Rebuilding a comstar information thread
Page 1 of 3 1 2 3 LastLast
Results 1 to 10 of 27
Like Tree5Likes

Thread: Rebuilding a comstar information thread

  1. #1
    Junior Member Honeybadgers's Avatar
    Join Date
    Jul 2016
    Posts
    29

    Rebuilding a comstar information thread

    *gasp*

    (insert comments regarding "DO NOT DISASSEMBLE" here)

    I'm going to update this post as I find information.

    As I said in an earlier post, I found a spoke had one minutely loose rivet when changing tires. it did not move left or right, only in and out by less than a millimeter.

    We all know that honda has said DO NOT DISASSEMBLE the wheels, but I wondered why. is it assembled under tension and I'll be unable to keep the rim true? Do I risk catastrophic failure of the wheel? If so, what part is going to fail?

    My logic is as follows; firstly, honda's main reason for telling people to not disassemble the reason is twofold. It means that when they do lose a rivet, you have to buy a new wheel. Great for honda, bad for us now that they don't make the bloody things anymore, and it assuages them from liability were someone to do it themselves and get something wrong. Second, the only thing we ever hear about failing is the rivets themselves, never spokes or rims. So now that our bikes are breaking the 40 year age barrier, the weak link most likely to cause said "failure" is these aging rivets themselves, and they're all essentially time bombs. Third, a bolt is stronger than a rivet when properly measured, assembled, and if loctite and safety wire are involved. If safety wire can keep a jet engine together, it should do marvelously for a wheel spinning a hundred times slower. And lastly, if something is to start failing, it won't be a bolt. It will be the spoke or the rim itself.

    Hairline cracks will be plainly visible and will progress to very noticeable vibration in that much rotational mass, and if the unthinkable were to happen and a spoke breaks away, it has a supporting spoke on the other side, the spoke will only straighten and will still clear your forks, chain, and swingarm. This situation is insanely unlikely if you check your wheels for fracture regularly anyways, as you should already be doing when you check your tire pressure, so I simply don't think that the bolt will differ in any way from the rivet. Even if a whole spoke fails bilaterally, the wheel is still supported through less than 144 degrees, which should not flatten the rim before you get it to the side of the road. that scenario seems just as likely, by the way, as it does with rivets in place.

    We've seen several people do this before, and I've yet to hear of a single failure, minor or catastrophic. Not many people have done it, which is why I think this needs to be further investigated, and as wheels get more and more rare, those rivets are staying the same age and eventually we won't be able to replace them. There are wire spoke conversions, sure, but for many people that's simply not a practical solution.

    So to that extent, I'm willing to put my rear rim on the line. I'm going to be extremely careful, only riding the bike at speeds under 30 for ~300 miles afterwards, and will be inspecting them before every single ride. Hardware I'll be using is grade 8 SAE or metric grade 10.9 (about identical to grade 8) The argument between grade 8 vs 5 in terms of strength vs ductility has been well debunked here: Grade 5 vs Grade 8 Fasteners - TineLok and Killing the myth of the Grade 5 bolt. I will also be drilling both the head of the bolt and the nut for 0.32 safety wire, tying each pair together and threading the tails through the centering holes in the spokes that were used for aligning the rivets on initial assembly. Once I have everything mocked up, I may also use loctite 648, which is likely insane overkill (i use it for press fit and flywheel bolts on rotax 990's, an extreme vibration application. usually requires a blowtorch to degrade the compound enough to loosen) As for the one detail of hardware I do suggest, go with aircraft grade. It is made from the same material as the equal grade from your local hardware supplier, but has more quality control tests to conform to. Every little bit counts, and you might spend a total of five dollars more. Once I know exactly what bolt I'm using, I will post its specs so you can easily cross reference them to an online supplier.

    Prior to the removal of the spokes, I checked the wheel's runout, and again after every single spoke's rivet removal. I found a 0.046in runout almost exactly opposite the loose spoke beforehand.

    My only concern regarding this is that the aluminum rivet had some ductility to it, which imparted a microscopic amount of flex to the spoke itself. I don't forsee it being a problem, but this should be the only actual difference between there being a bolt and a rivet in that hole.

