Sine Plate Roll Pin Enagement

[torinwalker]

Watched a video of a guy re-surfacing some v-blocks on a 6x6 magnetic chuck/sine plate and thought, damn - I want one of those. Shopping on ebay and kijiji, they're pretty expensive... then I re-discovered a 4x4 suburban mag chuck gathering dust on the shelf. Figured making a sine-plate for this mag chuck would be an interesting project.

But then my OCD set in and I'm stuck trying to decide on a method/mechanism for locating the roll pins. One the one hand, simple shoulders with a 45-degree chamfer (first image) allow one very easy vertical and horizontal placement of the roll pins secured with screws at 45-degrees. The chamfer doesn't really contribute anything except give one a flat spot to drill if coming in from the bottom, and doesn't even touch the roll. On the other hand, I've seen a number of designs that use chamfered slots where the pins rest against the chamfered edges of the slots, fastened with a vertical screw. This second method (see second image), in my humble (but inexperienced) opinion seems less precise in that the height of the pins as well as the distance between pins is determined by the distance between chamfers which would be pretty fussy to dial in for both height (relative to the other pin) and distance between them.

Method 1 - I'll call it the "shoulder" method - screws driven at 45 degrees into the roll pin - and my current preference given the ease with which pins can be aligned for both height and distance.

Pros: Contact points normal to the two directions for which I'm trying to achieve accuracy - one determines height from the plate surface, the other determines distance from other roll pin. Dead simple, and dead nuts accurate placement of the pins. Both can be milled and ground without angling the part.
Cons: Requires milling fastener holes at 45-degree angle which isn't so bad, but requires angled setup and increases difficulty of "manufacturability" (A minor point really. Not really sure why I'm whining about it.)

Method 2 - "Chamfered slot" method - Not fond of this design, but many sine-plate makers use this.

Pros - fasteners are drilled straight down (or up) into roll pins, leaving more plate surface area for mounting holes. Simplifies manaufacturability? Why do so many make theirs this way?

Cons - height of pins varies depending on distance from chamfered edges and thus requires more fussing about to get the slots (pin heights) and distances exactly equal (drawing exaggerates this to make a point) - I am really not fond of this design

The first method allows for very precise location of the pin, but awkward placement of the fasteners whereas the other is easier, but requires very careful grinding of the 45-degree chamfer to achieve accuracy which, as you might have guessed, requires a sine plate (or an angle dresser, neither of which I have.)

There are other methods, such as bolting on pivot plates, but I'm not considering those designs.

Has anyone here tried both approaches? Can anyone see other advantage to method #2 that I have failed to see?

 

[torinwalker]

Wow. No comments at all? If this were the Machinist page on Facebook, I'd have 149 highly opinionated remarks, 17 put-downs, 6 guys flaming each other over lord-knows-what, 3 memes, and and at least 2 pictures of women's breasts by now. Are we Canadians really so shy? TLDR?

 

[PeterT]

I'm no expert in this area but it seems like the angles you have capturing the cylinders might be hard to cut & dress accurately vs. some alternatives. Anything that affects center to center distance of the cylinders (laterally or vertically) will affect the resultant angle on a given thickness gage block all other things equal. Now if you build it and it comes out to some arbitrary oddball number like 5.006" vs 5.000", then its probably not a big deal to calculate your own custom table on that basis to tweak the required stack-up. It wuld be unique to your plate but different than what you would find in Machinery Handbook. I haven't sat down to see how much variation equates to how much angle difference. But I think if you made at least one of the notches a perpendicular/square cut, you would stand a better chance of controlling the variables better.

The other thing I notice is some of these plates have integral arms to preserve the position. Seems like some are light duty for example surface grinder. Others are a bit more robust like what might be intended for mills. So that might impact your design too. Somewhere in my travels I saw this discussed at length. Tom Lipton (oxtool) maybe?

Maybe this will provide some insight.

