Welcome to our resource for MG Car Information.
|
MG Midget and Sprite Technical - Uprated Front Stub Axles
http://autos.groups.yahoo.com/group/spridgets-aus/ There is a thread on this topic running on the Oz board. It may be of interest to many of you. |
Mike Allen |
That's interesting Mike. Here in the states most of the racers use a modified stub axle where the spindle is actually machined out of the upright and a new spindle made from 4130 steel is pressed and welded into place. Seems to be basically bullet proof. A description of the part is in the Winner's Circle catalog. http://www.spridget.com/catalog/25_FRONT_SUSP_SWIVEL.asp |
Bill Young |
Yes, that is interesting. I'm replacing the front brakes & hub bearings this weekend on my '78 spridget. Right side bearings are good, left side has too much play & an occasional rumble...no telling what I'll find. Thanks...interesting reading. Dave |
Dave Rhine |
I encountered the "bearing" and "stub axle" problem while restoring my '74 midget. Trying to get some answers to the problems I found eventually led me to some in-depth research of the problem over several months. I have only a rough draft report of what I found but because of the seriousness of the situation I found here it is. You will notice there is a different possible cause of axles breaking. http://smithtr6.com/bearings.htm |
tomshobby |
I was going to take part in the topic on Yahoo but found it confusing to navigate. At least I posted a link to my findings here. |
tomshobby |
curious tomshobby, Did you use the little spacer between the taper bearings....we had some lively discussion on this last summer...it was under the heading of "40 degree angular bearings" but I cant remember the heading title exactly. I want to say it was around "may perhaps april" interesting read, I dont think Ive met anyone that has fractured a hub before, yes Ive heard of it but never seen it,,,,Was it perhaps the wheel width that contributed? prop |
Prop |
Hi Prop, I read that thread and as many others on different forums that I could find. I will not use the spacer with the taper bearings. The first reason is that it would be impossible to correctly adjust them with a spacer. Have replaced thousands of taper bearings and never even seen such an application. The second reason is that if the inside taper bearing fits properly, I believe it will provide better support than the original setup. The diagrams in my article should demonstrate why. Not claiming to have all the answers and always interested in the opinions of others. In fact I ask some questions in my report, one being about the 40 degree angular contact bearings. I have contacted many bearing manufacturers and suppliers. All I can say is that it was a very disappointing experience. |
tomshobby |
""Have replaced thousands of taper bearings and never even seen such an application."" THOUSANDS Ah! Good grief, you must drive a midget 24/7/365...., Ive only replaced mine once. LOL. Sorry I dont mean to be beligerant. In fact I dont have enough book knowledge or hands on to offer any real comments or suggestions What where you disappointed about on the 40 degree angular bearings The one thing I stumbled across on the 40 degree angular bearings was to use the brass races, instead of the peidiamite races, the brass is more cabable of handling the needs of our cars much better, where as the pediamite is great for sub zero enviorment aka refrigeration, croosive chemicals, and continuous running power tools. not to mention the brass races are a 1/3 the cost of the pediamite races in the 40 degree angular bearing. Have you considard having a set of stainless steel 4130 hubs custom made...that would solve a lot of issues Id think there are 2 people on the board that are making them...you might try starting a new thread asking who the people are...One person I know is Deb evans with pristiest race engines, she has devoloped a really nice front hub Her email address is (pr.engines@tiscali.co.uk) (Sorry for the future spam my sweet lovely Deb) Unfortantly I cant remember who the 2nd person was, but I do remember that they have an outstanding reputation and are well worth a call....I want to sat it was peter burgass. If you want those code numbers for the brass race 40 degree angular bearings let me know..it will take a couple of days before I can dig them out...but I have posted them in the archives....under something like "Prop is an enginerring God" Prop...Yeah its Hard to belive I suffer from a debilitating deases called "Humble pie -litesis" |
Prop |
I just skimed over your artical...vary good. I have a differant thought as to why there is a vast differance between hubs, that goes hand in hand with my newest food for the brain adventure The Hubs are cast Iron....The optiumim word being "CAST" As to mean a mold was formed and molten iron was poured into the mold. considering this was most likely done in the UK where there are varying degrees of temp, humidity, and weather over a span of a year, id think you would get all differant sizes of the same mold for the hubs exampple if you where to make an aluminum cly head cast out of aluminum during the summer along the equarter you have to only allow for a 9% shrinkage of the cly. head as it raltes to the actual size,,,but if you have the same outside structure in hard winter enviorment you have to account for as much as 20% shrinkage in relation to the correct size....Thats alot. Granted I dont know if there is srinkage in iron, or even what effects enviorment has on iron casting...But I think you get my drift and where that discussion can end up I offer up My EN40B crankshaft of evidence that metal casting for moving machinary was not a strong suit for humanity 50 years ago ...when I had my crankshaft indexed, it was shocking how far off it was ...No way could those tolerances have worked in a new toyota engine of 2010, yet perfectly fine for a 40 year old engine...BTW if you read the Bible, visard has a nice artical about crankshaft history and science and the huge difficulty BMC faced. I have to belive the same issues of forging and casting was appearnt in many other aspect of car part manufacturing aka...hubs and spindles. I just think on a mass production scale where these hubs where used on many differant cars, and being cast followed by a slight amount of manufactued with human hands machine work...your gonna have some variation from one hub to the next, remember these are not rolls royeces, but a mass produced sports car like the VW Bug. the object was not about fine labor intesive assembly, but how many can you make in an hour with in spec. A note...even tho Im playing with the idea of a making custom cly. head Ive only recently started reading about fondry/casting principles and havent touched foundry/casting scence high school metal shop, So Im a long way from giving advice or participating in a well thought out debate of the principles of manufacture, casting, and foundry, that applies to this thread...Its just a personal frame of thought I have for why there is so much variance between hubs. Prop...Now Taking cash orders for my new hemi aluminum cross flow 8 port clyinder head for the A-seris BMC 1275 engine to come out in 2025 |
Prop |
Prop, Glad you looked at the article. Like I said, I found some answers and raised more questions. In the process found a solution for my particular situation. I knew from the start that I could find a solution for myself. The reason I kept going with the project was because I was hoping to find a universal solution. One that would make it possible for people to find "off-the-shelf" replacements readily where ever they might be. The answer might be from the replacement axles being made. If they use a 1mm radius there will be a much better choice of bearings. Or it might come by machining the hubs like I am doing. In the meantime I do wonder how many Spridgets might be on the road with bearings that do not fit the stub shaft radius. |
tomshobby |
