I finished slotting and doming the fork blade and successfully brazed the dropout in to the blade over the weekend. Without actually logging the time, I think I spent about 10 hours over a week and a half sawing, filing, sanding, and filing some more. I hope that I learned enough to cut that time down significantly for the next blade.
Is this worth it? I think it's a good experience, but in the future I'll be going with the socket-style dropouts. For the sake of continuity, I'll probably use slotted rear dropouts, but future frames will be socketed.
It was very encouraging to find out that I can braze with bronze using the hand-held torch. I'm sure that I'm eating up fuel like nobody's business, but in a pinch it's very capable.
Step-by-step walkthrough with photos will be forthcoming once I start the second blade later in the week.
Monday, July 19, 2010
Tuesday, July 13, 2010
More Right Tools
I purchased a "light-duty" vise from the local Ace Hardware store. It weighs about thirty pounds, and I carried it home in a backpack (with an extra canister of MAPP gas) on my Raleigh.
It is also completely covered in grease. I can understand lubricating the moving bits, but this thing looks like it was dropped in a vat before shipping off to the store. The screw, both handles (for the jaw and the rotational lock), jaws, and the flat surface on the anvil were all coated in a thick ooze.
After a quick wipe-down (now I need to go back to Ace and get more rags), I used the vise to pinch the slot on the fork blade closed. It was much easier than trying to use the C-clamp, but I encountered a few issues-
First, the vise is just resting on my workbench. I still need to hold on to it if I need to apply any sort of torquing force to it, such as trying to turn the handle to close the jaws on a stiff piece of steel tubing. This is just going to be unavoidable for now, but I may drill some holes in the workbench to semi-permanently mount the vise. I'll need to be able to remove it if I want to move the workbench around.
Second, the slot did not close evenly. One "tang" on the end of the blade moved much more significantly than the other. I don't think this is a huge deal, but it does mean that the end of the fork blade will likely be a little uneven. When I opened the slot up again with the file, it took more material off of the less pinched side of the slot, so maybe it will even itself up if I try to close it up again. I may be able to adjust the position of the blade in the vise to favor the unbent side of the slot.
Work has become difficult because a) we've hit full-on New England Summer and it's hot as balls and b) there is zero ventilation in the basement. Sweat just pours off me and there's no air circulation to help it evaporate. The file is frequently binding in the slot, making it slow going to get it opened up again.
Slowly but surely, work moves on. I'll take photos of the second blade as I slot it.
It is also completely covered in grease. I can understand lubricating the moving bits, but this thing looks like it was dropped in a vat before shipping off to the store. The screw, both handles (for the jaw and the rotational lock), jaws, and the flat surface on the anvil were all coated in a thick ooze.
After a quick wipe-down (now I need to go back to Ace and get more rags), I used the vise to pinch the slot on the fork blade closed. It was much easier than trying to use the C-clamp, but I encountered a few issues-
First, the vise is just resting on my workbench. I still need to hold on to it if I need to apply any sort of torquing force to it, such as trying to turn the handle to close the jaws on a stiff piece of steel tubing. This is just going to be unavoidable for now, but I may drill some holes in the workbench to semi-permanently mount the vise. I'll need to be able to remove it if I want to move the workbench around.
Second, the slot did not close evenly. One "tang" on the end of the blade moved much more significantly than the other. I don't think this is a huge deal, but it does mean that the end of the fork blade will likely be a little uneven. When I opened the slot up again with the file, it took more material off of the less pinched side of the slot, so maybe it will even itself up if I try to close it up again. I may be able to adjust the position of the blade in the vise to favor the unbent side of the slot.
Work has become difficult because a) we've hit full-on New England Summer and it's hot as balls and b) there is zero ventilation in the basement. Sweat just pours off me and there's no air circulation to help it evaporate. The file is frequently binding in the slot, making it slow going to get it opened up again.
Slowly but surely, work moves on. I'll take photos of the second blade as I slot it.
