zygomorph

Last year I did (most of) the GAP/C&O on my beloved Miyata 615gt. My partner crashed into my rear wheel twice (Velocity NoBS 36h with a hub running an 8sp cassette), so it got replaced with a Sunrim CR18 with a Shimano Deore FH-M756 hub with a disc rotor mount (I mean, whatever).


For years I had assumed (based on nothing) that my rear dropouts were 130mm spacing and that I was jamming a 135mm hub in (no interest in cold setting). But after struggling an newly unusual mount to get the new wheel in and out, coupled with the spokes de-tensioning lately, I got a a truing stand, a spoke tensionometer, and some calipers and I went to work.


As it turns out, the Miyata's spacing is 126mm! And I don't remember exactly what the new hub was spaced to, but it was definitely more than 130mm. The cone nuts themselves are quite long, and the spacers quite few. However, I was able to get the O.L.D. down to just a smidge over 130mm and it's much *much* easier to get the wheel in and out now. I redished and tensioned the wheel so that the drive side spokes are now at 96kgf. It feels great in the saddle!


However, the dishing on the drive side looks... you know. Almost vertical. Sorry I don't have an actual angle, maybe I can just do some trig and figure it out but I'd have to measure/look up so many different things, I'm lazy, and I think my questions are general enough that it's not really that necessary.


So! Questions:

- Is this tension okay? (96kgf per DS straight gauge 2.0mm spoke on a 36h double walled rim, I couldn't find any recommendations for this specific rim)

- Is there maybe something about the design of this hub (that it's disc-specific?) that keeps it from being more optimally re-spaced for 130mm? There seems to be a lot of wasted space on the NDS that I would have preferred to use to keep the dishing less extreme but the design of the cones and seals were the limiting factor here.

- Even if I could move the hub further to the NDS would this do anything notably sub-optimal to the chainline?

- If/when I'm replace the wheel (maybe I'll build one myself!), are there recommended parts/setup for 8sp in a 130mm spacing?

- Was it irresponsible of the shop to shove a nearly 135mm hub into a 126mm frame without telling me?


Secondarily, I put in an "x-treme range" (or something like that) Shimano cassette and did the upside-down b-screw thing and... it basically works. With the new dishing/spacing, however, if I'm not in the granny gear when I go into it, the derailleur veeeeery gently pings the spokes. I've adjusted the limit screw to minimize this, but I assume it's because the cross-chaining pulls the derailleur inboard.


Anyway, could I use something the Wolf Tooth Roadlink to clean this up a bit? A huge pie plate just in case? The derailleur is the original Deore long cage that came with the bike. I still use the downtube shifters in friction mode.


Which is all to say I'm in the process of building up an all-terrain bike for doing more dirt touring/camping from a new Ritchey Ascent! Woot! So when that's done maybe I can stop trying to force the Miyata to be so heroic.

zygomorph

By "pulls the derailleur inboard" I think my theory is that the cross chaining pulls the *bottom* of the derailleur outboard, forcing the upper part of the cage to twist inboard just enough to sometimes gently graze the spokes.

FBinNY 

I'm not proponent of tensions higher than necessary, and 95kgf is normally OK for a 2mm spoke. However, the issue is on the left side, where tension will obviously be lower. Based of the dimensions my off the cuff estimate would be somewhere near 50kgf, which is inadequate for a 2mm plain gauge spoke.

Consider carefully by degrees, ie. 1/4 turns all the way around, as needed to bring the right side tension to about 110kgf, Then tighten the left, again by degrees to redish the wheel back to center. Note, you want to do all the tension increase on the right (tighter) side first, then finish using only left (slacker) side spokes to correct dish and alignment.

For future reference, consider using butted spokes, at least on the left side to lower the minimum necessary tension.

zygomorph

Thank you, that's great to know! Especially re: butted spokes. I suppose I'd want to read some detailed books on the subject before building up my own.

I just now measured my NDS spoke tension at just over 63kgf (according to the Park Tool app), is this still lower than desired?

Bill Kapaun

I made a guestimate of 131mm spacing and with that hub, NDS tension would be 58.5% of DS per Spokcalc. Therefore, 90 kgf DS would be about 52.6 kgf NDS. (a 2.5mm OC rim would increase those numbers about 11%)
110kgf would give us about 65 kgf on the NDS. I think that's about the number we prefer for a minimum?
The CTF's of a disc hub actually helps in this case.
An offset spoke bed rim would really help.

zygomorph

The measured tension for my NDS is currently 63kgf. I'm happy to go up just a smidge if that'll make a stronger wheel.

I'm surmising that we're not really concerned about the angle of the drive side spokes so long as the tensions are good?

FBinNY 

I'm a bit concerned about your numbers. The ratio seems too narrow, but it's easy enough to cross check.

The ratio between right and left tensions is equal to the ratios of CTF distances.

So, measure both with a ruler - 65 minus flange to frame, and compare. If all the numbers jive, your OK but still might to add some tension.

However, if the don't jive then your tension messurements are off and adding tension isn't optional.

Bill Kapaun

ALL of them or just 1?

Bill Kapaun

If the tensions are well balanced, the angles have to be. There really is no 1/2 dead.

FBinNY 

I hate to be a PIA, but this kind of stuff makes me crazy.

How can you calculate a "guestimate" to 3 figure precision? It's a mathematical oxymoron

zygomorph

Yes, all of them are *close enough* to 63kgf for the purposes of this discussion.

