Monday, December 31, 2007

Updated Sites

No photos yet - sorry. I've been busy updating this and the Cycles Noir sites. This blog is just about done - although I have to figure out how to get my counter back. The website is coming along fine too - you can check it out on the link at the right.

In the mean time, Happy New Years to everyone!

Thursday, December 27, 2007

Catch up again

Over the last two weeks, the day job has been very busy - meaning no time in the shop. Then of course there was Christmas - which was great, but didn't allow for any building. So, today, I got back into the shop and working. I'll try to get some pix tomorrow to illustrate. Anyhow, here's what's up.

I'm back to working on Paul's track bike. The path-racer is so close to being done, but I decided that it's inspiration (the white tires) didn't look right on it. I've ordered some new Grand Bois to replace them, but I can't finish it until they arrive. That said, the Honjo hammered fenders look great against the Burgundy paint. The seat is a B17 special in dark green with copper rails. The handlebar tape will be green cotton shellacked, so as to accent the saddle. Final pictures should be ready in about 2 weeks.

Therefore, back to Paul's bike. He likes all things Italian and has been steadily collecting old Campy components. The hardest so far has been the hubs. He needed 32 hole (to match some NOS rims he has) and wanted high flange. These are a bit hard to find - but we did it.

Anyhow, the lugs are old Cinelli pressed steel. The bottom bracket is is a cast Cinelli road bracket - the one with 'spoiler' on the bottom. Of course, that isn't quite right for a track bike. Out came the hack saw and the cable eye is gone. That and a bit of work with the file and a rotary stone had things cleaned up. There are also cast in tunnels (canals?) for the cables. So I filled these using some Allstate Nickel-Silver filler. This filler has the characteristic of being Eutectic. What this means is that once it melts and rehardens, it takes significantly more heat to re-melt. So, it starts to be workable around 1200 degrees Fahrenheit, but it requires a bit over 1700 degrees to remelt. So what? Well, we'll be reheating this BB to install the Seat Tube, Down Tube, and Seat Stays. It sure would be nice to make sure that these processes don't damage the filler in the tunnels, eh? And my brass (bronze, alloy, whatever) filler works below the remelt temp of this Nickel-Silver so I can use it or silver based fillers to finish off the BB. At this point, the shell has been soaked and is ready for a little wire brush to clean off any remaining flux.

I also finished cutting miters on the main triangle today, and did a basic test fit. The lugs are close to ready for brazing and the fit up looks good. I had to adjust the lower head tube lug a bit, and may yet be doing the same for the upper and the seat cluster. The latter two haven't been checked yet for angles (the seat and head tubes are parallel - so this is basically to ensure that the top tube is level).

Adjusting the lower head lug was challenging. Basically I'm adjusting from about a 73 degree angle to a 75 degree angle. This means that the top of the lug at the back of the head tube needs to move forward. However, on this lug, this area is rather short. There isn't a lot of material to bend. I think that I have it looking good and fitting ok. But it took a lot of work with a ball-pein hammer. This lead to some additional file work to make sure that it had a smooth surface.

The pressed seat cluster had the standard hollow ears - which I always consider a possible weak point. My solution is to braze tube inside them (for the binder bolt to pass through) and then fill the lug up around the tube. Here is another good use for Eutectic filler. It's too easy for everything to get up to temperature at once, and therefore have the filler melt out of one part while trying to fill another part. Particularly on the inside (against the seatpost) there isn't much support. In fact its like a big square window into the backside of the ears. Moreover, I've already widened the slot (in part to get a nice square looking edge on the ears), leaving even less support. Working a spot, and letting is solidify
and then moving on, its possible to fill and get the filler to stay in place very easily with the Eutectic filler.

Anyhow, this is done and been soaked. Again, it may need a quick shot with the wire brush, but basically its time to clean up the filler with a file. First will be the inside of the tube so the seat tube will fit inside. Then I'll do around the inner edges of the ears, giving them some definition. Apart from this, the seat cluster appears ready to clean up and install.

Summing up, the angles of the seat cluster and upper head lug need to be checked and adjusted as necessary. The upper head lug needs a little filing and sanding. The bottom bracket needs a bit of filing and sanding - mostly in the hard to reach spots. And then we can fit everything in the jig, check numbers one more time, clean, flux and braze the front triangle. Yea!

Again, with luck there will be some build photo's tomorrow. That's all for tonight. Cheers!

Tuesday, December 11, 2007

What's up now?

Things have been busy, even though I've had too little shop time of late. So let's catch up a little.

The path racer is back from paint and I've been slowly building it up. After looking at the paint, it wasn't possible to go with a simple curved fender, so Velo-Orange shipped me some nice hammered Honjo's. The rear is fitted and looks great.

Usually with Honjo's, there is one eye-bolt per stay - centered on the fender. With the hammered fenders, there are two grooves running the length of the fender, and each is pre-drilled for an eye-bolt - so the stay is attached with two bolts set at close to a 90 degree angle. Eye-bolts always fit snuggly (there isn't a lot of take-up available so they need a tight fit). And the stay, where it bends around the fender, gets distorted a little so that the bolts fit even more tightly. Fitting two in together is a tricky job. I still have some final fitting to do so that everything is good and straight. That said, even with just the one fender mounted, its obvious that the hammered fender was the right choice.

The crank arms are old Stronglight 49's. Nice cranks. Came with a TA track chain-ring (1/8"). It took some experimentation, but a Sugino Mighty Comp (at least that's what the seller called it) with a 113 axle worked great. Perfect chain-line, the right crank sits right up close to the BB with the chain-ring bolts running a circle around the end of the BB shell. The BB is interesting, being composed of a pair of Sugino 75 NJS (says so right on them) cups with an axle that looks just like the 75 NJS track axle except that its longer and doesn't have the NJS label. I suspect that this will prove to be a good and durable BB, and plan to purchase some more. And, these cranks/BB set up yield a tread (or Q-factor) of 132mm - which works will with my knees.

After mounting everything, it became clear that the chain-ring was severely warped. I tried to straighten it, but after checking it on the alignment table - it was clearly shot. Once again Velo-Orange to the rescue! Hopefully my new ring will be here by the end of the week. In the meantime, an old Stronglight road chain-ring is in use for fitting everything together. It's pretty with their typical star pattern.