I am curious as to how many stub axles have broken on Spridgets, Morris Minors, A30s, a35s etc etc!! I have not come across any actually so is it a problem? except for those used in extreme conditions on uprated suspensions and huge tyres and brakes? Certainly there is a problem of modern bearing fit that the bearings available today are NOT made to the same spec as they were when BMC ordered them, even though they have the same number However using the 40 degree angular contact bearings gives us the ability to specify EXACTLY the specification we need to suit the Spridget front wheel hub and thus mimic the original design correctly see here: http://www.mgcars.org.uk/mgcc/midgetreg/Frontwheelbearings.shtml I think this thread has become confused The 40 degree angular contact sorts out the problem of poor fitting out of spec modern bearings and is a CORRECT replacement for the original high spec bearing. If stub axles are required to be uprated for high performance use then that is a different problem |
Robert (Bob) Midget Turbo |
after studying the link above, which nicely explains the importance of making sure that your ball bearings are face machined 40 degree bearings, the writer does admit that an alternative would be to install tapered roller bearings. he points out the differences between the two setups. The key point is that if you use the original type of bearing you must get the face machined version, for the original set up to work as intended. In EITHER case, it is important to make sure that the bearing inner race is not a smaller radius than at the root of your stub axle or they will interfere. Because this radius is important to the strength of your axle, you need to buy bearings with larger radius (not cut the axle). John Twist in his University Motors video, explains the importance of the spacer and the nut torque to put the stub axle into tension. Note that the example he uses is an MGB, which has tapered bearings AND spacers. http://www.youtube.com/watch?v=RJYIXZtnIiw From his explanation, it seems that changing the front bearings from the original ball + spacers to a tapered without spacers would expose the stub axle to greater stresses (because the stub would no longer be in tension). This is not to say that tapered bearings could not work, but that this difference also needs to be addressed when doing so. If I remember correctly, modern cars have the bearings flipped around so that the important stub axle tension is created by means other than shimmed spacers. Norm |
Norm Kerr |
Robert thanks for posting that link, I was looking for that last night to post. IMHO I thik what we are seeing is an evelosion of the bearing...1st it was the orginal designed by BMC, then a cheep ill fittting knock off, then the taper which from the people Ive talked to that hae it swear by them, and now the 40 degree anguar, which appear to be the cats meow, Im looking forward to installing mine at some point Prop |
Prop |
Norm, that is a great explanation of MG's spacer use. But it does not address the radius problem of the Midget axles. I am also interested in finding reference to the dimensional tolerance between the hub spacing and the inner spacing for the ball bearing application and if the 40 degree ball bearings require a tighter tolerance than the 20 degree bearings. |
tomshobby |
Tom, Robert sent me the RHP drawings for the bearings supplied to Austin for the application and the radius specified for the inner race of the inner bearing, where it butts up against the shoulder, is 2mm minimum. No wonder that the bearings in that chaps article caused problems if the radius was less. I've looked at some Porsche and VW taper roller bearings and the matching radius was much larger, maybe 4mm, presumably so the stubaxle radius was similar to reduce the stress raiser. |
David Billington |
Hi David, I was able to look at some old RHP catalogs but was not able to find that information, thanks. Could you a copy of that information, I would much appreciate it. The radius is definitely at the crux of the problem. The bearings being sold for our cars seem to have a smaller radius and the manufacturers and suppliers do not seem to care. I believe if that issue could be resolved all else would start to fall into place. For those not aware of the radius problem you can check out the drawings in the link I posted earlier in this thread. |
tomshobby |
Tom, If Robert doesn't have a problem with my sending them to you I can do so easily, but I would need an email address. |
David Billington |
tom(at)smithtr6.com Thanks David, it would be a great help. And pass on my thanks to Robert. |
tomshobby |
This IS an interesting thread! I just finished a brake job on all 4 corners, in the process fitted new front wheel bearings (ordered from Moss...thought they would be good quality) & now have as much, if not more play, than with the original 32 year old, worn out & rumbling bearings! I wish this thread had been posted before I ordered my new parts. I haven't driven the little beast yet, but I'm sure I won't like the results. Thanks for the post Robert...the 40 degree angular may be the direction I go now. Dave |
Dave Rhine |
I have one of the same questions that Robert has. That being I have also only heard about stub axles breaking. I would be very interested in hearing first hand experience of one breaking and where the location of the actual break was. |
tomshobby |
Hi Norm, I am the writer of that link on wheel bearings. I am sad that I have given the impression that tapers are a creditable alternative. Originally tapers were an alternative because they removed he play from the front wheel bearing that allowed those of us in the UK to pass our annual inspection. That said I believe that tapers without the central spacer would significantly weaken the total design of the hub. Thus originally IMO it was a better option to use standard 20 degree bearings and machine the spacer to take out the play caused by non face adjusted bearings. However with face adjusted bearings being available in 40 dgree spec then we should no longer have a problem of accurate fit and thus all other alternatives should be refused en bloc. By the way Norm most modern cars use double row 40 degree bearings that obviously are the same deign as our original Spidget but come complete in one housing and thus spacing is done within the bearing. This I think proves the original concpt was the best one. It is understood I hope that if the radius of the stub axle was too large then the 40 degree bearing would not fit correctly and would actually become heavily loaded, is this happening? Tom Dave will send you the information you require, he is a far better engineer than I will ever be. |
Robert (Bob) Midget Turbo |
Thanks Dave and Robert for those drawings. It answers some of the confusion that I encountered. I had the opportunity to look through a couple of old RHP catalogs and they only offered the LJT25 bearings with the 1mm inner radius and mot the 34LJT25 with the 2mm radius. Part of the trouble I have had identifying which bearings to use has been the reluctance of the bearing companies to give out information. I did try to get information including the geometry of the NSK and SKF bearings you named in your article, Robert. Robert, maybe you can answer the main question I had, do these bearings have the 2mm or larger radius to fit the stub shaft radius? If they do you might want to include that in your article. Another question I have about the 40 degree bearings is this, do they require a closer spacer tolerance when fitting. I might mention that the only, and only, reason I gravitated toward taper bearings was that I the only bearing I had found that fit the radius was the Koyo bearings. I had even spent an afternoon at Peter's shop, World Wide Auto, measuring radii of several stub shafts and test fitting different bearings on them. We found out that none of the bearings we had, with the exception of a pair of original Sprite bearings, fit, period. This I know, if I had a set of bearings, and especially taper bearings, that I had purchased and installed I would certainly remove the hub and make certain they fit the stub shaft correctly. If they have the space behind the bearing like my drawings show the results could be very unpleasant. |
