Making a rip fence system.

[dave pant]

My uncle owns a machine shop and I usually end up there when coming up with some sort of plan.

My new plan is a super smooth and accurate fence for my table saw.

I am doing this because I already have the stuff and I am guessing it will cost me about 75$ to do.

Anyways I am using Thomson linear ball bearings and case hardened 1" diameter shaft on the front and back of the saw.

One thing I been thinking about was the length of the fence itself. I think the Biesemeyer fence is like 42" long and overhangs over the back of the table.

Why is this? I was planning on making mine only 34" long because my saw is 27" deep.

Since this system I am making is going to be pretty sweet I wanna make sure I do it right and am able to use it on any saw for the rest of my life heh.

Should I make my fence 42" and if so why?

Thanks =)

Dave

[david brum]

Longer fences are especially helpful when cutting large sheet goods. As the bulk of the wood gets past the end of the fence, it gets harder to keep the wood from tilting away and ruining the cut. The longer the fence, the easier it is to make those straight cuts. I personally wouldn't mind if my Shop Fox fence were even longer.

[Mark Rakestraw]

+1 on what David said for sheet goods. Then ripping solid wood however sometimes tension is released and the board starts to move beyond the cut. In this case it's helpful if the fence stops immediately after the blade, other wise it pushes off the fence and ruins the cut. European fences and the old Unifece were adjustable so you could move the fence fore and aft. I have a version of that on my saw and really like that feature.

[dave pant]

Thanks guys I think I will just keep my fence at 34" then. I dont usually cut big sheets and this seems like it would be ok for both types of work.

[Mike Schuch]

+2 except I really like having a fence that extends in front of the table too when I am cutting 8 foot sheet goods. It makes for a MUCH straighter cut.

If it were me I would make a fence with a 34" face that can be easily removed and replaced with a 60" face for those occasions that long work pieces present themselves.

What are you going to use for locking the fence in place for a cut? Are you going to have a lock on the front and back rails? How do you plan to adjust the fence for square? I assume you will have two linear bearings spaced some distance (maybe a foot or so) apart on both front and rear rails to maintain square? If so what would be the pro's and cons of just having the whole fence be a foot wide?

With linear bearings on the front and back it seems like the difficulty will be making the rails perfectly parallel once mounted on the saw so the fence doesn't bind.

What kind of saw is the fence going on?

How long do you plan to make the fence? Are you going to use open carriage bearings so you can mount and support the 1" rod at various points along its span? The rods might also make a nice mounting point for movable front and rear support tables?

[dave pant]

Ill post this picture. They say a picture is worth a thousand words right =)

187260

1.I would have to buy 2 more bearings if I made 2 fences.
2. To lock it there is a clamp like thing that when I turn a knob it locks down on the rail.
3. I am just locking the front bearing on the rail.
4. I will adjust the fence square to the table and then to the front and back of the blade.
5. I am only using 1 bearing on each end of the fence.
6. The fence is only going to be 2.5" thick and 3" tall with extruded almuminum on the top for accessories.
7. I am using round shaft for the rails these are pretty tolerant of misalignment. As long as I am really close they should glide smoother than any fence made.
8. The fence is going on a 12" old Craftsman floor saw.
9. I plan on making the fence 34" to take it off I will just slide it off the end of the table.

[Chris Parks]

I think the fence needs to be modular to set up the alignment. My idea would be to make the carriages each end and then attach the fence to these with an adjustment of some kind between the fence and the carriages. One end has the adjustment the other end must have a pivot, not to do this would defeat the use of the bearings, it would make a really sweet system. The fence face could be made either extendable like on some compound sliding saws or a quick release attach system for different lengths.

[dave pant]

Im not following you totally but I think I kind of do. The picture that I made doesnt actually have everything on it. I only put 1 bracket on each end but actually it will have 3 that are adjustable for hieght.

[Chip Lindley]

Einstein said: "Everything should be made as simple as possible, but not simpler." If you are adding two linear bearings to the front rail, you are on the right track.

As with the Biesemeyer T-square fence, the wide stance of the front angle provides lots of stability, registering the "Tee" squarely against the guide tube. Also, the outboard ends of the "Tee" have adjustable pads for squaring the fence exactly; not just almost!