Re: the dish angle, my understanding is that more dish (a steeper drive-side spoke angle) results in a weaker wheel.

zygomorph

Cool, I'll do this in a bit. The wheel is currently off the bike and I'm doing other things at the moment and can't get my hands dirty (uh, like, typing on a laptop). I assume you suggest "flange to frame" because I couldn't find any easy way to measure with a ruler or calipers to the lock nuts. OTOH it still looks as though I'd still have to remove the cassette to access the drive side flange.

To confirm, I would be doing these measurements to confirm whether my spoke tensionometer measurements are accurate?

Aside from my tool being wrong, or the Park Tool app being wrong, what might cause this situation?

Bill Kapaun

Front wheels (non disc) have no dish and are considered stronger than rear.
The "less" angle (like rear DS) are more prone to collapse from a side thrust for one example.

Now let's just simply add "straight down" weight to the bike.
The rear DS spokes are going to fail first, since they already have 50-70% more tension to start with than the NDS.
A (non disc) wheel will have both sides fail at the same time at a noticeably higher weight since both sides started equal.

Bill Kapaun

I should have said "presumed". After that, it's the calculator output that I rounded down to 3.
If I were your employer, could I make out your pay check with just 3 significant digits?

FBinNY 

Yes, no and maybe. This statement is an example of a wrong conclusion based on only a partial understanding.

Imagine a.tripod set up so one leg is near vertical and the other 2 splayed out at a wide angle. Then, yes the camera is more prone to being knocked over toward the more vertical leg.

BUT

a wheel isn't like a tripod. In a wheel the rim is braced in either direction by the spokes coming from the other side.

So, while the rim is closer to the right flange, it's braced by the left side spokes which have a favorable angle. OTOH the right spokes have a less favorable angle, but the rim has more room to move.

So, while narrower flange separation makes a wheel more likely to taco, the asymmetry in and of itself is less important.

In any case, normal side loads are low enough to prevent issues, and radial loads are unaffected. If this weren't true many millions of bikes would be intolerably defective.

FBinNY 

It's a point that separates BCE folks like myself and AD people. (before computer era / after digital).

One of the fundamental rules for calculated results is that the result cannot be more precise than the worst of the inputs. Digital computer programs without rounding spit out credible precise results, but that doesn't make them meaningful.

FWIW I had very few jobs in my life, most of which had me as the employer. While in college I worked for UPS which used digital time clocks with 100 "minutes" per hour so it was easier to calculate pay.

zygomorph

Anyway.

Instead of disassembling my wheel to measure the CTF distances, I found somebody else who already did it. https://www.kstoerz.com/freespoke/hub/97. I have no reason to believe at this point that my hub flanges are like, bent in some impossible way so I'll assume these apply to my own hub.

The ratio of the NDS to DS CTF measurements is 1:1.63.

The ratio of the NDS to DS tensions -- as measured on my wheel as it is now -- is 63kgf:95kgf per spoke, or 1:1.52. If I very carefully remeasured to, say, half-mark readings (the most recommended by Park Tool for their TM-1) it might get closer...?

However, is this as far off as you fear? I mean, I'm perfectly willing to accept that this wheel might well just be a relative POS (there's a slight hump in the rim at the join, for example). It still operates fine for my daily commuter purposes. I might not want to trust it on a big fat tour, though.

FBinNY 

Don't overthink it. It's a bike. Not a moon lander

It's fine as is. There's always a margin of error, so close enough is good enough.

So, while some more tension MIGHT be desirable, it likely isn't necessary.

FWIW the issue with inadequate tension is an increased probability of spokes loosening over time. So, keep an eye on it, and if it does loosen, you'll know to go higher when you retension.

zygomorph

LOL did *you* just say that to *me*? But seriously, if you say "the ratios of A:B should be the same as C" I don't know how close is close enough. If you say "the NDS spoke tension seems low" I don't know how low is too low or what is desirable. This is why I ask.

Thanks for all the insight, anyway.

Bill Kapaun

+1
IF it's an old wheel, the nipples may be corroded enough that they are effectively "locked".
My first experience with Trig was using a slide rule.

zygomorph

Per my original post, I removed spacers and trued and re-dished the wheel with a truing stand and a spoke tension meter because 1) I had spokes coming loose and 2) the OLN was uncomfortably large for my frame. None of the nipples were seized or stripped. I wanted to confirm whether the tensions were reasonable, which you two have helped to confirm. Thanks!

zygomorph

I could lie and say my first experience with trig was as a Pythagorean initiate. Or with a Documenta Geigy. But I won't. It was with a TI-30.

FBinNY 

There's errors in everything. The tension meter may be slightly off, the CTF numbers likewise, and minimum tension is purely a rule of thumb because we don't know what the wheel will be subjected to in use; bad roads, heavy rider, etc.

Also, unlike a moon lander, there'll be plenty of opportunity for remedial action if/when it's needed.

ALWAYS consider the application:

Airplanes have to be built to higher standards of reliability because pilots can't pull over and call AAA if an engine overheats.

As for minimum tension for a 2mm spoke, I look for anything above 60kgf. I also try not to exceed 110kgf for the tighter side. If I can't meet both conditions, I split the difference, and get both as close as I can.

Bill Kapaun

Their numbers disagree slightly with Shimano's numbers. Maybe one is measuring to the "true center" of the flange and the other isn't?
It turns out Spokcalc lists a 45mm PCD vs 61mm for the other 2. So, obviously errors are to be found.

BTW, many rims have a slight flat spot at the rim joint. As long as it's minor, ignore it. Trying to "true it out" screws things up more than it helps.

zygomorph

Thanks. Unfortunately that document is illegibly small. Perhaps as small as the measurements in question.

Yeah, as soon as I noticed that the flat spot was around the join I undid my efforts to "fix" it as I figured it was an anomaly of the manufacture.