    I have extracted the aluminum rivets using a 1/2 drill bit on my press to remove much of the rivet head (using a center punch to get things aligned) and then a 3/16 to drill down into the rivet body, and then a punch and hammer to simply tap the rivet out. This required a few firm taps, but resulted in a clean extraction without drilling into the spoke on accident. The locked bolts on the hub were tight, so for this application, I left them in place. Someone had apparently put a fairly clean weld along the spoke overlap on the hub side, and since it's rock tight and still has the original bolts, I'm not going to bother with disassembly there. But removal of those bolts involves simply breaking them off, more or less, and replacing them with normal bolts. Again, use the right size hardware and consider loctite and safety wire.

    Measurements of the runout after each spoke were identical, 0.046, all in the same location. with the rim freely rotating on the spokes, the rim has not lost any trueness, and the old rivets slide easily through the holes when aligned to the spokes (a drill bit or 1/4 bolt fit nicely into the inner two alignment holes to hold it square for that test. Thus, the spokes are not tension set and the rim remains as true as it was. This rebuild won't be adding any static stresses to the spokes as a result.

    I measured the rivets and have run into a problem. The rivet barrel is 7.25mm in diameter (0.285in). That doesn't exist as a bolt. So I have two real choices. Either bolt it together with 7mm hardware and not have a flawless fit (alternatively I could attempt to use a helicoil as a jacket to the bolt as a spacer, but I think the diameter would then be approaching 8mm,) or center everything very carefully and drill each hole out to 8mm. I'm really torn over this, because one will reduce the structural integrity of the spoke bilaterally by 3mm, and the other has 1mm total play. This is where I'd really appreciate input, I'll also be reaching out to some aerospace engineering forums for input (and will post their responses here)

    I also stripped the gnarly black paint from the rim, so it'll be silver, though not insanely clean, since I'm not going to put hours into cleaning a rim that may not work out.

    Stopping by the aircraft hardware supplier tomorrow. Will update.
    Last edited by Honeybadgers; 08-23-2016 at 11:34 PM.
    bilbikek411 likes this.

  2. #2
    Member trilobitnz's Avatar
    Join Date
    Oct 2013
    Location
    Nelson, New Zealand
    Posts
    63
    Time and time again at work we see people come in and ask for a 'stronger' bolt ie highest tensile strength possible for their car, bike etc...

    The tensile rating of a bolt is exactly that, it tensile strength, the power it gas to resist being pulled or stretched. An application where high tortional load eg/ something that is required to be torqued up and therefore a higher clamping load is where 10.9 or 12.9 (grade8 and 9) should be used. There is often more surface area in a high torque joint which helps elliviate pressure or force coming perpendicular to the bolt.

    8.8 (grade 5) is used where the required torque wont be succicient to strip the thread, but also where there is a varying and constant side load (force coming perpandicular to the bolt).

    The metric system of rating also tells you what forces the bolt is capable if holding.
    Hopefully this can help:
    These are grade 4.6 and 8.8. The first digit relates to the ultimate strength of the material, whilst the second is the ratio of yield stress to ultimate strength. Thus grade 4.6 bolts have an ultimate material strength of 400N/mm2 and the uield (or proof) stress is 60% of the ultimate strength. Simlarly grade 8.8 bolts have an ultgimate strength of 800 N/mm2 and a ratio of yield/proof stress to ultimate strength of 80%. "
    http://www.asengineering.co.nz/whats-the-difference-between-46-and-88-bolts

    Basically that second number is what we're interested in at the moment. The easier way to look at it is that it tells you how much stretch/deformation it has before it fails, so at 80% strength it will stretch 20% before it snaps.
    10.9 (grade 8) and 12.9 (grade 9) has 90% strength which means they will only stretch/deform 10% before breaking.

    This means that a sideways force only has to prove less tortal movement in the higher tensile bolts before they fail. And it doesn't have to all come at once

    I have sold bolts for cranes, bulldozers, custom hot rods, cars, bikes, and kids scooters. I have seen first hand what happens people ignore our advice and put grade 9 bolts in the flywheel of their car rather than use G5 as per what is in there stock then try to sue us when those bolts have snapped and ruined his engine. Kids scooters have 8.8 (G5) bolts for their axles that strip threads from over tightening them, but if they use a 10.9 or 12.9 (g8, g9) they snap after 2 days of use going over bumps.

    I may seem to be ranting here but using a higher tensile bolt in this application where there is not much surface area being joined to provid enough surface friction to help take the the contant side load these bolts will be under.

    I sell bolts as my job. For 5 or so years and there is a reason for the different grades other than cost. I have read your posted links and am not convinced they are right. The tests they proved didn prove anything other than the high grade bolts are 'stronger' whch of course they are, but not under every operational condition.

