Strange Frame

Status
Not open for further replies.
dennisgb said:
Al-otment,

I do have a number of people who have the right equipment, but set-up could be complicated. The question is how much to spend. I do plan to put a mill in my shop but not anything big enough to handle the Norton. Maybe when I tear into it I will figure out that a little bigger machine could be of value...but those are slippery slopes...things seem to get bigger and better and start to break the bank.

Have you ever wondered how they fabricated the frames at Norton in the first place? You have to believe they had fixtures during welding that placed the parts in the right place...maybe they took them out of the fixtures and they changed shape...but if that was the case they should have been checked and fixed...maybe the specs were just to wide and they didn't know it.
Click to expand...

As far as I'm aware, they used the common practice of bolting every component in a jig (all mounting holes pre-cut) and then welding the whole assembly. As olChris mentioned before, as the weld cools and shrinks every thing then distorts in relation to the amount of weld - and there's the problem. It makes sense that any welded frame should have critical mounting holes and axes cut after welding, in relation to the steering axis. I don't know all the reasons for the way the frames turned out, tolerances, quality control and cost are obvious factors, but judging by the amount of threads on forums regarding Commando handling, there's a lot of frames which require correcting.

When you start bending a frame to get one thing in line, you're going to set up stresses elsewhere in the frame which can put something else out of line. It's sometimes a necessary process where crash damage is evident, but wont be accurate enough. The assembly then needs checking on a surface table.
 
Just curious Aloment.. Surface table as you previously mentioned ......... Granite ??? stone??? I dont understand or have i misread/interpeted wrongly.. If so why stone ??? A few photo's would be nice..
 
olChris said:
Just curious Aloment.. Surface table as you previously mentioned ......... Granite ??? stone??? I dont understand or have i misread/interpeted wrongly.. If so why stone ??? A few photo's would be nice..
Click to expand...

The one I've got is granite, they seem to be more readily available than cast iron and obviously don't rust. After several billion years of pressure granite is apparently a very stable material. Only photo's I've got are on my website http://www.parallelengineering.co.uk/ (work in progress on site) I'm hoping to make a living at something I would enjoy more than the current job. I'll have to get a camera for next years holidays :D.
 
Al-otment said:
As far as I'm aware, they used the common practice of bolting every component in a jig (all mounting holes pre-cut) and then welding the whole assembly. As olChris mentioned before, as the weld cools and shrinks every thing then distorts in relation to the amount of weld - and there's the problem. It makes sense that any welded frame should have critical mounting holes and axes cut after welding, in relation to the steering axis.
Click to expand...

This is what I imagined was the process.

Being and enineeer I see some problems with this. The frame is tubing...so all the critical relationships are controlled by the jig and the welding. The front stem tube is oriented by the welds so the only machining that could be done is the vertical relationship not horizontal (faces of the tube). Same with the swing arm tube only opposite. What would make the most sense to me would be the frame would stay in the jig while cooling so the positions would hold...but not only is that not pratical, it could impart stresses in the weldment. The frame could be heated to reduce stress afterwards, but not sure that would work because it may cause the positions to change. I honestly think the fabrication process combined with "floating" on the iso's results in an inherent problem, because the tolerances we are asking for are just too tight for the manufacturing process. This could explain why measured frames are always different...too many variables. Heat, time, material, operator...
 
Hi Dennis,

I doubt if leaving the frames in the jigs to cool would solve the problem. As the molten weld pools shrink then they set up stresses within the parent metal. The spine tube is a good example of this. It is bowed downwards due to the several inches of weld connecting the bracing tube to the spine tube underside. I'm sure all welded frames distort, it's just that the Commando's the only one with the iso's which highlight misalignment.
 
Let me share a lifetime of metal manufacturing with you. The frames would not have been machined after welding, main reason being cost, but just as importantly, not rigid enough to mill without resonance. All the lugs would have been machined ahead of time, then bolted into a jig. The welding process creates shrinkage, SOME of which is defeated due to restraint of the fixture, and some remains when removed from the jig. That shrinkage is compensated for, very easily, when designing the jig. If, for example, 6" is required between iso mount tabs when complete, the jig would measure 6.060" (or whatever yielded the correct as-built dimension) and account for the shrinkage. Simple. Fast. Easy. Repeatable. Cheap.
 
concours said:
Let me share a lifetime of metal manufacturing with you. The frames would not have been machined after welding, one reason being cost, but just as importantly, not rigid enough to mill without resonance. All the lugs would have been machined ahead of time, then bolted into a jig. The welding process creates shrinkage, SOME of which is defeated due to restraint of the fixture, and some remains when removed from the jig. That shrinkage is compensated for, very easily, when designing the jig. If, for example, 6" is required between iso mount tabs when complete, the jig would measure 6.060" (or whatever yielded the correct as built dimension) and account for the shrinkage. Simple. Fast. Easy. Repeatable. Cheap.
Click to expand...