Another BIG concern (for me) is that there seems to be no ready way to add a self-adhesive rule and pointer on your Thomson fence to measure rip width quickly and accurately.

The fun is in the experimenting. Make your linear bearing Fence as you see fit. The obstacles you run into will have to be overcome, OR the project will have to be scuttled as close, but no cigar! Just food for thought.

[dave pant]

Another BIG concern (for me) is that there seems to be no ready way to add a self-adhesive rule and pointer on your Thomson fence to measure rip width quickly and accurately.


I have a Digital Wixey scale for measurements.

[Chris Parks]

Second time around, hope I don't lose it this time. The bearings have to be concentric to the rails to move freely but the fence needs calibration to align it to the blade an mitre slot. The chances of doing this while building it and then bolting the assembly are somewhere between nil and damned all so it needs to be allowed for. As a ruke you need a bit of toe out on the outfeed side so the timber does not become trapped between the fence and the blade.

Make the bearing carriages with adjustment on the out feed one that can move the fence sideways an appropriate amount and the other end will have to have a pivot because you don't want to pull the bearing carriages out of line with the rails. The pivot will have to be lockable after adjustment or it will move all over the place, DUH. It only needs to be locked at one end. The bearing carriages may need wiping seals to keep the crap out, suck and see I guess but I would put some in, they only need to be felt seals not neoprene as the felt ones will have less stiction.

[bob hertle]

One comment Dave,

You need "wheelbase" at the front, IOW, 2 bearings spread apart by a reasonable, say 10-12" distance. It will do 2 things: Make it a whole lot easier to square it up, and provide moment resistance, or the ability to stay in square after you smack the far end with a sheet of plywood. Don't rely on a single Thompson bearing to do this!

Regards
Bob

[David Keller NC]

Dave - Can't answer the question (rip fence length) for Biesmeyer's design, but there's a feature to a uni-fence that you might consider copying since you're making your own. A uni-fence allows sliding the fence perpendicularly to the front edge of the table. By doing this, you can have the back end of the fence terminate just before the blade, which means that you can use the fence without extra clamps/jigs as a stop when cutting the shoulders of tenons (or making a number of small cut-offs of exact length). This feature also allows one to use an auxillary infeed table, with the extra fence length on the infeed side, which makes aligning glued-up panels and sheet goods for the intended cut very accurate indeed.

[Jamie Buxton]

One additional feature that is really useful is the ability to easily remove the fence. I use my table saw for cross cuts (with a cut-off box), and the fence needs to come off for longer pieces. With your choice of linear bearings, I think they'd have to stay in place, and the fence would need to separate from the subassembly that holds them. But when it goes back on, it would need to return to be exactly parallel to the blade. Ideally, the removal and replacement would be done without tools, in just a couple seconds. The Beisemeyer and Unifence designs do.

[dave pant]

I think Im with ya this time Chris =).

Well this is what my plan was.

currently I know the blade and miter slot are parallel.
I was just going to sit the fence in the spot where it belongs which is butted up against the fence.
Then slide the bearings under and bolt then to the fence in that spot.

@Jamie -> My fence will be a little more annoying because I will have to slide the fence off the end of the rails instead of just lifting it off.

[Greg Peterson]

Getting the fence set parallel, or for whatever degree of toe out you require is a delicate procedure. On my T2 fence, the difference between parallel and .001" off parallel is about one quarter turn of the allen screw on the base. Requiring the attachment point of the fence to the carriage assembly to perform as a alignment point can likely be accomplished. But it is equally likely you will find the alignment procedure challenging. If you do decide to leave this element of your design intact you should consider some mechanical means of positively locking these two units into position. Do not rely on a couple of nuts to hold the fence in position indefinitely. As you push your stock along the fence, the further this pressure is applied from the carriage, the lateral stress placed on fence/carriage joining surface increases (think mechanical leverage like a cheater bar on stubborn bolt). The fence will gradually slip out of alignment.

My initial impression of your design is there are to many moving parts. I also question whether or not saw dust will impact the smooth operation of the fence.

I am confused as to how you are able to support the rail in the center without interfering with the bearing assembly.