Wheels are Velocity Razor rims laced to Miche high-flange track hubs. The front is laced trued and tensioned. The rear, well, I've decided that the drive side needs shorter spokes. It's nice when all the spokes are the same length, but things got pushed a bit far, so I'm waiting for my new spokes to arrive. Should be here by the weekend, so I can finish the rear wheel. In the meantime, another wheel is standing in for fitting purposes.

If look at prior posts, you'll see this bike (sans paint) more or less built up around the tires that started this whole project. I may be replacing them with a pair of Grand Bois. The white tires feel a bit over the top for a claret metallic bike with hammered fenders. It's hard to be sure until we try them on the new wheels, but I'm leaning towards the Grand Bois right now.

I polished up a beat-up old aluminum seat-post for this bike. Up close, it still looks kinda rough, but from 2 feet away, it looks nice. Anyway, it'll do until I find an appropriate replacement. The headset is a Stronglight A9, and looks good on the bike. The steerer was left long so to fit brass bell on a spacer below the lock nut. In order for the bell to clear the headset, a 10mm spacer was required below it - so the steerer is really 15-17mm too long. That's ok because the stem is a Nitto Technomic, which is long to begin with. Having the long steerer hides more of the stem and makes everything look better balanced. If parts come in timely, this bike could be ready for pictures by the weekend!

I'm still working on Duane's bike. The key thing is polishing the head-tube. I'm down to 400 grit, at which point its possible to start to make out reflections. Still it needs to be worked down to 1200 or maybe 1500 and then polished. So, there are a few hours to go yet before final fitting of lugs and tubes prior to brazing. In the meantime, this is setting on the side as I have a more urgent project.

A friend and bike mechanic asked for a track frame, which we started this week. He's been collecting vintage Campy and Cinelli parts, and mostly has what he needs. I've found some Columbus SLX to fit, so we're building with that and an old set of Cinelli pressed lugs. He loves the Cinelli airfoil BB, so even though that's a road BB, we're going to use it. That means cutting off the cable stop, filling the cable grooves, and closing the cable eyes in the airfoil. This is more than my usual amount of prep work for a BB, but should be an interesting mini-project.

So far, the lugs are rough prepped. The still need their angles set and the edges squared, but they've been filed and are close to their final contours. The other night Paul cleaned, checked, measured and marked the main tubes, and tonight we started a couple of miters. First Paul did one with hacksaw and files. He got chance to see how hard it is to create a good fit. Mind you, he's good with tools and has a good eye, so if he felt challenged, anyone would. I finished that one up - just a little tuning really, and we moved on to the next one. This was the bottom of the seat tube. For it, I demonstrated the hacksaw, grinding wheel, file, method. It went much more quickly as the grinding wheel rapidly took us close to the template all around. So now we only have three more miters prior to fitting and tacking the main triangle. There will be no braze-ons, so this should go together relatively fast.

In other news, I received a set of the new ('08) White Industries H2 hubs today. They look lovely, and weigh in at about 300 grams for the pair. They will be paired with a set of Velocity 650B Arrowhead rims. Along with 32 14/15ga spokes per wheel, and brass nipples, I'm hoping to have these wheels come in around 1500grams total. Which would be quite a feat!

To go with these, a SRAM Force group is on order. It would be nice to use Red, but out-of-pocket expenses must be managed. Red can go on a customer bike. Anyhow, the Force group is sans brakes. Most likely these will be cantilevers - still trying to decide on which to use. Its likely that what ever gets picked will be lighter than even the Red brakes - but will accommodate fenders and wider tires.

Also in the package from white is one of their Cranksets. It looks lovely, and I want to see how low a tread I can get with it. It might be the modern replacement for the TA Cyclotourist. One disappointment is to find that the chain-rings only come in black. Apparently, the bare aluminum isn't durable enough for long life and requires a hard anodize finish to last. Old chain-rings didn't need this, and must have been made from a stronger alloy. I wish I knew how to get White to make their rings from a stronger allow to so that I could have polished rings.

By the way, if you've never looked at these, they're kinda of unique. The large chain ring has deep slots running in five radii from the center. The small ring bolts to these slots, so an infinite number of small chain-ring BCD's can be used. Kinda cool, and it allows for small chainrings too! Very nice.

It's late, that's all for tonight. More soon.

Saturday, December 01, 2007

Blue Bike Done

I've been a bit busy. Got the shop re-organized in preparation for another binge of building. Also got the blue frame built up. Take a peek (click on the pic to enlarge):

Sunday, November 18, 2007


My computer started to die early in the week. I've been struggling since then to move data and programs over to a new machine. The work is almost done, but has everything, including blogging, way behind schedule. I have received the path racer frame back from the painter. I'll have photos of the frame soon, and built up within 2 weeks.


Thursday, November 08, 2007

Checking back in with the world

OK, I've been gone for a while. Lots going on, much of which will provide fodder for the blog.

I'll have finished bike pictures of the green and blue bikes very soon, I promise. For today, we'll have something different to see.

What does that look like? If
you said a GAST DOA-V722-AA, you were right. And in that case you must be much smarter than me.

So what you say? Well, I came to the sad realization that my current vacuum pump was too small to do adequate work making wet wrapped vacuum-bagged joints on a bicycle frame.

There are a couple of ways around that obstacle. It's possible to buy a little device which creates a vacuum when hooked up to the hose of a compressor for about $100. Or you can buy a good vacuum pump. In my experience these tend to run from about $500 and up. Or, you can go to trusty eBay and find someone selling a nice industrial quality vacuum pump, used, for reasonable prices.

It took three or for tries, but I purchased this one for $47. It'll draw over 25" of mercury, which should be plenty of pressure for the joints I'm going to do. In fact, for a small object, where the carbon is being formed over Styrofoam forms, this is probably too much pressure. It would tend to distort the form during the cure.

The basics of the process go like this. I'll miter carbon tubes to fit together correctly. Then I'll glue them in place with an epoxy, using the frame jig. This will create a straight carbon frame, but no where near enough strength in the joints to be used as a bicycle.

From here, I will wet out layers of carbon fiber with a high quality epoxy. These layers will be wrapped in various patterns around the joint. Along the way, the carbon fiber will be oriented to use its strength (pulling or stretching) not its weakness (compression or pushing). This could then be allowed to harden - but it won't.

