Page 4 of 7 FirstFirst 1234567 LastLast
Results 46 to 60 of 101

Thread: saw plates old VS new

  1. Quote Originally Posted by David Weaver View Post
    I think if the discussions go on long enough, someone will come forward, though we're only going to be able to do it on the big saws with enough plate thickness (that's OK). My friend changed jobs and I lost my ability to hand things to him to take to the lab at his work. I never even thought about having saws tested back then, we were interested only in chisels and plane irons, but interested enough even to strike japanese chisels.
    It would be nice if we could. I've got a Beverly #3 shear that cuts saw plates like they're made of paper. I could cut up the blades and re-mail pieces to a number of different people to test.

  2. #47
    Join Date
    Feb 2003
    Location
    San Antonio, TX
    Posts
    530
    I checked with my buddies in engineering and have access to a Newage ME-2RDB Rockwell hardness tester. This is the sister instrument to the Versitron that George used (same test, a little less automated).

    Just so everyone understands how Rockwell hardness testing works, I'll run through my understanding of the process.

    Rockwell hardness testing works by measuring indentation of a material by an indenter which depends on the scale and can be a hardened ball or a diamond.
    The indenter is preloaded with a specific force on the sample and is allowed to come to equilibrium. Then a test load is applied that idents the material under test. The load then returns to the preload value. The difference between the zero point with the initial preload and the depth after returning to the preload value is inversely proportional to the hardness. (harder will have less penetration).

    There are some measurement issues with this depending on thickness and how close to an edge you may get. Since the indentation deforms the material, the convention is that you need a sample at least 10 times as thick as the indentation. This means that for thin material it must be harder than a particular value for the test to be accurate regarding the scale. It may be acceptable for direct comparisons between materials of the same thickness, but relating to bulk scale can be problematic. One reference I saw gave the thought experiment of measuring the hardness of a cigarette paper (hey it was an older reference), you would see mostly the hardness of the anvil in the instrument. This may be where some of the issues with the discussion of Rob's data lies. Some of the fancier (i.e. Versitron) machines can interconvert between scales and use appropriate loading based on thickness. If the loading is excessive for the thickness it might be measuring the composite of the sample and the anvil. The other issue is measuring near the edge, since the material may deform and flow under load. There are some conventions as to how far from the edge is appropriate.(I need to look those back up)

    As someone who teaches how to make measurements (granted in a chemical setting rather than mechanical) we also need to be careful about our sampling strategy. Are we picking a representative sample that is typical and relevant. I understand the desire to do the testing under a handle where it may not be seen, but I imagine that the blades were probably held by that area when heat treating, so the thermal history may be different than along the tooth line.

    I guess the one question I have is what is the range of plate thicknesses are we talking about?

    I will try to make some arrangements to go and check out the instrument, and perhaps I'll grab a selection of plane blades to test to make sure I know what I am doing without worrying about thickness. And then I'll have a better feel for what I can test.

    John

  3. #48
    Join Date
    Mar 2004
    Location
    DuBois, PA
    Posts
    1,897
    My suggestion would be comparison testing against a known standard, with the standard shipped from one tester to the next, along with the samples. As far as differences in hardness, along the saw plate, what will affect readings will be what has been done to the saw plate after shearing from coil stock, for example, was the plate taper ground? Was the plate sanded by one of the "boutique" saw makers to approximate taper grinding of old? Is everyone here aware of tolerance ranges acceptable for hardness readings? (in my industry-powdered metal, apparent hardness on the Rc scale is +/- 5 points; I cannot answer tolerance range for particle hardness tolerances, though the numbers should be readily available in say, a volume of "Machinery Handbook").

    I have no desire to get involved in any testing unless variables are identified in advance and minimized. Additionally, what is the point of the tests?
    If the thunder don't get you, the lightning will.

  4. #49
    Join Date
    Jan 2009
    Location
    Williamsburg,Va.
    Posts
    12,402
    I wonder what the point is too. I have done it for so long that I can give a pretty accurate guess at the RC of a saw by filing with a new,fine cut quality file. I got good at picking out the most durable antique bitted plane irons also,from many years of using them on a daily basis.