Any accurate machining operation is dependent on the work piece being adequately supported/clamped to minimise vibration between the work and tool. I don't see why a production engineer could not have implemented a manufacturing process where welded frames were mounted on a jig fixed to a milling table, the frame being on it's side with the steering head axis as the datum, from which xy dimensions could be accurately (+/- 0.0005") established and machined. You sure it wasn't a lunchtime of metal manufacturing? :D.
 
Al-otment said:
concours said:
Let me share a lifetime of metal manufacturing with you. The frames would not have been machined after welding, one reason being cost, but just as importantly, not rigid enough to mill without resonance. All the lugs would have been machined ahead of time, then bolted into a jig. The welding process creates shrinkage, SOME of which is defeated due to restraint of the fixture, and some remains when removed from the jig. That shrinkage is compensated for, very easily, when designing the jig. If, for example, 6" is required between iso mount tabs when complete, the jig would measure 6.060" (or whatever yielded the correct as built dimension) and account for the shrinkage. Simple. Fast. Easy. Repeatable. Cheap.
Click to expand...

Any accurate machining operation is dependent on the work piece being adequately supported/clamped to minimise vibration between the work and tool. I don't see why a production engineer could not have implemented a manufacturing process where welded frames were mounted on a jig fixed to a milling table, the frame being on it's side with the steering head axis as the datum, from which xy dimensions could be accurately (+/- 0.0005") established and machined. You sure it wasn't a lunchtime of metal manufacturing? :D.
Click to expand...

Awe, shucks, you got me. I'm a poseur. :lol:

Perhaps you skimmed quickly and missed the "C" word.. COST. :oops: And frames machined to +/- five tenths? That is ROTFLMFA gold. Please, go on... :roll:
 
concours said:
Let me share a lifetime of metal manufacturing with you. The frames would not have been machined after welding, main reason being cost, but just as importantly, not rigid enough to mill without resonance. All the lugs would have been machined ahead of time, then bolted into a jig. The welding process creates shrinkage, SOME of which is defeated due to restraint of the fixture, and some remains when removed from the jig. That shrinkage is compensated for, very easily, when designing the jig. If, for example, 6" is required between iso mount tabs when complete, the jig would measure 6.060" (or whatever yielded the correct as-built dimension) and account for the shrinkage. Simple. Fast. Easy. Repeatable. Cheap.
Click to expand...

This assumes the same operator is welding at the same rate with the same materials (material composition varies), evironment is static and many other variables must be always the same...not sure that was possible in the late 60's and early 70's...today with robotic welding and temperature and humidity control along with incoming raw material inspection, then yes this would be possible. Fixturing should always allow for shrinkage or even growth in some cases we knew this 40 years ago when I started in manufacturing, but given the methods of the day, there still could be variation...then also consider that most materials once heated to the melting point can take a long time to stabilize...

The big difference here was pointed out by Al-otment in that the combination of a welded frame with iso engine and trans suspension exhasperated tolerances which would likely have been adequate without the iso's (rigid mounting). Given that the concept of iso mounting was relatively new, they probably didn't realise that the old method of fabricating the frames was too loose in tolerance.
 
dennisgb said:
concours said:
Let me share a lifetime of metal manufacturing with you. The frames would not have been machined after welding, main reason being cost, but just as importantly, not rigid enough to mill without resonance. All the lugs would have been machined ahead of time, then bolted into a jig. The welding process creates shrinkage, SOME of which is defeated due to restraint of the fixture, and some remains when removed from the jig. That shrinkage is compensated for, very easily, when designing the jig. If, for example, 6" is required between iso mount tabs when complete, the jig would measure 6.060" (or whatever yielded the correct as-built dimension) and account for the shrinkage. Simple. Fast. Easy. Repeatable. Cheap.
Click to expand...

This assumes the same operator is welding at the same rate with the same materials (material composition varies), evironment is static and many other variables must be always the same...not sure that was possible in the late 60's and early 70's...today with robotic welding and temperature and humidity control along with incoming raw material inspection, then yes this would be possible. Fixturing should always allow for shrinkage or even growth in some cases we knew this 40 years ago when I started in manufacturing, but given the methods of the day, there still could be variation...then also consider that most materials once heated to the melting point can take a long time to stabilize...