tomshobby |
Would somebody please direct me to a written manufacturer's description of what "face adjusted bearings" actually means by the people who make them? I've never heard so much blather that entirely misses the critical point. FRM |
FR Millmore |
Tom, You should be able to get any dimensional information from the bearing manufacturers web site such as http://www.skf.com as it's crucial for the design process that it be known. Regarding variants such as larger radii then if they are not mentioned then a call to the bearing manufacturers technical support may prove useful. Last time I needed some technical information for SKF bearings used on the rear hubs of Ford Sierra/Granada I called SKF and they provided all the usual technical information for the bearings even though they were a special for that application. I had been asked to do a life calculation estimate for the bearings when they were used in a race car application. |
David Billington |
Go on then lets hear some real blather FRM and explain what the crucial point is? What a dick head! If you don't understand something then don't assume it is not important. The expression FACE ADJUSTED is the whole reason modern bearing equivalents do not fit! obviously I think the explanation on our website is easy to understand for anyone from an engineering background so what is your problem English? Engineering? |
Robert (Bob) Midget Turbo |
Robert, I think the face adjusted bearings are the ones mentioned in the bearing catalogues that are supplied as matched pairs so the inner and outer races butt against each other and when done up the pre-load is appropriate. In the spridget case the races are separated by the hub and spacer but the distance of separation of the inner and outer races is the same so the pre-load is maintained. Interestingly having spent the last hour looking for the crucial details of the critical radius of the mentioned NSK 7205BEAT85SUN bearing or the SKF 7205BECBP I haven't found it. The SKF site lists the bearing but seems to indicate the radius is about 1mm minimum. The NSK site lets you download a replacement bearing catalogue that lists the 7205BEAT85SUN but doesn't give the dimensions. Do you have the details of the 7205BEAT85SUN that indicate it has the larger than normal radius to suit the spridget application. |
David Billington |
No I don't David but it is a crucial point and one that I will ask my NSK engineer about ASAP. Is it fair to say then Dave that the radius on the stub axle is 2mm? and that basic 20 degree bearings are appropriate to that and indeed have a radius above 2mm? |
Robert (Bob) Midget Turbo |
Robert, Good thing you have a close contact with NSK. I would say regarding the radius issue that it only indicates that the largest radius on the stub axle at that point should be 2mm to match the minimum of the bearing inner race. The design should have the inner race abut the face on the stub axle rather than the radius touching. What the minimum radius will be would be down to the designer and production engineer. The larger radius the better within reason as it reduces the stress concentration at that point and so helps reduce the likelihood of a fatigue related failure. That said people are hanging much wider and far grippier tyres on the corners than the designers probably ever dreamed of so the likelihood of long term failure is increased. What were the original cars designed for, 100k miles maybe. I suspect many components have passed that point several times without problems, but it is worth while having these components crack checked occasionally when off for a recon or service. |
David Billington |
Robert, Peter C and I measured several stub axle radii and only one of those we had was 1mm. The rest were more and most were 2mm. David, I had the same problem trying to find the geometry for those bearings. Hope you have better luck. In the absence of a declared "face adjusted" bearing I have used an indicator with a flat plate to measure the inner and outer races. Not perfect but I am sure I was in the .0005" range which should be close enough. |
tomshobby |
Based on Peter and Tom's measurements of that large number of stub axkes, then, it sounds like we need bearings with a minimum inner radius of >/= 2 mm. |
Norm Kerr |
Hi to those interested. I can not comment on the actual stub axle dimensions that have been produced by BMC for all those years. However I note from the drawings supplied by BMC to RHP that they have asked that the bearing radii discussed by ourselves should be "no less" than 1mm. Which is fortunate because having discussed this issue with my friend at NSK he states that there is a standard between the inner diameter of bearings and the radius in question. Off the top of his head he thought that the bearings of the size for our stub axles would be no less than 1mm. He then checked the catalogue for our bearings and he gave me a figure from the catalogue of 1.1mm minimum Now that should mean that the engineering is correct however if Tom has measured the radii at 2mm then that is a concern. Tom do you trust your accuracy? Because if your figures are right then the original factory fitted bearings would have had a problem I think? My mate said we should be able to find a catalogue on the web so I think I will try. |
Robert (Bob) Midget Turbo |
Robert I read the drawing for the 34/LJT25 as requiring a 2mm radius for the one in question although the manner in which it is specified seems unusual. |
David Billington |
David Can you interpret this for us, well me anyway. :-) http://www.tec.nsk.com/Handbook.asp?menu=6,0,0,0&PageID=/TolerancesAndAccuracy/ChamferDimensionLimits1.html |
Robert (Bob) Midget Turbo |
Norm, although we measured several axles we only actually measured a very small percentage of all axles made. Still it is worthy to note the findings simply because they were random from different years. Peter and I agree that the best would be replacement ball bearings that would fit correctly. Lacking that the next best would be taper bearings that fit correctly. We also agreed that the first test of fit is at the stub axle radius. The only bearing, either ball or taper, we could confirm would fit is the KOYO taper bearing. I contacted both NSK and SKF and could get no information either in the form of geometry or in writing to confirm what the radius of their product might be. That is why I am anxious to learn what Robert can find. Here it is in a nut shell. If the inner bearing cannot fit against the flat of the spindle shaft it will not have the full support to help prevent the axle flexing at that point. Also adding the inner spacer under those conditions would stiffen the shaft and seem to focus any flexing at that point. So, no matter what bearings are used they have to fit the inner most radius. |
tomshobby |
David, you read it correctly. Though in practice it is usually not an true radius but a relief made with a ground cutting tool. I had the idea of actually "hard machining" the required radius on a CNC machine. However after contacting Timken I was advised that machining or grinding might form stress cracks so I gave up that idea. I apologize for going back to this link again but there are a few ways to work around the problems and reasons they were discarded. http://smithtr6.com/bearings.htm |
tomshobby |
Robert, That page could do with some further explanation I think. As I see it it just gives information relating to the r and r1 specifications, those being the nominal radii of the race. The ra being the maximum allowable radius of the bearing housing or shaft abutment to clear. I think that this provides some information but you still have to look up the r and r1 values for the bearing in question in the specific details for the bearing in question for that table to be useful. Haven't made any sense of the "r or r1" column in the middle of the table yet. |
David Billington |
This is an example of an original bearing from my '74 Midget.
