@@ellisbriggsbikes I'm surprised the first step isn't making sure the bottom bracket shell is perfectly centered. That seems like the most likely source of error when flipping the frame on a surface plate.
@@tedwingate the bottom bracket has no bearing frame tracking, if it was pronounced it would lead to some biomechanical problems but it would have to be more than a mm or so. When aligning the main tubes to the surface plate the aim is to get them all parallel to the table. We're not looking for the centreline at that point. As you can see at the end of the video the wheels are in track at ground level, which is the end game if you think about it? It's easy to get confused with other measurements along the way though.
@@ellisbriggsbikes I'm not confused at all, I'm saying it's the main source of error when Paul flips the frame on his surface plate to double check and it's no longer centered on his measuring tools.
Nice video. I took a framebuilding class from Doug Fattic in the past year or two. He apprenticed at Ellis Briggs in the 1970s. His method of aligning a frame looks almost exactly the same as yours. I wonder why... 🤔 It's not just Ellis Briggs that's been aligning frames like this since 1936. Don't forget all the framebuilders who apprenticed with you and went on to their own careers as framebuilders.
Yes Doug is a good friend and mentor of mine. Its great to know there are framebuilders that have followed our lineage thanks to Doug Fattic, who has made that possible.
This seems like a topic that could be explored further. Perhaps a discussion of which characteristics are the ones of primary importance, which points on the frame define the reference plane or surfaces that other points will be measured relative to, and how each measurement is referenced to the pertinent plane or point. It would be interesting to determine just how much error is really tolerable by the user too. i.e. what is "good enough"? Goodness knows that people manage to ride bikes with some real alignment issues, but I would guess that a few mm of error in most dimensions is precise enough.
Its dificult to say how much is an issue. Like you say people can ride bikes which are far out of alignment. But I like to think that when you are on the limit going road a corner at speed that it does make a difference to how the bike handles. I will certainly look at doing another video exploring this topic. Probably one of the alignment process on a new frame. Thanks for watching!
@@ellisbriggsbikes "Its dificult to say how much is an issue" is a very true statement, and applies to much of the cycling industry, as well as consumer goods in general. I'm a retired engineer, and have had to design items to meet very precise specifications. The company has had to spend quite a bit of money for sophisticated test equipment in order to be sure that the gear was good enough. In most consumer goods, there's no specification and the customer is usually not willing to spend enough to justify a lot of research or test gear. Every so often, someone runs a few tests and actually shows that our old ideas were without basis, such as the move away from 23mm tires to somewhat wider ones. In principle, someone could run tests and demonstrate that alignment beyond a certain level is wasted effort and expense, but who would spend that money... especially when advertising is cheaper and probably results in more sales? Having said that... I'd be curious about the alignment of some of the better frames of the past. I have modest expectations for my lovely 1974 Raleigh/Carlton International, but I suspect/hope that my '82 Raleigh 753 Team frame (built in Ilkeston by the SBDU group) would do well.
@@ellisbriggsbikes Just watched Paul's video... very practical and it looks fast, which suits a production environment. I was hoping to see some definition of key points and then ensuring that these were co-planar. The use of various points on the OD of the tubes at various points along their lengths fell short of this. This presumes that the tubes are round and straight and aligned on the centerline of the bike... and I don't know that this is the case. Still... probably good enough?? I've spent many years learning how my assumptions weren't true when designing stuff, as well as learning about the various errors in my measurement methods. It's fine to fully understand the errors and then decide what shortcuts are justifiable, but I'm not convinced that Paul has done this. OTOH, I don't suppose his customers have any complaints.
Love your videos. Would like to see a series of videos of you building a frame from scratch please. I didn’t mind the music but it was just a bit loud so was harder to hear you as someone else has said.
Yes, I think you are right, it was too loud. I am working on a new gravel frame at the moment but i'm going to complete the footage of the whole process, before I make some videos about it. So keep you eye out!
What is that length of aluminium please, looks remarkably like a length of framework for installing solar panels , if it is I don't blame you, I've used a 4 meter length to help build a garage for a friend and it is dead level puckery poo..
is checking the bottom of the wheel the way to go , would it not be better to put a straight edge at the half way height of the wheel and check there ?