A bearing assembly that simply rode on the top half of the tube would be sufficient and would allow for quick and easy removal of the fence.

How do you plan on locking the fence into position?

I certainly don't mean to discourage or poo-poo your ambition. Go for it, especially since you have ready access to a machine shop. You have the 'right stuff', by which I mean there is certain profile that ascends to hobby/craft of woodworking. Many talents find a happy home in this hobby/craft.

Regardless the degree of your success in this project, you will undoubtedly apply the insights gained to future endeavors. One thing is certain, not building this would be a bigger failure than building it only to find out the design has some flaws.

[dave pant]

Thanks for the reply Greg but why wouldnt I just make it perfectly parallel to the blade? I am not familiar with using any toe out when aligning a fence.

The main part of the saw is steel bar that is 1X2" in the middle which will have a countersunk hole for the bolting to the bearing I dont know how it would ever move after being tightened.

I am not sure if I explained it very good on what I am doing but the only moving parts will be the 2 bearings moving on the rails.

These bearings are rated at 780Lbs a piece they are used in making CNC machines and I am hoping 2 of them would easily support the weight of the fence.

I agree I wish I could just lift the fence off instead of sliding it all the way to the end to take it off however they dont make that style bearing =(. I suppose I can live with that since I rarely take the fence off anyways.

I am using an aluminum bracket thats connected to the front bearing that has a hole in it for the shaft and a slit at the bottom where a bolt will either compress the hole smaller to tighten or let it open back up to relieve pressure when I want to slide the fence.

[keith micinski]

.001 or .002 toed out at the back doesn't affect your cut any and gives you a little play so that you don't get a piece trapped in between the fence and the blade or excessive burning. I really think you should take a look at the Delta unifence. The main front bracket that has a removable fence face is absolutely what you should be doing. Plus again you can have different length fences if the need ever arises. You could do this by using a t-handle to bolt the 1x2 tube to the bearing instead of a counter sunk screw. Also you should build in adjustment to the fence for the future because things move and wear over time. Trying to build everything square to a thousandth is going to be impossible.

Last edited by Greg Peterson; Today at 12:13 PM. Reason: I sounded like a grumpy, critical old man.

This is easily the funniest thing I have ever read in this web site. Thank you immensely for writing this.

[dave pant]

Ok Keith I might do that toe out thing it makes sense to me. You just gave me a great idea though. Since the walls of the fence will be made out of MDO or MDF with phenalic covering it I could just make 2 of them and they also will be just screwed into the main beam of the fence. So if I wanted a long fence I could just unscrew the short fence off and screw on the long one =). That way I wont have to make an entire new fence and just attach the longer one. OR! I could just make some pilot holes and put some brass inserts in the normal short fence and just make it so I could screw the long fence onto the short one if I ever needed to.

[keith micinski]

Another thing, and I am not trying to hassle you, but after looking at your picture again I now understand how you are going to lock the fence down and it looks like you are going to have to use a wrench to tighten and loosen the bolt to crimp the bracket around the tube. This seems like a really big hassle and I wonder if you need to line the fence up if it will move at all while you are tightening this up. I guess you could use another t-handle to lock the fence down but I think you should look into some sort of a cam system like all other fence's use. Also I love my Wixey read out and wouldn't do with out it but I use my regular tape to get the fence close and then dial it in with the Wixey. Plus I use it as a back up Reference to the Wixey just in case there is a calibration problem, which did happen to me recently.

[dave pant]

I have a 1.5" diameter rubber coated knob that I plan to use not a wrench. I have used
this knob before on a similar project using 80/20 extruded aluminum I dont think that it will be a problem. I only have to turn it like 1 turn to lock or unlock. BTW your not being a hassle at all! I appreciate your input =)

In fact! you just helped me realize that the knob will be too wide and I will have to make the clamp even with the front of the bearing or it wont fit =)

There is only 1/2" of clearance from the center of the hole in my clamp to the edge of the bearing!

Thanks!