Wednesday, July 7, 2010
The Right Tool
After a long holiday weekend full of family, I was ready to get back to working on the bike. After getting the blades bent, the next step is to slot and dome the blade ends.
There are two types of dropouts - socket and slot style. The socket style is easier to work with by far, simply slide it on the end of the fork blade and start brazing. Some of them even have small recesses where you wrap the solder wire around; slide it over the blade and heat the joint up, and the solder will flow from its holding spot into the joint.
I went for the slot style because Talbot recommends using them for your first frame (for some bizarre reason). There is significantly more work involved, and it seems much easier to screw things up.
The process of slotting the fork blades involves cutting a slit down the smaller end of the blade (this is the round end if you have oval blades) and then filing it open to accept the dropout. You can perform a variety of finishes on the end of the blade; Talbot describes how to dome the blade, but there are other versions that he does not describe.
The first step is to cut the slit. The cut must be lined up with the bend in the fork blade or else the dropout will be installed askew. Install two blades side-by-side in your hacksaw, and then carefully cut a 1/2" length slit at the narrow end of the fork blade. I found this very difficult to get started, since the saw kept chattering and the two blades were bowing against each other.
After the slit has been roughly cut with the hacksaw, you must now start filing. If you don't have a warding (flat) file, now would be a good time to get one. I spent about two hours with a small flat needle file before I gave up - the needle file is simply not designed to remove the amount of material that was left. The hacksaw produced a slit about 2.5 mm thick, while the dropout is 4.5 mm thick; I managed to open the slit to about 2.8 mm before I gave up.
The next day, I stopped by the local hardware store and picked up an 8" flat file with a double-bastard cut. The file is about 3.5 mm thick, which works out perfectly for this application. It took another hour and half or so to open up the slit from 2.8 mm to a size that would accept the dropout snugly. You will be filing for quite a long time; turn on some music and drink plenty of water.
The next step is to press the slit closed and reopen it by once again cutting and filing. Here is where I've currently hit a wall. Talbot uses a vise to close the slit; I don't have one, so I tried for about 15 minutes to close it with a large cast iron C-clamp. The clamp, unfortunately, has way too much play in it - the side of the clamp attached to the screw tends to skew, and it's attached to a ball joint that allows the flat part to slide around the blade.
I got the slit closed maybe a fraction of a millimeter, but this method simply isn't working. I tried closing it using the split tabletop on the workbench and a ratchet clamp, but only succeeded in putting a dent in the MDF tabletop and breaking the ratchet in the clamp.
I'm going to have to buy a vise to finish the fork blades. I shouldn't need to mount the vise on anything to get the slit closed, but this is a good excuse to get a workbench to install the vise on.
No pictures yet, but when I get do the other blade I'll document it better.
There are two types of dropouts - socket and slot style. The socket style is easier to work with by far, simply slide it on the end of the fork blade and start brazing. Some of them even have small recesses where you wrap the solder wire around; slide it over the blade and heat the joint up, and the solder will flow from its holding spot into the joint.
I went for the slot style because Talbot recommends using them for your first frame (for some bizarre reason). There is significantly more work involved, and it seems much easier to screw things up.
The process of slotting the fork blades involves cutting a slit down the smaller end of the blade (this is the round end if you have oval blades) and then filing it open to accept the dropout. You can perform a variety of finishes on the end of the blade; Talbot describes how to dome the blade, but there are other versions that he does not describe.
The first step is to cut the slit. The cut must be lined up with the bend in the fork blade or else the dropout will be installed askew. Install two blades side-by-side in your hacksaw, and then carefully cut a 1/2" length slit at the narrow end of the fork blade. I found this very difficult to get started, since the saw kept chattering and the two blades were bowing against each other.
After the slit has been roughly cut with the hacksaw, you must now start filing. If you don't have a warding (flat) file, now would be a good time to get one. I spent about two hours with a small flat needle file before I gave up - the needle file is simply not designed to remove the amount of material that was left. The hacksaw produced a slit about 2.5 mm thick, while the dropout is 4.5 mm thick; I managed to open the slit to about 2.8 mm before I gave up.