Well if you think about it at the bottom where the wheels touch the ground is the end game. In theory it's possible to be inline at the axles but the wheels could still be tilted. Having said that what you're describing does have to be measured as part of the whole alignment but it is only one of the variables.
@@ellisbriggsbikes yep , but take it to the extreme and the two wheel bottoms become a point ,,, its hard to get two points that are not in a line , haf way up the wheel and that same line touches at four points , three will be in a straight line by default , making the deviation at point four much easier to measure ,,, basic engineering geometry , just my thoughts. Then again , I dont build frames , well not bike frames .
@@tomthompson7400 it wouldn't take into account if the wheels were tilted in opposite directions. Scenario 1 - Imagine looking from the back of the bike, the wheels could both be vertical lines and your measurement at the height of the axle would be fine. Scenario 2 - Imagine looking at the bike from the rear again. this time the the wheels are not perfectly vertical, one wheel is tilted +1 deg and the other -1 deg. The wheels still cross at the centre where you are proposing your more accurate measurement, but they are not actually in alignment.
I did a video on what alignment really is, this is the follow up I promised. Here is the first video th-cam.com/video/YTXGi94D5eM/w-d-xo.htmlsi=I-NgK3TAb7NgKC6M
Bosch 40 X 40 profile has a longitudinal straightness tolerance of 1.5 mm per 2000 mm, so your final alignment check with feeler gauges is pretty much useless.
Yes, I considered that. Literally the first thing I considered when I was handed the profile. I was actually surprised when I measured it on the surface plate and couldn't get a feeler gauge under it. Although I admit I didn't bother past 0.1mm. so I guess I must have got lucky. If it had 1.5mm over 2 metres that would have been significant! Did you really think I hadn't checked that? :-)
Given you do don't check for parallelism between the rear wheel (which isn't even true, as can be seen when you spin it in the video) and the frame your method for checking coplanarity between the wheels doesn't really work, since the front wheel is mounted in the fork and can rotate to parallel with profile. The frame could easily be angled at the rear axle toward or away from the profile without it being easily visible or measurable. Over an estimated wheelbase of 1500 mm, they only need to be out of alignment by .024 degrees to get an out of plane measurement of .63 mm. Also, is the front wheel in dish/true to better than 0.63 mm at the point where you measured the gap? And to what tolerance is your surface plate calibrated if we're using it calibrate the profile? ;) @@ellisbriggsbikes
The whole point of using the profile was to make it easier to explain. But I did set it up at least 4 times and had similar results each time. Wheels are dished and true to within to within 0.15mm. It's easy enough to check the wheels for parallel on the surface plate with some guage blocks but I don't think you'll find enough of a discrepancy. If you measure each variable of frame alignment using the surface plate as datum, you can see which frames are going to have an issue at the wheels. I already knew this frame was good and the results are repeatable. When you are riding the front wheel will also turn to be parallel, so I don't see your point there. Even still if the rear wheel is toe in or toe out it will give you an offset even if it is parallel. Also if the headtube tilted. The results were similar to the other method I've used in the past, which involves mounting frame fork and both wheels on the face plate. In this arrangement it is possible to isolate which variables cause the discrepancy. Surface plate is calibrated to grade 3, which is more than good enough for a bike frame. Perhaps I'll do a more technical video for those who are interested.
Repeatability is completely irrelevant if your measurement equipment isn't good. You're just repeating the incorrect measurement...@@ellisbriggsbikes I'm going to have to disagree on the trueness of the wheels, see 3:11 in the video. And even if they are within 0.15 mm, the best tolerance you can ever state is to within +-0.3 mm. That's already half of what you claim before one adds anything else to the tolerance stack. Was the measurement with the profile to measure parallelism or coplanarity? These are different things. For a grade 3 surface table of your size the tolerance is around .05-.06 mm, how thick were the feeler gauges you checked your profile with?