[keith micinski]

Well I really will like to see this when you are done but I have another thought. For this to slide smoothly you are going to have to have the front and back rails in perfect alignment all the way from one side to the other and I wonder how easy that is going to be because I bet if I went out and measured my top it wouldn't absolutely square which is what you'll need. If you have access to a machine shop why not mill a reference face on each end of the 1x2 tube that lock onto the top of the bearing so that you know the bearing's are in perfect alignment with each other? Then you could use a set screw to hold the fence faces themselves on and they could be removable that way.

[dave pant]

I dont think it will need to be perfect because even if it was off lets say 1/4" off (which is ridiculously bad lol) the front bearing would just always be 0.25" in front of or behind the rear bearing but they would still slide exactly the same distance when the fence moves. Also even if one of the rails were higher than the other it wouldnt matter much except my fence wouldnt lay flat to the table because the shaft is round. What does matter is that the 2 rails are equally distant apart from front to end. I am hoping my cast iron table top is square.

[Chip Lindley]

Guys, Dave will have to figure this out for himself! Research and Development is time-consuming, and sometimes has it's dead-ends or outright failures. Edison had many failures trying to perfect the incandescent light bulb. But, when the tungsten bulb stayed LIT for hours and hours, it was all worth it. Edison was on unknown ground; your Thomson table saw fence is rather like trying to re-invent the wheel.

Bill Biesemeyer did it best, and most import fences stole his simple design. I believe we all know how the T-Square fence works. Another great fence was the Paralok of the '80s/'90s. It's inventor used the principle of the drafting machine, with thin steel cable on pulleys, clamped to front and back of the fence. It worked great; kept perfect accuracy; but had too many working parts, and could be problematic to install. The Biesemeyer won out because of it's simplicity and cost-effectiveness. Delta's all-aluminum Unifence is also widely accepted, but Delta bought out Biesemeyer! That says much about the T-Square fence.

Attaching a Wixley to your Thomson may be simple enough. But, several have complained in this forum about it's flaws. Your experience may be all good! Best of Luck.

[keith micinski]

Both tubes will have to be the same height relative to each other end to end or it will twist, and I would be wiling to guess your going to need some shim stock because the top being perfectly square seems unrealistic. Please take a lot of photos during this process so you can post them on this board. It sounds like a really fun project.

[Scott Rollins]

One comment Dave,

You need "wheelbase" at the front, IOW, 2 bearings spread apart by a reasonable, say 10-12" distance. It will do 2 things: Make it a whole lot easier to square it up, and provide moment resistance, or the ability to stay in square after you smack the far end with a sheet of plywood. Don't rely on a single Thompson bearing to do this!

Regards
Bob

What Bob said is true. Thompson Bearings are "self-aligning". I am a Machine designer by trade. The minimum number of round rail bearings is three in any design with two rails and two with one rail. This is due to their self aligning nature. You will need to use two in front and one at the back for your design to be ridgid and smooth with the clamp in between the front bearings. Based on a 30" rail to rail centerline you will need the front bearings at least 10" apart. 1/3 of rail to rail distance which is our internal rule of thumb. If this cannot be satisfied then a profile rail is required-read much more expensive. A "plain" bearing is used on most table saw fences because it is cheap and accurate. The Beis fence uses this style. A central clamp pulling against two plastic bearings-lots of friction and aligned fence. Its hard to beat a proven design.

[dave pant]

Yeah right now I only got 2 bearings coming but my buddy bought 2 also because he is thinking of doing something similar. I may just buy one of his off him and put 2 in the front like you said and put the clamp in between.


Guys, Dave will have to figure this out for himself!
Im not sure what that means but I am just asking for suggestions and opinions.

[Chris Parks]

If I were doing it and I might having seen the idea I would start with a bigger rail, 2'' I reckon. I would machine two flats on that rail one to be top and one to be placed where it does not interfere with the bearing rollers. One of these would have a tape in it for optical setting and the other would have a tape (right word?) for your Wixey or DRO of choice. The whole lot could be built with the DRO as part and parcel of the assembly. Due to the lack of friction it should have a micro adjuster which would work really well and the anchor for the micro adjuster could be rare earth magnets on a flip arm to release it, flip it up, move fence, flip down and use the micro adjuster to get the DRO reading, correct if need be and lock the fence. Dream time but it sure would be the ultimate fence.

