ARB, how to eliminate it and yet still, have anti backlash.

[Breadfan]

In reading the recent thread about ARB failure, I'd like to propose my idea I've had for quite sometime to get rid of the ABR all together. I've thought about this but I have never seen it mentioned? I understand the concept of having two dogs, two of course is better than one. But why don't we tweak that and eliminate the ARB altogether? My idea is if you have two, or really, three dogs all contacting the gear at different times, you'd still get zero backlash, or close enough to it. In other words with each micro turn of the handle, one of the dogs would always drop down and be engaged. I don't see why this wouldn't work. It seems as a rather simple concept to get rid of a crappy bearing that continues to fail. It's very much like the bearings in the videos of the other thread posted, just a mechanical way of doing it, so to speak.

[Gfish]

So, you'ed need a bearing or bushing to go into the space where the ARB was removed from. I have one reel where the ARB has delrin bushings fore and aft of it on the crank/gear shaft for stability. There's triple factory dogs down around the main gear. Would a longer single delrin bushing be enough? Would it be strong enough and durable? Or, would you need a roller bearing in there? Fit could be a problem if a roller bearing was substituted.
That reel's a conventional. What about spinning reels with those large diameter ARB's. To my knowledge, some higher end spinners have those + a back-up dog, many don't...

[philaroman]

I've imagined a reel designed from the get-go, w/ 4 dogs in a bracket (2 alternating pairs)
could be a perma-peened assembly for mid-price & take-apart for high-end
that way a robust 10-tooth ratchet would provide 20-point A/R double-dogged at every point

What about spinning reels with those large diameter ARB's. To my knowledge, some higher end spinners have those + a back-up dog, many don't...

well, how brilliant is my idea for that scenario...  esp., huge Shimano spinner ARB's?

[JasonGotaProblem]

As you add more teeth/dogs, the size of each tooth and dog must decrease unless the reel size increases.

Extrapolate that out a bit and you end up with very small teeth, whose strength and reliability will eventually proportionally decrease to the point where the ARB surpasses it. Not saying that any of the above ideas approach that, just that there is an upper limit to the benefit of another tooth and/or another dog. 

Also a big part of why the industry standard has shifted to ARB is that they arrive pre-assembled and get installed with only a handful of screws, which might be able to be fully machine-installed. as opposed to a complicated mechanical AR system which must be painstakingly assembled by a skilled craftsman. I'm gonna make a blanket statement of questionable veracity that every dog you add to a reel adds $50 to the final total.

Or just fish an old fashioned reel? Plenty of old mitchells in circulation.

[oc1]

I prefer to not have any anti-reverse for light tackle casting and have several reels with the ARB removed and replaced with a normal ball bearing.  There is usually a pretty close match for inside and outside diameter available at Boca Bearing.  Like Gfish says, you will probably need to fashion a bushing to fill up the extra height.

[jurelometer]

This has actually been covered a lot.  There are even some equations that will help you calculate the force, and some explanation of how to design the system for maximum effectiveness.  A bit of searching should yield some interesting threads

The first step is to make the pawl (AKA ratchet) as large a possible.  The second is to ensure that the dogs and pawls stay aligned.  Easy to design into a new reels, much harder to fix aftermarket. The actual surface contact area is not nearly as important as mainlining alignment.

The next point is to decide how much backlash and noise is acceptable.  Folks dislike any backwards handle movement and the clicking. More dogs and stronger springs = more and louder clicking.

As to why roller clutches are so common:  It is easy to design a reel with a roller clutch (AKA ARB, AKA one way bearing).  And they are a cheap commodity item.  It probably costs less to get a reel to market with just a roller clutch than a dog/pawl.  And they feel better at the shop, so they are easier to sell.  There probably is not that much of a disadvantage using them for most freshwater fishing scenarios.  Economics and market preference is driving the popularity.

The problems with ultimate strength and corrosion of roller clutches are more specific to saltwater.

As for the noise, there are silent dog designs (ears, or the Avet thingy), but they come with their own drawbacks.

I mostly agree with Gfish's comments:  If you are doing this for yourself, you might find that you can remove the roller clutch and rely on the dogs already  in a saltwater reel.  Assuming we are talking about the common "bearing" style, roller clutches are not designed to be load bearing (they won't lock/unlock properly if bearing a load), so you can try  replacing the clutch with a bushing or bearing  that fits a tiny bit on the loose side if there is a need to have some support for the shaft in that spot.  I vaguely remember reading here that Accurate did just that with a nylon bushing on one of their models.  Don't know why they ditched the roller clutch.