    It is far better to have an 8.8 (g5) bolt stretch and be see to be wiggly before it fails, like your rivet, rather than have it fail catastrophically with little to no warning.

    My 2c. sorry if I havent explained myself very well, easier talking about it face to face than writing!
    Last edited by trilobitnz; 08-24-2016 at 12:38 AM.

  3. #3
    Junior Member Honeybadgers's Avatar
    Join Date
    Jul 2016
    Posts
    29
    Quote Originally Posted by trilobitnz View Post
    Time and time againcat work we see people come in and ask for a 'stronger' bolt ie highest tensile strength possible for their car, bike etc... The tensile rating of a bolt is exactly that, it tensile strength, the power it gas to resist being pulled or stretched. An application where high tortional load eg/ something that is required to be torqued up needs highr torqued 8.8 (grade 5)
    Not sure what you're saying. The grade 8 has a higher tensile strength as well as a higher shear strength. The only time where softer vs harder metals and fatigue rates are concerned is in truly extreme situations that the bike will never be subjected to (like project azorian, where that steel was near freezing, under obscene load beyond its design limits, and under millions of pounds of pressure) I honestly doubt the grade 5 bolts would even cause an issue (that 85,000PSI load rate of a grade 5 vs the 120,000PSI rating of a grade 8 will never even be remotely reached,) but since I'm already building the wheel to overkill specs. I don't plan on a crazy bolt torque number to prevent loosening, I need to do some more research on the torque number I actually want to use (and will update the main post when I know) and will use the loctite and safety wire as insurance against the fasteners coming undone.

    Shear maths.

    Grade 8
    A = Cross-sectional area of the fastener size (since bolt bodies/shanks have circular cross-sections, use area of a circle) = Pi x r2 where R (radius) = .250/2 = .125, therefore A = Pi x (.125)2 = .0491 square inches (in2)
    Capability in shear = 91,000 lbs / in2 x .0491 in2 = 4468 lbs

    Using the same .250-inch diameter grade 5 fastener results in the following:
    Capability in shear = 75,000 lbs / in2 x .0491 in2 = 3683 lbs

    Even a grade 5 bolt will have unbelievably superior shear resistance to those aluminum rivets.
    Last edited by Honeybadgers; 08-24-2016 at 12:11 AM.

  4. Remove Advertisements
    HondaTwins.net
    Advertisements
     

  5. #4
    Member trilobitnz's Avatar
    Join Date
    Oct 2013
    Location
    Nelson, New Zealand
    Posts
    63
    Sorry hit post by accident halfway through. See edited post

  6. #5
    Member trilobitnz's Avatar
    Join Date
    Oct 2013
    Location
    Nelson, New Zealand
    Posts
    63
    What im trying to say is use g5, not g8! The metal will fatigue under non extreme rates, and before it gies, it wil tell you!

  7. #6
    Junior Member Honeybadgers's Avatar
    Join Date
    Jul 2016
    Posts
    29
    Quote Originally Posted by trilobitnz View Post
    What im trying to say is use g5, not g8! The metal will fatigue under non extreme rates, and before it gies, it wil tell you!
    That is all well and good, but the bolts will not be under severe torque load. Sure, the concept of a stretch bolt has certain important uses, but all of those scenarios you described are the results of overtorqued applications. If the person using that grade 8 in his flywheel had used grade 5, he would have sheared them when he was tightening them. Hell, an LS1 only has a flywheel torque spec of 44lb/feet. That should never break a grade 8 at operating temperatures and vibration levels. Now, if he put them on with an impact gun, then we're talking about over 100 pounds added to however much the actual hammer strikes were at. Impact guns break bolts more than bolts fail on their own.

    We're not talking about an application that has fatigue numbers anywhere near approaching the limits of a bolt. The aluminum rim would deform or the spoke would fracture before the grade 5 even began to stretch. Remember, the material in there before was -aluminum-

    The major applicable difference in grade 5 vs 8 is not application, but cost. if a grade 8 bolt is 25 cents more expensive than a grade 5, and a manufacturer uses millions of these bolts a year, you can bet your butt they're going to be using the minimum safe number they can. It's really not a matter of strength in any but high stress applications. Fatigue is only a problem when the bolt is put under extreme or repeated flex (at figures way higher than the aluminum rims would tolerate) or extreme temperatures and pressures.