 

[torinwalker]

Actually, thanks, I watched that video some time ago (part of the reason I'm getting interested in sine bars). His straight shoulders are very easy to dress square and as he annotates, they are easy to get on target for both height and distance. That is what I was alluding to in my first image. I've removed the 45-degree cut for now to highlight the similarity. Like in Gotteswinder's design, using a simple right angle to capture each roll-pin is, I think, much easier to get dead-nuts accurate. In contrast, the other designs I mentioned use chamfered slots (or even v-grooves) that must be dressed at an angle, and one must be careful to dress precisely the correct width on each slot to set the roll ping at the same height and distance apart. I can't seem to figure out what advantage those have over the ones in the video, or the one illustrated below. That's why I posted - hopefully someone might be able to identify advantages in the other design that I cannot see. Suburban Tools, for example, makes their sine plates using the chamfered slot design, but they don't explain why they chose that design.

 

[PeterT]

Hard to know. The 3 facet slot removes less depth from the plate & I guess leaves more meat to the plate. OTOH after watching their video series, I suspect they are rough milled, sent out for hardening & then (in-house) surface ground everywhere. So hard to imagine heat treat distortion is a factor. Suburban is top shelf stuff with all kinds of different grinders, so unless you have grinding equipment capable of finishing this particular geometry, maybe opt for the square design you can maybe better control dimensions?

I cant visualize any other advantage between the 2. They both look like they have same movement range & overall setup footprint based on similar block thickness.

ps - notice in Stefan's part-2 vid he threads the cylindrical pins in from an angle to suck them into tangent surfaces simultaneously. That's maybe another consideration for you. What will you choose for a pin material that you can do some basic machining & then harden without distortion? If you have precision hardened pins, this wont be possible without annealing. If this is just a project for home use & ends up within the tolerances you work with, annealed drill rod might be acceptable? Although I've noticed when lapping stock that the stated tolerance can be not just diametrical but elliptical shaped. Generally its pretty on-spec stuff though.

What were you planning on making the block itself from?

 

[PeterT]

More room for stack up?

 

[PeterT]

This (Tom Lipton) vid isn't the one was I trying to recall, but another take on a specialized sine block with some cool design features. It kind of depends on what you want to do with the tool. Is it just to set up stock in a mill vise at shallow angles dictated by your gage blocks? Or if you take it a few steps further, it could actually encompass a mounted mini precision vise & then you are set up to grip stock at any chosen angle. The one in the video looks blocky but he menstions he can hold it at 90-deg which actually gives you more utility range.

I've heard these are not good for machining setups like a mill. They may drift without support & probably not cool to leave your gage blocks taking the load. Surface grinding is smaller down forces I imagine. I have a Chinesium angle block set with notches to hold rectangular stock at angles. They are quite low profile so good for mill vise work, but are kind of limiting in other setups

 

[torinwalker]

It's not clear from the 2D shot, so here's a 3D image to show the clearance for the blocks, and those little ovals on top are 45-degree drilled and counterbored holes for the screws that "suck up" the roll pins into those 90-degree corners. Only three shown because when I started working on the others, I started thinking of other ways to fasten the roll pins to the places (e.g. chamfered slots) and that's when I posted my question. I'm on the fence about the design. The T-shape of the top (sine) plate is needed to give it shoulders for roll pins, and leaving all that meat there gives it extra rigidity. The bottom was milled out to be complimentary and, because it too needs shoulders for the "shoulder" roll pins. Still investigating different designs.

This is going on the surface grinder to hold a mag chuck and will eventually be used to fabricate precision v-blocks For milling, I'd rather use a heavy-duty angle plate. You're right - don't want gauge blocks taking downward force from cuts.

I have a Boyar-Shultz H612 surface grinder, a tool-post grinder for cylindrical grinding (which is how I plan to make the high-tolerance roll pins), and I am working on getting a little muffle furnace for heat-treating. My interest for making tools rather than buying them is growing. I'll probably make everything from 4140 because it's cheap (compared to tool steel), readily available, and I happen to have a bunch of it already.

