I confess that math is not one of my favorite subjects. With that in mind, I have been compensating for the inevitable blade tilt in my gen 2 clamp by setting the angle independently for each side of the blade using my angle gizmo. I understand that the position of the arm mounts are not going to be identical on the calibrated bar. But from what I am reading in other posts, I get the uneasy feeling that my solution may be a bit too simple, easy and wrong. Can someone please point me to the formula that is correct as far as compensating for blade tilt is concerned.
Hey there Alan and welcome to the forum. Unfortunately Im not a big fan of that particular calculation either. It always leaves me wondering if I did it right. Tom and others have explained it at length in posts. Something about measuring both left and right blade angles, dividing by 2 to get the center point angle(lets say 3 degrees off center for example). Then you would add 3 degrees to the stone angle on the far side and subtract 3 from side blade is leaning towards?
PLEASE dont take that as correct instructions and sorry thats the best I can do. Im really more of a visual learner and a simple diagram(hint hint) would serve wonders to clearly demonstrate the concept of proper compensation for a tilted blade angle in the vice.
Thanks for reposting the video Josh. The first time I did it was for a full flat ground blade. To double check that the math made sense I laid the blade on my countertop and measured the angle of the blade from the counter top. Long story short the angle of the blade checked out with the measurements in the vice which was good to know before I stared grinding away.
Thanks for the video, it gives me a good idea of how to compensate for cant. But how do I account for a flat grind blade that tapers from spine to bevel? If I lay the angle finder on the side of the blade, since it will not give a true 0.00 degree vertical?
One thing just pointed out to me is that I had my math wrong in the video… someone pointed this out to me about a year or two ago and I edited the video with a caption to correct the math. So if you have a browser w/ an add on that disables the captions you will miss it. Just wanted to note this! Thanks and glad it’s helping.
Actually, Josh has two math errors in the video. He subtracted the left reading (+0.70) from the right (-4.00) when he should have added them. The object is to find the angular displacement between one side of the blade and the other. Imagine that you have the AngleCube on the left side, reading +0.70. Now rotate the 'cube to the left to -4.00. To do so, you have to rotate the 'cube 4.70 degrees, not 3.30. This means the median (centerline) is displaced 2.35 degrees from each face. If the blade were perfectly vertical, it would read +2.35 on the left side and -2.35 on the right side. Since the left side reads +0.70 instead of +2.35, the centerline is canted 1.65 degrees to the left. To compensate, you deduct 1.65 from the left setting and add 1.65 to the right.
On another issue, the Gen 2 vise uses face-to-face contact in gripping the blade. If one or both of the jaws is not parallel to that face of the blade, you are in danger of losing grip and the blade might move on you. When you are adjusting the jaws, try to keep them both parallel to the blade faces.
The Gen 3 vise has features to avoid that shortcoming. It makes contact with the blade at two points, roughly positioned at the upper outside corners of the vise. This “point contact” and a designed-in forward-to-back flexibility allows the vise to accommodate both vertical and horizontal variability in the blade’s contours. For those who question the use of such small contact points, rest assured that the principle is technically correct and born out of good mechanical design principles.
The force holding one body against another, preventing lateral movement is dependent on which of two modes of contact exist between them. They are either in “traction” with one another, wherein there is at least some physical engagement between them (think tires on concrete), or in “slip” where one body can slide across the face of the other body (think tires on ice), if enough force is applied.
Metal against metal is generally a “slip” situation. The force required to move one relative to the other is dependent on (A) the pressure applied forcing one face against the other and (B) the coefficient of friction between them. The size of the area is irrelevant.
In the case of the Gen3 vise, making the contact points larger would have added no benefit and simply would have made adaptability to various contours more difficult. Making the contact points smaller might have increased the point-contact pressure enough to have left marks on the blade. All in all, well done.
OK now I am more confused. This is certainly due to my own math deficiencies more than TCMeyer’s explanation. I’m begging here now, is there a way that someone can put this issue in a step by step format that people who are obviously math challenged can apply to their sharpening routine. I sincerely apologize for being so dense.
Bill: You are right about the included angle being the same even if the blade is tilted to one side or the other. The need to keep it centered lies in (A) the desire to have the bevels appear to be equal and (B) the need to be repeatable - able to find and match the bevels without going thru and extended Sharpie-matching process. I use the “measure and compensate” method every time I touch up my Delica 4 and never fail to hit the angles within 0.1 degrees.
Which brings up your statement about angle cubes not being accurate. You apparently have a bad one, as 15 degree variability is 10 times anything that I would consider bad. My Imaging AngleCube will never be off by more than about 0.1, when used properly. The video above demonstrates the right way. Place the cube against the stone, take a reading after about three seconds, then tip the cube away and replace it, taking another reading. Do this at least three times and choose an angle midway between the highest and lowest readings. If any of my readings falls more than 0.2 degrees from the opposite extreme, I assume the readings are invalid and start over, after trying to eliminate anything that might have contributed to the excess. Lifting and replacing the entire stone, cube affixed thereto, adds the variability of the system mechanics and your technique (how much and where you apply pressure) to the mix. This reduces the effectiveness of your 'cube and its potential accuracy.
As to the question of how accurate your angles must be, I can say that errors of more than 0.1 degrees are easily seen after the first few strokes. You’ll see that the new scratch pattern seems to not touch the entire bevel. There will be residual scratches from the previous grit showing, either at the bottom of the bevel or at the apex. Such an error will just require additional work to remove those scratches made by a coarser grit.
Alan: This isn’t a math puzzler by any stretch of the imagination. Any child who’s had 5th grade arithmetic should be able to execute the simple math here.
Measure the angle of each side of the blade.