First of all, there is no way to wet out the carbon fiber thoroughly without giving it too much epoxy. This adds excess weight, which isn't what we want to do with carbon fiber. Also, epoxy isn't all that strong on its own. If there are thick gobs and goobers of epoxy between the layers of carbon fiber, its likely to weaken the joint. Finally, as much as we try, there can be voids between the strands of CF where we didn't penetrate with epoxy during the wetting out - these areas will be weak.

This is where the vacuum pump comes in. We'll layer the wrapped joint with a material that will release from the epoxy after everything is set up. There are a variety of alternatives for this. A very smooth material, say like mylar, can give a very finished finish to the CF. That's nice, but no ideal for our case. I'm going to start with Teflon coated polyester cloth.

It just occurs to me that Teflon isn't a great material for the world, so maybe I'm going to have to look for an alternative in the long run.

Getting back on track, this cloth breaths. Behind the cloth will be a layer of batting. It looks sort of like the stuff people put on table tops to look like snow, before they set up a winter scene with figurines. Anyhow, this joint, wrapped in polyester cloth, with batting over that, gets sealed into a large plastic bag. At the back side of the batting (and potentially in more than one spot), a hose (or hoses) are attached to the bag and then to a vacuum pump (this is where the GAST DOA-V722-AA comes in).

Turning the pump on creates a (partial) vacuum inside the bag. Outside the bag, where there is no vacuum, we have the weight of our earth's atmosphere pushing on every side of everything it touches. If our skins were hollow (implying a vacuum inside), they would be squished down to a nasty little lump.

In the case of our bike frame, the carbon tubing provides a strong stiff platform for the vacuum to push against. So, the bag is pushed against the batting, which is pushed against the Teflon fabric, which is pushed against the set joint. In a process not unlike squeezing a little packet of ketchup, the wet epoxy wants to get out of the way of all this pushing.

It gets squeezed up through the layers of carbon fiber, and through the Teflon fabric, into the batting. Generally, the epoxy is too stiff to push through the batting and into the pump - which is a good thing. And, because the batting has enough strength or loft to keep from flattening out, it is home to a vacuum that the epoxy wants to fill.

In this way, the layers of CF are pressed closely together where the epoxy can effectively bind them and allow the CF to be the strength of the joint. Meanwhile, excess epoxy is removed from the joint, helping keep its weight to a minimum. And, with a little luck, most voids will be located somewhere with epoxy behind them. As the epoxy is sucked out to the batting, it will go through these voids, leaving enough epoxy to make sure that these areas of the CF are at full strength.

I seem to have a machinist lined up now to make my lug spigots for the bottom bracket. Hopefully before Christmas, I'll have my first prototype carbon frame with wet-wrapped vacuum-bagged joints. And while the GAST may look mundane, this is why I'm excited that it arrived. And that's all for today.

Friday, October 26, 2007

Friday Update

Wednesday night I did not mount fenders. Instead the rider took the bike for a test spin. By the way, its the blue bike listing below under "Sneak Preview" Anyhow, she was extremely happy with the ride. Good on that.

Now I've hung the blue bike up for a few days. Tomorrow is our annual ride up the Des Plaines river trail to the Wisconsin border. This is fun, and often wet. While the trail isn't paved, much of it is cinder, so it doesn't require anything like a mountain bike. 30mm tires and fenders will do the job. Nonetheless, a number of folks will ride mountain bikes because their road bikes can handle wider tires.

For myself, I'll be riding my test hack - which is why I've put the blue bike aside. I had some nice Michelin 28s, which were really more of a 30 or 32mm tire. They weren't great on the road, but they were fine on the trail. Unfortunately, one of them wore out. They got replaced by a nice set of Hutchinson's with a folding bead. These are fast tires, with 80psi in front and 85psi in back. But, they're too narrow for wet weather on the trail. It's been raining this evening and some form of rain is being predicted for the morning.

I went scouting out wider tires, naturally looking for a deal. I found some Panaracer Pasellas (sp?) with a folding bead for less than $20 each in a 700x35c. These are pretty highly regarded tires, reasonably light and good rolling. The question was: could I mount them and fenders on the hack?

Meanwhile, on my last ride, the hack was slow and draggy. The culprit seemed to be the bottom bracket. The crankset is an FSA Gossamer compact double with external cup bearings.

So, BB, tires, fenders, have taken all my shop time since Wednesday.

I pulled the cranks, sure enough the left bearing was toast. I dug up a Truvativ crankset with external cup BB from inventory. I mounted the cups and tried to push the Gossamer through. Oppps, I pushed the press-fit bearing out of the left cup. It turns out that FSA and Truvativ use different specs. The Truvativ left bearing has a smaller hole for the axle, which necks down to fit. The FSA axle doesn't neck down. As best I can tell, Truvativ uses the same spec as Shimano, but apparently FSA doesn't.

Hopefully I haven't damaged the Truvativ BB, but I wasn't going to use this crankset (carbon arms) on the hack in the mud tomorrow. This lead to some thinking. I got out a nice old Stronglight 93. One of prettiest cranksets ever made. It's rather dirty, and has some oxidation - but its cool.

The Stronglight is a true low-tread (low Q) crankset. And it needs a longer axle than any of the BB I have in stock - so that means a trip to pick up a new BB. Also, with 10speeds on the back, It didn't make much sense to me to leave the the large (52t) chain ring on. Instead the small (47t) ring got moved to the outside, and mounted with track chain ring bolts. On this went a set of bright red Look Keo pedals.

The normal wheels on the hack are a set of Record 10 hubs laced to some MA3 rims. Smooth rolling and dependable. I also have a set of wheels that matches this except that the hubs are Chorus. I mounted the new tires to the Chorus wheel set - so we can easily switch between to the two tire sizes. Unfortunately, the Chorus wheels (an eBay find) had their own problem. In particular, the cassette lockring won't engage to thread onto the cassette carrier. I can see anything wrong with the cassette carrier, and tried three different (brands or) lockring with the same results. Eventually, I gave up for now and swapped tires on the Record rear hub. Along the way I cleaned all the grit and dirt out of the cassette - so it looks so much better now next to the Stronglight crank.