 

[torinwalker]

Thanks, I saw that video too. A Sine dresser is also an interesting idea. I saw a Yuasa radius and angle dresser going for less than $200 on ebay that looked like it could be cleaned up, but now looking at that video again, I could build a little sliding rail and do something very similar.

 

[PeterT]

Nice. You are equipped miles beyond my shop. I look forward to what you ultimately come up with. I think making your own tooling & jigs has to be one of the more rewarding aspects of the hobby (or business if it turns into that).

ps you might like Doug Ross stuff. His YouTube is kind of sparse still, but I started following him on another forum (Homemadetools,net) where he has posted some really cool tooling.
https://www.youtube.com/channel/UCDD5ZTKuj841Str6iW5XuUg

 

[torinwalker]

As I post this link here, I realize this post (one I researched to look into slotted plates) and your youtube channel are the same guy. Here, Ross makes a sine plate like the Suburban style except he calls the slots v-grooves and mentions how difficult it is to match up the roll pins. Thanks for pointing this guy out. When I first looked, he was just some guy making a sine plate. Now that I've seen his video channel, I realize how much deeper he is in the rabbit hole. This is good stuff but very difficult to achieve until one is practiced. Don't let all the "fancy" equipment fool you. I'm nowhere there yet.

http://www.homemadetools.net/forum/sine-plate-63316

Just yesterday, I was grinding my magnetic chuck nice and slow at .05" per pass, trying to get all the little -2's (tenths) out when as I approached the last quarter of the plate, the wheel started taking off a lot more material. At first, I wasn't sure why. I kept feeling the plate which was still cool to the touch (it might have risen by 2 or 3 degrees from when I started), but on the end of the run (toward the front of the chuck), I measured a drop-off of -5 in the front-center, -2 at the front corners, -8 at the rear-center, and -2 at the rear corners. Turns out that after years of sliding back and forth in the center, this grinder is cupped so that as I approach 1/2" of the extreme edges, the table lifts up causing more material to be ground away, leaving me with a 5x10 or maybe a 5x11 grinder rather than a 6x12. So long as the work is smaller than the reduced envelope, I think I'll be good.

On another occasion, I realized mid-way through a grinding operation that I forgot to prime the automatic oiler. As soon as I did that, the table rose by a few tenths making the wheel cut deeper part-way through the job. Never do that... complete the job, lube the table, then start again.

I'm still trying to achieve perfection on the table, but I haven't got the knack of it yet. Hopefully I won't have ground the chuck down to the nub before I figure this out


Torin...

 

[torinwalker]

I'm so ashamed... after all this talk of making my own sine plate, investigating different designs, drawing up CAD models, and even pricing out metal, I saw this on Facebook:

Facebook Find

Yes, indeed, that is a Suburban 6x6 "SineSet" sine plate. He asked $60 for both the sine plate and Bridgeport handle, so figuring this (including shipping) is cheaper than the 4142 material I priced out, it was a no-brainer to buy instead of build.

Received it yesterday evening during mail call. Pulled it out and boy, is it sweet. Some numskull very sloppily engraved his name in several places, and apparently used a very pointy instrument to tap the baseplate around, but otherwise solid, smooth, no dings or burrs despite being shipped exactly how pictured with the handle sitting directly on top of the plate with no padding between.

Nameplate Wiped Clean

Then I discovered why it was only sixty-bucks. It wouldn't open! Figuring it might be corroded, I disassembled. Nope. Everything is ok - The plate-hinge interface has no observable corrosion and is smooth to the touch. The roll pin hinge shoulders rotate silky smooth and have very impressive tight-tolerance fits. One of them was a little sticky at certain angles (the reason will become evident later) but not enough to prevent the plate from opening. Cleaned and re-lubed with spindle oil then reassembled in exactly the reverse order. Still wouldn't open. I suspected the screws next - they did seem a little long - and pulled one out to compare its depth against the pin - it looked like it would hit the center pivot. Put it back but loosened all the hinge screws and it opened. Removed them again and lined them up. One of these screws was taller than the others and missing the flat end grind. I started looking for the missing shortened screw.