|
tomshobby |
Tom, I've seen many bearings where the radii were as machined and the OD, ID and face details ground so know what you're saying. Did Timken give you any detailed reasons why something like grinding was an issue likely to cause stress cracks. That had occurred to me as a method to form a larger radius and can't initially see a problem if the race is kept cool. |
David Billington |
David, this is an excerpt from the email I received from a Timken engineer. "In response to your question regarding tapered roller bearing hardness, the hardness spec is 58 to 63 Rc. Now I must advise you do not alter the tapered roller bearings in anyway. If you do you run the risk of creating stress risers in the product which may lead to cone cracking or rib separation." I no longer considered it an option from that point because I was looking for a wide application rather than just for myself. |
tomshobby |
Sorry gents I now see from the RHP drawing that the radius in question is as you rightly state no less than 2mm I will confront my friend with this situation and see what he makes of it. In the past he did state that manufacturers like to keep this radius small so that the thrust face is kept large. Then on the other hand you need a fillet in the corner of the axle. Bad news this engineering.. :( |
Robert (Bob) Midget Turbo |
David - r/r1 (LH column)define the standard minimum clearance limits, as described in "Remarks", for different bearing d. d columns are bearing diameter "r/r1 radial/axial" columns define the maximum material that can be missing along two different axes, due to the actual machined "chamfer", aka "r/r1". This is equivalent to the Cartesian dimensioning both in this drawing and the BMC drawing. There is no reference on the drawing for the RH column ra. I think this is an error, and that "ra" is the nominal standard radius, and that "r/r1" is the actual radius as machined. The designation "ra" should be at the top of the LH column, which is similarly boldfaced. The notation "max" in the RH column makes no sense to me here; I suspect it should be "min", since satisfying all the other criteria allows of larger radii. I really hate errors in catalogs, especially catalogs with critical info. I am trying to follow this in a BCA book that has dozens or hundreds of such errors. BTW, there is no mention whatsoever of "face adjusted bearings", which is why I requested that somebody supply me with the true text of any book that does mention it. I trust nobody's reading, including my own, any more than I trust the catalogs. The response has been overwhelming indeed. FRM |
FR Millmore |
Hi FRM, While this is important to final outcome of this discussion it is somewhat a side issue unless we can begin to find bearings that fit the inside radius. A face adjusted bearing is one that has been surface ground so the surfaces that contact the hub and spacer are flat and parallel to each other when both races are in contact with the balls as they would be when installed. The question I have however is what is the allowable tolerance while these bearings are actually under load and is it even tighter tolerance for 40 degree bearings over the original 20 degree bearings. The reason for this question is that a steeper angle would allow more vertical movement with the same lateral tolerance. But this seems a digression from the more immediate thrust of this conversation. |
tomshobby |
One more thing about the face adjusted bearings. They are more expensive. While working on this subject a few months ago I realized a way non-face adjusted bearings could be used and marketed by a supplier. But that is a subject that can wait until needed. |
tomshobby |
Tom I am confident that if the angular contact face adjusted bearing will not fit the radius correctly then no off the shelf bearing will either. To increase the radius on that face of the bearing I believe would require a "special" to be made but let's see what develops. Regarding the cost of the Face adjusted bearing I am pretty certain that if a major supplier such as Moss were to carry this instead of a cheaper 20 degree fit and hope example (which by the way has the same radius)that the price would be reduced significantly. I will pass on FRMs comments to the directors of NSK (formally RHP) that their charts are rubbish, I am sure they will check with you FRM and put the charts right. LOL. |
Robert (Bob) Midget Turbo |
The funny thing is that I was going to machine my hubs last weekend to accept the KOYO bearings and I was not able to get into the shop. Now I am glad I did not and am hoping we find a good answer. |
tomshobby |
Great thread this is! I had no idea of the scope of this problem...installed new front hub bearings last week & had more play than with the original worn out bearings...I thought I'd done something wrong. I even disassembled both sides & double checked my work...all was correct & properly installed. You fellows keep this up, hopefully there'll be a solution to this problem. |
Dave Rhine |
FRM, Thanks for that information, the centre column makes sense now. I forgot to mention before that the text indicating the dimensions are in um (microns) must be in error and should really be mm from what I can see. I think the LH column may be the r/r1 nominal and the centre the r/r1 maximum, the RH column being the ra maximum (maximum housing or shaft radius). If this doesn't get resolved with an off the shelf bearing with suitable radius I can see myself machining the spacer down and using shims, the trick will be determining what they should be before assembly. Tom, Regarding modification, maybe a specific question regarding whether the radius in question being increased would cause an issue. Robert, Can you ask your mate about whether larger radii on certain features is a standard option and whether increasing them by say grinding would be an issue. |
David Billington |
Dave Rhine, I have been thinking the same thing. |
tomshobby |
Dave B, Not sure I understand, which radius are you referring to? The one on the bearing? |
tomshobby |
Tom, Yes I was referring to the particular radius on the bearing being increased to that required. As far as I can see there is only one radius which is critical to the application, the others are more or less standard and not critical as they don't butt against anything like a larger radius. I was thinking of the angular contact bearing rather than the taper roller. I'm not up on how the bearing races are manufactured so not sure why Timken might suggest that increasing that radius might cause a cracking issue, enquiring minds would like to know. All this discussion does help to highlight just how big a subject engineering can be, we're just touching on a small part and it has gotchas. |
David Billington |
Dave, I was not able to get any response from NSK or SKF and have no idea what the effect may be on a ball bearing race. Maybe Robert could get that information from his friend. |
tomshobby |
The guys at BMC must have had faced the same issues as we are, because, did you notice in John Twist's video (attached to 1/16/2010 post above) his sample axle was from an MGB, and IT has a radiused spacer inboard of the bearing. So, the MGB guys have no issues with their bearing's inner radius interfering with anything because that additional spacer provides them a nice, sharp corner for it to fit in to. I guess that our cheaper, smaller suspension didn't warrant that additional piece, in the BMC engineer's minds. It is great that you guys are working on this, hopefully an appropriate bearing will be found, or a specially made one commissioned, then folks like BMH and MOSS can purchase them in enough quantity to (hopefully) keep the price somewhat reasonable for all of us. In the meantime, everyone take good care of your bearings, because if they fit, they are pretty valuable for now! Norm |