Both wheels fit on the bike. They need a little pop to get past the brake shoes, but not enough to interfere with the angle of the shoes. The rear tire has a good fit all the way around except the chain stays. I tried to get my crimper in there to make an adjustment but it just doesn't fit. The tire appears to be not quite true (the rim appears fine) for some reason. I pumped it up to 120psi, and then the trueness returned. This isn't a realistic riding pressure, but it does increase the diameter of the tire enough so that it drags on the front mounting screw for the rear fender.

After returning to normal pressure, the snakiness returned to the tire. Argh!! Anyhow, careful adjustment makes sure that the tire clears both stays without issue. If the mud were to get deep, however, it could cause a problem. But that should be an unlikely scenario on this trail.

With the additional traction provided by these tires, it became clear that the front brake needed a tweek. Tightening up its cable made a big difference in the front stopping power.

I pulled out an old set of hack fenders and started to mount the front. It was a little bit tighter than I like, but not too bad. Then I tried the rear. Nope, not a winning combo. The fender is too narrow and when combined with a being a bit close to the tire, it just isn't ago.

I have a set of Velo Orange brand aluminum fenders size appropriately for these tires, and pulled out the rear. A bit of fiddling, tweeking, cutting, filing, drilling, futsing, and so on later, and I had a workable rear fender. The fender line isn't perfect, but it's not bad.

So now the hack has a plastic front fender painted white, and an unpolished aluminum rear fender. Not a fashion statement, but definitely a look that stands out.

Hopefully everything is snug and will hold together for the ride. More on this later.

It's late and I need to pack, so we'll see you around.

Wednesday, October 24, 2007

Blu Bik

It's been a busy few days with lots going on in the day job.

Still, I've started the build on the blue bike. The rider tried drop bars but decided she wants to stick with upright ones, so that's how the frame was built - giving it a bit longer top tube.

The fork is mounted using a nice Tange Levin headset (very durable and classic looking). I first grabbed a Nitto Technomic stem, not really thinking. Put it in the steerer and push down. Whoops, it won't go very far. This is a small frame with a short steerer. The bottom of steerer tubes is butted (its where most of the stress gets concentrated), and a stem won't fit through the butt. Because this steerer is so short, the butt is nearly half the tube and a tall stem like a Technomic ends up sticking way up in the air like a flag.

I have a nice SR forged stem in stock, this has more traditional proportions. Substituting it in gives just the vertical fit I'm looking for. Into this I mounted a Nitto Albatross aluminum bar. This is a big wide bar which the rider likes, even though it's more than is really necessary for the size of loads she'll be carrying.

For the seatpost, we're using a NOS Sugino fluted one bolt. It has a classic look, but is a bit easier to adjust. The lug design looks nice against the front of the post. On top of this we selected a Velo ladies saddle. I can' t say much about this except that its supposed to be comfortable to the rider. She wasn't interested in trying to break in a Brooks.

Brakes are Cane Creek cantilevers, which are now mounted on front and back. With their tension spring adjustments, these are so much easier to balance than old fashioned brakes. I toyed with the idea of using a TT style lever mounted in the end of the bars. But we're using cork grips and I don't like the idea of carving them up to make passage for the brake cable. Instead, we're using a nice old set of Sun Tour MTB brake levers. The rear is connected, using a single long casing routed through a top tube tunnel. It looks like it will wrap nicely under the handle bar tape, routed along the front side of the handlebar. This will give the rider a couple of more hand positions for longer rides, including one that gets her down out of the airstream.

As you may assume, given that the rear brake is adjusted, the wheels are mounted. The rear wheel needs a little work, based on the test fits prior to paint. It is an old 120mm hub with a new longer axle, that is re-spaced to 130mm. The spacers need to be adjusted to move the hub further toward the non-drive side of the dropouts, otherwise the derailer can't move the chain over to the small ring. As part of this, the wheel needs to be redished. So mounting and adjusting the derailers needs to await this change.

Even with all that's left to do, this is starting to look like a real bike now. Tonight I'll pull the wheels and mount fenders, then connect the front brake and finish the handlebars.

Pix soon.

Sunday, October 21, 2007

Wednesday, October 17, 2007

Late breaking news

Just heard from Duane at Chester Cycles. I'll have two frames back from paint on Sunday. Pix Soon!

New Rule!

New Rule: Never say you're getting into a rhythm. It's only tempting fate.

Since the last posting, work's dumped on me. That said, I did a little more work last night.

The front and back dropouts on the Surprise Bike are traditional Campy forged. No fancy stainless steel, just good strong dropouts. Because of the
horizontal slot on the rear, the bike will be a good candidate for a fixed conversion - which is another factor in choosing these DOs. Also, they are very classic and this will be a classic looking frame. So Campy it is.

Last night I brazed the rear DOs to the chain stays. I wan
ted to do these with brass filler. 1) there is a lot of space to fill and brass is way cheaper than silver; 2) when the seat stays are mounted, the temperature will be too low to bother the filler in the chain stay joint. The latter is important because with a high-mass DO like the Campy, much of the heat gets directed to it rather than the stay. If you over do it (without burning flux or anything) it carries this heat some distance and can cause silly problems. Therefore, it's brass on the first joint with the DO and Silver on the second.

I was low on flux and tried to pick some up at the local welding supply. I asked for Gasflux blue paste, but they didn't have any. Instead they offered me another brand, saying that all blue fluxes are the same. I should have just ordered some more from Henry James.

So, i
t turns out the new flux isn't paste. Its a dry powder. Powder may be an exaggeration. The grains are about like course sand. Anyhow, I tried mixing some with water. Nope, its just like wet sand in a bucket of water. It doesn't absorb the water no matter how much I mix it. Adding more flux doesn't help. So I set this aside. Fifteen minutes later, it was one solid mass. Hmmm??? Maybe with some heat it will melt and can be applied like paste flux.

can says to dip a heated stick of filler into the can and then braze with the fluxed filler. That doesn't sound good enough for me. So, I dipped the parts into some degreaser and then dipped them in the can. This seems to work to get a decent coating on everything.

So, the parts go together and the chain stay gets mounted DO-end up in the
vice. Then I heat the filler and dip it in flux and get to work.

ne thing I'll say about this flux, it covers the work pretty well. Sometimes, a flux can drip off of a small area of the parts being heated and that area gets a burn mark. Not so for this stuff.