The next day, I stopped by the local hardware store and picked up an 8" flat file with a double-bastard cut. The file is about 3.5 mm thick, which works out perfectly for this application. It took another hour and half or so to open up the slit from 2.8 mm to a size that would accept the dropout snugly. You will be filing for quite a long time; turn on some music and drink plenty of water.
The next step is to press the slit closed and reopen it by once again cutting and filing. Here is where I've currently hit a wall. Talbot uses a vise to close the slit; I don't have one, so I tried for about 15 minutes to close it with a large cast iron C-clamp. The clamp, unfortunately, has way too much play in it - the side of the clamp attached to the screw tends to skew, and it's attached to a ball joint that allows the flat part to slide around the blade.
I got the slit closed maybe a fraction of a millimeter, but this method simply isn't working. I tried closing it using the split tabletop on the workbench and a ratchet clamp, but only succeeded in putting a dent in the MDF tabletop and breaking the ratchet in the clamp.
I'm going to have to buy a vise to finish the fork blades. I shouldn't need to mount the vise on anything to get the slit closed, but this is a good excuse to get a workbench to install the vise on.
No pictures yet, but when I get do the other blade I'll document it better.
Wednesday, June 30, 2010
Blades: Bent
I headed over to Royal H Cycles in Somerville yesterday, where frame builder Bryan was able to bend my fork blades. He actually has a bending tool similar to what I was planning to build, which was encouraging to see. I will have to try to finish my own bender and try it out on some scrap tubing.
One piece of advice Bryan gave that I had not fully considered is the trail of the bike. Trail is defined as the horizontal distance from the axis of rotation of the front wheel (the center of the hub) to the extended centerline of the steerer tube. Proper trail allows the front wheel to act like a caster; hitting an obstacle that changes the direction of the front wheel won't suddenly change the direction of travel of the whole bicycle. A shorter trail makes the bike more responsive but will lead to instabilities - track bikes will often have short trails, since there aren't things like potholes or sticks to run over on the track.
With my desired head tube angle and rake, the bike was going to have a relatively short trail; UBI apparently recommends somewhere in the vicinity of 55-65 mm, and my design was floating around the 35 mm area. To solve the problem, I'm going to reduce the head tube angle to 72 degrees, which BikeCAD says will give me 58 mm of trail. Since I still want the same frame size (especially the head tube, since I've already cut the steerer with that dimension in mind), this will effectively lengthen the fork blades by about 10 mm and the down tube by about 85 mm. Convenient, since I haven't cut any of these tubes yet.
A more significant problem will be sourcing a head tube lug with a 72 degree angle. Most of the lugs available are angled at either 73 or 74 degrees, so finding one "off the shelf" may be difficult. One option will be to purchase a 73 degree lug and open it up a little bit by hand.
Next steps are to cut the fork blades to length, slot and dome them to accept the dropouts, and then file the fork crown tangs and braze everything together. I'll have some photos of Royal H soon.
One piece of advice Bryan gave that I had not fully considered is the trail of the bike. Trail is defined as the horizontal distance from the axis of rotation of the front wheel (the center of the hub) to the extended centerline of the steerer tube. Proper trail allows the front wheel to act like a caster; hitting an obstacle that changes the direction of the front wheel won't suddenly change the direction of travel of the whole bicycle. A shorter trail makes the bike more responsive but will lead to instabilities - track bikes will often have short trails, since there aren't things like potholes or sticks to run over on the track.
With my desired head tube angle and rake, the bike was going to have a relatively short trail; UBI apparently recommends somewhere in the vicinity of 55-65 mm, and my design was floating around the 35 mm area. To solve the problem, I'm going to reduce the head tube angle to 72 degrees, which BikeCAD says will give me 58 mm of trail. Since I still want the same frame size (especially the head tube, since I've already cut the steerer with that dimension in mind), this will effectively lengthen the fork blades by about 10 mm and the down tube by about 85 mm. Convenient, since I haven't cut any of these tubes yet.