Norm Kerr |
Norm, I also noticed that it was of an MGB. The Timken rep I was in contact with recommended a shim or spacer behind their bearing to take the space. I was not very excited about that method for a number of reasons. One was that if it slipped into the radius much it could position to cut the seal on the outer edge. I also thought it might slip in and effect the bearing alignment slightly. Both being possible because it would be so thin compared to the MGB spacer which was obviously much thicker and looked to actually provide the surface for the seal. I had other reasons also so discarded the concept in favor of other possibilities. BTW, I do very much agree with your last sentence. |
tomshobby |
Can't add alot at the moment but have sent the info to NSK. But can I make the following comments. David not sure machining the spacer is an answer anymore. The basic 20 degree bearing will suffer the same problem at the axle thrust face as the 40 degree bearings. All! the spacer at the thrust face would cure the problem if it indeed exists. It would be interesting to push a bearing onto a stub axle and see if a feeler guage would go down the side? :) What bothers me is that I installed 2 new sets of the basic 20 degree basic bearings 15 years ago and whilst they had a bit of play in them they did not fail due to radius interference neither has the 2 new face adjusted bearings which naturally fit the hub correctly and so do not have play. Still not convinced the problem exists but I am waiting for answers from my friend. |
Robert (Bob) Midget Turbo |
Robert, I agree we are at place where we need more information. I have some coming also. I have two questions at this time. One is if anyone knows why the radius in question ranges from 1mm to 2mm on different cars and if there is any documentation as to which cars might have which radius. The other question I have is if there are bearings with the 2mm radius. I appreciate the views and participation in this discussion. |
tomshobby |
Mother Mary french kissing Judas... Are you guys still bent on this topic....Come on, yeah there are differances from one hub to the next...just buy some bearings, stick them in if they dont fit then thake the hubs and bearings to a machine shop and have them massaged till they do fit...end of story Accept the fact your NOt going to find a universial bearing to fit perfectly every hub regardless how far out it is, After all some midget brakes works and others not so good, some peoples lever shocks work perfectluy others dont, some peoples rear oil seals work and others not so good. But this brow beating is getting old Prop |
Prop |
Sorry Prop, didn't know your approval of the topic or discussion was required. But your assessment of the situation will probably end up being spot on. |
tomshobby |
AAAHhhhhGGGggg....Now I feel like the heel of the BBS, Oh man! Guys the comment I made about an hour and 1/2 ago was out of line ...I want to take it back, I certianly dont feel that way, as the primary purpose of this BBS is collabeeration on cool midget projects, Im completely in the wrong for stiffleing creativity and conversation. Im off to bed to put this complete waste of a day to an end and all the mindlss total BS that came with it This thread may have gotten a bit numbing on my part, But in no way should I have stomped my foot down on the back of its neck.... Guys please continue the conversation, It really does benifit us all as a whole. Prop...I'm n a debt of gratitude to all that will give me a slide on this one. |
Prop |
Prop, likewise, I should have held my fingers off the keys. I did not really believe you meant what you said. This just has been a long project and a very frustrating one. I just want to be as certain of the facts as possible that everyone can understand what to look for when replacing their bearings. Like I have said, I have a solution for myself. The problem is that there are a lot of folks with less experience, and I don't mean that in a bad way, working on their cars and I am thinking of them. I actually think the information Robert is after will pretty much wind things down. |
tomshobby |
It was asked earlier in the thread if anyone experienced an axle failure. I have never seen one and wonder if this was a problem before BMC uprated the front axles. Former part number was 601/602 and the later ones were 746/747. I might be off on the numbers, my books aren't available, just my memory. Is this only a concern for race cars or have street cars broken axles? |
J Bubela |
This started because Bruce Miles announced on the Spridgets-Aus site that he had a cracked stub axle and was looking at getting new uprated ones made: http://tinyurl.com/ydczjvx So there is at least one in this world that has failed (in racing). It seems what Bruce is chasing is already produced in the US by Winners Circle. If they have a market for their product there must be other failed stub axles out there. Alternatively it would be funny if that market is built entirely on a rumour that the originals are not up to the job. Perhaps Bruce's example is a world first. Now to return to bearing availability, why can't we solve the problem from the other end and produce stub axles to match available bearings? |
Mike Allen |
I have only seen one stub axle actually fail: that was on a street driven Bugeye that was also used for autocrossing. The stub broke in an autocross, and the wheel parted company with the car. This was a drum brake car at the time, later switched to disk brakes. This back about 1970 or so, by then most racers knew enough to check for cracks, so severe problems were not common on track. The cars were much newer then, so fatigue problems would occur only on cars tha had been used pretty hard. I have, however, seen many stubs crack where the axle portion meets the upright portion. Rough or damaged filets seem to crack more often, but even the best will sometimes fail. Back in those days, we would replace a cracked stub on a street car since it would go many miles without another check, but for a race car we would just note the problem and be sure to crack check the stubs on a regular basis. The crack would first appear as a short fine line in the filet radius, usually first on the underside of the axle, then another would appear on the top side. The crack would lengthen and widen as time went by: we would replace the stub when one of the cracks reached 1/4-1/3 of the circumference. We found that, like halfshaft, stub axles failed more often on cars that were prone to hitting curbs and taking off-course excursions. While I was racing, I had a large box of stubs magnafluxed, and more than half had some cracking. I sent some of the cracked ones to Winner's Circle to be converted to their racing spec. One of my cars has the Winner's Circle stubs, the other has a similar set done by a friend with a somewhat diferent approach. No further problems with either. I crack check the stubs on all my cars when I acquire them, and any time the hubs are off. I use a dye penetrant kit that is available at most welding shops. I still have a few sets of 'good' stubs on the shelf, so I tend to replace cracked units when cracks first appear. - Bill |
Bill Gavin |
Hi Bill, When I first noticed that a bearing I was installing did not fit against the upright portion because of the radius I did not realize there was a stub failure problem. But I did immediately see that the space between the bearing and the upright portion of the stub axle caused a lack of support and would allow the stub to flex and possibly cause such a failure. A discussion with a Timken engineer verified my concern. That is the exact reason I have been so doggedly pursuing this subject. I again refer to the drawings in the article on the link I posted earlier in this thread. The drawings are verifiable and measurable fact. For that reason I feel they are a good reference point in this discussion. I also wonder how many are unknowingly driving with ill-fitting bearings in their cars. |