Moreover, the dipping process makes it pretty easy to add flux as desired to
the hot parts. Adding a paste flux to hot parts is a no-no. As the water boils off, the temperature of the parts plummets which can affect their temper.

From here, the brazing went pretty well: plenty of penetration in the joint,
not too much excess brass for later clean up. The same went for the second chain stay. I let things cool, and over dinner the stays went into the washtub with hot (too hot for me to put my hands in) water. Usually, if no flux is burned, this (about 45 mins) is plenty of time for the flux to wash off. Not so with this stuff.

After another 30 mins of soaking, there was still a lot of excess flux hardened onto the parts. Maybe an overnight soaking would have worked better. Anyhow, I tried taking them back into the shop and chipping the flux off with a welders hammer. This often works easily after the flux has soaked for a while. Not with this stuff. It's tenacious. They should probably use it to coat the ceramic tiles on the Space Shuttle.

Using a combination of a sanding drum, the welders hammer, and an old bastard file got most of the remnants off. Then it was off to clean up brass.

been working on some ideas for a signature style for finishing off the ends of stays. I haven't seen any examples that I can remember of the style that most appeals to me - perhaps because it's a lot of work.

After slotting the stays, I file the ends in a semicircle with the slot being the equator of the semicircle. This differs from a traditional bullet end, where the profile as seen from the side & the top is similar. With my style, the tubing is not bent inwards on the end. When viewed from above, the surface of the tube forms a straight line all the way to it's end. Whereas on the bullet style, the end of the tube is hammered inward, so that even from above it looks like a semi-circle.

nce this is brazed up, I file off the excess so that this semicircle presents a square face to the world. Pictures will be clearer, so here they are (Click to Enlarge):

The reason that these are time consuming is that the further the edge of the tube is from the dropout, the more slumping/lumping the final filler will do in that space. With this style, there is a large open area between end of the stay and the DO. So, there is lots of post braze filing.

Next I have to prep the BB Shell so we can start to assemble the main triangle.

By the way, I hope everyone loves this little composition of pictures. Formating pix to page is a real PIA with blogger!

Tuesday, October 09, 2007


Not building frames full time, it can sometimes be hard to get into a rhythm of work. Lately, however, thats coming easier - leading to more progress, faster.

No pix yet, but tonight I finished the penultimate miter in the main triangle. After fitting up in the jig, I'll miter the bottom of the down tube. The rest are all done based on measurements from BikeCad and either miter templates provided by same, or scribing the profile of the inside of a lug. Anyhow, a test fit up looks good so we're almost ready to jig.

After the miters, it's important to take care of any fittings that go on the tubes. In this case, that means H2O mounts on the seat and down tubes, and a brake casing tunnel in the bottom of the top tube. The H2O mounts are pretty simple. Measure, mark, drill a pilot hole, drill out one size larger, then drill the final hole. A little clean up with a deburer, file and sandpaper and they're ready to prep for brazing.

The tunnel is a bit more work. First a mark the tube where I want to entrance and exit to be. These are approximate locations. Then its a similar process to the H2O mounts to drill holes - only this time there are more bits involved because the holes are bigger. Then its time to cut a length of brake line - this needs to be longer than the distance between the holes. Using a tube bender, I tilt each end about 45 degrees. The bends are slightly closer together than the holes in the top tube - so that the tunnel will stand away from the walls of the top tube once its brazed in place.

Next comes a large (12" I think) bastard cut round file. I use this holding it with two hands and cutting with the last 6" or less of the file. Using this I can elongate the hole s in the top tube. With a bit of back and forth to test fitting, the holes get finished off and the ends of the tunnel shortened until I can just pop it into place. Then it's time to pop it out of place so we can prep for brazing.

All three tubes spent some time in hot water with Wisk to remove any cutting oil left over from drilling holes. Then a dip in a degreaser just to make sure all is well. After this, I check that the edges of the holes are clean of burrs. Finally, the tubes get sanded with 80 grit on the inside and outside. The inside can be a bit of a trick to sand. I ended up making a flap sander for this purpose,. It's a piece of copper tube about 3/8" in diameter. One end has a slit cut in with a hacksaw. Then this was squeezed part way closed. Into the slit go pieces of sandpaper. Usually, I pull a 3" strip off a roll of 2" wide paper. This piece gets split lengthwise, and the to pieces are placed back to back - then shoved into the slit. The back to back thing makes sure that a rough side always faces the tube, no matter how the paper is put into the tube.

Anyhow, this gets chucked into my portable drill and shoved up the tube. Zing! The inside gets clean.

Then its time to flux everything up. I used Freddy's Stainless Light tonight. It was a little dry, so water was added before zapping in the microwave. The heat usually makes it easier to spread onto the tubes, and have it stick. I like the flux to have the consistency of slightly loose mashed potatoes. Done right, not to much falls off in the application process - and what stays there drys onto the tube if its allowed to sit for a little while. This makes it easier to keep the flux where I want it once the torch starts heating the tubes.

The H2O mounts are surrounded by decorative reinforcers. I try to get these straight before the flux drys. Its much easier this way than trying to poke them into position after things are heated up. Using a small tip and a soft but medium to large flame, I warm the area until the flux starts melting. As the flux gets clear, its time to add silver. I start by pointing at the actual threaded mount, and applying silver to the edge of its flange on top of the reinforcer. I point the tip of the flame into the threads (which have been fluxed) to draw this silver around the mount and down into the tube.

I don't like to apply too much filler here, because its easy to leave a glob on top of the reinforcer while pulling silver around the bottom. So I then apply filler first to the edge of one end of the reinforcer, and then to the other end. The end of the filler wire gets touched up against the edge of the reinforcer that is closet to me, while I direct the flame the other side of the reinforcer. This pulls the silver from the point of contact, under the reinforcer, to the far side. A little waving of the torch makes sure that silver is making it to the side points of the reinforcer. The same process is used at the other, end, but less filler is necessary because the first step filled more than half of the reinforcer. At this point, its important to look to see if there are gaps, globs, or other issues than can be best cleaned up now while the joint is hot. On both tubes, things looked great, so it was time to move on.

