machine it by down-feeding the head, since the way of the platen (=table) have more wear, resulting in the platen moving up and down while traversing.

The setup consists in clamping down the vertical bars against aluminum blocks that provide the proper clearance between the sole and the table.

There is one peg stopping the straightedge in the direction of the cut and three pegs with side screws pushing perpendicularly to the cutting direction. Only a couple of shims were used to position the sole perfectly parallel to the table and, if I recall correctly, all it took was ~0.025”.







The tool you see in the toolholder is just a 3/8” HSS blank that will be replaced by a round-nose tool and was there only to verify the distances during the setup.

The cutting head was adjusted at an angle very close to vertical to allow clearing most of the bottom by removing the least material. Again: the raw casting is essentially resting on its vertical bars. Their surface is not guaranteed, or even expected to be perfectly perpendicular to the sole and we are talking at most about less than 2°.







The big Tuckahoe show will start tomorrow (here you can find the program and more details) and we will be rather busy. Therefore, I cannot guarantee that I will update this thread in a timely manner.

I'm also starting another thread dedicated to the show in the Antique Machinery and History forum.

Paolo
 
Quite neat, Paolo,

Are you machining the "top and the side (as it sits), then flipping it over and doing the other "top", so you get the two sides parallel and perpendicular to the bottom when in the normal position?

Good luck with the show.

- Leigh
 
Thanks, Leigh.
Yes, we will be machining the bottom, rotate the cutting head to machine the dovetail side at roughly 40-45° in a single setup. Then we will flip it over for the other side. Functionally, it wouldn't need to be with the edge parallel to the other edge, but we will keep it reasonably close, so that it will also look better.

Paolo
 
Quick report

This evening I've managed to take the first couple of passes, removing most of the skin from the bottom (apologies for the bad tint of the pictures, I'll fix it when I have more time).





Following Denis' suggestion, before starting machining, I changed the setup so that the straightedge would rest exactly on three points, replacing the U-shaped hold-down with a regular one and a pile of washers under the camelback rib.
ll the rest of the setup remained the same. This is the first time I'm using pegs with side screws and I really like them. After the first pass I noticed that it was cleaning more toward one end than the other: paused briefly and adjusted the screw where it was not cutting and took up the slack in the middle one. No need to release the hold-downs for this simple type of adjustments.



That's everything, for now.

Paolo
 
Yesterday we managed to finish machining the bottom and the 45° side of the straightedge. I do have mix feeling about the choice of using a single setup for both sides: it worked great for the show since I couldn't take deep cuts and got to run the planer longer, rather than spending time in the setup.






The only problem encountered so far is that that the corners of the casting are a bit hard and it worked much better to switch to a carbide tool for all the roughing and use the round-nose HSS tool just for the finishing passes (not visible in these picture, I did place stacks of washers between the bed and the straightedge to contrast the downward cutting forces - they will be visible in the videos I'll upload next week)



Now the casting is already set up for machining the other side. If the rotary phase converter is powerful enough to start the Dill Slotter that you can see next to the planer, we plan to clean-up the ends with the slotter.



The weather forecast call for good weather and max 83° for both today and tomorrow. Probably, we will set up a hands-on scraping demo outside the museum and that could take care of the rough-scraping of the straightedge.

Paolo
 
Small suggestion- if you still have it set up to do the 45- machine a small flat where the 45 meets the bottom. Imho, easier to scrape and more durable. Ie, \_|

I'm really enjoying your pics!

Lucky7
 
Thanks, Lucky7 for your suggestion. It is indeed what I did: I hate scraping knife edges. ;-) The flat is very small, but enough to make the edge blunt.
It is also true that it is easier to create a flat between the 45° and the bottom starting from a sharp edge, than the other way around.

My apologies, but tonight I do not have enough steam left to start uploading pictures. But I will anyhow provide an update and clarify better what I posted last night while already half asleep.

