Machine sanding, question on grinding out coarser scratches

B

bndshardr

Joined
Oct 16, 2013
Messages
33
When I hand sand a blade, I alternate directions as I progress to finer grits, this allows me to see the coarser scratches, and know that I got 'em all before going to the next finer grit. That's not possible when grinding bevels on the grinder and having all the scratches in the same direction makes it difficult to see deeper ones. Ultimately, they tend to reveal themselves after spending a bunch of time hand sanding. Do you guys have any tricks that you use to know when to make the jump to the next belt?

Thanks

bdnshardr
 
One little trick I picked up from Caffrey and I think many of the guys out there know it, so I think it's okay to mention, is that you should try skipping to a fine grit right away. As an experiment just see what happens if you hog to 36 grit and jump right up to a new 220. Maybe 120 would be better or maybe you can go straight to 400. Just try it.
 
I use 3.5x reading glass's holding the piece at eye level, rotating in the light until I see the hidden little buggers ;0)
 
I clock my blade back and forth when I flat grind. Not only does that allow me to see deeper scratches, it prevents grit from falling into a deep scratch and making it deeper. This leads to smoother belt finishes and reduced grinding time. Obviously this doesn't work for hollow grinding.
 
Nathan- I love this: "Build a man a fire and he'll be warm for a night. Set a man on fire and he'll be warm the rest of his life."
Sorry, back to grinding....
 
Nathan, would you mind explaining what you mean by "clocking the blade back and forth?"

Michael, what is the jumping right up to a higher grit supposed to do or how does it help? Do you mean if you do it that way that you may not even have to bother with working up though the grits, as long as using a fresh 220 belt etc.?

hah.sorry, but I'm curious about this subject as well :D

-Paul
www.youtube.com/Lsubslimed
 
Hi Paul. Those 36 grit scratches won't be able to hide even if they are running the same direction. Just try it.
 
I do my coarse grinding with 60 grit belts, at least for now, I'm a newbie and would rather burn through a couple of belts than a couple of blades when it comes to coarse grinding. I have tried jumping from 60 to 220, and that has worked okay. I burn through more 220 belts doing it that way, but they're cheap in comparison to 120 cubitron's. I guess my question goes further than the coarsest belts. Beyond 220, I currently go to Trizact A45 and A30. I suspect that once beyond machine grinding, I'm seeing 220 machine marks that need to be hand-sanded out, even though I've machine sanded to a 600 equivalent with the Trizact's. So, is there anything I can do to detect and correct those marks prior to hand sanding? Or, just take the finer machine grits beyond where my gut tells me they need to go before switching to hand sanding?

Thanks again.

bndshardr
 
I've been at this for what I can say is "several" years and still see myself doing a poor job of chasing the scratches from one size to another. I do know at time I over use any size it kicks me in the back side. Frank
 
Last edited:
coldsteelburns said:
Nathan, would you mind explaining what you mean by "clocking the blade back and forth?"

-Paul
]
Click to expand...

Clocking means to rotate in a plane around an axis like the hands on a clock, in this case meaning the knife is rotated in a plane so the flat surface remains parallel to the platen while allowing the actual grinding direction to change.

I do most of my finish grinding lengthwise with the top wheel removed. That platen is a plane so it can be approached from any direction and still be the same, geometrically speaking. So the same principals that apply to hand sanding, such as changing the direction of sanding, apply when flat grinding. This gives a better finish in less time and fewer belts. Hollow grinding, obviously, is a different can 'o worms.
 
I don't go quite as far as Nathan, but do something similar by shifting my angle now and then. Obviously at the plunge that's not going to work, but past that I don't stick to the exact same position every stroke.
I also jump from 36 grit to 60 then 120 fairly fast. 36 grit gets me down to "ok, it's got a bevel mostly flat and mostly up the blade, I've got 75% of the metal removed." 60 flattens that out and gets me into the actual ballpark for the bevel height but still some actual metal to be removed, I'm not worried about exact grind height or edge thickness yet, just that things are consistent. 120 is where I start making it look like it will be when it is done. Bevels get ground to almost the height I have in mind, edge thickness should be right about there too. If they aren't coming together properly I make my adjustment here. By that point there's no way there are any scratches bigger than 120 grit and I have enough metal left to remove that they're going to go away too. I switch to 220 to finalize the bevels positioning and from there it's "finish" where I run through progressively finer belts till I'm wherever I plan to start handwork or leave it as is. I do mostly belt finish or scotchbrite finishes these days anyway.
The key for me is that each step removes enough metal that I'm not worried about scratches, I'm still removing significant metal right until I hit 320 or 400 grit. The amount of metal is less and less, but so is the potential scratch size I'd have to worry about. Also, mistake size, which is how I got into this routine. I never want to be in the position where one extra pass, or a bit too much pressure on that pass can over run where I want to finish. I would rather make three our four extra passes with the next finer belt than risk screwing up the knife by over grinding.
 