There are plenty of examples here of folks adding extra dogs to reels, but once you get beyond two dogs, IMHO it is more of an exercise in self-gratification than functional improvement   Often the extra dogs are cut to allow them to engage (sort of) simultaneously as opposed to alternating to decrease backlash.  But that is a whole 'nother can of worms.


-J

[boon]

ARBs do a really good job, and in my experience very seldom fail. Virtually all reels with one have a backup dog anyway, so if it does die while you're on the water and you don't have a backup reel, you fish without an ARB for a few hours and have a reel that is functionally no worse than the thousands of reels out there with a single dog AR system.

And with all of that said, "modern" reels with no ARB (2-speed Talicas, Makaira) have negligible backlash with a silent 2-dog AR system.

Hunting for a way to delete the ARB feels like a solution to a problem that isn't really a problem (other than the fairly niche use-case of wanting to have no AR altogether), although the engineering would be novel.

With all of that said, I'm biased - the only reels I have that go "tick" any more are electrics, and that's more or less because they're all that is available.

well, how brilliant is my idea for that scenario...  esp., huge Shimano spinner ARB's?

To my knowledge, none of the big beasty spinners (Stella/Saltiga/Makaira) have ever had issues with going backwards.

[jurelometer]

ARBs do a really good job, and in my experience very seldom fail. Virtually all reels with one have a backup dog anyway, so if it does die while you're on the water and you don't have a backup reel, you fish without an ARB for a few hours and have a reel that is functionally no worse than the thousands of reels out there with a single dog AR system.

And with all of that said, "modern" reels with no ARB (2-speed Talicas, Makaira) have negligible backlash with a silent 2-dog AR system.

Hunting for a way to delete the ARB feels like a solution to a problem that isn't really a problem, although the engineering would be novel.

With all of that said, I'm biased - the only reels I have that go "tick" any more are electrics, and that's more or less because they're all that is available.

well, how brilliant is my idea for that scenario...  esp., huge Shimano spinner ARB's?

To my knowledge, none of the big beasty spinners (Stella/Saltiga/Makaira) have ever had issues with going backwards.

We need to be a bit more specific about the type of  reels, clutches and and other terminlogy. 

Shimano (bless their heart) decided to reuse the ARB acronym.  A-RB  is not an anti-reverse bearing, but more marketspeak (Anti-RUST bearing) for an allegedly more corrosion resistant ball bearing. 

The typical off-the-shelf  generic roller clutches (AKA one way bearing, AKA ARB) used in most reels are made from 440c stainless.  440c is used because it is can be hardened, and the hardness is needed for the rubbing/scraping that goes on when the rollers jam against little ramps in the "race" to lock the clutch.  While 440c is quite corrosion resistant in fresh water, it corrodes fairly easily in saltwater.   

Folks here generally try to thread the needle and lubricate with  light oil or film, using  just a little bit to minimize corrosion without giving up too much stopping power.   The reel companies sometimes recommend zero lube and sometimes a small amount of oil. No matter how you do it, these clutches are corrosion magnets unless you are willing to live with compromised stopping power.  In another thread, we calculated the torque at full drag on the clutch for a popular saltwater reel, and looked up the specs for the same part from a reputable manufacturer.  The torque generated was pushing the limit on a full spool with a dry clutch .  Get spooled a bit, the drag goes up, and the clutch will slip.  Which is what many folks reported, and why these reels usually now have "backup" dog/pawl systems ("overload dog"  is probably more accurate).

440c is also used for ball bearing in reels, but these bearings can be fully lubricated, usually packed in grease if the bearing is not used for casting, lowering the risk of corrosion.  If you lubricate/protect  a roller clutch with a dry film lubricant, the lubricant is rubbed off quickly from the locking action.  Oils decrease the friction that is needed for the clutch to hold, decreasing the maximum torque. Greases are even worse.  In addition to decreasing friction, the grease inhibits the rollers from moving up and down the ramps potentially leading to greater slippage or failure.   


I can think of  three failure modes to  consider.  One is when the  clutch is overloaded and slips.  While it is true that reels with backup dogs will not end up with a handle spinning backwards, the rubbing/scraping from slipping rollers is  damaging the clutch.  These parts are designed to lock, not to constantly slip under load.  The backup dogs will prevent continuous slip and the clutch will probably survive a decent period of usage with an occasional slip, but I would be more nervous if it happened frequently.  And if it happens frequently, the clutch is not doing much for you anyways.