@@alexv7349 you haven't shown anywhere where the measuring equipment isn't good enough. You pointed to the profile, which has been measured on the surface plate, then you pointed out the surface plate which has been calibrated, then the wheels which are true and in dish. You've just made a lot of assumptions. I'm not sure what point you're trying to make really? Obviously any bike in the real world can only have wheels as good as you can make them..I accept there will be some error but not enough to make a significant difference.
I don't think Paul Brodie does 'Yorkshire' and I'm fairly sure EB frames are straight or you and your predecessors wouldn't have lasted ten minutes as builders. Having said that, I might track down a straight edge and have a check : -) PS. love the video.
Darn shame the background music was so annoying. It was in some areas louder than his narrative. I would love to have watched the rest of the video but couldn't get past the annoying music.
Yeah i think I over did the music a bit. Thanks visiting anyway. I've toned the music down now! th-cam.com/video/yUkgdC2MgL4/w-d-xo.htmlsi=tZt8_3emaL5HbcrO
Paul has put up an invaluable resource with his videos, he's a pleasure to watch and learn from.
Yes he certainly has, I think ive watched all his framebuilding videos. Great channel
@@ellisbriggsbikes I'm surprised the first step isn't making sure the bottom bracket shell is perfectly centered. That seems like the most likely source of error when flipping the frame on a surface plate.
@@tedwingate the bottom bracket has no bearing frame tracking, if it was pronounced it would lead to some biomechanical problems but it would have to be more than a mm or so.
When aligning the main tubes to the surface plate the aim is to get them all parallel to the table. We're not looking for the centreline at that point.
As you can see at the end of the video the wheels are in track at ground level, which is the end game if you think about it?
It's easy to get confused with other measurements along the way though.
@@ellisbriggsbikes I'm not confused at all, I'm saying it's the main source of error when Paul flips the frame on his surface plate to double check and it's no longer centered on his measuring tools.
My apologies! I got the wrong end of the stick there! Yes your absolutely spot on with that observation
Very cool! .. music is unnecessary and just makes it harder to watch/hear.
Ok, thanks for the feedback. I've done some videos with and some without.
Got to say I like the music. Adds depth and texture.
Is that music? I noticed some sort of notes playing but I didn't notice any music. Otherwise an interesting video.
So informative, I love these videos of Paul's!
Great stuff ………..absolutely love it. Thanks for both the pure enjoyment and education Paul 👍
Love the content.
Music is deafening.
Never in doubt!
Nice video.
I took a framebuilding class from Doug Fattic in the past year or two. He apprenticed at Ellis Briggs in the 1970s. His method of aligning a frame looks almost exactly the same as yours. I wonder why... 🤔 It's not just Ellis Briggs that's been aligning frames like this since 1936. Don't forget all the framebuilders who apprenticed with you and went on to their own careers as framebuilders.
Yes Doug is a good friend and mentor of mine. Its great to know there are framebuilders that have followed our lineage thanks to Doug Fattic, who has made that possible.
Yes, when I wrote my comment I didn't realize you'd taken Doug's class. He mentioned it in a note I saw later on Classic Rendezvous.
I would say that amount of track error was highly acceptable. I wonder what the limits are for an autoclaved carbon frame and forks?
Of the few carbon frames I've measured, the error was more like 5mm between the front and back wheels.
This seems like a topic that could be explored further. Perhaps a discussion of which characteristics are the ones of primary importance, which points on the frame define the reference plane or surfaces that other points will be measured relative to, and how each measurement is referenced to the pertinent plane or point. It would be interesting to determine just how much error is really tolerable by the user too. i.e. what is "good enough"? Goodness knows that people manage to ride bikes with some real alignment issues, but I would guess that a few mm of error in most dimensions is precise enough.
Its dificult to say how much is an issue. Like you say people can ride bikes which are far out of alignment. But I like to think that when you are on the limit going road a corner at speed that it does make a difference to how the bike handles. I will certainly look at doing another video exploring this topic. Probably one of the alignment process on a new frame. Thanks for watching!
