What's new
What's new

how long to scrape a dovetail slide, like a compound?

durableoreo

Plastic
Joined
Jun 5, 2020
How long would it take an experienced scraping hand to fit a small dovetail slide? In my case, it is 10" long and 4" wide. There is about 1" of bearing on either side and the dovetails are about 0.375" deep, straight gib.

I've already done a lot of the work, only the gib left to fit. The duration of this project has me doubting my competence. For example, it took 5 hours to scrape the male side of the dovetail for alignment (was tapered 0.0018 over 10") and bearing. Today I spent 4 hours on the bearing of the non-gib side of the dovetail, mostly an angular adjustment. This is all with a carbide-tipped tubular scraper, 100% elbow grease.

Is this slow? Are chicken sounds the solution? Would a power scraper help?
 
There are a bunch of variables, but I find the Biax easier to control with very short strokes and faster for small details like dovetails. The time it takes very much depends on the starting condition and the accuracy you are aiming for.

In general it seems that most people starting out try to scrape parts that are too worn and should be machined or ground before breaking out the scraper. With careful milling on an accurate machine you can get something like a dovetail slide really very close and finish it up in a few hours by hand.
 
I'd be interested to see other's viewpoints of speed using a power scraper vs hand scraping. (paging Mr King)
First time I used one where much stock needed removing it was ... wow.
By hand I do not think your time is bad at all but I do not do this everyday for a living. I would likely be longer than your times.
How much time of that 5 hours was blueing and checking and how much was pushing the blade?
Does anyone use power to get it close and then do the end finish stuff by hand?
Bob
 
Last edited:
Anything more than about a thou, requires machining to get it close. Also badly worn assemblies often need to be cut 5-10 thou deep before the slideways clean up from the wear marks. In some cases it's worth simply manufacturing new parts rather than trying to save old ones. This permits the leadscrew assemblies to be re-centered concentric with the nut axis. This slide had about 20 or 30 hours of work done on it. Also the lower part of the crossfeed slide was missing, that had to be manufactured as well.
 

Attachments

  • PW_2.JPG
    PW_2.JPG
    1.6 MB · Views: 43
  • PW_6.JPG
    PW_6.JPG
    1.6 MB · Views: 44
  • slide_complete_1.jpg
    slide_complete_1.jpg
    109.1 KB · Views: 44
  • final_assembly_2.JPG
    final_assembly_2.JPG
    243.8 KB · Views: 44
On flat surfaces, it's not too bad to take off several thou by hand. You just get on it, and the CI dust is flying. Longish strokes, maybe a couple cm or so.

The trick is first, to swap directions every pass, and not scrape in the same path you did the last time. Also, to see when you need to stop and begin with the short strokes. If you are not careful, you will have ridges and valleys that are very hard to get rid of.

You may be able to get what you need done in less than the time to set up the job on a mill. And you will not be likely to take off too much. It's pretty easy to mill off too much unless you are very fussy in the setup, and "fussy" takes time.
 
It's like painting - 90 percent of the work is in the prep. Fussy for sure. My problem as a novice was maintaining good geometry when trying to scrape off a lot. And as mentioned, for a fairly worn dovetail slide, it's not uncommon to see the 'one or two thou' problem eventually be revealed as 5 or 10 thou to really restore good geometry. For this milling the setup really requires a tenths reading test indicator and a machine that has enough travel to make all the cuts without refixuring the workpiece halfway through. That slide shown above is modestly sized (for a 7" bench lathe) and the work was roughed out on an FP2, and that had barely enough travel to accomplish it.
 
How long would it take an experienced scraping hand to fit a small dovetail slide? In my case, it is 10" long and 4" wide. There is about 1" of bearing on either side and the dovetails are about 0.375" deep, straight gib.

I've already done a lot of the work, only the gib left to fit. The duration of this project has me doubting my competence. For example, it took 5 hours to scrape the male side of the dovetail for alignment (was tapered 0.0018 over 10") and bearing. Today I spent 4 hours on the bearing of the non-gib side of the dovetail, mostly an angular adjustment. This is all with a carbide-tipped tubular scraper, 100% elbow grease.

Is this slow? Are chicken sounds the solution? Would a power scraper help?
Before I answer I want to here more about you and your ability.

A couple of things first. You say you used a "tubular scraper" Anderson makes a tubular scraper and in my opinion it is a lousy scraper to scrape anything accept paint. They are aprox. 24" long or more with a round wood handle. Did you sharpen the carbide blade that came with it? Those scrapers come with a silver soldered on and there not sharp and are flat No radius no rake. Last week 2 of my student were given an Anderson Tubular scraper and I had them sharpen the blades to a 60 radius and 5 degree rake on both sides. Answer that first and was this your first time scraping something?