For the H2O mounts, I used a 56% Silver alloy. This gets very wet when heated and fits into small places well. For the brake casing tunnel, I use something called Filet Pro, which is a special formulation (I think high nickel content) for making silver based filets. It will wet out and I can pull it into tight places, but this takes work. On the other hand, it has a relatively wide band of temperatures at which it is plastic. And if the tubes aren't over heated, its possible to set a 3/8-1/2 inch length down and melt it a bit at a time by moving the flame around. Where the fit is tight, more heat pulls the filler through. Where a filet is needed, less heat is applied to the filler. But, where the filler needs to pull up on to the exiting tunnel, heat on the tunnel with an occasional dip onto the filler (to soften it up) pulls the edge right up. The net of this is the filler can vary from internal to filet and back as it works around the tunnel.

Don't get me wrong, file work will be necessary tomorrow to dress up the tunnel to tube transition - but it will be much easier than if there was a large filet all the way around.

Well, that's all I got done today. Well the above and soaking the flux off the tubes. So tomorrow I'll inspect my work and do any necessary filing. Then it will be time to get out the jig. The breathing holes need to be drilled in the head and seat tubes. Then test the fit up of tubes and lugs in the jig. If all is well, the tubes can be preped and fluxed and then it will be on to brazing. So there should be lots to report on in the next few days.


Monday, October 08, 2007

New Efforts

Well, two frames are being painted and should be back and built up in the next week or two, and the Path Racer is ready to go into the paint shop when I pick up the others. So its time for a new project.

I'd like to start on the bike for the winner of this years MS ride, but I'm still looking for the machinist who will turn down some lug spigots for that build. So, it's on a short hold.

As a consequence, I've started on a surprise bike. It has somewhat odd dimensions (one of the bene's of a custom build) - by memory they are 59 c-c seat tube & 57 c-c top tube. It's for someone long-legged.

This frame is using a set of lugs left over from when the Match frame shop closed. At one time they produced the Schwinn Paramount, and these are the lugs used on the Paramount.

I thought (memory is a tricky thing) that these used three main tubes all of 28.6mm diameter. Usually, either the top tube is a smaller diameter
than the rest, or the down tube is a larger diameter than the rest. Which would make these lugs even more unique.

Alas, my memory is bad. These use a standard 1Xoversize tube set (31.7 dt, 28.6 st, 28.6 tt). Which is fine, I have the necessary tubes in stock, including a stainless head tube. When polished up, a stainless head tube really stands out + it really shows off lugs with fancy shapes and these Paramount lugs are definitely fancy.

Mostly this will be a pretty straight forward design. It's designed for a 700cX28 tire, although a 30 or 32 will probably fit as well. The chain stays are a medium length at 415mm. Not short, but not long either. I would
have made them longer but with this rider, the weight tends to sit forward and I didn't want too much weight shifted to the front wheel.

Angles are 73 degrees parallel with about 51mm trail. This is a traditional
racing trail from back in the days when they used wider tires, like the ones that will be used on this bike. It should be able to handle a front bag, but probably not more then 5 lbs or so in the bag.

With the 28mm tires, fenders will fit, so it will come with mounting points.
Naturally it will be set up for two water bottles and a pump, and it's going to look something like this. Click to enlarge the picture. Two cautions: 1) don't pay too much attention to the numbers. They don't always represent the measurement that you would assume. 2) Colors haven't been selected by the client yet - these colors were just to give me something other than black & white to look at.

At the end of the day, this should be a pretty good road bike, but it will have the edges softened just a little to make it more flexible in use.

This picture gives a hint at what the head lugs are like. Note the head tube still needs a lot polishing.

See ya next time!

Wednesday, October 03, 2007

Details details

Updated Oct 6

Back to work this evening. Yesterday I built fillets up on the head tube lugs similar to the seat cluster as seen in prior posts. I'm sorry, but you'll have to wait until tomorrow for pix. Anyhow, these will show how the raw filler looks after washing off the flux. The trick is to melt the filler enough to flow and bond, but not enough to 'wet out' and want to fall with gravity or the pull of heat. So the raw filler looks kinda smooth and lumpy.

This smooth and lumpy filler needs to be made into a finished joint, which is what I was doing tonight. 3 sizes of half round files, 5 sizes of full round files (including regular and chainsaw), plus a few other odd files and various sandpaper are the tools for this project. First, any odd nodules need to be 'knocked' off. A heavy work glove on the left hand allows it to be used as a guide for the bastard (course) files used here. Then a large half round comes in handy to begin shaping the contours.

In an ideal world, filler gets laid down perfectly symmetrically on the joint. But in reality,
some of the shaping is to find the lowest common denominator. At the center, the lug spreads the farthest as does the filler, at the sides the lug is narrow and fairly flat, so filler isn't needed. In between, the size of the concave face of the filler varies infinitely - so the contour has to vary infinitely in shape from side to side. And, the shapes should be pleasing to the eye, without abrupt changes. Net net, there is a lot of eyeballing going on. Even when the shape looks good, and I've sanded it smooth, I'll find problems and have to go back to filing. This often happens 5 or 6 times on a fillet. But patience is rewarded with a nice looking fillet that will redefine how the lug looks under paint.

I'd like to say that the joints are all done. They
are for tonight. But probably, by tomorrow, I'll decide that they need a little extra work.

I also brazed the rear fender mount tonight. This is a piece of stainless tubing with water bottle mount brazed to the end. The tube has bent with a nice arc near the end. At first the
plan was to mount this to a chain stay, but tonight, the right solution presented itself. I trimmed off most of the straight portion and mitered the end to fit the seat tube. Mounting it about an inch above the point of the BB socket leads to having the fitting perfectly positioned for the fender. And the profile is of a graceful arc coming out of the seat tube, ending just above the chain stays.

Most of the flux came off with a wet rag. The rest is soaking should be gone by tomorrow.

At present, I'm thinking that this frame needs a pump peg. Several options present themselves: a) mount it just above the fender mount on the seat tube; b) mount it high on the left-side seat stay, allowing the bottom of the pump to rest on the track fork. The more that I think about it, the more I am leaning toward this latter option, but tomorrow will tell for sure.
Well I tried fitting the pump in between the left track fork and the seat cluster. Guess what? It fits in place beautifully without a peg of any kind.