@@ellisbriggsbikes "Its dificult to say how much is an issue" is a very true statement, and applies to much of the cycling industry, as well as consumer goods in general. I'm a retired engineer, and have had to design items to meet very precise specifications. The company has had to spend quite a bit of money for sophisticated test equipment in order to be sure that the gear was good enough. In most consumer goods, there's no specification and the customer is usually not willing to spend enough to justify a lot of research or test gear. Every so often, someone runs a few tests and actually shows that our old ideas were without basis, such as the move away from 23mm tires to somewhat wider ones. In principle, someone could run tests and demonstrate that alignment beyond a certain level is wasted effort and expense, but who would spend that money... especially when advertising is cheaper and probably results in more sales? Having said that... I'd be curious about the alignment of some of the better frames of the past. I have modest expectations for my lovely 1974 Raleigh/Carlton International, but I suspect/hope that my '82 Raleigh 753 Team frame (built in Ilkeston by the SBDU group) would do well.
@@ellisbriggsbikes Just watched Paul's video... very practical and it looks fast, which suits a production environment. I was hoping to see some definition of key points and then ensuring that these were co-planar. The use of various points on the OD of the tubes at various points along their lengths fell short of this. This presumes that the tubes are round and straight and aligned on the centerline of the bike... and I don't know that this is the case. Still... probably good enough?? I've spent many years learning how my assumptions weren't true when designing stuff, as well as learning about the various errors in my measurement methods. It's fine to fully understand the errors and then decide what shortcuts are justifiable, but I'm not convinced that Paul has done this. OTOH, I don't suppose his customers have any complaints.
Love your videos. Would like to see a series of videos of you building a frame from scratch please.
I didn’t mind the music but it was just a bit loud so was harder to hear you as someone else has said.
Yes, I think you are right, it was too loud. I am working on a new gravel frame at the moment but i'm going to complete the footage of the whole process, before I make some videos about it. So keep you eye out!
@@ellisbriggsbikes cheers I will look forward to that.
Watch out for next weeks video, I have most of my footage filmed for the gravel frame i just built
What is that length of aluminium please, looks remarkably like a length of framework for installing solar panels , if it is I don't blame you, I've used a 4 meter length to help build a garage for a friend and it is dead level puckery poo..
is checking the bottom of the wheel the way to go , would it not be better to put a straight edge at the half way height of the wheel and check there ?
Well if you think about it at the bottom where the wheels touch the ground is the end game. In theory it's possible to be inline at the axles but the wheels could still be tilted.
Having said that what you're describing does have to be measured as part of the whole alignment but it is only one of the variables.
@@ellisbriggsbikes yep , but take it to the extreme and the two wheel bottoms become a point ,,, its hard to get two points that are not in a line , haf way up the wheel and that same line touches at four points , three will be in a straight line by default , making the deviation at point four much easier to measure ,,, basic engineering geometry , just my thoughts. Then again , I dont build frames , well not bike frames .
well strictly speaking it is taken a couple of inches up the rim so it is 4 points.
@@ellisbriggsbikes totally agree ,,,but easier to measure and thus more accurate half way up the rim , hence my original post
@@tomthompson7400 it wouldn't take into account if the wheels were tilted in opposite directions.
Scenario 1 - Imagine looking from the back of the bike, the wheels could both be vertical lines and your measurement at the height of the axle would be fine.
Scenario 2 - Imagine looking at the bike from the rear again. this time the the wheels are not perfectly vertical, one wheel is tilted +1 deg and the other -1 deg. The wheels still cross at the centre where you are proposing your more accurate measurement, but they are not actually in alignment.
Great challenge! I suspect Brodie's comments were aimed at someone far less true than yourself.
Yes I'm sure your right! But I love a challenge!
Didn't you raise the same point a few months back with one of your videos?
I did a video on what alignment really is, this is the follow up I promised.
Here is the first video th-cam.com/video/YTXGi94D5eM/w-d-xo.htmlsi=I-NgK3TAb7NgKC6M
Bosch 40 X 40 profile has a longitudinal straightness tolerance of 1.5 mm per 2000 mm, so your final alignment check with feeler gauges is pretty much useless.
