Skew block query

[Derek Cohen (Perth Oz)]

Is it reasonable to beef about one degree of error? Discuss.

Alf

The question to answer is whether you are able to obtain the type of shaving I get in my picture? Another question is 'are you satisfied with the performance of your #140?' If either of those answers is "no", then send it back!

The motivating factor behind my mod to the Stanley was that it couldn't cut for toffee. It was just not doing what it had been designed to do, and it was obvious where the problem lay when you looked at the blade/bed relationship. Now the #140 cuts superbly. It is a pleasure to use. You should be experiencing the same. I know that other LN #140 users stated that theirs cut as mine does.

That one degree does not sound much, but something is very wrong. Hmm a thought - that is one degree over a span of about 1-3/4"? Is that a lot or a little?

Regards from Perth

Derek

 

[Shady]

If it's a manufacturing error, it is indeed a lot: think of a one degree error over the average kitchen cabinet's length.....

 

[MikeW]

...Not like this then:

I apologize for such a long series of posts. Your picture does show the difference quite well.

But, is the blade projecting out of the sole the same--i.e., is it the same depth from one side of the blade to the other?

Even with just enough pressure from the lever cap to just stop it dropping off, I had a noticable difference in projection, and not enough leeway to compensate via lateral adjustment. Believe me, I tried...

...I slide the blade so that the right edge is against the removable fence...

Er....? :? Explain that again, slowly. Don't get it - sorry.

The lower part of the blade which is wider than the upper portion is what I push against the removable side. It will slide over against it, but it is easy to move the back, thinner portion of the blade to far causing the leading edge of the skew to pivot away.

Alternately, you can remove the side and, with that side of the plane flat against your bench, wiggle the blade so that wider portion of the blade is also flat against the bench.

Also, and I imagine you've done this, without looking at the mouth/blade gap, go for just an even projection. Use a small double square set so you can check each side for evenness. You probably should make it way more projected than you would ever use just to make it easy to adjust for an even projection.

Then look at the gap between the blade and front of the mouth. If there is a difference, it can be caused by the blade not being correctly ground. But only if it is significant (whatever that means) does it indicate there's a problem.

I'm sure you've already done these things and I'm sorry that you are having this problem with the plane.

You really ought to contact LN if you cannot get it adjusted to your satisfaction. It could well mean there is something wrong with the angle the bed is ground from the leading side to the trailing side. Which is weird geometry in itself. (Try making a skewed plane...)

The one degree you mention in a different post truly affects the whole blade/mouth issue as well. If you think the 140 is hard to get the skew angle correct, try the 98/99s--even Leonard Lee thinks they are about the hardest...

Sorry I cannot be of more help than that right now. My brain hurts

 

[MikeW]

If the blade is ground so that the projection of said blade is now even across the sole of the plane, but the front edge of the mouth is not equidistant to the blade edge, does that mean the bedding is at fault?

Alf, the answer is ... yes. Yes, I think the bed is skewed. It is in cases like this that epoxy is helpful - but you should not do this, and instead send it back to LN. (I used epoxy on my Stanley to close the mouth as much as square the bed - I would have sent it back to them, but the warrantee had run out ).

Regards from Perth

Derek

The bed is suppose to be skewed--or at an angle. Higher on the leading edge to lower on the trailing edge. The geomtery isn't too hard to understand. But from the two attempts I've made at making a wooden skew smoother it is difficult to achieve!

Basically the bed angle should be the difference of subtracting the front to back bedding angle from the skew angle.

In the case of the LN 140, subtracting the 12 deg bed (back to front) from the skew angle of 18 deg leaves 6 deg. angle right (leading) to left (trailing).

In practice, because there is the whole sine cosine math thingy, it is actually not an even number, if I remember correctly. Maybe if I was better at math my attempts at making one would have been more successful

 

[Shady]

Mike, you're right on the math: I actually knocked up an excel spreadsheet a couple of years ago, giving effective attack angle for a variety of skew angles - it's a nice sine wave. tangents are the actual multiplier, from recollection. I might reconstruct it if people are interested enough -useful when building and testing skewed planes....