The second failure mode is grease or heavy oil inhibiting the clutch action.  If the problem occurs locking, then the clutch will slip prematurely.  When unlocking, turning the handle could get a bit stiff or sticky when first turning.  In both cases, the load would not be distributed across all of the rollers evenly, increasing the risk of damage.

The third failure mode is corrosion causing a bearing failure. They can corrode and freeze in either the unlocked (most common) or locked state, or can break apart while in use.  Not too pretty.

These type of bearings are usually placed on the handle shafts of conventional reels, where the gear ratio is working against the clutch (6:1 gears means 6X leverage on the clutch).  In the better designed spinning reels, the clutch is located right next to the spool, ahead of the gears, making the clutch more effective.

And for at least some of the super-spinners, these off-the-shelf clutches are not used, correct?  The roller clutch is an oversized, specifically designed unit, so they have an opportunity to make better tradeoffs for performance and corrosion resistance.  I don't have any first hand knowledge here, as I have an allergy to spinning reels 

I can't argue with someone else's preferences around features vs reliability. I  personally try to avoid roller clutches as much as possible, but sometimes you have no choice, and have toyed with the idea of modifying the reels to only use dogs myself.

-J

[boon]

For reference, I'm completely disregarding Shimano's marketing term, when I say ARB I refer to an Anti-Reverse Bearing.

Shimano has, for a long time, used varying grades of a custom mechanism that I guess you could call an anti-reverse bearing in their spinning reels. It is a large diameter assembly that typically acts on a sleeve around the pinion gear, thus preventing both the geartrain and rotor from going backwards. They are relatively corrosion-resistant, except when abused, and seemingly offer far more torque resistance than most conventional ARBs, unless over-lubricated. The Stella (until the latest iteration) gets a bespoke version of this mechanism, and it is widely regarded as both extremely reliable and very effective.
The Makaira uses a large conventional ARB, although as you have highlighted, it acts against the pinion/rotor instead of the drive gear, so the torque multiplication is not present as it is in a conventional reel.
The Saltiga... I'm not sure about the current one, but the previous model used a large off-the-shelf ARB from NSK which operated in an extremely similar manner to the Shimano mechanism.

I've literally never noticed a reel with an ARB-based AR go backwards on me - and it's worth noting that I almost exclusively fish with reels that used an ARB. Nor have I had one of the bearings go bad. Fish your reels within their limits and look after them properly and it seems to be almost a non-issue? All of the failure modes you identified above (which I entirely agree with) seem to be completely avoidable.

[oc1]

try  replacing the clutch with a bushing or bearing  that fits a tiny bit on the loose side if there is a need to have some support for the shaft in that spot. 

On conventional, without that support the peened gear shaft is taking all the lateral load when you are cranking it.  That peened post is fragile enough even when supported by a bushing/spacer.  Without some support out there you would be asking for trouble.  Reference here the shredded gear thread.

[Breadfan]

I was thinking about what I could do to my ABU 7000's to take up the slack. they move a little less than a 1/4 turn before the dog engages. They are from the mid 70's so they don't have an ARB, just a bushing. But in reality, it doesn't hurt anything. I put my rod in sand spikes and it makes no difference, the fishies hook themselves. Just an annoyance.

[jurelometer]

try  replacing the clutch with a bushing or bearing  that fits a tiny bit on the loose side if there is a need to have some support for the shaft in that spot. 

On conventional, without that support the peened gear shaft is taking all the lateral load when you are cranking it.  That peened post is fragile enough even when supported by a bushing/spacer.  Without some support out there you would be asking for trouble.  Reference here the shredded gear thread.

I haven't tried it myself, but my theory is that the best chance for success is to replicate the original fit.  On a star drag, it is likely that the OD is a loose fit to the sideplate to allow the ARB to slide up and down the shaft when you tighten and loosen the drag.  On the ID , the bearings are designed with the fit on the loose side, and the bearing gets locked by changing to a tight fit when the rollers are wedged between ramps on the outer "race" (not really a race), and the shaft.  Similar to the OD problem, unless there is a sleeve design, the ID fit has to allow for the ARB to slide up on down th shaft for drag adjustment. 