tomshobby |
How many Tom? Well everyone who has ever changed their front wheel bearings with a replacement bearing not in a BMC box. Unless their stub axle does not have a 2mm radius. If it is 1 mm then they would have no such problem as I understand it at the moment. :-) |
Robert (Bob) Midget Turbo |
That is how I see things Bob. At this time the only bearing I have found with a 2mm radius is the KOYO but in order to have the KOYO fit properly the hub needs some work. An opinion that I share with at least a few others including Bill Young is why don't at least one or two bearing manufacturers just admit the error and make proper bearings for our cars. When I informed one supplier the response was, "I have been selling these same bearings for over 30 years. There never was a problem." The way I see it is if someone had a problem would they even put things together and even think the bearing might be part of the problem. |
tomshobby |
I think that there is not a problem that is noticable. As I said I have fitted to my midget 4 new sets of bearings without a decernable problem so unless something broke why would anyone believe something was not right? At the moment if things are as we imagine (and I am not convinced of that yet, but do see the arguement) then the bearing face will be butted up to the edge of the radius and everything else will be fine. We have to assume hat this must be the case with thousands of these suspension types yet no problems are being reported, Why not? Can you suggest a reason as to why no problems? |
Robert (Bob) Midget Turbo |
Been away for a couple of days so just catching up. In my opinion the radius in question should not be in contact with the stubaxle fillet radius as that prevents the race seating against the shoulder properly leaving a gap and also possible causing some slight expansion of the inner race. Over the years I have fitted a number of QH bearing sets to my car, none recently, and not had any problems but then I wasn't expecting any as the bearings were supplied for the application and I assumed correct. I would be very interested to know the results of someone measuring the nominal radius of the radius in question, as supplied by the current usual suspects, to see of they are supplied with the correct >=2mm radius but with poor control of fit up. I can check one of mine tomorrow possibly, as I have radius gauges, and see what they are. They would have been changed around 1996 before the car came off the road. Don't know if QH still do them but I have contacted them in the past and they provided some information about stuff. I wonder if they still supply the bearings if they could provide the specifications to which they supply them. Might be worth a catalogue perusal and if still supplied a few phone calls. |
David Billington |
That would be interesting Dave if you noted the details of the bearing and the radius plus the fillet radius of your stub axle. I would look forward to that |
Robert (Bob) Midget Turbo |
I've got one loose on the floor and they're easy to pull the hub on, the steel wheel hubs anyway, so I'll check that tomorrow. Dinner awaits. |
David Billington |
I can see a few reasons for not having problems. One is that the car happened to have a smaller radius on the stub axles in the first place. Another is that the driver's habits do not make much demand on the stub axle. A third is that enough time has not passed with the axle under load. A fourth is what size tires are on the car and what is the wheel offset. Then comes the question of spacer or no spacer and what are the stress effects of each with bearings that fit and those that don't. Again not as straightforward as it might seem. Especially with the information that Bill Gaven supplied as to the position of the cracks. Because of the widespread exposure of this topic I have decided that if someone does not step forward with a 2mm radius bearing by this weekend I am just going to machine my hubs for the KOYO bearings. I am still hoping that a correct bearing appears but after searching for months I really want my car back on the floor. |
tomshobby |
And so it goes, just as I start to give up hope, someone opens a window. Looking forward to what you find David. |
tomshobby |
Dave B, If its any help I should have a couple of pairs of stub axles here also if measuring them would help. What gauge is necessary - can something be improvised ? R. |
richard boobier |
Just seen this thread and I think some people are overlooking a point about the radius. Everyone seems to think it is a full radius and this is usually not the case i.e. a 2mm radius usually due to tolerances does not always extend the full 2mm up the faces and could cause a problem on some bearings depending on the tolerances being worked to. In other words a radius machined to the minimum and then the face ground to the maximum will reduce the size of the radius. Having worked in the bearing industry some years ago I would agree with the comment about trying to machine hardened bearings and stress fractures. We did do it with certain materials for special uses but the standard way is to turn the radii before hardening and then grind the faces, o/d and bore afterwards leaving the radius a turned finish. Trev |
T Mason |
Richard, What did we do before the internet, wiki to the rescue again. I have a set of radius gauges as here http://en.wikipedia.org/wiki/Radius_gauge which allow a comparison of internal and external radii. If you don't have that sort of thing than the internal radius on the stub axle could be compared using a 4mm drill shank which will be very close, the shank usually being very slightly smaller than the drilling end to be pedantic. Maybe use it to drill a hole also and file back to form a gauge for the external radius on the bearing. The difference between a common garden bearing with about 1mm radius and the 2mm required in the RHP drawing should, I think, be easy to spot. Obviously any other drills can be used to do a quick check to get an indication of the closest size. My gauges are in 0.5mm steps so will see what is closest tomorrow. |
David Billington |
Dave, Will try to find the stubs in the garage ! Have sets of fractional ? metric drills in the workshop. Assume best fit to radius with back lighting etc. Thanks, seems simple when you mentioned it ! R. |
richard boobier |
Reporting back people. I've pulled my steel wheel disc brake hub and I found that the radii being discussed on my stubaxle doesn't fit 2mm well at all, it's larger. The 2.5mm radius gauge seemed the best fit, 2mm not a good fit at all, 2.25mm better but too small, 2.50mm seemed the closest, 2.75mm similar to 2.5mm, 3mm showed daylight in the radius area. I don't think the radius on the stubaxle is true but maybe like the bearings where it approximates a radius. Picture of current bearing fitted, I believe an old QH bearing as I always got them from my local motor factor and they supplied QH. |
David Billington |
Now for the bearing radius. Camera can be a pain as it doesn't like to focus on what I want to photograph, if it's metal, but here I got focus and added my fingers and gauge into the shot. That's a 3mm radius gauge which seemed to be the closest fit to the radius in the bearing I have fitted. Can of worms springs to mind, did the bearings vary between the drum and disc brake variants. These bearing have a much larger than standard radius at this point but I think this is common for car hub bearing applications where the radius in question is a stress raiser due to the cantilevered nature of the design. |