[dave pant]

Chris you really make me wanna do this micro adjust thing too but I dont really have any good idea on how I would want to do it. You really cant machine off the top of the bar for 2 reasons. 1. Its super hard and my uncle would kill me for even trying heh. 2. I am pretty sure there are bearings that ride on the top of the bar.
Hrmm Ill think some more on this though thanks =)

[keith micinski]

I was also going to mention using a micro adjuster on it. The Wixey works really well and would excel at having a micro adjuster on it. I love the idea of machining a flat on the top for a tape. That is exactly how a tape system should be done if you wanted one which I think you should. One question I do have is where are the brackets going to go that hold the rail to the saw and not interfere with the roller bearings?

[keith micinski]

Also I am to fired up about your fence because its 1:30 in the morning and I came home to check and see if there were any updates on the thread. The micro adjuster would be basically like this one. http://www.rockler.com/product.cfm?page=16755

[dave pant]

I followed the link to that micro adjuster. I see it might be something I do after I get the main parts done and see what I have to work with a little better. The way that one works is it sticks to the rail but my rail is round and I dont think it would work. I maybe could make some sort of thing that surrounds the round rail that has magnets on it too.
I decided not to do any more work until the bearings come in and thats not till tuesday or wednesday =(. I am afraid of wasting materials by trying doing work before I even get the stuff.

[Chris Parks]

I keep losing replies to this thread, it must be cursed. Dave, you are using a hardened ground bar but I would start with something cheaper. A round form is fairly stiff but a hollow round form is stiffer again. So to make things simple and cheap I would use a 40mm rear axle out of a race kart, very stiff and would only need to be fixed at the ends. The roller carriers to ride on this I would make in a half round form so the fence could be lifted on and off. You can buy single rollers that can be put into the half round, six each would do it, two top, two front and back. This would also enable intermediate fixing if it were necessary to join sections to make the fence longer. The carriages would be quite long maybe four inches or even longer.

The flip stop would be shaped to the bar and a few rare earth magnets put in it, you would only need a couple as the friction would be so low it will move very easily. As you pointed out, get the basic fence built and see what needs to be fixed then add the tricky bits. I would build this but I refuse to start any more jobs until I finish the one I have on the go now. As an afterthought try and use a cam clamp as a fence lock, a screw lock will drive you nuts.

[Brian White]

Isnt it going to be impossible to remove the the fence even from the end without removing the rails attachment to the saw ?

[dave pant]

Isnt it going to be impossible to remove the the fence even from the end without removing the rails attachment to the saw ?


I dont see how I cant just slide it off. Only issue would be the bolt that runs through the lock and that would just unscrew.

[keith micinski]

I see what Brian is saying. It looks to me like each end is going to have a bracket on it mounting the round bar to the saw itself. How is the Bearing going to slide past this bracket. Doesn't the bearing encompass the entire round bar? I guess I'll have to do a google search for a Thompson bearing when I get home to get a better idea of how it is going to work. So, I just looked and I see they do make open bearings. http://www.thomsonlinear.com/website/com/eng/products/linear_guides/linear_bearings/super_ball_bushings.php

[dave pant]

187465
yeah it has a gap

[Chris Parks]

It can slide off but much easier if it will lift off. To slide it off you need to lower the blade below the table, not a big issue just another inconvenience and I like things to happen easily. Dave, please let us know how it goes and what issues you strike in building and use as the theory usually gets modified when the machining starts.

[dave pant]

I know I wish I could just lift it off but I am not sure about what I have to work with on the bearings until I get them. However I dont need to lower the blade because I can slide the fence off either side.

[keith micinski]

I still say you should do something like the Unifence does except from the top of the bearings. That way you can just unscrew two handles on top and lift the fence faces and the 1x2 tube off.

[Steve Campbell]

Dave I would think if you have two bearings that the best way to use them would be to put them both on same end of the fence. Space them out ten or twelve inches apart. That way you really wouldn't need anything on the other end of the fence. Most all the T-fences out there don't have any rear support other than the table. Mine has small plastic rub strips and slides as smooth as silk. You can also get away from having to get the front rail aliened with a back rail.
There is always a way to build a better mouse trap, but I have to agree that it is hard to reinvent the wheel.

