locker support questions

[I saw that!]

.....

[mmcmac]

[B]Is there supposed to be almost no play in the locker support, needle roller and clamp plate assembly? There's a slight amount of play between my rollers and locker supports that I notice with them off the saw. The black oxide finish is worn off the rollers at the contact areas, but they don't seem scored using the fingernail test. I'm not sure whether the rollers are harder steel than the supports, but there's a very shallow groove worn in each of my clamp plates.

I see some play between the locker support (#71) and needle roller (#15) in my cabinet shop refugee bt3000 but not much in the other two units. That as well isn't the greatest design Ryobi could have mustered but at least the components are steel and replaceable. Unfortunately there isn't much room for improvement in there (literally) and it seems living within the existing design may be the only practical choice.

It might be possible to press a sintered bronze bushing into the support but it would be preferable to create a new locker support with more area to securely retain the bushing. Another possibility is using inexpensive drill rod of slightly greater diameter than the original needle roller and ream the original locker support for tight fit to the rod. It appears you could build up the locker support only in the area where it retains the needle roller, doubling its thickness and giving it more wear surface. Eg: spot/tig weld a same thickness "patch" over the locker support needle roller hole and drill/ream that out to a tight fit with the original needle roller. If excessive play should develop in the future, replacement with a slightly larger drill rod should yield double the wear life.

[mmcmac]

The needle roller number changed when Ryobi made the switch to non-countersunk locker supports and pan head screws. Both early flat head screws and later pan head screws are 10-24 x 5/8" size.
It makes me wonder what about the roller changed -- material hardness, perhaps?

Possibly, or the source for (or type of) component may have changed.

I may act as volunteer guinea pig since the replacement parts aren't very expensive. If so, I'll post my findings here.

I'm starting to suspect play between the pin and locker support may less likely be caused by usage wear but is rather due to sloppy manufacturing tolerences. The joint should not see much motion in typical use and is for the most part a statically loaded bearing surface.

I haven't disassembled the pin/support for examination but from what I can see the support is stamped sheet and I fear the pin hole was created in the same stamping operation. Dowel pins are hardened, ground, precision components intended to be press fit seated in a hole reamed to exact size. I'd hazard Ryobi sized the support hole for use as it exits the stamping operation without further machining, which would push the error preference to an oversized pin hole.

In any case an oversize locker support hole can be resized after slight buildup via hard solder, brazing, or attachment of an patch. Another possibility is peening the area around the locker support hole slightly and uniformly on both sides of the support which will close up the hole for finish sizing. Hardened dowel pins are quite inexpensive. A quick check found a box of 100 0.25" x 0.75" precision ground pins for under $14.

I should be able to simply flip my clamp plates over to the virgin sides, which will take care of play due to the worn grooves. Alternatively, I could use thin shim stock between the plates and rollers. So, I'm not too concerned about new plate availability after all.

It seems a bit unusual for play to exist between a clamp plate and pin. If it does it could be due to the pins rotating somehow causing them to wear into the Al casting. Or the saw could have sustained a drop at some point causing the weight of the motor/arbor to loosen the clamp mount. In any case it should be an easy fix.

I don't want to do too much removal/reinsertion of screws on this saw, though. The screws are the self-tapping variety and about the last thing I need is stripped or badly worn hole threads in the aluminum.

It also appears the partially cut swarf and/or deformed threads are often relied upon to act as a thread locker of sorts. I'd prefer to replace them with a conventional fastener rather than trying to run the reinserted screw in the already cut threads, and use a real thread locker such that the steel screw only needs to be snugged in a soft casting thread.

[mmcmac]

My plate screws had worked loose, allowing the rollers to rotate against the plates and table pocket, thereby digging into them.

The threads in the table casting look to have some galling. One possibility would be to screw in 10-24 studs with a medium strength threadlocker and seat the plate with a nut and lock washer. That removes future Al thread wear as a concern but the studs are still replaceable if needed.

Correct, but early supports have machined countersunk holes. This is probably one reason Ryobi switched to regular holes and pan head screws. The supports could then be stamped in one operation with one jig in one configuration = quicker, cheaper.

But unfortunately not "better". At least the conical countersunk holes offer some inherent indexing precision of the locker support on the locker bracket. Mounting of the latter version with punched cylindrical holes and pan head screws may require alignment care during assembly to minimize perpendicularity error between the blade and table rails.

The supports are inexpensive, too. Repair operations on the supports wouldn't be worthwhile to me, as I can get a set of new supports plus rollers for a few bucks total -- far less than the labor value involved in repair.

My primary concern is NLA creep of repair parts, increasing captive market price creep for that which is still available, and to a lesser extent lead time for availability. But I'd as well buy whatever is reasonably available at a given time.

It may all be moot anyway. Having examined the parts and test fit them, I'm much less concerned about excessive play. As it turns out, the play I felt at the support joint was basically lateral (orbital?) wiggle (not the term I'm looking for, but I can't think of it right now, hopefully it makes sense); I didn't have the parts held in place as their normal operation would have them.
There's almost no play when trying to move the rollers exactly perpendicular to the support holes, and there appears to be no ovalness of the holes.