Who taught you to scrape or are you self taught? Where did you hear about someone doing a chicken noise? How did you sharpen the blade? Do you have a straight edge? Granite plate?? How did you check the dovetails, dowel pins and a micrometer? Show us some pictures of the compound and your tools.

Rich
 
I'd be interested to see other's viewpoints of speed using a power scraper vs hand scraping. (paging Mr King)
First time I used one where much stock needed removing it was ... wow.
By hand I do not think your time is bad at all but I do not do this everyday for a living. I would likely be longer than your times.
How much time of that 5 hours was blueing and checking and how much was pushing the blade?
Does anyone use power to get it close and then do the end finish stuff by hand?
Bob

I'm going to go a bit off topic now and tell you about the tools I would use so other readers can learn.
Bob I never scrape a 3/8 wide dovetail with a power scraper. I use a hand scraper with a 5/8" wide blade ground at a 60 or 40 radius blade (Biax 15-150 or a # 15 blade) I grind the top corners on a bevel to reach into the bottom of the vertex or narrow end. I use a professional made hand scraper made by DAPRA. or one can use a home made one with a lot of flex. My hand scraper is about 21" long when I'm using a 15-150 blade or 18" with a #15 blade.

A 1" wide Sanvik hand scraper wwould be OK too but I would grind my inserted claamped carbide blade the same way with thee corners ground with a bevel side It's late. I'll add some pictures tomorrow.
 
For that slide shown above, it was a HSS cut-off blade, worked up on a grinder to get into the corners, and that was jammed into the end of a short piece of dowel and clamped in place. Mr. King would have barfed up his lunch if he saw that. I *never* figure out how to do small dovetails right. I try to rough them out as slowly and smoothly as possible to minimize the agony.
 
Before I answer I want to here more about you and your ability.

A couple of things first. You say you used a "tubular scraper" Anderson makes a tubular scraper and in my opinion it is a lousy scraper to scrape anything accept paint. They are aprox. 24" long or more with a round wood handle. Did you sharpen the carbide blade that came with it? Those scrapers come with a silver soldered on and there not sharp and are flat No radius no rake. Last week 2 of my student were given an Anderson Tubular scraper and I had them sharpen the blades to a 60 radius and 5 degree rake on both sides. Answer that first and was this your first time scraping something?

Who taught you to scrape or are you self taught? Where did you hear about someone doing a chicken noise? How did you sharpen the blade? Do you have a straight edge? Granite plate?? How did you check the dovetails, dowel pins and a micrometer? Show us some pictures of the compound and your tools.

Rich

Grizzly class B granite plate.

I had to scrape in the straightedge first. It's 50 points/in. I used Grahm's napthol red for contrast, wiping most of it off with paper towel, then burnishing the high spots on the surface plate. With the right lighting, I could see what to do.

IMG-7074.jpgIMG-7073.jpg

The scraper is the one you described. I have a 1-1/4 in brazed carbide blade that I've sharpened to 115 mm with the usual 3 degrees. I also used a 3/4 in blade that's sharpened to a 75 mm radius and -4 degree rake on the end of the blade. I've also dressed the adjacent side (not the end) with 3000 grit diamond. The smaller blade is beveled to get into these small dovetails.
IMG-7069.jpgIMG-7070.jpgIMG-7071.jpgIMG-7072.jpg

The 1-1/4 blade is sharpened on both sides
IMG-7077.jpgIMG-7076.jpg

No, this wasn't my first time scraping something. I've been working on small projects for several years. For example, I scraped a fixed tool post for this lathe. I tuned up a 3" kurt-type import vise, getting everything flat and parallel. I've also straightened out a 6" rotary table and an adjustable angle plate. I've scraped the carriage of this lathe to the hardened ways. Oh, and the striaghtedge used in this project.

I'm self taught via the internet, so it's your students that were my teachers. I've read Connely, too.

I sharpen the blades with a 8-in diamond "lap". Someone on eBay is selling these machines for people who like to polish stones. I couldn't even get a price on an Accu-Finish at the time. So it's a little hacky but gets the job done. The 3000 grit wheel works for touching up the edge. I established the radius with a 360-grit wheel. There's no guides but estimating 3 degrees is easy.