Let's start with some explanation: we machined the straightedge on a 99 years young planer, which does a relatively nice job, but has still a few limitations partially due to the design and partially due to the age (read wear). The cutting heads of planers and shapers are pulled down by gravity: unless you have the head virtually locked, the leadscrew has the function of pulling against gravity and not pushing the tool into the cut. That is the reason why planer and shaper tools have significantly more pronounced rake angle than comparable lathe tools, since they need to drag themselves into the cut (with the help of part of the weight of the cutting head, of course), instead of being pushed into it.
Of course, the more pronounced is the rake angle, the more fragile is the cutting edge and the more frequently you need to resharpen your tool. The round-nose tools I ground for this job were somehow a compromise between the optimal rake and durability.
Unfortunately, this makes these tools behaving differently with material of different hardness and tend to try to float above the surface if the material is too hard.
In the casting the four corners were slightly thinner than the rest (we're talking about ~20-30 thousands, the casting overall was impressively straight, compare to others I have seen elsewhere). Therefore, I think two things have happened: the corners of a rather thin casting like this would cool down faster than the rest. At the same time, being the corners a bit thinner, my machining resulted in removing just the skin and little material below.
The material in the corners of the machined straightedge is significantly harder than the one toward the center, but it is still scrapeable. A ridge next to the vertical edge is harder than the surrounding, but softer than the corners.

If I had milled the straigtedge with a carbide face-mill, instead of planing, I would have likely never realized of such differences in hardness.

Bottom line, the cutting tool and cutting head deflected enough to create a 0.005" ridge near the side and ~0.010 bumps in the four corners.

In order to spare one of the round-nose tools for the finishing passes, we decided to use carbide tooling for roughing of the two sides of the casting, followed by the round-nose tool for only the finishing passes.

The current status is that the machining is done and We've starting scraping down the "ultra-high areas". I'm almost done in flattening such spots in half of the length of the straightedge, I have sped up the operation by doing "precision roughing" with the Biax set for ~1/16" stroke length.

After machining, the Featherweight weighs just over 25lb.

Paolo
 
Paolo remember you can grind the Biax Carbide to a 2 degree negative and it cuts better then the recommended 5 degree's on soft material. Be sure to use a 60 radius to. I would also lengthen the stoke to 1/4".

Being the worldwide instructor for the BIAX factory we only advise customers to set the scraper to 1/16" when your scraping 40 PPI. While your scraping it and you start to get a 5 PPI print take a photo. Note the hinge and PPI please hang it up on a rope and ring it the way I instructed you in our class. Then clean the surface and recheck the PPI and print, then take a photo and share it with us.
 
I need to apologize again, but also tonight I'll give only a brief report: After removing the ultra-high areas by both drawing a map with test indicator and Sharpie and by spotting with my HKA-18 using copious Canode ink, today I blind-scraped the casting uniformly with the Biax till I removed most of the machining marks.



NEX_6_DSC00816.jpg


Then, still with copious ink, I did spot it with my 36" granite straightedge, making five independent prints: four each covering one of the four corners, and the last spotting in the middle of the casing.












All the prints are fairly in agreement, revealing that the straightedge is higher on both ends, but without ultra-high spots anymore.

I have brought it back home with me, but I'm not sure when I will have any time for scraping it. However, during the week I should find a few minutes to go through the pictures and editing the videos.

Richard: yes, I know it would be an option. But, yesterday and today at the show, regrinding the blades wasn't an option.

Thanks!

Paolo
 
Thanks, Lucky7 for your suggestion. It is indeed what I did: I hate scraping knife edges. ;-) The flat is very small, but enough to make the edge blunt.
It is also true that it is easier to create a flat between the 45° and the bottom starting from a sharp edge, than the other way around.

My apologies, but tonight I do not have enough steam left to start uploading pictures. But I will anyhow provide an update and clarify better what I posted last night while already half asleep.

Let's start with some explanation: we machined the straightedge on a 99 years young planer, which does a relatively nice job, but has still a few limitations partially due to the design and partially due to the age (read wear). The cutting heads of planers and shapers are pulled down by gravity: unless you have the head virtually locked, the leadscrew has the function of pulling against gravity and not pushing the tool into the cut. That is the reason why planer and shaper tools have significantly more pronounced rake angle than comparable lathe tools, since they need to drag themselves into the cut (with the help of part of the weight of the cutting head, of course), instead of being pushed into it.
Of course, the more pronounced is the rake angle, the more fragile is the cutting edge and the more frequently you need to resharpen your tool. The round-nose tools I ground for this job were somehow a compromise between the optimal rake and durability.
Unfortunately, this makes these tools behaving differently with material of different hardness and tend to try to float above the surface if the material is too hard.
In the casting the four corners were slightly thinner than the rest (we're talking about ~20-30 thousands, the casting overall was impressively straight, compare to others I have seen elsewhere). Therefore, I think two things have happened: the corners of a rather thin casting like this would cool down faster than the rest. At the same time, being the corners a bit thinner, my machining resulted in removing just the skin and little material below.
The material in the corners of the machined straightedge is significantly harder than the one toward the center, but it is still scrapeable. A ridge next to the vertical edge is harder than the surrounding, but softer than the corners.