That's all for tonight. Pictures tomorrow.

Monday, September 24, 2007

More Bits and Pieces

Today was good for progress on some small bits of the Path Racer. For starters, I drilled and tapped mounting holes in the track forks.

You can see how the bolt fits into the curve at the end of the stay. I'll be fabbing up some custom connectors for the fender stays that will fit in place of the bolt.

You can also see how on, the chain stay, I've cut back into the pin (nail) used to hold it in place while brazing. I'll have to work that a little differently next time.

By the way, there is a weird shadow in the picture that makes the bottom edge of the track fork look carved out. It's not, instead it forms a straight line up to the bend,and then tapers gently towards the stay.

I also tried out some new filler today (from Fred Parr) which is designed to build up fillets. The lugs have a very abrupt transition from one spigot to the next, and I
wanted to try smoothing the transition out. Here's an example from the seat lug. Take a peek at the edge of the outline - that looks nice to me.

Based on the results, I'm going to try the head lugs too.

Finally, I spent some time playing with polishing crank arms. Here are three arms in various states of polish:

Sorry for the glare, but I wanted to get in close with the camera.

That's it for tonight.


Sunday, September 23, 2007

Odds 'n Ends

Since dropping frames off for paint, I've been moving a little bit slowly in the shop and getting in a few more miles on the road. The next two weeks may be more of the same as we are going out of town next week for a long weekend to attend a wedding. Still, there are a few things to share which you may find of interest. Hence, Odds 'n Ends.

First my Stein Stronglight crank-puller arrived along with a VAR that works on modern cranks and older TAs (another unique crank thread). So the Path Racer is fully torn down and I'm working on final surface prep, but probably won't get this to the painter for 2-3 weeks.

Using the Stein/Stronglight puller, the fit is very tight. So much so, that I stopped installing it, and tried the VAR/TA. The TA is too small. It threads in nicely, but has a very loose fit, which worries me about damaging the threads. So, I reverted to the Stein/Stronglight and used a wrench to turn it in. The crank pulled out easily and it wasn't too difficult to remove the puller. But a small sliver of aluminum fell out while removing the puller, which has me a little concerned. I think I'll contact Stein before using this on my Stronglight 49 crankarms.

Having pulled the Stronglight 93 arms, I tried doing a little polishing on the right arm. Most of the face cleaned up nicely, but these was a spot that wouldn't come out (I'll have to post some pictures). My guess is that something got on the arm and caused some corrosion, leaving a small area with a pitted surface. I tried sanding the area, and then filing it lightly. But neither will get to the bottom of this pitting without changing the contour of the outer face of the crank. I guess I'll have to do some research on methods to repair this surface.

A new set of Velo-Orange fenders arrived this week (for the Path Racer). They appear very well formed, and of reasonable weight. The stays are good, as is most of the hardware. On the Velo-Orange site, it's noted that the fork attachment isn't as nice as the one that comes from Honjo. But, I find that, half the time, I need to fabricate a custom attachment anyhow, so that doesn't concern me very much.

These fenders cost about 1/2 of the comparable Honjos, but they come unfinished. The raw aluminum sheet is formed, packaged, and shipped. I did some test polishing on one of these fenders to see how hard this would be. I put a buffer in the drill press, running at its highest speed (3150 RPM). First I tried Simichrome polish. I does a nice job, to a point. The fender become shiny. But, there are some typical imperfections in the aluminum that it doesn't easily address.

Next up was some buffing compound (designed for steel). I have several, including one that is courser for cutting and one that is finer for polishing. Even the finer compound did a better job than the Simichrome at removing the surface imperfections. And, I couldn't see improvements in the finish after applying Simichrome on top of the area that had been worked with the buffing compound. My guess, is that it'll take me about an hour per fender to polish them to a satisfactory state.

I was concerned about the durability of the polished surface - would it oxidize, discolor, whatever? After doing a bit of research, it appears that a thin oxide layer quickly forms. This layer protects the underlying aluminum and generally looks fine. So I'm going to try going without any overcoat (Shellac, clear-coat, whatever).

Well, that's it for now. See ya next time.

Thursday, September 20, 2007

Tuesday - D-Day

Tuesday was my D-Day. I had an appointment to drop two frames off with Duane for paint. Only there was a problem with the Path Racer.

Stronglight cranks (until recently) used their own proprietary crank pullers. I used to have one and believed it was still around. I could remember what it looked like and approximately where to find it in my old tools and components.

So, post shakedown, I started taking the Path Racer apart. When it was time to pull the cranks, a quick search found my Stronglight puller. Only, it was a Sugino puller. URGH!!!!! So the cranks are still in place (everything else is off). Apparently Stein is the only tool maker to still offer a Stronglight puller, thankfully they do. So I ordered one and a TA old-style puller as well. But until they arrive, the bike can't go to the painter.

Therefore it was time to come up with plan B. My first-ever frame was built without a jig or alignment table. The trick to this is to put a beam across the seat tube and either the head tube or down tube. Because the seat tube doesn't generally have the same diameter as these other tubes, shims must be used to keep the beam parallel with the centerline of the frame. Then the dropouts can be located laterally from the beam with a ruler.

I apparently calculated my shims wrong and ended up with a mis-aligned rear triangle. URGH!!!! From this I went on to build my next bike nice and straight. Now my frames rarely need a tweek of anything on the alignment table. Even when a tweak is required, it is in the build sequence and not afterwards. Knock on wood that things continue this way.

About 4 months ago, I cut out the rear triangle from frame numero uno and started over. This time withe nice stainless Henry James drop-outs. Even though this largely completed the frame, it's been sitting and waiting for final bits and cleanup.

Meanwhile, I have assembled a straight leg fork for this bike, with a polished crown and polished tips.

On Sunday I began final prep on this bike so that it could go to Duane's with the Randonneur. This included: thinning lugs; mounting a brake bridge on the seat stays; cleaning lug edges; polishing the dropouts; filing and sanding excess filler around the H20 mounts, double checking alignment, honing the seat and head tubes, milling the head tube, etc.

This kept me busy through Monday evening. Tuesday if was off to Duane's. Jody was there to help with color choices. Janet had already chosen a blue in the decals for the rando bike. Duane will use a slightly darker shade to outline the fancy Prugnat lugs. It should be pretty when done.