The big difference here was pointed out by Al-otment in that the combination of a welded frame with iso engine and trans suspension exhasperated tolerances which would likely have been adequate without the iso's (rigid mounting). Given that the concept of iso mounting was relatively new, they probably didn't realise that the old method of fabricating the frames was too loose in tolerance.
Click to expand...
So, have you ever welded a tubular frame? Of any type?

Yes, back before robots, us old guys would actually be able to weld repetitively quite well, if you can believe it. They were known as skills, as opposed to theory.
 
Al-otment said:
concours said:
Any accurate machining operation is dependent on the work piece being adequately supported/clamped to minimise vibration between the work and tool. I don't see why a production engineer could not have implemented a manufacturing process where welded frames were mounted on a jig fixed to a milling table, the frame being on it's side with the steering head axis as the datum, from which xy dimensions could be accurately (+/- 0.0005") established and machined.
Click to expand...
Click to expand...

I agree...this is basically what the guys did on the "WSC" frame. They made the measurements and fixtured the frame in a mill and machined the surfaces.
 
concours said:
Yes, back before robots, us old guys would actually be able to weld repetitively quite well, of you can believe it. They were known as skills, as opposed to theory.
Click to expand...

I have been in the machine design and manufacturing business for 40 years, so yes I have welded tubing both square and round. I do all my own welding on my projects.

I totally understand what you are saying, and I don't think we are disagreeing...I think the main point is the tolerance required to insure "perfection" with the iso's just wasn't possible.

Manual manufacturing of weldments is/was pretty much the only way to do things for years, and I have a number of friends who did this for their entire working life, and yes they all are very proficient welders. Artists in many ways in terms of quality and speed. No question that they could weld a motorcycle frame in a jig and get consistent results...but I still think they could make 100 frames that would be different enough in tolerance to create the problems we see on the Commando because of the floating iso suspension system.
 
dennisgb said:
concours said:
Yes, back before robots, us old guys would actually be able to weld repetitively quite well, of you can believe it. They were known as skills, as opposed to theory.
Click to expand...

I have been in the machine design and manufacturing business for 40 years, so yes I have welded tubing both square and round. I do all my own welding on my projects.

I totally understand what you are saying, and I don't think we are disagreeing...I think the main point is the tolerance required to insure "perfection" with the iso's just wasn't possible.

Manual manufacturing of weldments is/was pretty much the only way to do things for years, and I have a number of friends who did this for their entire working life, and yes they all are very proficient welders. Artists in many ways in terms of quality and speed. No question that they could weld a motorcycle frame in a jig and get consistent results...but I still think they could make 100 frames that would be different enough in tolerance to create the problems we see on the Commando because of the floating iso suspension system.
Click to expand...
What are those problems?
 
concours said:
What are those problems?
Click to expand...

Basically what we talked about before in this thread. Holding the parts in a fixture, even with tolerance built in for shrinkage, and a qualified welder (this is the one variable that is difficult to control because there must have been multiple welders and technique is not perfectly repeatable), you still have changes that occur due to heat, depth of penetration, rate, material variance and environment that will allow the finished product to vary.

I still think the thing we are struggling with is tolerance. A weldment for a support frame for a hopper has a wider tolerance than say a precision machine surface. In cases where precision is required there is almost always a secondary machining operation to get the precision. In the case of the Commando in order to insure consistent tracking and front to back wheel alignment it is approaching that level of precision in terms of the finished frame. What we all have figured out 40 years later is that the frames were not very precise...and that effects those variables that mess up the handling of the bike.

When they made the featherbed frames this wasn't an issue...but with the Commando it was.
 
dennisgb said:
concours said:
What are those problems?
Click to expand...

Basically what we talked about before in this thread. Holding the parts in a fixture, even with tolerance built in for shrinkage, and a qualified welder (this is the one variable that is difficult to control because there must have been multiple welders and technique is not perfectly repeatable), you still have changes that occur due to heat, depth of penetration, rate, material variance and environment that will allow the finished product to vary.

I still think the thing we are struggling with is tolerance. A weldment for a support frame for a hopper has a wider tolerance than say a precision machine surface. In cases where precision is required there is almost always a secondary machining operation to get the precision. In the case of the Commando in order to insure consistent tracking and front to back wheel alignment it is approaching that level of precision in terms of the finished frame. What we all have figured out 40 years later is that the frames were not very precise...and that effects those variables that mess up the handling of the bike.

When they made the featherbed frames this wasn't an issue...but with the Commando it was.
Click to expand...
I challenge the notion that standard process deviation in frame manufacturing at the time is causal to "messing up the handling".
 