A more significant problem will be sourcing a head tube lug with a 72 degree angle. Most of the lugs available are angled at either 73 or 74 degrees, so finding one "off the shelf" may be difficult. One option will be to purchase a 73 degree lug and open it up a little bit by hand.
Next steps are to cut the fork blades to length, slot and dome them to accept the dropouts, and then file the fork crown tangs and braze everything together. I'll have some photos of Royal H soon.
Tuesday, June 29, 2010
Fork Blades
I was expecting to have my fork blades bent over the weekend and writing up the process early this week, complete with color photos and maybe even an informal interview with the frame builder who said he would help me.
None of this happened.
Instead, I waited around all Saturday afternoon for a call, only to never receive one. I sent the guy an e-mail Monday morning asking what happened, and his reply told me that he would be available today, Tuesday - just drop the blades off at his shop. So, there will likely not be any photos or talk with the frame builder and I have no insight in to when this work will actually get done.
I realize that asking to bend my fork blades is pretty small potatoes compared to purchasing a full custom frame, but there is a certain amount of professionalism that is lacking here. Even a phone call trying to reschedule would have been fine with me.
Best case scenario - I get the blades bent today and everyone goes home happy. In reality, I'm expecting that it will take a few days and I won't get to continue work until after the July 4 holiday. If everything conspires against me, the work won't get done, I won't get my blades back, and I'll be out twenty bucks.
None of this happened.
Instead, I waited around all Saturday afternoon for a call, only to never receive one. I sent the guy an e-mail Monday morning asking what happened, and his reply told me that he would be available today, Tuesday - just drop the blades off at his shop. So, there will likely not be any photos or talk with the frame builder and I have no insight in to when this work will actually get done.
I realize that asking to bend my fork blades is pretty small potatoes compared to purchasing a full custom frame, but there is a certain amount of professionalism that is lacking here. Even a phone call trying to reschedule would have been fine with me.
Best case scenario - I get the blades bent today and everyone goes home happy. In reality, I'm expecting that it will take a few days and I won't get to continue work until after the July 4 holiday. If everything conspires against me, the work won't get done, I won't get my blades back, and I'll be out twenty bucks.
Wednesday, June 23, 2010
Cleaning up joints
I spent about an hour in the basement yesterday cleaning up the joint I made over the weekend. The solder material is much softer than the steel, which makes removing the excess pretty easy. I used a triangular mill file to remove the bulk of the excess; the points on the file worked great at keeping the joint clean and sharp.
After removing most of the extra solder with the mill file, I alternated between a medium-grade emory paper and a square needle file. The emory paper was better at removing material from the steerer tube, while the needle file was good for removing material from the fork crown and keeping the joint profile square.
I noticed a small amount of pitting in the solder while I was cleaning it up, and I'm not sure what to make of that. I was removing material from the "bottom" of the weld - where the solder flowed out of the joint - which is actually the top of the fork crown. I think, but am not positive, that the pitting is probably due to the fact that I tried reheating the joint in some places to get more solder in and fill in the small section of the joint that didn't have a large glob coming out the bottom. I could try to file the joint down some more, but this will all be hidden by the headset, so I'm not sure it's really necessary.
After finishing the outside of the joint, I took a round file and cleaned out some of the junk on the inside of the tube. I also had missed some flux that had dripped into the steerer threads, so I took out the wire brush again and cleaned that up. I think this joint is now "done". There will be more work to do after the fork blades are attached.
One thing I realized - after the bike is finished, there's going to be a large hole in the bottom of the fork crown where the steerer tube ends. Do people normally seal these up somehow? Without a fender, a lot of grime and muck is going to get thrown into the fork. I've never noticed a hole on any other bike I've owned, but maybe I'm just not attentive enough.
