Radial Arm Saw Cut Tolerances

After a discussion of the inherent roughness of radial arm saw cuts, a woodworker gets his unit tuned up for fine cuts on melamine. October 14, 2009

I recently purchased a used 14" Delta radial arm saw for the purpose of crosscutting melamine. This saw is quite heavy duty and can crosscut over 29". Iím wondering what the side to side tolerance of the cutterhead should be (there is currently about 1-2mm movement as I pull the motor across the track)?

I have adjusted the tension on the bearings, but the blade is still wobbling - causing the back of the blade to chip parts of the cut. I am hoping with four new bearings I should result in zero lateral movement of cutter, resulting in clean cuts with my new Forrest melamine blade. At the end of the day, I am hoping that this saw will result in excellent crosscutting accuracy - assuming it has new bearings and is correctly tuned. I would appreciate any comments from others with this class of radial arm saw.

Forum Responses
(Cabinetmaking Forum)
From contributor P:
If the back's chipping on one side or the other, the blade's not sitting parallel to the arm and can be adjusted. It may have been a matter of taking out the bearing play unequally. A millimeter or so of slop is pretty typical with that machine, and there are good reasons that this isn't the preferred way to cut melamine accurately. I've seen a few installations where guy wires and turnbuckles were run at 45 degrees from the front of the arm back to the wall to remove any lateral play in the arm. It helped get some of the play out, but also removed the ability to cut angles.

From contributor G:
Typically a radial arm saw is a rough cut machine. You can tune them up pretty good, but it will never be a perfect cut or a perfect 90. I can't see cutting melamine on RAS. I have a hard enough time getting a chip free cut on both sides on the TS using a melamine blade.

From contributor X:
It sounds like excessive end play in the motor creating the problem.

From the original questioner:
I just replaced a bad bearing in the motor so there is no play there. It almost feels like there are dips in the track - where it feels a bit rough when I pull the saw out. However, the track is in mint shape with no such wear, and I have even rotated the slides to expose a new unused portion. I think the cutter-arm bearings are the problem. It almost appears that there are gaps or wear inside the bearings. They were too tight when I checked so maybe this helped wear them out?

I think there should be some benefit to cutting melamine with a 14" blade, as the exit angles would be less severe than say a 8" or 10" blade. Same principle as comparing router bits to wider shaper cutters.

I have no concern that I can't get this saw set to a perfect 90 degree. I have done so successfully on my previous RAS, but the benefit with the large 14" saw is that it should hold the setting for a long time, unlike a Sears RAS which can be fairly easily thrown out of alignment with some kickback or running something into the overhead arm. So far I have spent almost $500 on new bearings and a 14" melamine blade. I sure hope it doesn't disappoint - haven't received either yet. Currently this saw is only suitable for cutting solid lumber (and very thick lumber at that).

From contributor A:
Contributor G hit the nail on the head. Most radial arm saws are used for rough cutting of wood. Some shops set them up for specific tasks like dados/rabbets. Melamine is one of the most frustrating products to cut. I think you would be hard pressed to get a very good cut on a brand new 14" Delta even after you spent a few hours trying to calibrate it. If your parts are less than 12" in width, I suggest you purchase a brand new 12" sliding compound miter saw. Forrest melamine blades are excellent, however a great blade doesn't fix an inaccurate machine.

From contributor F:
I have to agree with the other guys. You'll be hard pressed to keep tight tolerances on an RAS. Itís just not the best machine for what youíre doing. How are you doing the rest of your cuts - like rips? Are you also using a tablesaw? If so your money may have been better spent on a decent sliding attachment?

Having said that, with the new bearings in and everything tightened up you'll probably get her cutting pretty good. I have a Craftsman in my shop that's just used for rough stock and I can't imagine trying to get it accurate enough for any finish work. I've received good results with the Forrest melamine blades and I think you'll be happy with the cut they produce.

From contributor P:
You may improve your exit angle, but any deviation in blade flatness bearing runout, or motor shaft out-of-true will translate to greater wobble in a 14" blade compared to a smaller diameter. I get my best results running the smallest diameter blade available. I don't know that I've seen a motor without some lateral runout. Itís not an issue in a belt-driven machine, but may explain the dearth of direct-drive precision machinery.