Play axially along the pins shouldn't be much concern as cutting operations where this is a factor are rare. Play radially around the pins (perpendicular to the pin axis) will result in left-right-up-down blade motion and would probably be otherwise chalked up as cut roughness due to blade vibration. I'd extend the blade fully and put a dial indicator on the side of the highest projecting tooth while rattling a locker bracket around its pin to determine whether pin fit is an issue.

Also, although the roller wear as shown in the pics is extremely mild (seems to be only the black finish rubbed off), the roller can simply be reversed if desired, with the "fresh" end in the support hole. That does nothing for hole wear, but it could help reduce play a little, if need be.

I don't see any pin wear of significance. I suspect any locker support hole slop started with an intentionally oversize punch during manufacture and was probably exaggerated by vibration during use against the irregular shear/fracture profile left from the punch operation. The irregular wear marking on the pins appears to underscore this as a cause. A reamed to size hole would have performed far better from a technical perspective, but at additional manufacturing cost.

I don't generally use anaerobics unless I'm certain I won't be touching that fastener again for a very long time, if ever. Gunked up threads can be a royal PITA to clean and sometimes even to break free w/o causing damage.
In this particular case, I may upgrade to the 15A motor at some point and even though I'd normally opt for lockwashers, the fact that the clamp plates float pretty much demands Loctite in the bosses.

Thread lockers are available in various strengths. The familiar Loctite 242 blue is probably a reasonable starting point. Other products are available which don't produce a hard bond (eg: Vibra-tite VC-3) and are intended mainly to eliminate fastener loosening due to vibration.

[mmcmac]

I don't have the supports (yet), but I'll go out on a limb and give Ryobi the benefit of the doubt and assume they punched them pretty closely to the diameter of the screws. If they're off a little, the errors should be rectifiable via parallelism adjustment of miter base, rip fence and miter slot table (if installed).

Indeed minor blade/rail perpendicularity error can be compensated out for most cutting operations. Doing so does result in a slight parallelogram relationship between the rails and blade/SMT/rip_fence. For most usage this probably doesn't matter yet it would be more pristine to avoid the error propagating to other alignments. Ryobi probably dismissed this as acceptable given the dowel pin seats aren't further machined after casting, nor do provisions exist for adjusting the locker support location on the locker bracket to zero this error.

the amount of play is very low, even on these old supports. So yes, machining of the holes would have tightened up the roller contact across the length of the hole, as there'd be no secondary shear/fracture, but as you said it would have cost more.

A trivial approach to remove the play in the stock locker supports is attached. I suspect the slop in my cabinet shop casualty bt3000 may be pronounced enough to warrant the tweak. The saw however is currently disassembled for a complete rebuild, but I'll capture a few pictures when I get around to addressing that issue.

The secondary slide arm functions just to prevent vertical motion of the locker support under vibration and essentially carries no significant weight. It needs only to be snugged to remove joint play. I'd fabricate it from brass sheet and should it loosen due to wear, it can be trivially readjusted. But given the fix and a coat of a MoS2 lubricant on the dowel pins they are likely no longer a maintenance concern.

[mmcmac]

Before the play compensator slide, I'd sooner do as you suggested earlier and just enlarge the roller/pin hole, reducing the new, larger roller on the end that belongs in the main table pocket. If you're into machining, you probably already have various sizes of bar/rod in inventory and a metal lathe. This would seem a little less work than the slide.

The opposing slide arm was suggested as it can be adjusted for future wear without component replacement. As the tools needed to fabricate it from brass plate are minimally a hacksaw, drill, and tap the fix should be applicable for a wider group vs. that requiring machine tools. Also given the minimized depth of the pivot pin axis relative to the table surface, the result doesn't leave us much daylight in the locker support pin hole circumference. Getting a brass/bronze sleeve of substance mounted may require a brazing operation to insure integrity of the modification. So from a practical perspective it may be reasonable to live within the existing design.

The minuscule amount present in my components should be eliminated as soon as the blade contacts the workpiece. The locker bracket should be thrust up and back, securing the [seemingly unworn] support hole firmly against the roller. There should be no wander or vibration (due to these particular components) during actual cutting operations.

The force developed by the blade during a cut will result in an opposing torque seen by the locker bracket centered about the arbor. This torque appears to work against at least three thresholds. First would be the spring flex of the shim/gibs seen in the "ungibbed" fold of the shim. This will rotate the guide holder in a motion opposite of blade rotation but won't alter arbor elevation. If the tangent cutting torque is great enough the guide holder can lift vertically a distance equal to the elevation screw backlash. If the cutting force exceeds this it would cause the locker bracket front support to thrust upward and the rear support to thrust down. So in practice only the front support/pin play should be significant, although depending on the variability of the cutting force both supports can rattle about their respective pins due to stresses developed during the cut operation which create forces perpendicular to the blade.

In the cabinet saw patient, I measured via dial indicator the play in the locker support around the pins and see about 5mils of lateral movement. While I'll likely zero that out, I did observe flex in the locker bracket casting less than this play but still more than I'd anticipated. After the saw is reassembled I'll make static measurements to determine how significant the error contribution is.