 

[Alf]

The question to answer is whether you are able to obtain the type of shaving I get in my picture? Another question is 'are you satisfied with the performance of your #140?' If either of those answers is "no", then send it back!

Oh it works now (I'd have been looking for a replacement in a heartbeat if it didn't), but I dunno, it does bug me having to fettle a L-N and still end up with it not "perfect". Sort of feel it goes against nature somehow... :lol: but there seems to be this uncertainty over whether what I have isn't as good as I could expect to get anyway, which makes returning it - however bugged I am - actually into a decision not to have one at all. I suppose it does come down to asking L-N if this is is within the specs they consider okay, and if so at least I know I don't have a pup. Edit: I've emailed to ask. Heck, I've stirred up this pot, I better get it settled one way or the other. :roll:

I apologize for such a long series of posts.

No worries, Mike. It's very helpful in checking I haven't boobed somewhere.

Your picture does show the difference quite well.

But, is the blade projecting out of the sole the same--i.e., is it the same depth from one side of the blade to the other?

In that picture, yes. That's after I adjusted the angle of skew in order to get the projection even.

...I slide the blade so that the right edge is against the removable fence...

Er....? :? Explain that again, slowly. Don't get it - sorry.

The lower part of the blade which is wider than the upper portion is what I push against the removable side.

Ah-ha! Side, not fence. All is clarity. Thanks.

Cheers, Alf

 

[MikeW]

Mike, you're right on the math: I actually knocked up an excel spreadsheet a couple of years ago, giving effective attack angle for a variety of skew angles - it's a nice sine wave. tangents are the actual multiplier, from recollection. I might reconstruct it if people are interested enough -useful when building and testing skewed planes....

Hi Shady, I would like a copy if it isn't too much trouble. I have a reference book if I could just find it. But seeing how we have well over 3000 books in this house and only about 500 are out of their storage boxes--I shudder to think what that would entail...

 

[MikeW]

...In that picture, yes. That's after I adjusted the angle of skew in order to get the projection even.
Cheers, Alf

Ok. If you were to move the blade around to where you have an even gap between the leading/trailing edge and the front of the mouth, it looks like the trailing edge would not make contact, correct?

In looking at the picture again, something bugged me. It appears that the blade has a camber to it. It appears as if the leading edge is ground back further than the trailing edge.

My completely scientific method of determining this was to hold a piece of paper up to my laptop screen . But doing that both the front and back edges of the mouth are in agreement with the paper's edge. But the blade curves.

Oh it were so simple.

 

[Shady]

No probs Mike - wait until tomorrow: I've not got my textbooks in the house, but being in the computer department, and having the applied maths department on the floor below me at the university kind of helps for this sort of question... :wink:
(edit - paper against the screen?? - brilliant!!!)

 

[Alf]

Ok. If you were to move the blade around to where you have an even gap between the leading/trailing edge and the front of the mouth, it looks like the trailing edge would not make contact, correct?

Correct.

In looking at the picture again, something bugged me. It appears that the blade has a camber to it. It appears as if the leading edge is ground back further than the trailing edge.

That'll be because I lost the will to live and didn't sharpen the new angle right the way across. It's most of the way and I've made allowance for the bit that wasn't. It's sheer laziness; I intended to finish it off once I reckoned I had the right angle, but then I got to wondering about the parallel business and thought I wouldn't bother until I'd decided if it was going back or not. Let me stress this: I have allowed for this in my measurements. The majority of the edge is the "correct" angle.

My completely scientific method of determining this was to hold a piece of paper up to my laptop screen . But doing that both the front and back edges of the mouth are in agreement with the paper's edge. But the blade curves.

Nope, the blade doesn't curve. If anything it's meeting at an angle, albeit a very slight one, about four fifths of the way to the trailing edge, which could give the impression of a curve, but it isn't. I used the Veritas honing guide and am as absolutely certain of that as it's possible to be. I'm totally confident that is not the problem. Not often I'm so certain of anything, but of this, I am.

Cheers, Alf

 

[bugbear]

Hmm, I think BB may have a point here... Is it reasonable to beef about one degree of error?

I think the angular error in the bed will be MUCH less than this; the angle of the blade is a secondary, consequential thing. LN machined the bed.