    I understand 1070 spring steel is available from India,at 52 RC,while the 1095 is more like 48 RC. Why this is,I don't know. Hardness by itself is not the end all and be all,though. 1070 steel at the higher hardness will not have the carbon content to have particularly good wear resistance. The best combination would be 1095 at the formerly usual 52 RC.

    The same sort of thing applies to chisels. Vanadium content steel chisels can be hard,but some do not like them.

    So,all these tests may or may not have real relevance. What is needed also is the chemical analysis of the steel involved. It would not necessarily be of value,with the mix of spring steels now on the market,for a custom maker to even guarantee the hardness of his saws. I would not care for the 52 RC 1070 steel,for example. But,that might be what you were getting. The country of origin would also have to be guaranteed. Then,who would prove it?

    Some good spring steel can still be had from Austria if you pay the higher price for it. If I wanted to get back into saw making,I'd opt for the Austrian product. Or,an American product if available. I haven't bought spring steel for a long time,since we had to buy a high minimum purchase in the first place,we had a good stock pile. I had to buy the .042" steel 12 1/2" wide to get enough width to make proper Kenyon style crosscut and rip saws. We paid a cut off fee for every gauge also. I couldn't get 12 1/2" wide .042 spring steel in the available small packages of 6" wide blue shim material.

    John,My Versitron hardness tester was a manual one,not the latest model by any means. I would pull the handle forward to indent the diamond into the steel,then let up on the handle till the needle stabilized at the RC number the steel represented.

    Being in a museum,I had access to Federal surplus equipment. I got this valuable tester for $50.00 because they did not know what it was,and were too lazy to research it. It was the best surplus bargain I ever made.

    On the other hand,they would try to sell WELDED bandsaw blades for just about retail,when you didn't even know their length,or in many cases,exactly what the blades were made of:Welded on HSS teeth,for example. They let a couple of 10' tall stacks of white pine shelving wood sit outside until it,over the years,turned into tired little heaps of brown rot. They would put things outside if they didn't get the price they wanted in a certain length of time,and just let them rust away to nothing. Would they REDUCE the price? NO. They'd rather let the stuff rot.
    Last edited by george wilson; 01-15-2015 at 8:50 PM.

  5. #50
    I've got two plates that can be struck. One is a 1900 or so (perhaps a little before) d8 that has been sharpened its share.

    The other is a #7 (shown with no handle, I have the handle and thus know what it is) that is dark, but both are fresh enough to be struck for sure.. The 7 has not been sharpened too much, but it's an old saw that I purchased years ago for $20 on ebay. It was the very first saw I sharpened almost a decade ago (before I knew what I was supposed to do) and the teeth would need major reshaping just for me to sell it for a pittance, and I've go no interest in that. It is perfect for this purpose, because it's been filed (I don't think the bottom one has by me, but it's clear it has by plenty of other people, and it's got no broken teeth).

    P1080053.jpg

    I don't know what the point is too much other than getting relevant readings at the teeth of the saw. I think we'll be able to do that with this exercise, with proven era saws (not too late, not too early) and with no financial risk.

    I can joint the teeth off of these plates before I send them.
    Last edited by David Weaver; 01-15-2015 at 8:39 PM.

  6. #51
    Join Date
    Nov 2007
    Location
    NW Indiana
    Posts
    3,078
    WOW...there is a whole lot going on here. If I were going to do a study like being suggested, I would do some things first.

    1. One would be to make absolutely certain that everyone who is interested is comfortable with where the testing is going to be done. To figure this out, one needs to be pretty certain about exactly how the saws were produced and the thermo-mechanical processing of the saws. You want something as close as possible to the teeth and yet not be in the area where there may be uneven work hardening from the straightening or tooth setting or anything else. (in reality, does it really make any difference what the hardness is except on the actual tooth of the saw.)
    2. As mentioned, you need to make certain that the surface condition of all the samples is the same.
    3. I would want a chemical analysis of each saw being tested. Yes, I know that they are supposedly 1095 but the range in chemistry for this grade is quite large...Carbon 0.90 to 1.04%, Mn 0.30 to 0.50%, the level of silicon is not specified. This range of chemistry could make a difference in hardness. The ability to run accurate chemical analysis 50-100 years ago was not all that good which means that the steel maker was doing the best he could but there was a lot of variation of chemistry.
    4. If you test newer saw blades, do you even know what the grade of steel is? If I was looking to make saw blades today, I would take advantage of some different grades of steel to get better properties.