David Billington |
Thanks much Dave, It may be a can of worms but at least we are shedding light on the subject. |
tomshobby |
Dave, I have two sets of stubs and hubs - both disc brake models - do not know years, but one has early steering arms so pre about 1971 and one post. Have not made gauges for bearings as yet but they have a very generous radius similar to yours. The stubs do not seem to have true radius - looks more like a hand ground form tool ? Therefore difficult to get accurate results, the best fit seemed to be 3.5mm-4mm drill shanks i.e 1.75-2mm radius. Need to acquire some gauges ! R. |
richard boobier |
That is staggering Dave and worth knowing. In this day and age it is fast becoming apparent that perhaps a spacer is on the inner axle to clear the fillet is the only answer? |
Bob Turbo Midget England |
Does anyone have a new set kicking about so the radius detail of the bearing supplied these days can be checked. |
David Billington |
Bob, I had considered a spacer. Some things to consider are that when I installed new, rotor, caliper, and pads a 1/16" offset from the bearing caused the rotor to rub the caliper. I realize not all cars are the same but there is a limitation. Another thing is to consider any effect on the seal and its mating surface. That would mean a spacer Would have to be thinner than 1/16" and have a small enough outer diameter to not interfere with the seal even if it did not center correctly behind the bearing. I also only compared a 2mm stub radius with a 1mm and 1.5mm bearing radius. That already required a theoretical 1mm and .5mm spacer and with real world wear and whatever else applied on my car an actual .023" and .020". Now Dave found a 2.5mm radius which would require a thicker spacer. It still seems to me that at least one of the bearing companies could just step up and provide a bearing that actually fits. |
tomshobby |
Dave the only ones I have are taper roller bearings. One KOYO at 3mm and two Timkens, a 1mm and a 1.5mm. |
tomshobby |
I have boxed standard bearings (20 degree) at work that I can measure on Sunday morning if my mate does not come through before with an answer? Tom as things stand if we are using bearings with 1mm radius then they are already pushed out a little so I dont see that as a major issue but I take your point. If the bearing already has a 1mm radius then a further 1mm of spacer could surfice? |
Robert (Bob) Midget Turbo |
Peter Caldwell just contacted me. He has several different bearings and asked if I would come down to his shop, 5 miles from me, next Tuesday so we can measure them. There is more that we will also be discussing that may have some import to this subject, promises to be an interesting morning. Bob, hope your mate comes through. It would seem his company could be very well regarded by a lot of people if they would work with us, even if only behind the scenes, to find a bearing that fits correctly. Bob, I listed a few workarounds in the article I posted and mentioned some of the possibly less desirable effects. Doesn't mean they are not usable, just not high on my list of choices. Then again we may have to settle for a less-than-perfect solution. |
tomshobby |
OK here is a couple more bits of info. I asked my friend the following question regarding the original inner bearing number 34/LJT25 ""Steve You will see that the original drawing that you gave me some time ago does specify this radius as 2mm do you think that the bearing was made special and had this larger than normal radius? His answer was ""Bob, Yes this is a "Special" bearing. The standard is LJT25. The 34/ before it means it is special.(The34th modded version of LJT25) The special features being the face adjustment and this radius which clearly I failed to spot when we talked about it before! Steve. Then I asked the following ""Thanks Steve Someone has suggested that the KOYO taper roller has a 3mm radius at that point! do tapers have a bigger radius? Bob The answer ""No they do not, they are the same. do not forget the 2mm is max on the stubshaft and the 1mm is min on the bearing so there will be a lot of instances where there is no conflict perhaps? If there is a problem I do not see why a 1mm spacer behind the bearing will not solve the problem. Steve. So there I have it!! The only answer I can see at the moment is to use a spacer as I am totally commited to the 40 degree face adjusted bearing as it does the job exactly except for the radius. The only problem then that I have is how to keep the spacer central? |
Robert (Bob) Midget Turbo |
Having searched earlier for the 34/LJT25 some sites do list it as an automotive special, possibly not surprising. If some company is still making them to suit the BMC vehicles with the larger than normal radius then possibly they can make it to the required tolerances, the question would be who is making them and what cost the correct tolerance. I'm not up on the production details of bearings but I presume that as RHP made these as a special with a larger than normal radius then they could do it consistantly as reject bearings wouldn't comform to the normal bearing spec. BTW searching for LJT25 does get a lot of options here http://www.ahrinternational.com/oemnames/OLD_RHP.shtml so they may have made many specials by the look of it. The outer bearing is as far as I can see fairly standard without any odd large radii so can be sourced from the standard bearing range with the appropriate designations to match the original 3 MJT17 BTW the search turned up these guys that list as having 2 34/LJT25 in stock. http://www.crewe-eng.co.uk/bi/SAR%20Bearings%20Impex%20Stock%20List%20Sept%2007.txt |
David Billington |
Copy and paste that last link it works. Related to this I have wondered why BMC, or whatever they were called at the time, specified the bearings like this. Maybe as a cost saving as it may save time doing delicate adjustments of the hub on the assembly line. Possibly the added cost of the bearings to a tight requirement was offset by savings elsewhere. Many cars these days use a cartridge bearing which operates in a similar manner. Vaguely related, a mate that used to work for Scammell was doing research on one of their gearboxes designed about 1935 IIRC and he realised that the main gearbox bearings were metric in an otherwise inch product. He tracked down the designer, he was still living at the time, and asked what the reason was. Cost!, anti-friction bearings were a European invention and developed in metric sizes so always dominated the production, inch based series being more expensive due to lower volumes. That last bit being said I've never seen anyone query why the spridget rolling element bearings are all metric, cost most likely. |
David Billington |
I don't agree entirely with that Dave. The outer bearing would also be a special as it too would need to be face adjusted. I assume based on the inner bearing number that the outer bearing is MJT17 and the 3 being the 3rd modification, no doubt the face adjustment. |
Robert (Bob) Midget Turbo |
Robert, I've not checked that but thought that it likely that a standard bearing, with appropriate (standard) designations to control the required features (the tighter the control the greater the cost normmally), would match the 3 MJT17. The main thing was that the outer bearing didn't seem to have any unusual radii, you then need to know the designations to specify the relationship between the inner and outer mounting face relationship which seem the critical factor in this case. Any feedback from your mate in this case would be appreciated. |
David Billington |