If I had milled the straigtedge with a carbide face-mill, instead of planing, I would have likely never realized of such differences in hardness.

Bottom line, the cutting tool and cutting head deflected enough to create a 0.005" ridge near the side and ~0.010 bumps in the four corners.

In order to spare one of the round-nose tools for the finishing passes, we decided to use carbide tooling for roughing of the two sides of the casting, followed by the round-nose tool for only the finishing passes.

The current status is that the machining is done and We've starting scraping down the "ultra-high areas". I'm almost done in flattening such spots in half of the length of the straightedge, I have sped up the operation by doing "precision roughing" with the Biax set for ~1/16" stroke length.

After machining, the Featherweight weighs just over 25lb.

Paolo

Paolo, I think the likelihood is that, as you surmised, you have really not gotten completely through the skin of the casting. It is more or less the equivalent of the moderately hard skin that is commonly encountered on hot rolled steel. I have not noticed the problem you did likely because I do use a carbide face mill on a 35 ton horizontal mill and I take a 20 or 30 thou initial cut and usually remove an eighth of an inch or and usually significantly more before doing final passes. That is why my finished SE's usually weigh in close to 21 pounds or so. For each 1/8" removed from the sole you take off about 2.5 pounds of iron. Side note: I designed the SE with what I thought was an allowance of a solid extra 1/8" of iron in the casting to provide plenty of meat for truing up the casting and getting into nice buttery cast iron for scraping. But, many people seem to prefer to remove the absolute minimum material to allow clean up of the casting and then begin scraping at that point. Since each rescrape of the casting would be removing only a thou or so of material in most cases, I guess they expect to live long long time! :)

The cause of the skin on cast iron castings most likely (cannot claim certainty on this) is the initial freezing of the iron as it first comes in contact with the mold sand. The sand is initially very cold relative to the molten iron and causes immediate freezing of a very thin shell of iron. As more iron flows in and time passes (fraction of a second) the sand starts to warm and freezing slows dramatically. There are numerous dramatizations of the freezing sequence of castings available on the web for those interested.

I need to apologize again, but also tonight I'll give only a brief report: After removing the ultra-high areas by both drawing a map with test indicator and Sharpie and by spotting with my HKA-18 using copious Canode ink, today I blind-scraped the casting uniformly with the Biax till I removed most of the machining marks.



NEX_6_DSC00816.jpg


Then, still with copious ink, I did spot it with my 36" granite straightedge, making five independent prints: four each covering one of the four corners, and the last spotting in the middle of the casing.












All the prints are fairly in agreement, revealing that the straightedge is higher on both ends, but without ultra-high spots anymore.

I have brought it back home with me, but I'm not sure when I will have any time for scraping it. However, during the week I should find a few minutes to go through the pictures and editing the videos.

Richard: yes, I know it would be an option. But, yesterday and today at the show, regrinding the blades wasn't an option.

Thanks!

Paolo

Wow, Paolo, you are making rapid progress. Looking good! Thank you for your very thorough write up and pics. Such a complete report requires considerable work in addition to all the work of planing, scraping etc. We all owe you a big thank you. It sounds like you put in an exhausting long weekend at volunteering at Tuckahoe.

Denis
 
Besides the scraping it's nice to see some pics of a belt-reversing planer with the relative sheave sizes giving a graphic sense of how the cutting and faster retract stroke times are managed.
 
You would think the planer would plane much straighter than that. I have a hard time seeing your cutting tool digging in then easing up at the end of the cut and beginning of the cut. Was your granite straight edge checked straightness. It's acting like the middle of the granite straight edge is "dished" from wear.
 
You would think the planer would plane much straighter than that. I have a hard time seeing your cutting tool digging in then easing up at the end of the cut and beginning of the cut. Was your granite straight edge checked straightness. It's acting like the middle of the granite straight edge is "dished" from wear.

The granite straightedge is perfect. The problem was mainly my setup: I wish I had more variety of tooling to clamp-down the straightedge, since the position and distance of the T-slots and peg-holes aren't very helpful for somebody incompetent like me for setting up a straightedge.

Although I did add to my initial setup two nice toe-clamps to the ends of the curved back, I believe that the casting flexed a bit during the cut and, in a way or in another managed to rock a few thousands. However, I was mostly distracted by the "hard bumps" in the corners (yes, there, everything moved leaving those high spots). Moreover, the bluing was done after a few passes of hand and power-scraping, while holding the straightedge with the WorkMate bench, something not totally rigid, especially when I was scraping near the ends. Personally, I'm not surprised at all that I'm low in the middle, where I could assess heavier scraping marks.
In all honesty, I'm not too disappointed and I rather prefer the casting to be low in the middle and bring it into full contact that way, so that I avoid any false reading due to rolling a convex item.