Here is an example that illustrates my point:  There is a problem on some star drag reels where you cannot adjust the drag tension under load.  This is because there is no ARB sleeve, and the  ARB has locked to the shaft.  You have to turn the handle, ulnlocking the ARB and allowing it to slide down the shaft, at which point the change in clamping force is transferred to the drag stack (another reason to just say no to ARBs)

Because of the loose fit and the lack of a true outer race, ARB's are not designed to be load bearing. Of course, using th ARB to support th shaft would not be the first example of a reel company disregarding  basic mechanical design principles, so I do agree that it might ot  be advisable to  leave a hole where the ARB used to be if the shaft is not supported on the handle side by a bearing or bushing. And there probably needs to be something like a spacer to transfer the load from the star to the drag stack.  My vote is still for  a bushing with an ID and OD that corresponds to the original part. 

BTW, If the reel is a lever drag,  a tighter fit might work.

There is another thread here somewhere where I do my best to describe how these ARB gizmos work.  Also has a diagram.

-J

[jurelometer]

For reference, I'm completely disregarding Shimano's marketing term, when I say ARB I refer to an Anti-Reverse Bearing.

Shimano has, for a long time, used varying grades of a custom mechanism that I guess you could call an anti-reverse bearing in their spinning reels. It is a large diameter assembly that typically acts on a sleeve around the pinion gear, thus preventing both the geartrain and rotor from going backwards. They are relatively corrosion-resistant, except when abused, and seemingly offer far more torque resistance than most conventional ARBs, unless over-lubricated. The Stella (until the latest iteration) gets a bespoke version of this mechanism, and it is widely regarded as both extremely reliable and very effective.
The Makaira uses a large conventional ARB, although as you have highlighted, it acts against the pinion/rotor instead of the drive gear, so the torque multiplication is not present as it is in a conventional reel.
The Saltiga... I'm not sure about the current one, but the previous model used a large off-the-shelf ARB from NSK which operated in an extremely similar manner to the Shimano mechanism.

I've literally never noticed a reel with an ARB-based AR go backwards on me - and it's worth noting that I almost exclusively fish with reels that used an ARB. Nor have I had one of the bearings go bad. Fish your reels within their limits and look after them properly and it seems to be almost a non-issue? All of the failure modes you identified above (which I entirely agree with) seem to be completely avoidable.

I have seen plenty of ARB carnage, and I am not even a repair pro that sees lots of broken reels.  With fly reels in salt, it is a question of when, not if  the off-the-shelf ARBs will fail.  The sealed fly reels with factory-only service are the worst, as you have no idea regarding the condition.  On conventional reels, especially star drag, the ARB is located at a point of water ingress, and generally cannot be completely sealed off due to other design limitations.

If you think about it, pulling off a locking clutch design that works by jamming rollers against a shaft is a somewhat tricky piece of engineering.  These off-the-shelf gizmos only work if the rollers can move freely and the springs are the right tension and everything is properly spaced so that the rollers are far enough from the shaft that they do not interfere with forward rotation, but close enough to the shaft that one or two are lightly touching the shaft at all times, allowing the the locking process to initiate when rolled in reverse.

So we have a device with tight tolerances, limited load capacity, made from a material that corrodes in salt, and  cannot be properly coated with a corrosion inhibitor and still maintain it's function.   What could possibly go wrong ?   

Thanks for the info on the proprietary roller clutches in the higher end spinners!  I will agree that it is possible to build a reliable and strong roller clutch if the reel type and particular design has enough room to  fit a larger mechanism.  This makes sense.  If you go to the bearing company sites, they have some pretty heavy duty roller clutches.  But none of them are very small.

I am still of the opinion that  the common off-the-shelf ARBs are undesirable  for most saltwater applications.

-J

[Gfish]

Agree. Though generally the spinner ARB's are easier to access for me to clean/lube than those of the conventionals/baitcasters, so I don't fret about saltwater ingress as much. Those of the Shimano spinners are well made and not as hard to rebuild as they look.
After reading a tutorial from John Tuttle on the Abu Revo Toro Beast, about the system for releasing the drag adjustability while under pressure, I had to have one. A rather complicated system, but works. Up until then, I didn't even know about ARB drag lock-up, till I lost an expensive rig as my kayak was gettin wind pushed away from an accidental turtle snag-up, while using a Lexa. "Hey, how come the drag didn't loosen-up?"