After removing most of the extra solder with the mill file, I alternated between a medium-grade emory paper and a square needle file. The emory paper was better at removing material from the steerer tube, while the needle file was good for removing material from the fork crown and keeping the joint profile square.
I noticed a small amount of pitting in the solder while I was cleaning it up, and I'm not sure what to make of that. I was removing material from the "bottom" of the weld - where the solder flowed out of the joint - which is actually the top of the fork crown. I think, but am not positive, that the pitting is probably due to the fact that I tried reheating the joint in some places to get more solder in and fill in the small section of the joint that didn't have a large glob coming out the bottom. I could try to file the joint down some more, but this will all be hidden by the headset, so I'm not sure it's really necessary.
After finishing the outside of the joint, I took a round file and cleaned out some of the junk on the inside of the tube. I also had missed some flux that had dripped into the steerer threads, so I took out the wire brush again and cleaned that up. I think this joint is now "done". There will be more work to do after the fork blades are attached.
One thing I realized - after the bike is finished, there's going to be a large hole in the bottom of the fork crown where the steerer tube ends. Do people normally seal these up somehow? Without a fender, a lot of grime and muck is going to get thrown into the fork. I've never noticed a hole on any other bike I've owned, but maybe I'm just not attentive enough.
Tuesday, June 22, 2010
On Bending
I think I have a lead on getting my fork blades raked. It's probably not going to happen until this weekend, so this week will probably be a little slow - I can finish the steerer joint, but after that I have to wait on the blades before I can continue work on the fork. In the meantime, let's look at the different options I've considered for bending blades.
Talbot makes no mention of how to rake fork blades, because he was able to purchase pre-raked blades for the frame he builds in his book. I have not found any source of bent fork blades, which means that it's up to the builder to do this work. Without a large work shop and heavy machinery, I'm at a bit of a disadvantage.
The first plan I had in mind was based off of this builder's fork bender. I went to the local woodworking shop and got a heavy piece of maple. Cutting it with a jigsaw, I realized that the wood was slightly warped, meaning that the saw did not make a flat cut. I could have solved this problem if I had a wood plane. A second problem arose - how do I cut a groove in the wood to accept the tubing? I was unwilling to purchase a router (or router table), and hand methods would have taken a considerable amount of time - not to mention being inaccurate. So, scrap this idea for now. I may continue to work on it for future bikes.
Some more research put me on to the idea of a conduit bender. These tools are used to shape electrical conduit, and it looks like recumbent builders have put them to use increasing the rake of previously fabricated forks. Off I went to Chasen Industrial Supply, who have a terrible website but a really fantastic set of folks on staff. Unfortunately, after talking the idea over with the folks at the store, they convinced me that I was very likely to crush or kink the tubing trying to use a conduit bender; conduit is much thinner and more malleable than the tubing for a bicycle fork, and the benders just don't seem to be up to the task. (Although, if you have any hard experience successfully bending a straight fork with a conduit bender, let me know!) They did have the Dykem I was searching for, though, so the trip was not all in vain.
When I got home, I called around to some of the bike shops and fabricators in the area. Independent Fabrication told me that they only use straight blades (the rake is achieved with an angle at the crown). I also called the Broadway Bicycle School in Cambridge - they have an open shop that one can rent by the hour. They don't have the tools, but they did suggest calling Geekhouse Bikes in Allston. I left a voice mail with them, but also surfed through their Links page to Royal H Cycles in Somerville. Bryan told me he'd be willing to rake the blades for me with his homemade bender for a nominal fee. This is the route I'll be taking for now.
What other options are there? Nova Cycle Supply sells a bender for $375, which is outside my price range for this project (especially considering the likelihood of being able to resell the tool is pretty minimal). Finally (and the option that I'm most attracted to but least equipped for), I could make a bender out of metal.