Hope this helps. Steve

[johnny means]

How stiff is 5' of unsupported 1" bar? My slider has a similar fence setup with a 2" bar and the folks at SCMI seem to think that a 5' x 2" solid steel bar requires more than 2 attachment points to avoid flexing. I would think that pushing on the fence in your design would probably bend the bolts that you have at either end of the fence. If you look at and round bar type fence you will see that they were designed to aloow for multiple mounting bolts. could you do a cam style clamp, which would leave the bottom of your clamp open to allow for multiple mounting points?

Also, one of your attachment points seems to be your extension table, not the sturdiest portion of your saw. Maybe you should use a 3x3 angle across the front of the saw to attach the bar to.

Wait a minute. Cam style clamp with an open bottom, 3x3 angle for bar mounting. I think I just turned your fence into another Biessy clone:D

[Chris Parks]

How stiff is 5' of unsupported 1" bar? My slider has a similar fence setup with a 2" bar and the folks at SCMI seem to think that a 5' x 2" solid steel bar requires more than 2 attachment points to avoid flexing. I would think that pushing on the fence in your design would probably bend the bolts that you have at either end of the fence. If you look at and round bar type fence you will see that they were designed to aloow for multiple mounting bolts. could you do a cam style clamp, which would leave the bottom of your clamp open to allow for multiple mounting points?

Also, one of your attachment points seems to be your extension table, not the sturdiest portion of your saw. Maybe you should use a 3x3 angle across the front of the saw to attach the bar to.

Wait a minute. Cam style clamp with an open bottom, 3x3 angle for bar mounting. I think I just turned your fence into another Biessy clone:D

I don't see why it should need intermediate fixing points, that might be necessary if the bar was a soft grade of material, maybe. Why do manufacturers use solid sections when a hollow one is not only stiffer but lighter is beyond me. Do you push down on your fence and if so why? other than that there is no load on it to cause deflection, slide and lock. A cam clamp would work given the right shape of the clamp so I don't see any reason to copy a Biessy.

[dave pant]

3 reasons I use the bar I use.
1.Its the proper bar recommended by thomson.
2. It was free
3. I need to tap holes in it and solid bar wont allow the bar to rock because the threads are deeper.

I dont believe just becuase the Beis is so successful that its the only way to do things. Lets not forget their design is in part because of cost too. For someone to make a fence with high precision linear bearings and shaft would cost 4-5X as much and not offer a whole lot of advantage over the beis design.

[ian maybury]

Not sure what the reality is, but I'd be a little wary of the possibility of there being some slop in a linear ball bearing that could add up to some inconsistency in fence alignment - depending on the the locking mechanism. Some Biesemeyer fence designs while using very basic materials avoids this sort of issue by clamping the sliding fence mount to the front face of a box section guide bar.

A linear bearing will probably have seals, what about friction in use?

[Harvey Melvin Richards]

Dave, I think you are on the right path by not trying to copy a Bies. If I were making the fence with the parameters that you've set for yourself, I would use slip fit dowel pins to align the fence with the Thomson bearings. By removing 4 fasteners, the fence would come off and leave you with a clear saw table with nothing protruding up.

[Mike Schuch]

It looks like everyone else is doing a good job in steering you in the right direction. I have two SSR20 (kind of shaped like an I beam) linear rails 7' long and bearings in the attic which were going to be my fence before I stumbled across a cheap Vega. So I have pondered your idea many times.

I agree on the T concept and it sounds like you are on board with that. Even sturdier than a T configuration would be a box configuration. 2 bearings on the front and 2 on the back. My thinking on the box configuration is it will keep both the front and rear of the fence from deflecting. Kind of a double T fence. My plan was to use a piece of C channel I have lying around (http://www.engineersedge.com/standard_material/Steel_channel_properties.htm) that is about 6" tall with about 1 1/2" flanges laying down with the 6" face against the table. Then just bolt the bearings to each of the 4 corners. Adjustment for square would be accomplished via a little play in the bolt holes on the C channel.

I like your lock, simple and effective.