I still see a couple slim possibilities to check but it does look like this is truly coming to an end. Bob, I once thought that a spacer, possibly better described as a shim considering how thin it will be, might be made with a slight rolled edge around the center to fit into the slight gap left by the mismatched radii of the shaft and bearing. |
tomshobby |
If it turns out that face adjusted bearings with 2mm radius are not available (hopefully they will be found, but if not), then, like stated above, we are approaching a consensus that the option that sounds most "doable" is a 1mm spacer, kind of like a small version of the MGB's one (the spacer is shaped to fit over the stub axle radius, which would hold it centered during assembly), like described by tomshobby above. If you adopt a spacer behind the bearing, would you also have to reduce the length of the spacer located between the two bearings by the same amount? Also, if you but a spacer behind that bearing, won't that shift the whole front wheel assembly outboard by that much? I am just thinking out loud here, so bear with me... Maybe if you add a 1mm spacer, and cut the original, middle spacer shorter 1mm and you counterbored the place where the inboard bearing goes 1mm deeper, then it ought to work while keeping everything else "as originally intended"? Ok so far, then here's a follow-up question for the panel: Moving that bearing 1mm outboard this way would then reduce, by that amount, the distance between the two bearings. I wonder how much margin there is in the design of the two bearings together to handle loading, can it allow a 1mm reduction (the farther apart two supports are, the greater their moment arm to handle loading, so generally, paired bearings are spaced as far apart as possible)? Maybe, though, there is more margin here, than there is in the other option, which would be to cut the stub axle fillet radius down by 1mm (and create a stress riser right at that most important point). My guess is that BMC did not adopt a spacer because they could have the bearings made special with that large radius, and so keep the two bearings that much further apart. When they designed the B axle, they had more room to work with and designed in the spacer from the beginning. One good thing, though, even if one did adopt a radiused spacer, and even if that did lead to increased bearing loading. The only down side would be increased bearing wear, which is pretty safe/progressive kind of wear/failure, compared to the other option of cutting down that fillet radius and running risk of a catastrophic failure under load (axle break off - only the caliper would keep the wheel from coming astray). this topic is endlessly fascinating! When this study is completed, someone ought to submit it to Practical Classics, Moss, BMH and Classic Cars magazines (oh, duh, and MG Enthusiast!) for publication and widely sharing in the midget community. Norm |
Norm Kerr |
Hi Norm, A spacer behind the bearing would in no way have any effect on the bearing fit inside the hub. I will try to explain. There are two conditions that determine the inner spacing. One is the spacer that provides the spacing of the inner bearing races and the other are the lands inside the hub that the outer races contact. The relationship of the axle stub in the spacing is only to provide compression to hold the bearings against the lands and the spacer. A piece of threaded rod could do the same. And like Bob indicated, a spacer/shim behind the inner bearing would have little if any effect on the bearing location on the stub because the stub and bearing radii would already cause the space to be there. Primarily all the spacer would do is to fill the already open space and stabilize the bearing to the vertical face of the stub. This in turn would provide a much greater support to the stub against flexing. Not having actually measured this is only an estimate. The difference could be compared to the force from a wrench with a 4" long handle and one with a 12" handle. By introducing a bearing that fit the radius or a spacer would be like reducing the force from the 12" wrench to the 4" wrench. Knowing this it can now be realized the importance of this radius issue. I could measure but it would only be an idle curiosity. |
tomshobby |
Correct Tom Norm if the spacer is cut down then the face adjusted bearings would not fit correctly and they would in fact become loaded up to the point of failure I suspect. The whole point of the inner spacer is as Tom described and it is essential that its length is exactly the same as he land machined into the hub that spaces the oute racs apart. Thus the land in the hub spaces the outer racers to the exact position relative to the inner races that are spaced exactly by the inner spacer. I think that 1mm extra track on a spridget is nothing and with production techniques at the time I would imagine the standard track varied by far more than 1mm LOL |
Robert (Bob) Midget Turbo |
What about the brakes if you space the hub out slightly. I wonder if the pistons have enough extra movement into the caliper to cope in the condition of new discs and pads. |
David Billington |
David, mine made contact with about 1/16" (.0625")offset. 1mm would only be about .025" |
tomshobby |
Dave got some standard bearings that would be sold by Moss MGOC etc but I understand how to measure the fillet on the axle but is there a way of measuring the bearing radius without gauges? I was going to get a set of tapers as well to measure. |
Robert (Bob) Midget Turbo |
Bob, this might save some time. The Timken 30205M is a worse fit. It only has a 1mm radius and is wider which is not good either.
|
tomshobby |
I just returned from a morning at Peter C's shop. It was an interesting time spent measuring and discussing different possibilities. We focused entirely on ball bearings today. It helped having so many different parts to compare while we were looking for answers. We compared a 34LJT25 and a 39LJT25 and found the 39LJT25 had a smaller radius, less than 2mm. We measured several different bearings and only found 1 that had a 2mm radius. The manufacturer of that bearing did not put any markings on it so we have no idea of who made it. I also wonder how good the quality would be if the maker did not have enough pride to identify themselves, from the crude machining marks I would not either. We did find a very good quality bearing that had a 1.5 radius and Peter is going to see if the manufacturer could make it available with a larger radius. |
tomshobby |
So in summary...... Are we saying that there is no commercially available bearing that fits the stub axle? Tim |
T Dafforn |
It would seem so. |
Dave Rhine |
Tim, at this point all I can say for certain as there are at least as many as there are unicorns. But to say there are none is not possible. I saw an original 34TJT25 in it's original package today. It has been saved for many years. The other with no markings was only a maybe-fit and the quality was not impressive, besides I have no idea when it was made or if it is even available. I am between two choices at this time. The KOYO I have been talking about and the high quality bearing I saw this morning. Peter is going to see if we can obtain that higher quality bearing with the needed radius. If not it would be a very good candidate to use with a shim. The third bearing, the one without identification, might be good enough for some but I would not use it. For me it is a trust issue when a manufacturer or a re-manufacturer hides their involvement in a product. |
tomshobby |
This thread was discussed between 14/01/2010 and 26/01/2010
MG Midget and Sprite Technical index
This thread is from the archive. The Live MG Midget and Sprite Technical BBS is active now.