Homemade benders that I've seen online tend to have a large radius curved metal block with a groove in it for the tubing - similar to what I was trying to make out of wood. They also have a long handle that is pinned to the block so it can rotate around the curve; the handle has a roller on it that guides the tubing around the curve. Making something like this would require a machine shop - at the very least a mill, which I don't have the money or space for. However, there is a new community workshop opening in the Boston area which should have the capabilities I would need.
I feel a little chagrined about not bending the blades myself, but I'm coming to terms with the idea that it's ok to outsource when you don't have the capabilities yourself (I'll let someone else paint the bike, for example). Maybe after visiting Royal H, I'll have a better idea of how to build my own for future projects.
I also wanted to share an acronym I learned from the framebuilders e-mail list - SOPWAMTOS. It stands for "Society Of People Who Actually Make Their Own Shit", and I'm obviously not following the spirit of it by not building my own tooling. One day!
Talbot makes no mention of how to rake fork blades, because he was able to purchase pre-raked blades for the frame he builds in his book. I have not found any source of bent fork blades, which means that it's up to the builder to do this work. Without a large work shop and heavy machinery, I'm at a bit of a disadvantage.
The first plan I had in mind was based off of this builder's fork bender. I went to the local woodworking shop and got a heavy piece of maple. Cutting it with a jigsaw, I realized that the wood was slightly warped, meaning that the saw did not make a flat cut. I could have solved this problem if I had a wood plane. A second problem arose - how do I cut a groove in the wood to accept the tubing? I was unwilling to purchase a router (or router table), and hand methods would have taken a considerable amount of time - not to mention being inaccurate. So, scrap this idea for now. I may continue to work on it for future bikes.
Some more research put me on to the idea of a conduit bender. These tools are used to shape electrical conduit, and it looks like recumbent builders have put them to use increasing the rake of previously fabricated forks. Off I went to Chasen Industrial Supply, who have a terrible website but a really fantastic set of folks on staff. Unfortunately, after talking the idea over with the folks at the store, they convinced me that I was very likely to crush or kink the tubing trying to use a conduit bender; conduit is much thinner and more malleable than the tubing for a bicycle fork, and the benders just don't seem to be up to the task. (Although, if you have any hard experience successfully bending a straight fork with a conduit bender, let me know!) They did have the Dykem I was searching for, though, so the trip was not all in vain.
When I got home, I called around to some of the bike shops and fabricators in the area. Independent Fabrication told me that they only use straight blades (the rake is achieved with an angle at the crown). I also called the Broadway Bicycle School in Cambridge - they have an open shop that one can rent by the hour. They don't have the tools, but they did suggest calling Geekhouse Bikes in Allston. I left a voice mail with them, but also surfed through their Links page to Royal H Cycles in Somerville. Bryan told me he'd be willing to rake the blades for me with his homemade bender for a nominal fee. This is the route I'll be taking for now.
What other options are there? Nova Cycle Supply sells a bender for $375, which is outside my price range for this project (especially considering the likelihood of being able to resell the tool is pretty minimal). Finally (and the option that I'm most attracted to but least equipped for), I could make a bender out of metal.
Homemade benders that I've seen online tend to have a large radius curved metal block with a groove in it for the tubing - similar to what I was trying to make out of wood. They also have a long handle that is pinned to the block so it can rotate around the curve; the handle has a roller on it that guides the tubing around the curve. Making something like this would require a machine shop - at the very least a mill, which I don't have the money or space for. However, there is a new community workshop opening in the Boston area which should have the capabilities I would need.
I feel a little chagrined about not bending the blades myself, but I'm coming to terms with the idea that it's ok to outsource when you don't have the capabilities yourself (I'll let someone else paint the bike, for example). Maybe after visiting Royal H, I'll have a better idea of how to build my own for future projects.
I also wanted to share an acronym I learned from the framebuilders e-mail list - SOPWAMTOS. It stands for "Society Of People Who Actually Make Their Own Shit", and I'm obviously not following the spirit of it by not building my own tooling. One day!
Subscribe to:
Posts (Atom)