    To do this right, there is a large investment needed in time and resources. It is not a simple thing. Is all of this worth it other than a point of interest. As I reread this thread and the other one, I find myself coming back to what George Wilson and others said is that they could tell a good blade by the way it filed. Not very scientific but probably the most useful comment in the entire thread.

  7. #52
    Join Date
    Jan 2009
    Location
    Williamsburg,Va.
    Posts
    12,402
    Not very scientific,true,but the human brain is still the most sophisticated computer out there. Provided it has a proclivity for the work involved,and gets enough practice and exposure to develop good sensory judgements. That was a mouth full!! The proclivity part is very important. I have seen craftsmen with huge amounts of years under their belts who were still not particularly good. Some never could learn to design well in a lifetime or two. Even if he had a very refined English accent (a commodity too highly thought of amongst the curatorial types in the museum.)

    And,then there is another vital ingredient: The passion to do something with your talents. But,that is getting into a whole other field. I have seen some young people with tremendous potential who just did nothing but get into trouble,do drugs,sell drugs,steal things,and likely who would have ended up in jail once they were on their own. Or some others I TRAINED (fortunately not too many),who drifted into mundane jobs just for the security of them,forgetting about talent I saw that they had. But,the smartest one I ever trained did that. Ending up as a civil servant,interviewing people for jobs. Waste,waste,waste. I hate to see such waste.
    Last edited by george wilson; 01-15-2015 at 9:25 PM.

  8. Well, let the debate begin.

    At first thought I would suggest that the plates be cleaned if not polished on both sides by an agreed method. This can be all or most of the surface or in patches. Chemical, mechanical or abrasive cleaning or no cleaning at all? I suggest this be done by one person, somebody who has expertise in preparing materials of this type for the testing we envisage. Areas unsuitable for testing should be determined and clearly marked so that they are not tested.
    Then I suggest that we agree on selections of regions to be tested - heel to toe at three or four distances from the tooth line for starters.
    Whatever the matrix size and dimension I then suggest that the plates be ruled with some kind of non defoming marking system, perhaps a paint marker.
    Depending on how many participate in the testing we should decide on how many replicates each tester should do at each testing region across the saw plates.
    A test block or two to be sent along with the saws is a good idea.
    Each tester should perform an agreed number of replicate measurements of both the test block(s) and each coordinate of the plate assigned for their testing.
    Somebody needs to collect the data from all of the testers, compile, analyze and report it. It would be best if at least two independent people could do this step to minimize the chances for Excel errors and so on.
    Then the results should be posted here.
    What do you think? Too complicated, too simple or have I overlooked some things?
    Last edited by Rob Streeper; 01-15-2015 at 9:40 PM.

  9. #54
    I'd say you strike them first, I'd be glad to tape spots about an inch above the tooth line.

    I can sand the surfaces clean (no burnishing will occur, no hardening, etc).

    Just these first two saws would get your started without having to buy a junk lot. They are pretty much all I have that's disposable. Ebay may be of help to add another one later.

    I don't think we need to know anything about the composition of the vintage saws, both of the ones i'm showing above are first quality models and not second line saws. I've put a file to one, and someone else to the other a LOT, so there's no reason to believe these two are defective.

    Preliminary results should dictate whether or not anything additinoal is necessary, especially given that your 1095 strikes have been very consistent and in line with 1095 specs.

  10. #55
    Quote Originally Posted by David Weaver View Post
    I'd say you strike them first, I'd be glad to tape spots about an inch above the tooth line.

    I can sand the surfaces clean (no burnishing will occur, no hardening, etc).

    Just these first two saws would get your started without having to buy a junk lot. They are pretty much all I have that's disposable. Ebay may be of help to add another one later.

    I don't think we need to know anything about the composition of the vintage saws, both of the ones i'm showing above are first quality models and not second line saws. I've put a file to one, and someone else to the other a LOT, so there's no reason to believe these two are defective.

    Preliminary results should dictate whether or not anything additinoal is necessary, especially given that your 1095 strikes have been very consistent and in line with 1095 specs.
    I think sanding or lapping is preferable at this point to electrolysis, acid cleaning or other methods. Given that the plates may be taper ground we need to carefully select testing areas too.