    The math is solid, and while your rationale is perfectly legitimate, if I was overtorquing the bolts to cause stretch, I made a mistake before I took to the road, and it's not the bolt's fault. Since I'm going for a whopping ten bolts, I'm willing to pay the few bucks more and have increased shear and tensile loads just in case there was a manufacturing defect that caused a slightly weakened bolt from the factory.

    I'm sorry to say, but in the applications we use bolts (not subjected to extreme pressures or temperatures, or load ratings anywhere near the failure point) then the higher grade bolt is simply a better choice as long as it's been torqued appropriately for the application. In our circles, it's simply an old wive's tale.

    And that's not even counting the insanely unlikely scenario wherein if the bolt shears, there is another bolt there. If that bolt shears, the rim is still fit to the spoke with a lip and you'd know immediately something was wrong and have plenty of time to get off the road. And even then, if something insane happened, you still have four other bolts supported angularly across about 140 degrees of angle, to support the rim long enough to take it off the road.

    little more practical barnyard example.



    I'll give ARP and BBS wheels (since I figure BBS would have a good idea regarding bolts in two piece race wheel applications) and see what they have to say.
    Last edited by Honeybadgers; 08-24-2016 at 01:22 AM.

  8. #7
    Member trilobitnz's Avatar
    Join Date
    Oct 2013
    Location
    Nelson, New Zealand
    Posts
    63
    Sweet. Regardless of which way you do it I am looking forward to seeing how it goes, and if all else fails, you could always use the hubs to convert to spokes as has been done previously!

    Matt

  9. #8
    Junior Member Honeybadgers's Avatar
    Join Date
    Jul 2016
    Posts
    29
    Left a message with BBS, but according to the guys at ARP, at the levels of stress that a wheel will experience from vibrations and shear loading, real fatigue resistance is going to come from the heat treatment of the bolt, so an aircraft spec 8/10.9 will absolutely outperform the 8.8. The only time that the failure attributes of the grade 5/8.8 will become apparent is in high temperature/pressure or high vibrations at constant high axial loads, and even then, the bend/break situation in a wheel is going to be moot. That grade 8 bolt will simply last longer, and if something fails, it's going to be the thinner metal spoke or rim long before a properly heat treated, reliably sourced grade 8. The main reason is that we're bolting together 1-2mm aluminum plates, not 1/4+ plate steel.
    Last edited by Honeybadgers; 08-24-2016 at 03:51 PM.

  10. #9
    Super Moderator longdistancerider's Avatar
    Join Date
    Oct 2011
    Location
    Vancouver Washington (text 408-239-9580)
    Posts
    21,590
    Play is definitely a bad thing. That leaves enlarging each hole. My gut feeling is that this won't affect the structural integrity of the rim enough to be of concern. Engineers know more about this than I ever will.
    Is it possible that there is a 7mm shoulder bolt that has a thick enough shoulder that can be re-threaded?
    Possibly use the 7mm hardware with a solid steel pin inserted between the bolts? You'd have to assemble the wheel and get any vertical runout squared away first, then drill the pin hole and glue the pin in place.
    Jim O'Brien
    1979 CM400T "road bike" modified for travelling
    1978 CB400T1 restored
    1972 CL350 survivor restoration


    https://www.hondatwins.net/forums/22...d-attempt.html and https://www.hondatwins.net/forums/1-...r-attempt.html
    Road Trip https://www.hondatwins.net/forums/11...here-i-go.html or "where's Jim now?"
    Member Map https://www.hondatwins.net/forums/dcMembermap.php

  11. #10
    Junior Member Honeybadgers's Avatar
    Join Date
    Jul 2016
    Posts
    29
    That's certainly an idea. But I didn't have time to run to the hardware supplier, but I think I may pick up some m8 bolts and simply measure the threads. Unfortunately, I don't think I'll be able to find any bolts with a grip length of about 0.5cm, so I may be stuck using a bolt that has threads as the tensioned structure. Not ideal, but again, I simply don't think that the bolt will be anywhere near a risk for failure. I think I may just enlarge the holes in the end. I've got a drill press and getting the wheel aligned exactly is surprisingly easy.

Page 1 of 3 1 2 3 LastLast

Sponsored Links

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •  

Search tags for this page

comstar wheel restoration

,

how to rebuild comstar wheels

,

loose rivets comstar rim what to do

,

painting comstar wheels

,

rebuild comstar wheels

Click on a term to search our site for related topics.