Paolo
 
Funnily enough I got the same pattern on my 1m long parallel flange straight edges off of the planer, when I planed it with the bottom face (sole) facing up. At the time I put it down to wear in the table/bed as, while I'd levelled the machine quite accurately, I hadn't made a light cleanup cut over the table. I know the previous owner had done but I don't know how well he'd levelled the machine first.

You could also see the tooth pattern from the drive in the blueing but it wasn't visible by eye or touch - probably only a couple tenths of a thousandth in it but definitely there. Straight tooth rack & gear drive.

PDW
 
For everyone's info I found this. Not knowing the exact size I said .062 x 3" x 36" and it said 1.7299 pounds. so if Paolo takes off another .125" then it should come down to approx. 22 pounds. I don't know if Dennis sends along instructions on how to machine his SE. I never have, but have been asked to. I assume people know how, but it might be worth sending this calculator and some instruction or links to everyone's pictures.

Metal Weight Calculator – Boston Centerless
 
See post 10 above concerning weight removed vs depth of cut.

I do do not tell people how to machine their straight edge as I have found that is a matter of personal taste. ( I assume most people know how to calculate the weight of metal removed based on the density of iron. I also have mentioned the weight they could expect to remove based on thickness milled off in several prior threads and have shown the calculations.)

When I do the machining, I ask them what target weight they would like and machine as little or much away to achieve that goal.

Because I am able to use a very large and tight horizontal milling machine, I can get flatness of less than .001" reliably though I quote a wider tolerance just to give myself leeway and figure if there is a small area higher than that, it can be scraped off very quickly anyway. I do take considerable pains to hold the SE in such a way as to not induce deflection and wait until the casting has cooled fully before taking final cuts. I put DTI's at each end of the casting as I clamp down to watch for fexion as clamping force is applied and leave them there to monitor for movement during the cut Also put a third one mid-bow during cutting.


Denis
 
All the prints are fairly in agreement, revealing that the straightedge is higher on both ends, but without ultra-high spots anymore.

I have brought it back home with me, but I'm not sure when I will have any time for scraping it. However, during the week I should find a few minutes to go through the pictures and editing the videos.

Richard: yes, I know it would be an option. But, yesterday and today at the show, regrinding the blades wasn't an option.

Thanks!

Paolo

Just a note: since I am fairly familiar with your 36" granite SE, I think it is very slightly convex, the reason I am thinking that t it is an almost perfect match to my 36x24 plate which is slightly concave to 1 or 2 microns. I know your 36x24 plate is slightly convex. The granite SE should almost spin on your surface plate. I know you will use multiple masters to check it out. What I am insinuating here, I would not solely rely on the 36" granite SE, even if you think it is a higher grade reference. But hey if you scrape it to 2 microns over the whole length, you did better than I would have.

edit: From the prints, the CI SE is a lot more concave than the granite SE is convex. Just wanted to point out the issue of reliance on one reference. But Paolo has a 6' granite SE as well or is it 5';). You will be able to trust that SE as a reference when he is done with it.

dee
;-D
 
Last edited:
Maybe machined straight edges tend to show high on the ends is because they have less reinforcement there so the tool pressure tends to push them away more there thus resulting in less material removed.
 
Maybe machined straight edges tend to show high on the ends is because they have less reinforcement there so the tool pressure tends to push them away more there thus resulting in less material removed.

Brian,
I think you've hit the nail on the head.
Here is a sketch of my original flawed setup highlighting the fixtures that were supposed to hold the straightedge against the cut pressures:

Machining_Setup_Flaw.jpeg

I realized pretty soon that there was a problem and I did address it by adding two toe-clamps at the ends.

Machining_Setup_Fixed.jpeg


I believe that the residual deviation from straight is mostly due to deflection of the casting and the machine.

Looking at the setups posted by Richard in this thread, I really like the use of a square tube that pushes the straightedge into the cut and has also a couple of hold-down clamps used to pull the straightedge against the very box.

As I've commented elsewhere, my challenge is to think about the setup 100 miles away from the machine, without a clear recollection of what it is available in loco (e.g. I completely forgotten that we had the toe-clamps). Then, getting to the machine after having planned the setup for a week or two, I got blind to the obvious.

Paolo
 








 
Back
Top