  11. #56
    Maybe do tests higher up the plate too? We don't know exactly what Disston did with their saws and a map of hardnesses might give us a clueto how they tensioned the plate.

    There is allready chemical analysis of Disston saws on the Disstonian institute website.

  12. #57
    Join Date
    Mar 2004
    Location
    DuBois, PA
    Posts
    1,897
    Again, what is the point of the test? Is it to demonstrate the variability of hardness in vintage saws? Is it to test the variability of a string of hardness test operators?

    If you want to learn the hardness of saw plates, simply find a person who will test the materials for you, using a tester that has been properly maintained and calibrated (ours is done quarterly), along with an operator must be familiar with the correct operation of the tester and depending upon the model of tester, selection of indent and weights (if needed). For the cost of postage and test blocks, a certified quality lab can probably provide you all of the readings you would want, but the question remains, what will be done with them? Is it simply to say "this vintage Atkins saw, with "Silver Streak" steel, had a Rc value of 52 at the sharpened down toothline" and "this Disston #16, with "London Extra Refined" steel had an Rc value of 56 under the handle"?

    I agree with a comment George made, in that he could tell the hardness of material when filing, though that came from years of experience. Since George has a hardness tester, maybe he can answer if he tested hardness on each saw before filing?
    If the thunder don't get you, the lightning will.

  13. #58
    Quote Originally Posted by Rob Streeper View Post
    I think sanding or lapping is preferable at this point to electrolysis, acid cleaning or other methods. Given that the plates may be taper ground we need to carefully select testing areas too.
    Rob, shoot me a PM with your address and I'll send these two plates to you after I sand them clean at the toothline. I'll send an area clean on both sides in the middle of the plates, too.

    Send your work address if you're worried about giving out your home address.

  14. Quote Originally Posted by Tony Zaffuto View Post
    Again, what is the point of the test? Is it to demonstrate the variability of hardness in vintage saws? Is it to test the variability of a string of hardness test operators?

    If you want to learn the hardness of saw plates, simply find a person who will test the materials for you, using a tester that has been properly maintained and calibrated (ours is done quarterly), along with an operator must be familiar with the correct operation of the tester and depending upon the model of tester, selection of indent and weights (if needed). For the cost of postage and test blocks, a certified quality lab can probably provide you all of the readings you would want, but the question remains, what will be done with them? Is it simply to say "this vintage Atkins saw, with "Silver Streak" steel, had a Rc value of 52 at the sharpened down toothline" and "this Disston #16, with "London Extra Refined" steel had an Rc value of 56 under the handle"?

    I agree with a comment George made, in that he could tell the hardness of material when filing, though that came from years of experience. Since George has a hardness tester, maybe he can answer if he tested hardness on each saw before filing?
    Hi Tony,

    Fundamentally I just want to know. By testing across the saw plates we'll produce information to inform the debate we've been having here. Much more importantly we'll have a benchmark, albeit for only two saws, that will help us understand how Disston used to produce saws. It is known that Disston hammered the saws but apparently nobody has a detailed idea of exactly where. Everybody knows that the saws can be filed which demonstrates that the teeth and some of the area around the teeth is in or close to the low 50's range of C scale hardness however nobody knows where the zones of hammered steel are. Are they just along the spine of the plate or do they extend to other regions? I'm willing to map these plates and try to throw some light on this issue.

    Cheers,
    Rob

  15. David,

    Sand the whole surface. I want to map them from tooth to back. A lot of work but I think the effort is worth it. I suggest using some wet-or dry and a thin hydrocarbon lubricant such as WD40 to help keep the surface uniform. I've found that sanding in this way produces a finer finish. In addition, I think that the teeth should be filed away. The reason is that the anvil of the C tester has a radius of about two inches. If I test close to the tooth line the points of the teeth will bear against the surface of the anvil because of the set and raise the saw plate above the anvil surface producing useless measurements. Let's get this agreed before you ship. PM sent.

    I use these in my surface prep. Expensive but they last a long time if used with light oil. For lapping oil I use Rust Oleum spray on rust preventer. Home Depot stores in my area sell this in the spray paint section for some reason. It's not stocked with the other spray lubricants.

    label side a.jpgdiamond side a.jpg
    Last edited by Rob Streeper; 01-16-2015 at 9:06 AM.

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •