Edge Inquisitor 3000 edge tester

HeavyHanded

HeavyHanded

Joined
Jun 4, 2010
Messages
6,642
Have been wanting to make an edge tester that I could come back to and get good repeat-ability either as a stand alone or after performing other cut tests to use as a baseline. I have no confidence in my previous attempts to stabilize the cutting tool or test media for this sort of thing, and so no confidence in the results of my tests or good feeling about the $ I'd spent on rope...so bring on the EI3000!

Much respect for the piles of rope cut by some of the forum members (David Martin, Jim Ankerson and many others) and their willingness to share the data from this work. I hope to make some contributions of my own as time permits.

Anyone have any idea what the most abusive rope is, or consistent methods of dosing the rope with an abrasive to speed edge wear?

Thanks for watching!

[video=youtube;2pxtGRaJT2I]https://www.youtube.com/watch?v=2pxtGRaJT2I&feature=youtu.be[/video]
 
This sounds better than my first choice, shaving live gorillas. Gorilla's do get a little miffed about it, but the consistency of their hair is quite uniform, making it all worthwhile. Yeti shaving would be good as well if you could find enough of the damned things. :)
Knife makers will do about anything to prove a point.

Fred

Its setting the clocks back in the fall that causes this irrational commenting on my part. It really messes with my head.
 
Fred, the timechange is a curse, has me out of my gourd for a week or two.

I don't make my own, but am planning on using this to test some custom knives and am hoping it gives me some sense beyond mundane usage of the qualities of the steel. Of course tests like this do not take into account the ease with which an edge can be sharpened etc, but you have to start (or finish) somewhere. And by loading it with various materials I hope to get some data re edge finishes and different media - rather than test to failure, check by poundage/ travel distance on the edge under load prior to severing/splitting etc etc. Right now I have experience with a bunch of steels, but is like describing different shades of blue - I can get my observations across and compare to others, I'd like to have more points to plot.
 
Well worth putting one together for those with more than a passing interest in edges, which you and I both possess.

We all tend to speculate and having something as a standard to compare would make the collection of data much more sound and worthwhile.

It takes me at least a month to convert to the time change. I don't wear a watch but am very time sensitive, always knowing the time within a couple of minutes. I think the change throws my mental clock into a dark hole.

Thanks for posting this, Fred
 
Hey Martin,

Let me see - base [check], arm [check], rope [check], counter weight [check]... Wow wow this is exciting -> a high performance trebuchet knives slinger with a lousy name ;)

For a complete waste of money try bamboo silk rope. Otherwise, soak a roll of paracord 550 in green paint. Either of these 2 will quickly degrade an edge in a consistent manner. It's quite difficult to draw cut a silk rope with edge finished with grit < 2000. LOL - at $1/ft, might as well cutting a dollar bill instead :p
 
Luong, you're a genius! A short length of frayed bamboo silk rope tied through the lanyard hole for a stabilizing tail and...once they falter on a cut test I can launch them over to the refurbishing bench 50 ft away.
 
Glass fiber rope is by far the most abrasive, like used for wood stove door gaskets.
A cheaper alternative would be twisting fiberglass insulation, but consistency might be a problem.
 
The EI-3k is starting to produce some data. By far the largest variable is the rope, as it runs along the density and diameter change as the splices come up etc. Still, I am pretty happy with how its working. Is now adjustable from 7 - 16 lbs with pre-set drawlengths as desired.



[video=youtube;XiWAd9RJhs8]https://www.youtube.com/watch?v=XiWAd9RJhs8&feature=youtu.be[/video]

And in context:

http://www.bladeforums.com/forums/s...sses-multiple-testers?p=15457355#post15457355
 
Yes, stove door gasket rope is much more abrasive. But have you noticed the price per foot for that stuff?
The sisal rope splices encountered are more difficult to cut thru. But it's a small area that can be skipped over. Still, you're comparing the data as a whole not getting too picky about small sections in the rope. Another way to cut down on rope cost is to only use half the blade. Tape up the other half. Then extroplate the amount of rope a whole blade would cut. When I do it, I'm after a end result number of cuts on 3/8" sisal rope at the various levels of sharpening. Then it's a straight comparison. i.e. this steel did this number of cuts on this medium. The area of isolating the force required will be a very useful number. DM
 
David Martin said:
Yes, stove door gasket rope is much more abrasive. But have you noticed the price per foot for that stuff?
The sisal rope splices encountered are more difficult to cut thru. But it's a small area that can be skipped over. Still, you're comparing the data as a whole not getting too picky about small sections in the rope. Another way to cut down on rope cost is to only use half the blade. Tape up the other half. Then extroplate the amount of rope a whole blade would cut. When I do it, I'm after a end result number of cuts on 3/8" sisal rope at the various levels of sharpening. Then it's a straight comparison. i.e. this steel did this number of cuts on this medium. The area of isolating the force required will be a very useful number. DM
Click to expand...

I intended to get black/white failure sets, but in reality I have moved to a pass/fail based on how many it misses entirely in a row. It can easily miss the odd cut or even two and just keep going for another 20 without a miss. At three skips on rope that seems of average thickness the warning buzzer sounds (in my head) and if cannot make a cut on the 4th the test stops. If the rope seems thicker than average I'll pull a little bit more through and retry one more time. Even after technically failing it might still run a few more, but I've noticed that it definitely begins to miss a lot more consistently as it gets close to failure, so is definitely dulling. Even after a technical failure these edges are still plenty sharp, just not sharp enough for this test.