I have an idea that those problems, in most cases fall within a range which is exceeded by rider capability and attitude. I've known one or two absolutely dreadful Commandos that nobody could ride but far more which owners have complained about and which other riders have said steered and handled fine.

In most cases, I suspect that the factory testers were such an experienced bunch, probably with a racetrack bias that they adapted to and allowed for most of the commonly found quirks and hardly noticed them. The factory probably didn't perceive a problem.

More owners seem to be worried now than they were back in the day despite the fact that there were probably more bikes being ridden harder then. I've seen dreadful pattern isolastic housings and lots of rebuilt disc wheels where the rim has not been pulled over anything like sufficiently....Slightly more than half a thou pissed as well.
 
Concours,

If you haven't read the article below, take the time to read it. I have read it multiple times trying to figure out a way to check and "fix" my bike. I still am not sure I can accomplish it...maybe we are slitting hairs...but the variances are real and they do effect the handling.

http://www.vintagenet.us/phantom/wsc.html

This might be interesting also...they talk about the reasons frames are not straight including methods of fixturing.

http://www.parallelengineering.co.uk/
 
dennisgb said:
Concours,

If you haven't read the article below, take the time to read it. I have read it multiple times trying to figure out a way to check and "fix" my bike. I still am not sure I can accomplish it...maybe we are slitting hairs...but the variances are real and they do effect the handling.

http://www.vintagenet.us/phantom/wsc.html

This might be interesting also...they talk about the reasons frames are not straight including methods of fixturing.

http://www.parallelengineering.co.uk/
Click to expand...


WSC may be riveting to an accountant, but as soon as I see statements like "turned out that the swing arm pivot was out of square with the steering head by a mere .029". " MEASURED HOW? WHERE? PROJECTED OUT HOW FAR? ANGLES ARE MEASURED IN DEGREES, MINUTES, SECONDS, NOT INCH DECIMALS. I discount the entire literary masterpiece as a flexure of many pieces of lovely technical verbiage arranged to "wow the jellybeans". Chock full of words and theory, but just like the 6:00 news, woefully devoid of substance.
I have access to granite surface plates big enough to host the entire bike, all manner of measuring tools, but, even if I had a crashed frame, would not put any effort into attempting to measure to that degree. The Acorn platen bench would be flat enough for the needs of measuring the frame.

How do you KNOW that "...but the variances are real and they do effect the handling." ? Just what ill handling attribute are you searching to cure?
 
Al-otment said:
olChris said:
Just curious Aloment.. Surface table as you previously mentioned ......... Granite ??? stone??? I dont understand or have i misread/interpeted wrongly.. If so why stone ??? A few photo's would be nice..
Click to expand...

The one I've got is granite, they seem to be more readily available than cast iron and obviously don't rust. After several billion years of pressure granite is apparently a very stable material. Only photo's I've got are on my website http://www.parallelengineering.co.uk/ (work in progress on site) I'm hoping to make a living at something I would enjoy more than the current job. I'll have to get a camera for next years holidays :D.
Click to expand...


That service that you intend to offer will be very handy to motorcycle rebuilders/restorers.... i assume that picture provided within your website is the "Surface Table".. I would have thought there would have been "fixing points" at various location around the table for push/pull equipment.... Obviously not, so when you find a "misalignment" in a frame how do to straighten it ????
 
concours said:
WSC may be riveting to an accountant, but as soon as I see statements like "turned out that the swing arm pivot was out of square with the steering head by a mere .029". " MEASURED HOW? WHERE? PROJECTED OUT HOW FAR? ANGLES ARE MEASURED IN DEGREES, MINUTES, SECONDS, NOT INCH DECIMALS. I discount the entire literary masterpiece as a flexure of many pieces of lovely technical verbiage arranged to "wow the jellybeans". Chock full of words and theory, but just like the 6:00 news, woefully devoid of substance.
I have access to granite surface plates big enough to host the entire bike, all manner of measuring tools, but, even if I had a crashed frame, would not put any effort into attempting to measure to that degree. The Acorn platen bench would be flat enough for the needs of measuring the frame.

How do you KNOW that "...but the variances are real and they do effect the handling." ? Just what ill handling attribute are you searching to cure?
Click to expand...

All I can say is that you don't need to worry about it then. If other's believe there is a problem and realize that there are methods to correct it, then they should be able to do that and to discuss it. I don't think you understand that degrees really have nothing to do with it, because the method is to square the surfaces not "twist" the frame back into shape.
 
Status
Not open for further replies.
Back
Top