I keep small flap wheels on hand to chuck in the hand drill. Works great at getting out those errant vertical scratches. Supergrit has a wide selection of sizes and grits, as well as non-woven flapwheels.
 
You can also do a quick dip in FeCl. It will darken the scratches.
As you grind, the deeper scratches will remain easily visible until they are gone. ;)
 
Karl B. Andersen said:
You can also do a quick dip in FeCl. It will darken the scratches.
As you grind, the deeper scratches will remain easily visible until they are gone. ;)
Click to expand...

I like that idea, Karl. I've tried Dykem, but it always seems to get pulled out of the deeper scratches before they're gone . . .
 
I've pretty much given up on a real good "off the grinder finish". I knock scale off with a 36, then to a ceramic 80 grit. Then to 120, 220 and 320. I use a A30 trizac belt run over the edge of the platen to true the plunge cuts up. After that I start hand sanding at 320. Sometimes I use an orbital sander with a mycarta pad with 320 sand paper glued to it to true everything up. After that it goes fairly quick hand sanding, a stiff backer and the blade clamped on an aluminum holder where I can get some pressure, and windex makes a good lubricant. Maybe if I was to upgrade to a variable speed motor for my grinder I'd have better luck, but so far works for me.
 
I also dip in the FC tank , rinse, and then go to the next grit. This is also a great way to watch where your grind is going, as you can see exactly what is being removed. On a dagger it works especially well for keeping the center line straight.
 
Remyrw said:
I don't go quite as far as Nathan, but do something similar by shifting my angle now and then. Obviously at the plunge that's not going to work, but past that I don't stick to the exact same position every stroke.
I also jump from 36 grit to 60 then 120 fairly fast. 36 grit gets me down to "ok, it's got a bevel mostly flat and mostly up the blade, I've got 75% of the metal removed." 60 flattens that out and gets me into the actual ballpark for the bevel height but still some actual metal to be removed, I'm not worried about exact grind height or edge thickness yet, just that things are consistent. 120 is where I start making it look like it will be when it is done. Bevels get ground to almost the height I have in mind, edge thickness should be right about there too. If they aren't coming together properly I make my adjustment here. By that point there's no way there are any scratches bigger than 120 grit and I have enough metal left to remove that they're going to go away too. I switch to 220 to finalize the bevels positioning and from there it's "finish" where I run through progressively finer belts till I'm wherever I plan to start handwork or leave it as is. I do mostly belt finish or scotchbrite finishes these days anyway.
The key for me is that each step removes enough metal that I'm not worried about scratches, I'm still removing significant metal right until I hit 320 or 400 grit. The amount of metal is less and less, but so is the potential scratch size I'd have to worry about. Also, mistake size, which is how I got into this routine. I never want to be in the position where one extra pass, or a bit too much pressure on that pass can over run where I want to finish. I would rather make three our four extra passes with the next finer belt than risk screwing up the knife by over grinding.
Click to expand...

This is a very clear explanation !
 
Every time one of these topics comes up I feel obliged to toss in my quote:
"Whenever you're sanding out scratches - you're not sanding scratches. You're sanding everything that's NOT a scratch". ;)

(See if that helps)
 
I don't hand sand
I use the progression of 36 60 120 grit Ceramic belts then use a 220 Cork belt loaded with green chrome then a 400 cork with chrome. Then my final satin finish is done by using Scotchbrite belts.

So I take it up to a higher finish with my cork belts and then step down some to a satin finish with the Scotchbrite.

I make all working knives. To me, nothing looks worse that a mirror finished blade that then has use marks on it.
 
You must log in or register to reply here.
Back
Top