Mike W...

Basically the bed angle should be the difference of subtracting the front to back bedding angle from the skew angle.

I don't think it's that simple.

http://nika.frontier.iarc.uaf.edu/~cswi ... 93#message

BugBear

 

[Shady]

Yup: Bugbear's link is right: I spent a happy 20 minutes with 2 lady mathematicians this morning, who both obviously thought I was nuts, but we eventually arrived at the formula for 'effective attack angle' if we know the face angle (ie bedding + grinding angle) and the skew angle.

Formula is: "Effective attack angle = Asin (sin bedding angle *cos skew angle)"

I'll run up an excel spreadsheet to make this a little more useful for us all later today...(wish it worked in native degrees instead of radians...) :wink:

 

[Shady]

OK - I have an excel worksheet and a chart showing the effect of skew on effective attack angle. Can anyone advise on the best way of 'showing it' here?

 

[Alf]

Shady, if they're not too hideously enormous (dial-up you see) email 'em over to me and I'll sort it out. alf AT cornishworkshop.co.uk

Cheers, Alf

 

[Shady]

Sent it - thx!

OK: hosted courtesy of Alf the magnificent techno-wiz, here's a quick excel chart showing the effect of changing skew angle on the attack angle. (click on URL at bottom)

Points to note:

1. I've just used a 'bedding angle' of 30 degrees. This is, of course, the sum of the body bedding angle, and the angle ground at the tip of the blade for a bevel up plane, or just the bedding angle for a bevel down plane.

2. The 'resultant attack angle' column is worked out trigonometrically from the other 2 angles for each cell in the worksheet, so if you want to play around with different seating/skews, just enter them in the first 2 columns, and the 3rd should give you the correct result.

3. Dunno how practically useful you might find it - good for plane making/tweaking, and for making skew blade sharpening jigs, in my experience. Enjoy!


http://www.cornishworkshop.co.uk/skew.xls

 

[MikeW]

...I don't think it's that simple.
...BugBear

Of course not...that's why I said there's the sine cosine thing. Gee, my head hurts again already.

But, not that it matters. Unless one is making a plane and wants to do the math for the math, it's all moot.

The easiest way for people making their own skewed-iron planes (non-production) is to build the body, stick the blade out the mouth and scribe it, adjusting the grinding as you get closer to the scribe line.

Works fine. I've had to do it that way.

 

[MikeW]

Hey Shady and Alf--thanks. This will make it so my head won't hurt so much .

One reason why this (the math) even matters is because I'm going to be making several sets (of either 3 or 4) skewed rebate planes as gifts for some friends this Christmas.

It will make it to where I can at least order the blades and have them close to ground correctly--at least if I make the bodies properly!

Again, thank you both!

 

[Chris Knight]

Shady and Alf,
Thanks.

The results would suggest that at anything like a normal skew angle the reduction in effective angle is not going to make a great difference to cutting performance. So - is there anyone who would care to explain why the skew cuts more easily (I presume the "slicing" action") or why the shavings from eg a 95 come out as spirals?)

 

[Shady]

Chris: I reckon there are at least 3 aspects to improved cutting performance 'on the skew':

1. The decrease in angle, however slight, will have at least some effect (worth noting how relatively small the 'normal' range of bed angles is to deal with exotics - people only seem to go from about 45 to 55 degrees - so maybe, although the values are small, the effect is significant.

2. I think Garrett Hack suggests that, because your effort is now going into a longer area of blade (the skew increases the length of edge in contact, all other things being equal), the 'force per unit length' equation changes in your favour. Seems counter-intuitive (ie, more fibres to cut, should be harder), but there we go.

3. I personally suspect, although I've never seen any evidence to back it up, that one of the ways it helps reduce tear out is because the fibres are not all being 'hit' by the blade at the same time. You're now attacking each bit along the grain at a different point, and maybe this reduces the degree to which it lifts/tears.

4. As to curls - just the geometry of the cut, I guess...

 

[Chris Knight]

Shady,

I can't really think of any other factors (your 3 was what I meant by slicing) and as you say 2 seems counter intuitive. I'll just characterise it as one of life's little mysteries