What's new
What's new

Which cast iron straight edge type is better?

SAG 180

Titanium
Joined
Sep 17, 2007
Location
Cairns, Qld, Australia
Which is a better type of straight edge for scraping use: The type shaped like an I beam or the camel back type?. I was under the impression that the I beam type was more rigid and less prone to ageing distortion, where the camel back has been around for ages unchanged and is more....traditional?.

I'm planning on making some castings and could make patterns for either type including variations on height etc. if there's a technical advantage (like two flat surfaces on an I beam).
 
This one is kind of I-beamy but what I really like is that one side has a flat with sides at 90 degrees, and the other is flat with two different angles along the edge for doing dovetails.

DSCN0028.jpg


I doubt a properly made camel back will distort any more than an I beam style. The advantage is that in longer versions camel back will (should) be lighter.
 
Screwmachine,
how do you handle that straight edge?

The camel backs I've seen had wooden handles fitted, and the box section straight edges in Moore's book had lifting points at the ends. That one has me curious:confused:
 
Screwmachine,
how do you handle that straight edge?

The camel backs I've seen had wooden handles fitted, and the box section straight edges in Moore's book had lifting points at the ends. That one has me curious:confused:

Well, I haven't actually used it yet as it's awaiting a rescrape ;). My main straight edge is a normal camel back, just a little longer than this one at around 20". That one doesn't have wooden handles, and I just try to keep my hands on it as little as possible. If I'm feeling extra-precise I lift it with rags. I know from reading Forrest's posts over the years how important keeping your body heat away from references is; but I've seen at least a dozen different smallish straight edges here and none have had insulated lifting handles.
 
I don't know every nuance of their history and use, but I see them as almost two different functions. I don't think 'tradition' plays into it. I think the arch is the right shape to support the load with the minimum mass...I just bought norb's 60" straight edge and man o man, minimizing mass is important! The parallel on the other hand can be used to spot but it is heavier....being parallel though lets it be used for things other than spotting... you put a level on it for example. I haven't heard the distortion thing, a properly made one should be stress relieved and stable.
 
24" & under, I like the parallel I beam style. It has more uses, especially since either side is a parallel reference. It is, of course, much more work to scrape accurately than a "camel" back and tends to weigh more for a given width x length. The camel backs are probably at least if not more stable, and lighter for the same overall stiffness.

Over 36", the camelback tends to win on the weight issue. But parallel straight edges were made up to at least 8' and probably longer, for tasks where the second parallel side is useful.

smt
 
24" & under, I like the parallel I beam style. It has more uses, especially since either side is a parallel reference. It is, of course, much more work to scrape accurately than a "camel" back and tends to weigh more for a given width x length. The camel backs are probably at least if not more stable, and lighter for the same overall stiffness.

Over 36", the camelback tends to win on the weight issue. But parallel straight edges were made up to at least 8' and probably longer, for tasks where the second parallel side is useful.

smt

Ok, that makes good sense, I'd only be after 24" anyway as that's the largest surface grinder I have access to at the moment. If I surface grind it I can get the parallelism in the ballpark before scraping. What angles are dovetails commonly made to? 60 and 45 degrees?. If I make the height to length ratio the same as a camel back with diagonal ribs and cut outs, I should have a comparable straight edge for stiffness...I hope.
 
Ok, that makes good sense, I'd only be after 24" anyway as that's the largest surface grinder I have access to at the moment. I

you're just using the grinder to get things close to parallel and then scrape? i hear what you're saying, but I'd be making the straight to the length i need the straight ...not what the grinder can handle. You could step a longer one along the table of a vertical mill and flycut.... then the constraint becomes the diagonal on the surface plate.
 
Not much point to many "better than" arguements. Every scraping problem seems to have a factor or two that makes you think: if I had a cameback (or parallel, or longer or shrter etc) straight edge, life would be much better. It might make scraping seem to go a bit faster as you tussle with concepts like the Ideal Straight Edge but the next job will make you wish for a different feature set.

Personally I prefer light, stiff, and flat as primary characteristics in a scraped reference. I've doodled away many an hour sketching lighter, stiffer, more better straight edges. The notions I keep coming back to is a camel back structure that's hollow and can be partly filled with water/glycol, has built-in adjustable ten arc second level/plumb vials, leather insulators on every place it's convenient for handling, has center-of-gravity lifting points that allows full rotation in 3 axes, and so-on.