Add the two numbers; result is the included angle of the blade's primary grind.
Then divide this included angle by two; result is blade's centerline.
Subtract the smaller of the blade side angles from blade's centerline; result is the lean angle (How the blade leans to that side)
Subtract the lean angle from your intended sharpening angle; result is rod angle for that side.
Add the lean angle to the intended sharpening angle for the opposite side; result is the rod angle for that side.
I guess by definition, what you’re describing is not a full flat grind (FFG). Most of the knives I sharpen are not FFG. If it’s not FFG, there’s no point in measuring the face angles, as they differ from the facets at the vise contact points. But the objective is to set your rod angles so that they are aligned with the centerline of the blade, regardless of the grind. Do what you gotta do.
So how would the math work if you had two positive numbers something like 1+ on the left and 6+ on the right? You would actually subtract them then, correct? Because this way the difference is only 5 degrees…I’m no engineer, one of you engineers should have made the video I will probably re-do it so as to not confuse people, only if one of you don’t want to take it on that is!
Bill,
It depends on what you are wanting as to how much the angles matter… if you can sharpen a knife on a brick freehand to this level of sharpness, that’s all anyone really ever needs. If you are going for a mirrored edge it matters much more, however, even when I’m mirroring I never use my angle cube after the first stone. from there I adjust using a sharpie (some also use the vsta) and a flashlight.
As far as FFG blades go… I don’t always measure the tilt of a blade, but when I do I will either do it on the bevel face (primary grind) or the flats above the grind. Sometimes if it’s a folder I already have the blade out of the handle and I will measure it on the flat of the pivot area. Some convexed fixed blades you can’t do this with so you will have to eyeball it though
Yes, +1 and +6 would mean the faces are displaced by 5 degrees. The centerline is 2.5 degrees and the blade is leaning 3.5 degrees (2.5 + 1) to the right.
The object of my method is to find the displacement between the two faces so that you know what the centerline is, so you can calculate the amount of cant/lean. In your video, you measured 3.3 degrees, with what you thought was a centerline of 1.65 degrees. The real centerline is 2.35 degrees (4.7 / 2). While your method ends up with the correct offset, for this clamping instance, you can’t use that centerline again when you remount the blade.
For example, if the next time you mount the blade, you measure +1.30 degrees on the left and you subtract that from your old 1.65 centerline, you’ll deduct 0.35 degrees (1.65 - 1.30) from your left rod setting when you should deduct 1.05 degrees (2.35 - 1.30). Your bevels will be off by 0.70 degrees.
So long as you use your complete procedure, you’ll be OK, but you’ll have an incorrect value for the primary grind angles, because your formula is convoluted.
Perhaps this is a big deal for me, because my EDC Delica gets touched up about once a month, and I depend on knowing that the primary FFG angles are 3.75 degrees inclusive. I just read the left face of the blade and bingo - I’m done. Whereas you (Josh) don’t think in terms of the centerline angle - you just depend on the procedure getting you to the correct answer (which it does) and I think this is confusing to those who are trying to understand what and how you are doing this.
A couple points/questions… Tom, So how would the math work if you had two positive numbers something like 1+ on the left and 6+ on the right? You would actually subtract them then, correct?
Yes, +1 and +6 would mean the faces are displaced by 5 degrees. The centerline is 2.5 degrees and the blade is leaning 3.5 degrees (2.5 + 1) to the right. The object of my method is to find the displacement between the two faces so that you know what the centerline is, so you can calculate the amount of cant/lean. In your video, you measured 3.3 degrees, with what you thought was a centerline of 1.65 degrees. The real centerline is 2.35 degrees (4.7 / 2). While your method ends up with the correct offset, for this clamping instance, you can’t use that centerline again when you remount the blade. For example, if the next time you mount the blade, you measure +1.30 degrees on the left and you subtract that from your old 1.65 centerline, you’ll deduct 0.35 degrees (1.65 – 1.30) from your left rod setting when you should deduct 1.05 degrees (2.35 – 1.30). Your bevels will be off by 0.70 degrees. So long as you use your complete procedure, you’ll be OK, but you’ll have an incorrect value for the primary grind angles, because your formula is convoluted. Perhaps this is a big deal for me, because my EDC Delica gets touched up about once a month, and I depend on knowing that the primary FFG angles are 3.75 degrees inclusive. I just read the left face of the blade and bingo – I’m done. Whereas you (Josh) don’t think in terms of the centerline angle – you just depend on the procedure getting you to the correct answer (which it does) and I think this is confusing to those who are trying to understand what and how you are doing this. [/quote]
[quote quote=30813]Alan: This isn’t a math puzzler by any stretch of the imagination. Any child who’s had 5th grade arithmetic should be able to execute the simple math here.
Measure the angle of each side of the blade.
Add the two numbers; result is the included angle of the blade’s primary grind.
Then divide this included angle by two; result is blade’s centerline.
Subtract the smaller of the blade side angles from blade’s centerline; result is the lean angle (How the blade leans to that side)
Subtract the lean angle from your intended sharpening angle; result is rod angle for that side.
Add the lean angle to the intended sharpening angle for the opposite side; result is the rod angle for that side.
Easy peasy. [/quote]
Tom, your instructions are amazingly clear and yes its “simple” but what I’ve found for myself and Im sure(?) many others is a diagram is worth 1000 words. Sadly computer drawing is not in my wheelhouse. I love the “Sweet Spot” diagram that was recently posted. Ive heard it described 100 different ways but there it is in a picture and my brain “gets it” instantly. More spacial/visual then verbal I guess…
I will probably re-do it so as to not confuse people, only if one of you don’t want to take it on that is!