Perhaps the same guy who scrawled his name all over the place, disassembled the plate at one point but reassembled it incorrectly. Looking at the first photo, the screw in the bottom right corner of the plate attaching the top plate to the rotating part of the hinge was swapped with the middle screw on the left holding the front roll pin in place. Where the front roll pins are solid, drilled full-depth and use regular length screws, the roll-pin hinge shoulders have the center of the roll pin hinge pivot passing through them and those screws need to be ever so slightly shorter so the screws don't intersect the pivot.

After reassembling with all screws in their proper place, the sine plate opened exactly as I saw in Don Bailey's videos on their modular sine-plate system video - tight enough to stay in in place wherever you put it, but not so tight that it can't be moved. Now, time to measure flatness.

On the surface plate, a test indicator shows .0000 at the back and .0002 in the front. More if I don't put my hand on it because there is a certain springiness to the plate. Another you-tuber mentioned Suburban plates are springy like this and need a little assist from light springs, or weight to keep them in place, so I guess this is normal, and II'm not bothered by it. Even though .0002 is within their spec, it still bothered me, so I re-checked the roll pin to see if perhaps it would behave differently if flipped around. The slight marks on the roll pin matched with the complimentary marks on the plate, so it was in the right position. I then checked the torque on the roll pin screws, and let up a little bit. The plate then closed without any springiness at all. Much better. I then re-torqued the front roll pin while under the indicator and watched it jump up and down by +/- 2 tenths. Like with the other screws, I found a torque that would keep them in place, but without using Abom79 strength... lo and behold, it measured perfectly flat across all four corners.

Rear corner at the hinge measures .0000

Front corner measures .0000 (forgive the parallax)

Next will be mating the non "SineSet" 4x4 mag chuck to this plate. Need to make a new baseplate for it matching the hole pattern above, pin and screw it to the mag chuck, then square it up.

Sorry to disappoint anyone who wanted to see a sine-plate build, but the sine plate was a means to an end - I got the sine plate to make precision v-blocks and other tooling, so stay tuned.

BTW, if anyone wants a badly brazed but original bridgeport knee handle and you're in the GTA, PM me. Yours for free. With a cut, drill, tap, and turn, it could be made quite new again.


Torin...

 

[Janger]

Look at all that CAD you did. Lots of good experience and practice there. Good posts and I like your screenshots. Can we see your surface grinder?

 

[John Conroy]

You got quite a bargain. Suburban makes nice stuff. I have one of their angle plates and it is very accurate.

 

[torinwalker]

Janger, it didn't take very long at all to draft those up. 20 ~ 30 minutes, so nothing lost. I don't know how many cad designs I've come up with only to abandon them because the design wasn't viable, moved on to other projects... if I had a nickel for every cad drawing...


The grinder is, I dunno, a bit of a sad story. Got it from a guy who I think - and I'm not certain of this - got very sick and had to move fast, and therefore dumped all his stuff on kijiji. Really nice though. Never did find out what was wrong with him, but if I had to guess, I'd say it was probably terminal. Anyway, he got it from George Brown college where it got good use but was never cleaned in the history of time. I can't complain. Got it for $800, came with a couple of wheels, some adapters, two mag identical mag chucks, and I had him throw in a 5C collet indexer and a bunch of metal/plastic stock. When I got it home, I took the table off and observed that it was dirtier than a coal mine, but the lube was still working (even though it dripped inside the cabinet instead of into the waste can behind the machine (?!) and the factory hand-scraping was still intact across all surfaces, so I think I got quite lucky.

I've had this machine for about two years now but it rarely sees any use. Only recently have I become interested in grinding precision parts. I don't know if these grinders are good, bad, middle of the road, but it's one of the more popular small grinders and has a large downfeed wheel with large graduations (.001" is nearly 1 cm of wheel circumference) and a tenths vernier allows one to very easily dial a tenth at a time. Being able to control down to the tenth was the most important factor when selecting a grinder.