From the original questioner:
I would love to have a slider, but don't have the space for it in my basement. I have been cutting most of my melamine with an Amana melamine blade on my unisaw. Not really able to cut full height end panels though (83x25), so I have to use a downcut blade on a jigsaw. Contributor P - good comment on vibrations being directly proportional to blade diameter. Not sure about the direct drive comment though, aren't vertical panel saws (like Streibig) and CNC routers all direct drive?

From contributor P:
I don't know the Striebig setup, but my Holz-Her 1265 is a belt-drive. I assume the CNCís are built to tighter tolerances than the average motor or may have some preloading to remove lateral play. You might want to check into one of the smaller vertical saws. If you can get it into your basement, it'll do a lot of precision cutting in a very small footprint.

From contributor B:
I got rid of my RAS last year because I couldn't get it to cut accurately enough. Cutting lumber is one thing, making precision cuts is another.

From the original questioner:
To contributor B: what model saw did you have, and did you discover what the route cause of the inaccuracies were? What problems were you having with yours? This 14" Delta looks bulletproof and should maintain its settings for a very long time. The only weakness that I can see may be in the table/backfence design common to all radial arm saws.

From contributor R:
My 14'' Maggi RAS will cut as close as the naked eye can see or my fingers can feel on a 16x24 framing square. You may need to shim those rails that the bearings ride on up or down or side to side, they donít have a worn out spot but they bolt to a cast iron arm that is less than perfect. Get some shim stock or cut up your feeler gauge and use it. One rail up higher than the other just a few thousandths translates into much more at the bottom cutting edge of the blade. Paralell, plumb, and square to machinist specs and you will wonder why every shop doesnít have one.

From the original questioner:
I never thought that the rails may not be seated properly. My RAS has four of them (top and bottom). I'll have to check with straight edge. Maybe I'll remove rods and clean any dirt from its seat. I had tried to get my caliper in there to measure if there were variations in the distance between the tracks but it wouldn't fit in there. I can see how small variations on bearing/track would be multiplied many times to the bottom of the blade.

From contributor V:
In terms of RAS only being for rough work remember that they take different forms. I think the first time any of us saw one of the ubiquitous Delta turret arm radial saws it was most likely in a lumber yard not a cabinet shop. In the early 90's I was in a used machinery warehouse and came across a Wadkin RAS with a 50 inch capacity. The radial arm telescoped on the end and inside was a very solid piece of rectangular steel mounted at an angle (similar to a diamond shape). It was quite obvious this saw had been designed with the idea of crosscutting sheets in lieu of a sliding table saw. I tried to make the end deflect but felt nothing. So a used Delta may not be the best place to start for an RAS with fine tolerances.

From contributor E:
I've also seen that Wadkin and like all Wadkin machinery it was built like a battleship! Not often you see it here though. I saw it in England and it was years ago.

One more small suggestion for your basement shop: I have had a shop in my very small basement for 20 years. I had the same issue and solved it with a very large crosscut sled for my unisaw. At first I made them smaller (24" wide and 36" long) and used Ipe for the runners. As time went on I found the need for cutting larger material. Eventually this sled evolved into a 36" wide and 65" long sled that I use ball bearing slides on (I think from Rockler) and support on the outfeed with roller stands. It can end-trim most full size doors and just about any size cabinet side. I do melamine with it too with very little chipping (Forrest blade). These days I have access to several large shops that I work with so I rarely use this setup, but it still works.

From the original questioner:
Just a quick update: I have received and installed new bearings for the 14" Delta RAS motor carriage and now it slides very smoothly, with no lateral movement. With the original bearings once I had tightened them to take out any side-to-side play it was quite rough (and required some force) to slide the motor. Now I have to reduce the spring return or it'll slam the motor into the column if I let go. Anyway, cuts are perfect on top of melamine (with no chipping), haven't received Forrest's melamine blade yet, but am hoping that it'll take care of the bottom of cut.


Installed Forrest HiATB blade on my 14" Delta and am now getting perfect cuts, top and bottom. Iím very impressed with the blade, and rigidity of the Delta. Don't know what life will be like on the blade, but I plan to send it back to Forrest when it gets dull. No bottom chipping if I move the cutter at a moderate rate. If I did it again I would probably get higher tooth count though. One thing I have noticed with the HighATB blade is the hook angle makes the blade want to self feed - not a good thing on radial arm saws. Just have to watch it a bit, especially when it starts to get dull.