Yes, I considered that. Literally the first thing I considered when I was handed the profile. I was actually surprised when I measured it on the surface plate and couldn't get a feeler gauge under it. Although I admit I didn't bother past 0.1mm. so I guess I must have got lucky. If it had 1.5mm over 2 metres that would have been significant! Did you really think I hadn't checked that? :-)
Given you do don't check for parallelism between the rear wheel (which isn't even true, as can be seen when you spin it in the video) and the frame your method for checking coplanarity between the wheels doesn't really work, since the front wheel is mounted in the fork and can rotate to parallel with profile. The frame could easily be angled at the rear axle toward or away from the profile without it being easily visible or measurable. Over an estimated wheelbase of 1500 mm, they only need to be out of alignment by .024 degrees to get an out of plane measurement of .63 mm. Also, is the front wheel in dish/true to better than 0.63 mm at the point where you measured the gap? And to what tolerance is your surface plate calibrated if we're using it calibrate the profile? ;) @@ellisbriggsbikes
The whole point of using the profile was to make it easier to explain. But I did set it up at least 4 times and had similar results each time.
Wheels are dished and true to within to within 0.15mm.
It's easy enough to check the wheels for parallel on the surface plate with some guage blocks but I don't think you'll find enough of a discrepancy.
If you measure each variable of frame alignment using the surface plate as datum, you can see which frames are going to have an issue at the wheels. I already knew this frame was good and the results are repeatable.
When you are riding the front wheel will also turn to be parallel, so I don't see your point there. Even still if the rear wheel is toe in or toe out it will give you an offset even if it is parallel. Also if the headtube tilted.
The results were similar to the other method I've used in the past, which involves mounting frame fork and both wheels on the face plate. In this arrangement it is possible to isolate which variables cause the discrepancy.
Surface plate is calibrated to grade 3, which is more than good enough for a bike frame.
Perhaps I'll do a more technical video for those who are interested.
Repeatability is completely irrelevant if your measurement equipment isn't good. You're just repeating the incorrect measurement...@@ellisbriggsbikes I'm going to have to disagree on the trueness of the wheels, see 3:11 in the video. And even if they are within 0.15 mm, the best tolerance you can ever state is to within +-0.3 mm. That's already half of what you claim before one adds anything else to the tolerance stack. Was the measurement with the profile to measure parallelism or coplanarity? These are different things. For a grade 3 surface table of your size the tolerance is around .05-.06 mm, how thick were the feeler gauges you checked your profile with?
@@alexv7349 you haven't shown anywhere where the measuring equipment isn't good enough. You pointed to the profile, which has been measured on the surface plate, then you pointed out the surface plate which has been calibrated, then the wheels which are true and in dish. You've just made a lot of assumptions.
I'm not sure what point you're trying to make really?
Obviously any bike in the real world can only have wheels as good as you can make them..I accept there will be some error but not enough to make a significant difference.
I don't think Paul Brodie does 'Yorkshire' and I'm fairly sure EB frames are straight or you and your predecessors wouldn't have lasted ten minutes as builders. Having said that, I might track down a straight edge and have a check : -) PS. love the video.
Judging by some of the frames that I see through the shop, I don't think theres enough emphasis on getting it right anymore. Thanks for watching!
Darn shame the background music was so annoying. It was in some areas louder than his narrative. I would love to have watched the rest of the video but couldn't get past the annoying music.
Straight talking video.
Can’t understand what you are saying due to the irrelevant and unnecessary music running in the background.
Yes perhaps I went a bit over the top with the music on that one. All though others have said the opposite!
Shameful? Explain?
Shameful at best
Please explain?
unwatchable and unlistenable awefull
Yeah i think I over did the music a bit. Thanks visiting anyway. I've toned the music down now! th-cam.com/video/yUkgdC2MgL4/w-d-xo.htmlsi=tZt8_3emaL5HbcrO
@@ellisbriggsbikes don't tone down, no music period.
@@hrxy1 I find it a bit boring without music tbh. Maybe I should do a poll on this