Today, was my third and finally successful attempt to grind the chuck. As I'm learning this, there are two camps on grinding a chuck - one says use small Y increments and let the leading edge of the wheel take the brunt of the grind allowing the rearward edge to clean up as it trails behind. The other says to use 50% of the wheel per pass and take only a tenth (max two) at a time. Trying both techniques, I discovered that the first method tends to leave my horizontal edges slightly curled upward relative to the center, and if I take more than a tenth at a time, the wheel glazes, increases heat, and by the time I reach the front edge of the chuck, it starts cutting deeper. While the finish looks better using the former technique, the latter gives better accuracy.

The map below is from the fourth or fifth pass at 50% wheel increment taking 2 tenths per pass max for the first three passes just to get the chuck more or less evened out, then the remaining couple of passes at 1 tenth per pass. Mapping is quite finicky when using a tenths indicator because even blowing on the indicator (literally) can move it a couple of tenths. I always start in the middle and go left, then right, always verifying as I pass back through zero, then north to the mid-way mark, map out that row, then do top, then work my way back down to the bottom. Since the tenths indicator bounces around quite a bit, I have to frequently revisit center to make sure I'm still on the mark.

The other important aspect is moving the indicator - when you're standing still, there's a certain stickiness that causes the needle to jump when it gets moving. One must make sure to very slowly accelerate to keep that needle on zero, otherwise you'll get a false reading. Anyway, with practice one begins to understand the numbers even when the needle jumps around a bit.

Mapping out the chuck elevation with a test indicator (Note: the map is measured in TENTHS, not thousandths.)

This sort of pattern repeated several times, every time leaving the edges curled upward until I discovered today a little technique that seems to work to correct the edges: For every pass, wherever there are sparks (at the left and right), I make multiple passes over that area until it totally sparks out. So for example, as I enter the left side of the chuck in the top left corner where it shows 2 tenths, I'll get some sparking until a third of the way in, then I'll reverse back to the edge, then pass over that area two, three, sometimes four times until no more sparks are produced. Then I'll advance over the center (0) area to the other side where it sparks again. A couple of passes over that area, then I return to the left side, decrement Y by 1/2 the wheel diameter (so .250"), and do it all over again.

It takes a couple of hours to creep down a tenth at a time, but the results are worth it in the end.

I had to buy a new tenths indicator on Friday because my current pocket indicator is... not as snappy as it should be. I bought it used for a song, but always had to dial it in more than half-way through its travel to get the mainspring to have enough tension to repeat properly (the mapping above was done with that indicator.) But the new Mitutoyo 513-403 is superb, 1/3 the price of the Interrapid, and I'm happy with the purchase. Using the new indicator today gave me more reliable measurements and actually confirmed what I was seeing. So after trying the aforementioned grinding technique where I pay extra attention to the edges, I ran one final pass over the chuck, measured again and got... if you can believe it... less than +/- 0.000075 across the entire chuck.

Of course that doesn't mean much because it will always read perfectly sitting directly under the wheel - the real test is to place several test coupons in each corner, and at center, and grind them all at once and measure. I'll be doing that tomorrow while my kids are downtown at their aunt's poetry reading. I'm really disappointed I'll be missing the poetry.

This is my completed mag chuck chuck, ground dry to less than +/- 0.0001. Now I'm ready to grind some parts.

New indicator to replace the old one, right.

Torin...

 

[torinwalker]

You got quite a bargain. Suburban makes nice stuff. I have one of their angle plates and it is very accurate.

Thanks - it was too good to pass up, even if it ended up in worse condition, it would still be worth $60. I was quite fortunate.

Something is going on in Windsor... perhaps a major manufacturer closed, or something, because suddenly Kijiji is filled with all sorts of precision tools. I've got my eye on a Suburban "whirly jig" with baseplate and tailstock for only $500, but I can't use it. I mean, I could use the whirly jig for cylindrical grinding on the surface grinder, but I'd have no use for the baseplate and tailstock. What a shame.

Torin...

 

[Janger]

Nice description of your surface grinder. Quite interesting process.