Have you considered using some angle iron or square tubing rails inside of your 1" linear bearing rails to hold the top pieces of your saw together and then adjust the 1" linear bearing rails off of these rails for parallel? This would also give you a place to mount a measuring tape. Maybe someone else already mentioned this... sorry if it is a repeat.

If you are going to use bearings that fully encircle the shaft and the shaft is mounted at each end how are you going to remove the fence?

Here is a thought... 2 bearings on the front, 2 on the rear. The bearings on each rail are connected via a 1" x 1/4" piece of flat bar 6" long in my case that has slightly oversize holes for alignment. Then the flat stock has 2 alignment holes at each end and a center threaded hole. The Fence itself would have alignment pins on each corner and wing bolts (or similar to your threaded locking handle) in the center to secure it to the 1" flat bar.

This would allow alignment via adjusting the flat bar. Two threaded handles would allow for easy removal of the fence. The alignment pins would allow for quick reinstallation of the fence without having to square the fence again. The top of the C channel would give you great potential for making jigs. Sacrificial faces of multiple lengths could be easily mounted and removed. The bearings would never have to be removed from the rail (This will make your life easier). Just a thought!

If you don't already have the linear rod look into the SSR20 linear bearings. They allow mounting points the full length of the rail.

[Mike Schuch]

It looks like everyone else is doing a good job in steering you in the right direction. I have two SSR20 (kind of shaped like an I beam) linear rails 7' long and matching bearings in the attic which were going to be my fence before I stumbled across a cheap Vega. So I have pondered your idea many times.

I agree on the T concept and it sounds like you are on board with that. Even sturdier than a T configuration would be a box configuration. 2 bearings on the front and 2 on the back. My thinking on the box configuration is it will keep both the front and rear of the fence from deflecting. Kind of a double T fence. My plan was to use a piece of C channel I have lying around (http://www.engineersedge.com/standar...properties.htm) that is about 6" tall with about 1 1/2" flanges laying down with the 6" face against the table. Then just bolt the bearings to each of the 4 corners. Adjustment for square would be accomplished via a little play in the bolt holes on the C channel.

I like your lock, simple and effective.

Have you considered using some angle iron or square tubing rails inside of your 1" linear bearing rails to hold the top pieces of your saw together and then adjust the 1" linear bearing rails off of these rails for parallel? This would also give you a place to mount a measuring tape. Maybe someone else already mentioned this... sorry if it is a repeat.

If you are going to use bearings that fully encircle the shaft and the shaft is mounted at each end how are you going to remove the fence?

Here is a thought... 2 bearings on the front, 2 on the rear. The bearings on each rail are connected via a 1" x 1/4" piece of flat bar 6" long in my case that has slightly oversize holes for securing the bearings to facilitate alignment. Then the flat stock has 2 alignment holes at each end and a center threaded hole. The Fence itself would have alignment pins on each corner and wing bolts (or similar to your threaded locking handle) in the center to secure it to the 1" flat bar.

This would allow alignment via adjusting the flat bar. Two threaded handles would allow for easy removal of the fence. The alignment pins would allow for quick reinstallation of the fence without having to square the fence again. The top of the C channel would give you great potential for making jigs. Sacrificial faces of multiple lengths could be easily mounted and removed. The bearings would never have to be removed from the rail (This will make your life easier). Just a thought!

If you don't already have the linear rod look into the SSR20 linear bearings. They allow mounting points the full length of the rail. If you already have the rods and bearings... never mind, they should work fine.

[keith micinski]

I kind of like the idea of both bearings up front like a t style fence but I must admit I was also wondering how much flex there would be with the bar being anchored only on the ends. So for that reason I think you should stick with your design. Hopefully you get building soon so we can see some photos.

[dave pant]

Im anchoring it in 3 spots on each side. Today I got the bearings and spent 3 hours just drilling 3 holes heh.

[keith micinski]

Isn't
the third spot going to interfere with the lock down mechanism you designed?

[dave pant]

It would have but I decided to do this instead. 187647

[johnny means]

Manufacturers use solid stock because it is much stronger. Hollow bar is not stiffer than a identical diameter solid bar, not even close. Hollow forms being more rigid only applies when equal masses and lengths are used. In other words pound for pound hollow forms are stronger but they don't compare to an equal sized solid form that weighs 50 times as much.