The feed tube I use is necked down a bit, so a splice can just pass, and if the rope is thin it will slide right through. Even with weight and draw length controlled for there are still a lot of variables, so as you say is really about producing some very general data. I suspect a great deal go into this, a straight blade might perform better than one with a pronounced belly, recurves haven't been tested yet.
 
That's an awesome machine. I've thought about something like this, but never came up with such a good idea.

Just thinking: I wonder if you could apply a very precise tension to the lever -- much lighter than your current setup -- so that you could use a lot less rope. You could also use two different tensions -- light and slightly heavier -- to see how quickly the initial edge lasts and how long the subsequent working edge lasts.

But, dang, good work.
 
Thanks for taking a look!

Honestly, I'm a bit surprised my initial napkin sketch needed very little modification to become workable. $15 worth of pine and the rest was hardware I had lying around.

The unit will go down to 7 lbs and I could get it to go lighter, but getting an edge under these constraints and with relatively normal geometry to cut the rope with less than 10lbs over 2" draw is not easy. Fresh off the block it could manage the odd pull at 8lbs, but wasn't consistent enough to run a trial.

Not visible in the video, the arm that holds the load has holes drilled from the pivot point out to the end. The same 15 lb kettlebell exerts from 7-16 depending and it seems pretty repeatable - dropped on a scale it comes back reliably to the target weight every time. If off its not by more than a few ounces. It likely shifts as it gets further away from the clamp, so a longish pull would need to be centered at the clamp site.
 
"is really about producing some very general data." Your wording. To me the information & data I amassed doing this cutting is specific. Ex. this knife steel cut 3/8" sisal rope this many times. When sharpened at this angle coming off this grit stone. Using 7-8 lbs. pressure maxium during the cutting. Then when sharpened to this other grit it made this many cuts less on 3/8" sisal rope. Same steel, same edge angle, same pressure. Just changing the stone grit. This is quite specific data. So, perhaps you were saying this in jest as an example. DM
 
Not in jest, this is some hard work on the digits to be goofing around.

Yet the variation in any rope sample make it impossible to say with precision what a given outcome has proven. What is the deviation from multiple tests of the same edge prepared the same way with different coils of rope? If it is 10-30 passes over 600 total then no issue. If is 50-150 passes over 600 total (or greater) then it drifts toward "general information" - still useful but not easy to replicate with confidence beyond a certain degree.


One of the coils I used had mineral inclusions - grit embedded in the rope that I heard and felt contact the edge.

Overall it creates an impression that is predictive but no guarantee by a long shot. To me that makes any conclusions drawn somewhat general even if the data were specific.
 
So, now you're talking about a whole different ball game.? When you're doing cutting with all those machines running how can you hear these little sounds and know exactly what it is.?
Whereas I was doing my cutting in a very quite shop.Nothing running and no distractions. DM
 
Most of the equipment was turned off, nothing but some cooling fans running. I don't think it an entirely new ballgame, I noticed the edge would sometimes miss a cut then continue untroubled for dozens more passes. The rope changes in density and diameter by percentages large enough to effect the outcome.

I actually felt the mineral hit the edge through the sled, and could hear it easily, both times. Was in a hurry, so grabbed a pinch of the cutoff fibers and couldn't find anything with a 2-3 second look I shrugged and kept going - the knife was still cutting anyway.

Now I realize I have to run same to same and see what an average deviation might be +or-. The numbers from a single test unless a landslide might not be very accurate (though still better than nothing). Predictive, but not a guarantee.
 
Several times you've made a video it's at work because the machines are loudly running?? Thus, I had a hard time hearing you talk. I could not conduct such items on company time. Ok, wash the rope and get it spotless and grit free. Measure the rope at each foot. Then you'll find something else to obsess over. These picky items are all part of the test. When skinning an animal it's much the same factors. Some of the hide is dirty other areas cleaner. Some areas thicker than other. Even chickens the same thing. Some here cut cardboard it's the same MO. But in the end the blade steel did 2 deer not just one. When Jim and I are cutting the blade is in hand. Thus we have much greater feel for what is really going on at the cutting edge. Whereas you have completely eliminated this telling factor. I never had a blade miss a cut. Except on the splice. So, you're introducing other factors and errors to deal with that we did not have. Going toward laboratory to satisfy your methodology. Whereas Jim and I went with a real world method. Which you're free to do but now I'm seeing even the parameters are changing. "Data in the real world beats lab data any day." From Jeff Hubbard, a 25 year veteran of Buck Knives. He knows a thing or two about knives. DM
 
I think what Heavy is saying is that the cutting medium is not consistent, so to that extent the results won't be consistent, either. That's a valid concern. One nasty piece of grit hitting Edge A can make it look far worse in comparison to Edge B, if Edge B had nothing but clean sailing.

It's not true that data in the real world beat lab data any day. Field data is certainly important and can often reveal strengths and weaknesses that specific lab tests miss, but field data are not consistent. You need something like a epidemiological study with lots and lots of data before you can get a true picture of what the field data are telling you.

I think Heavy is on the right track. The limitation seems to be the cutting medium. Anything super consistent in resistance to cutting is likely to be super expensive. The other way to solve the issue it to do many tests so that outliers are winnowed out, but that is also an expensive and time-consuming process.

Thanks to all you guys who put in the effort.
 
You must log in or register to reply here.
Back
Top