(see next post for more pictures---limit is 10)
 
Here's a picture of the straightedge toward the end of the process, BTW. This is enjoyable work. Eventually the straightedge was flat enough to notice irregularities in the surface plate. So it's pin pointed but only grade B. I think it was detectable because the deviation in the plate is less than a very thin layer of blueing. But without the ink, the hinge test is very sensitive. The viscosity of the Dykem Hi-Spot, rolled to a fraction of a 10th, produces much more resistance than cast iron on granite. That's my theory. Now I need to buy a better surface plate.
IMG-7035.jpg

Here's a view of the male side of the dovetail, before measuring with pins, trying to get it flat enough for a reliable measurement. The female side got the same treatment.
IMG-7048.jpgIMG-7054.jpg

Then, measuring for parallel with a tenths micrometer. This is after 3 rounds of step scraping.
IMG-7053.jpg

Then matching the angle of the male part of the dovetail to the female side. This is where I was having trouble. Bringing the angle of the male side down took many many passes. I never did get the bearing I was hoping for. Here's where I left off:
IMG-7064.jpgIMG-7066.jpg
 
First off, thank you for the photo's, you did a nice job documenting your work. Last week when I taught the Class at All World Machinery in Roscoe IL 2 of the students who work for Bourn & Koch of Rockford IL brought that same Anderson Scraper. The blade was sold flat tip and the sides were rough from the silver soldering process and the flux was like yours. You need to lap your blade down further toward the silver solder. I had them lap the sides like you did but all the way to the silver solder. I re-read what you wrote about lapping the blade tip. You said "I've also dressed the adjacent side (not the end) with 3000 grit diamond". Did you sharpen the end?

When you look at your scraping there are several scratches that are caused by the corner of the blade. You rolled the scraper and blade dug in on the sharp corner. Also your depth of scraping isn't deep enough, the minimum depth should be .0002" from the low spot to the high spot that should be right next to it. It looks like you over stoned it and the percentage of contact is to high. I would sure like to see the 50 PPI on the straightedge. I wish Jeff Thiele would write as he brought a lathe compound to the IL class as he had a dovetail scraped with a straightedge that had 20 PPI and 40% contact . He has a nice hand scraper he made.

I had those 2 B&K guys scrape using that Anderson hand scraper, they had 2 of them. I had them practice scrape with them for about 4 hours. Then I let them use 2 DAPRA scrapers with a 20-150 blade sharpened to a 60mm radius. They were so grateful as that Anderson is way to stiff and long. Super hard to scrape with them. You need to make a DAPRA type or buy the smaller Anderson that is 1" wide aprox 15" long and has a sliding clamp. That is the only Anderson in my opinion that is ok. I also had them use a # 15 and 15-150 Biax blade. I have to look through my photo's and will add them later.

You need to sharpen your blades to a maximum of 60mm and on the dovetails use a narrow width blade 5/8" wide or 15 mm. On narrow ways you use a narrow blade, not a 1 1/4" wide blade. I'm going to email Jeff and ask him to comment. I'll write a little later too.
 
Last edited:
To answer your question on time - I would think I could scrape one your size a day and 1/2 or 10 to15 hours. It being your first one 30 hours sound good. I don't want to hurt your feelings, but you need to learn to scrape the checkerboard pattern as it will look better and speed up your work. Also make of buy a better scraper. When I taught the Norway class 2of the students had those same Anderson scrapers. I always take several of the BIAX /DAPRA hand scrapers along I have different lengths for tall, medium and short heights. One size is not perfect and that one you have is way to long unless your 7' tall. In Norway thats all they had and we put those Andersons on an anvil and hammered them flat so they bent or were more flexible. The B&K guys had me mark in the DAPRA catalog what blades to buy. They were going to make Dapra types as they drew up a sketch of them. I would mail you one if you want to test one. Or message me your email and I can send you a drawing of one.
 
I re-read what you wrote about lapping the blade tip. You said "I've also dressed the adjacent side (not the end) with 3000 grit diamond". Did you sharpen the end?

Thanks for the feedback.

Yes, I regularly touch up the end. When I first started, I left the bottom face of the carbide rough. Unfortunately, that surface defines the section of the cut, so my first attempts didn't look very good---lots of scratches. Maybe the DAPRA blades are smooth enough it isn't an issue.

I'll pick up a DAPRA scraper. Here's the shopping list
  • Scraper with BIAX adapter P/N BB20016
  • 3/4 x 3.5 blade P/N 20
  • 3/4 x 6 blade P/N 20-150
  • 5/8 x 6 blade P/N 15-150
 
I don't like the adapter that comes with it. That makes the handle to stiff if you usea Sanvik insert. Keeith Rucker cad camed the handle and you can make one for under $50.00 - The Dapra one is over $150.00. You can made 2 lengths handles for the price of a new one.
 
Richard commented on the scraping I had done on a dovetail and the scraper I had made and wanted me to write something about it.

The following is what I have and how I use them. That does not mean it is the only way or the best way.