[Chris Parks]

Manufacturers use solid stock because it is much stronger. Hollow bar is not stiffer than a identical diameter solid bar, not even close. Hollow forms being more rigid only applies when equal masses and lengths are used. In other words pound for pound hollow forms are stronger but they don't compare to an equal sized solid form that weighs 50 times as much.

Not what I have been led to believe by engineering people but I will defer to your greater knowledge. It really is not a factor in the short lengths we are talking of due to no vertical loading. When was the last time a fence rail/slide was seriously loaded? I guess the real reason solid is used is that it is cheaper to manufacture.

[dave pant]

I dont think a solid bar is weaker but i do believe it has less strength per weight ratio. Think about it. A solid bar IS a a hollow bar with smaller hollow bars inside of it.

[Chris Parks]

The strength is irrelevant in your case due to no loads being applied. The only thing it must nor do is deflect under its own weight and that is not likely to happen. I would use a hollow section as I like to keep things light.

[dave pant]

I got Thomson hardened linear shaft for free so Ill use that =)

[keith micinski]

It would have but I decided to do this instead. 187647

Boy am I glad that I didn't waste a bunch of time trying to do a sketch up drawing of tha exact idea like I was going to. If you are going with a point in the middle it would tempt me to want to try and make it a t style fence and just go with a front rail. That having been said I think your way will work great and be a really cool fence.

[dave pant]

188017

Still have to build the fence part and do get some hardware.

[dave pant]

Ok I put the fence together and it didnt work out that great. I then got the 2 bearings from my buddy and added them and put a space of about 3 inches between them and it made a huge difference. I am very very excited to get this finished but I have to wait till monday. It moves so smooth and no deflection at all! Right now I am using some wood for brackets because I couldnt wait till monday to make some out of steel heh.

I also put on the Wixey and that thing is so sweet. I do see the need for a micro adjuster now though haha!

[Chris Parks]

Good to see, now we need to see photos.

[Myk Rian]

As with the Biesemeyer T-square fence, the wide stance of the front angle provides lots of stability,
That front guide rail surface needs to be MUCH wider for stability.
Quite frankly, you can find fences for the same or less than you plan/will spend.

[dave pant]

Nah I got about 80$ into this

[keith micinski]

Make sure you get some photos up I have been wondering all week long how this was going. To bad about the extra expense of the added bearings but it seemed like it was going to have to be more of a t style anyway so I don't think you were to surprised. Don't take this the wrong way but I hope you are planning on upgrading that saw next. It seems like such a waste to have a really nice custom fence on that saw.

[dave pant]

Yeah I have to admit I had to spend another 55$ on bearings so I am around 130 ish on money spent. Keith my saw works good but I do agree any saw upgrading I do will definitely be having this fence on it!
I probably wont have any pics up until monday or tuesday because I still need to go to the shop to make a few brackets. It doesnt open until monday.

[keith micinski]

What ever happened with this fence?

[Noah Katz]

I know I wish I could just lift it off but I am not sure about what I have to work with on the bearings until I get them. However I dont need to lower the blade because I can slide the fence off either side.


might be too late, but if the rail in front were low enough you could just rotate it 270 deg and it would hang down out of the way.




I don't see why it should need intermediate fixing points, that might be necessary if the bar was a soft grade of material, maybe. Why do manufacturers use solid sections when a hollow one is not only stiffer but lighter is beyond me. Do you push down on your fence and if so why? other than that there is no load on it to cause deflection, slide and lock.


1. Softness doesn't have anything to do with stiffness; hardened alloy steel is no stiffer than annealed mild steel, just stronger


2. Said before but I'll reinforce; hollow sections of the same O.D. are not stiffer (I'm a mechanical engineer)


3. There are loads - pushing the stock against the fence, stock binding when stress is released when the wood is cut, accidentally bumping the fence


Dave,


I'm surprised you think you need a microadjuster; I'd think with the low friction of your fence tapping would be quicker and easier.


It works fine on my Powermatic Biese clone fence, which has quite a bit of friction.