My most vialble notion revolves around a tubular steel fabrication with a segmented cast iron face. If my calculations mean anything, I think the weight of a 6 foot straight edge (normally around 120 Lb) can be reduced to about 60. If carbon fiber can be used somehow the weight can be reduced to maybe 30 lb dependng on how the segmented cast iron face is proportioned. All this is very temptng but expensive in time and material.

In the end we fall back on the eBay and Craig's List scraped tooling we can scrounge, doctor up, and recalibrate.

For the record you can scrape a straight edge twice as long as the diagonal of your largest reference flat but it's chancy. The longer it is the trickier to execute. You need an overhead crane and a low-rate spring that supports the full weght of the straight edgee as you take prints. If you have a differential electronic level or an optical collimator to track progress there's no practical limit to the length you can scrape flat and straight except patience and time.

To me, probably the best all around straight edge is that designed by Steven Thomas. It's just the right length (32"), very robust, and the reference face can be machinied to any angle set you wish or simply machined away to a standard flat face. That straight edge and a fitter's flat about 18" long is all you need for machine tools of a size that are tranportable intact.

Scraped reference toolng does not end with the Ideal Straight Edge. You aso need a couple of 10 arc second levels, a granite flat, scraped prisms and fitted angle blocks you cutom make for one application then re-make for the next, pins, dial indicating gizmos, and a host of other gewgaws and gadgets each of which will tell you one thing which i why you need many of these items in sequence and combination to tell you the full story.

If your ambition to scrape in your decrepit machine tool and you lust for a scraped straight edge, remember that a straight edge is only one hard-to-find tooling item you need in your arsenal.
 
Last edited:
Speaking of Stephen Thomas' straight edge castings, what is the current status of next availability. My last inquiry was answered with expected availability in Fall 2009, but I either missed the news or it didn't happen then. I've got a 32" grinder table I've been putting off until I acquire a longer straight edge. I can borrow a 48 inch beast, or try to cheat with my 24", but I won't do either if something 30" to 36" is foreseeable.
 
Forrest-

Thanks for the plug!

One clarification, these are aprox 30 -1/2" or a little longer, enough to trim to an honest 30" + (But not 32")

Cecil-

From about July or August on, I was strung along by the last foundry to cast a batch, then told they were going "dormant". I made arrangements with their sister foundry, who expected to cast them by early December so i could conceivably make Cabin Fever. They finally delivered in mid January. I promptly planed them and had them heat treated (thermal stress relieved" at Elmira Heat Treat.

Then I got swamped with Millwork (woodwork) my actual business. When the castings came back from HT, my (small shop) and the planer were in use for wood parts, plus I cannot run CI with a big wood job because the iron dust can contaminate the finish & cause weird stain effects. I also was too busy and machines were set up such that even with 3 table saws and and automatic beam saw in the shop, it was not practical to make crates. (In case you had not suspected, I'm a one-horse show)

However, I am sending the last of the wood work off by Saturday, and will be selling SE's the next week.

Rough planing straight edges:
smt_Whitcombblaisdell27.jpg

smt_Whitcombblaisdell28.jpg

smt_Whitcombblaisdell30.jpg


Planing wood parts (reeded columns) :D
smt_Whitcombblaisdell31.jpg

smt_Whitcombblaisdell32.jpg


The other factor is I made a new set of patterns to cast a "lightweight" version. I'm not sure how to go about testing? Send them out and let "caveat emptor???" Suggestions?

Here are some defective castings that were sectioned to show the difference

smt_straightedgecastings18.jpg


The one on the right (original version/last batch) is about 8 - 10 lbs heavier than the one on the left. (depending on width). The one on the right will accomodate angles through 45. The one on the left (lightweight) can accept a 45° with careful set up. But it was designed for minimum weight, and 50 or up would be better. 50° is the typical BP dovetail. There really aren't many 45° DT's out there on sizes larger than cross slides.

smt_straightedgecastings19.jpg


Same sectioned casting, showing cored out area. Unlike the heavies, these don't provide for a wide 90° angle on the back side. The dovetail is maybe a 1/16 - 3/32" wider on the lightweights, though.