Below are pictures of the hand scrapers I use. On top is a solid shank Anderson scraper. In the middle is an Anderson tubular scraper. The bottom scraper is a Dapra scraper with the plastic handle replaced with a wooden one. The picture does not show the 5 inch sanding pad that can be attached to make a “body” push comfortable.
P4010387_DxO.jpg

Here is a view of the modified handle of the solid shank Anderson scraper. The notches allow it to be gripped between my fingers as shown.
P4010393_DxO.jpg
P4050417_DxO.jpg

Here is the scraper I believe Richard was referring to. It is made up of a Dapra 130 M toolholder and a Dapra KL 170 toolholder extension, which has a wooden handle attached. The blade in the picture is one that I made. I also use it with a 6 inch Dapra blade, which makes it 3 inches longer I already had the toolholders, so it was easy to assemble to see how it worked. It would be easy to make your own and not use the Dapra parts. That would allow you to have a series of holes to secure the handle, and vary the distance from the handle to the blade tip.
P4010396_DxO.jpg
P4010398_DxO.jpg

The scrapers I prefer to use for small dovetails are the solid shank Anderson with the handle modified for a between the fingers hold and the Dapra 130M/KL 170 with the vertical handle. Both are short, and being closer to the work helps me see where to scrape. More important though, are the handles. The “between the fingers” hold especially positions the scraper the same way every time, and it becomes an extension of my arm. It is much easier to know and control the tilt of the blade in relation to the blued surface. A controlled tilt of the blade is what I use to scrape to the back of the dovetail. I get a better feel for the tilt of the blade from my right wrist than from the fingers of my left hand wrapped around the shaft.



Here are the blades I use for small dovetails. The upper two have the top of the carbide beveled to allow the cutting edge to reach the back of the dovetail. The bottom two are ones I made. The Biax blade and the middle blade have a radius of around 15 to 20 mm in the center and increasing to about R 10 to 15 mm on the edges. They started out at R 90 or R 60 and were ground down until they reached where I wanted. The shape is closer to an elipse.

The very small blade has a radius of around 10mm. The only time I use it is if I have a specific spot to pick off or where the other blades do not work well.

The rectangular pieces are shims which allow the home-made blades to be used in any of the handles. Clamping the blade with a tightening screw works best as I am both tilting the blade and moving it sideways. The tubular Anderson scraper can have the blade shift with the sideways movement. With the right shim the solid shank Anderson scraper can be made tight enough to work, but it may take several tries.
P4010401_DxO.jpg




The diagram below represents a 3/8 in dovetail with one of my dovetail blades at various angles. The large arrow is in the middle of the blade, and the small arrow is where the blade reaches the back of the dovetail.

A: The center of the blade reaches the back of the dovetail, there is no need to tilt the blade- but unless you move sideways as you go in, the scrape marks do not cross.

B: You can go in straight with no tilt – but the corner of the blade hits the back about ½ way through the dovetail. Because the sides of the blade are higher than the middle the blade will hit sooner than the diagram shows- and you could put scratches in the vertical surface. You could tilt the blade to bring the corner down- but the edge digging in is very likely. The scrape marks cross at 90 degrees.

C: You can go straight in and get more of the width scraped before you hit in the back. The scrape marks cross at 58 degrees. If you tilt the blade you can scape at 45 degrees all the way to the back- but tilt a little too far and you could get a scratch.

D: You can go straight in and cut a little further back than with C- but the scrapes now cross at 45 degrees. Tilt the blade and you can scrape at 45 degrees for 90 degree crosses. You can reach all the way to the back with less risk of the corner digging in than in C.

For scraping at 45 degrees using the side of the blade (C and D) you need a sharp blade so it will not want to skate out of the cut.

With a dovetail every cut at the edge is “into” the edge and I find those challenging

On a good day I can start the cut as in B, and during the cut tilt and rotate the blade progressing to C and then D.

Most days are not good days. The center is easier to cut -so I tend to end up with the rear 1/3 taking blue, a dip in the middle and the outer 1/5 taking blue. I will use D at 45 degrees for the rear 1/3. For the front edge I will hold the blade between my thumb on top with my finger underneath and up against the dovetail edge, and rock the blade until the contact point is right at the edge, then make a short stroke, and move to the next high spot.
P4050419_DxO.jpg
 
Jeff points out that there several ways to do it. I always say I don't care how you do it as long as the quality and time are close to the same. I was hoping Jeff would show the dove tail he scraped down in IL as he did a beautiful job. Keith is welcome to come to any of my classes -no charge as he is a journeyman scraper as he has rebuild many machines. He isn't afraid to share his opinion either. He one of my top 100 students out of approx. 40 years of teaching.
 








 
Back
Top