Anyway, how much risk of the webs "telegraphing" through the bases on these over time? I certainly can't afford to send them out and then pay freight to redeem them if people aren't satisfied. & I don't see much time this year to finish one for myself and do extended testing.

Subject to suggestions on the lightweights, I'll post a separate FS ad by Saturday with description & pricing.

Brian/Versamil, I don't think our communications system is working well, I have left several email messages including options since Jan.

smt
 
I have an 18" CI camel-back that I still have to scrape in. It has the flat bottom and the 50° bevel.

I also have a 24" black granite Rahn straight edge with no bevel. It is a camel-back though not a smooth curve and has holes for handles. Damn heavy.

I also have two 12" DoAll black granite parallels "master" grade. They are especially handy for use on a surface plate.

Naturally one needs to be careful how any straight edge is handled but a CI camel-back is easy to keep a grip on. The granites are much more fragile.

-DU-
 
Stephen,

I'd be happy to buy a hollow one, scrape it in and monitor it for stability. I am just a hobby scraper hand and there is rarely time pressure with my rebuild endeavors.

I'll see Brian at Versamil tomorrow and let him know the news.
 
This is all really good information guys, my main reason for making them here in Australia is the shipping and customs will more than double the landed price and I have access to non ferrous casting facilities anyway (Bronze, brass and aluminium). I understand if you can handle bronze then iron is as easy to cast.

I like Forrest's idea of fabricating a straight edge with a steel back although I wonder about how it will perform afterwards with regards to flexing and resonances.
 
I like Forrest's idea of fabricating a straight edge with a steel back although I wonder about how it will perform afterward with regards to flexing and resonances.

The real issue with any straight edge made from dissimilar materials is temperature distortion. Steel and CI have different thermal expansion coefficients. The straight edge would only be flat at the temperature at which it was scraped. Above or below that temperature you would get shear stress at the material boundary.

One way around this is to build a symmetric parallel. The cross section must be balanced across the neutral axis. If an orthotropic material such as honeycomb aluminum core is sandwiched between CI slabs the core will have negligible shear effect on the CI.
 
I have a steel 18" parallel with fairly wide and somewhat thin flanges (compared to typical iron parallels), which is a pleasure to use due to light weight. It is also very accurate and consistent. Scraping it flat took more time than an iron straightedge - steel is not easy to scrape!

The only drawback, which goes back to Alpaca's orig. comment to Screwmachine, is that it is very sensitive to body heat. Far more sensitive than an iron straightedge. When I handle it I put a pair of insulated gloves on.
 
My post referred to a "segmented" cast iron edge. Break up the reference surface into short plates having limited linear compliance and the difference in coefficient of linear expansion disappears.

In technical discussions like this one must assume that the writer may bave omitted details and refinements in the interest of brevity. There is a difference between steel and cast iron but the two terms represent broad categories whose coefficients of linear expansion vary and overlap to some extent. There are a few that coincide: 1040 a commonly available medium carbon steel available in tube has a coefficient of linear expansion of 6.3E-6"/degree F nearly identical to grey iron. Several malleable iron formulations are close to A36 steel.

Besides where is it written that only cast iron is suited for reference faces. Why can't the proposed straight edge be made of a homogenous material like structural steel? Steel may be scraped with greater labor and has the disadvantage of lower durability and a tendancy to dent. The payoff is a far lighter less temperature sensitive scraped straight edge.

Then there is stability. Fabricaton techniques and weld technlogy has advanced in the last 40 years. Stability is further inhanced if the hollow within a fabricated straight edge is partly filled with water/glycol which may be sloshed to distribute heat uniformly throughout the tool. A well designed and skillfully welded steel structure can be made much lighter and possibly stiffer than a casting if price is a secondary consideration. Careful metallurgy, post-weld heat-treatment, and stress relief will make the finished fabrication at least as stable as an iron casting.

There are trade-offs in any departure from tradition. If this means reliable scraped reference tooling almost immune to localised heat input that are half the weight of a comparable iron casting I would certainly give the matter serious thought.
 
Forrest....I must admit I had visions of making a segmented straight edge out of segments of barbecue plates or manhole covers brazed to a 1040 truss frame. :cheers:
 








 
Back
Top