One the other bike, Jody helped steer us to a nice sage green metallic. We're going with a fairly fine grain to the metallic, and using silver lug lines. The silver will hopefully work well off of the polished stainless and the silver lettering in the down tube decals.

So it's time to clean up the shop again, and then start work on Charlies little carbon racer.

I took one of the Nervex arms with the stripped out pedal threads and gave it a quick polish. It looks nice. Clearly a better polish job is warranted, although it won't be possible to make them like new as there are some obvious nicks in the arms. Meanwhile, I drilled it out at a 150mm length and threaded it for the pedal. Next the old end will be cut off, the shape cleaned up with a file and sandpaper and then a real polish job. I figure the two arms are going to take a couple of days by themselves.

That's it for today - see you soon.

Thursday, September 13, 2007

Path Racer Trial

As noted in the previous post, yesterday was trial day for the Path Racer. So how did it go? Generally quite well.

It's hard to over emphasize how big a 700Cx35x38 tire is. Someone who had seen the tires on the shelf, felt the need to comment on their size after seeing them on the bike. They aren't as wide as an old Schwinn, but they're tall. And with my weight on the bike, the bottom squishes way out. So much so, that I went back to the Berto chart of suggested pressure for tire size and bike/rider weight. All was correct so I pushed off.

As I've gotten older, my hands have gotten more and more sensitive while riding - to the point I can now claim to have Cyclists Palsey. The tingly/numbness can last for hours after a ride. On the black track bike (which can be seen at, I've switched to upright bars, and then had to add Oury grips to give me a cushioned perch. Otherwise, a 30 minute ride left my hands tingly for a couple of hours.

On the PR (Path Racer) shake down ride, I had some stiff Nitto drop bars. They weren't taped, but had a set of Oury's on the bottom for just-in-case. Generally, the bars are a bit slippery without tape and require a tight grip which can lead to cramping and exacerbate Cyclists Palsey. After two hours, my hands were fine. I felt like a young man again.

The PR's frame is 531 standard gauge in a 1.0/0.7/1.0 profile - so it's not light. And, while it isn't the stiffest frame I've ever ridden, it's not a slouch either. Those heavy wall tubes do their job. Combining these tubes with the tires (which I estimate at between 700 and 800 grams each) and you end up with a pretty heavy bike.

The heavy tires are felt under acceleration. The PR doesn't sprint as well as my road bike - not that I expected it to. But otherwise, it doesn't feel heavy under the saddle. The tires roll pretty well, considering that they aren't performance tires. Coasting isn't a problem, nor is maintaining a steady speed. And in some circumstances, its possible to go a little faster than usual because of the cushioning provided by the tires. So far - so good.

Handling is interesting - which isn't bad, just not quite what I expected. The tires provide a significant amount of stability. It can change direction quickly, that's not an issue, but it's as if the steering has a dampener in many circumstances. I need to decide whether to further rake the fork (and reduce trail), or leave it with the assumption that the next set of tires will be a little smaller/lighter.

Rolling down the road, the PR feels like its on rails. And this impression rises quickly with speed. Nonetheless, it's easy to turn the bars and this registers as an immediate reaction from the bike. Going around corners, the size of tire makes everything seem very planted. Lean angles that normally would feel 'at the limit' don't. But it is possible to steer to a tighter or looser line. On the other hand, leaning the bike takes more effort. Some of this may be due to the overall weight, and some to the increased centrifugal force from the big tires. The effect isn't unpleasent, just different.

Climbing was also interesting. I consider low-speed climbs to be a good test of effective bicycle geometries. Poor designs, tend to wander all over the road. The want to fall off of their line, and need a little burst of speed to be pointed back uphill. The path racer doesn't do this at all - so good. But, when standing, as the bike swings back and forth, the front wheel steers from side to side, about six inches either way from the center line. There is no loss of control - the wheel returns to straight ahead easily as the bike returns to vertical. But it doesn't want to track a perfectly straight line while standing. It may be that, with a loaded handlebar bag, this behavior will be dampened out. Time will tell.

Meanwhile, I have to mount that rear brake bridge and a couple of rear fender fittings, give it a final file and sand everything down. Tuesday evening it's scheduled to go to Duane's for some new paint. I'm still thinking about colors - so you'll just have to wait and see what we come up with.


CLB Brakes

Yesterday afternoon was spent on a two hour shake-down ride with the path racer. So far, it still only has the front brake - and that's worth some comment.

The brakes are some NOS CLBs that I picked up from Velo-Orange. They're sidepull brakes with about a 47-60mm reach. In todays terms, this is long-reach, although at the date of their manufacture (probably early 70's) they would have been considered medium-reach.

These are not the classiest brakes on the block. The finish is a greyish/silverish paint with a little sticker that says "CLB". The bolts that hold the brake shoes have an agricultural look with a hex head in which there is also a slot cut for Phillips screwdriver. Nonetheless, these brakes are full of good design and work really well.

First of all, if you looked at the pictures in the previous post, it's clear that they were designed to provide fender clearances. This is too often overlooked in the design of longer modern sidepull brakes.

In the head-on picture, the front arm has a little tab on the right. Behind that, on the rear arm, is a nylon bearing. This ensures that under heavy braking, when the arms flex, they won't bind up on each other. This should contribute to better brake feel/control - but I've never seen it on another brake.

What isn't visible in the pictures is how the back of the arms fit around the brake bolt. For the most part, the arms are in a single plane. However, at the brake bolt, the forward arm sweeps forward and across the rear arm as it reaches for the brake cable. The sweep is elegantly designed to be hollowed out with stiffening tabs that run from front to back. The rear arm also gets a hollow before coming to the portion that forms the bearing on the brake bolt. The result of this is stiff yet light arms. Good on CLB.

After looking at them closely, I'm going to have to weigh them on a digital scale just to see how they do.

All of this is for naught, of course, if the brakes don't work well. The first time the brake was applied, the bike was moving slowly and the braking seemed rather anemic. The next time, I was at speed and pulled hard on the lever. It took about a revolution of the front wheel to break through the old surface of the brake pad - then the brakes grabbed right-now! Having got the pads bedded in, they offer lots of stopping power and good modulation. Combined with the fender clearances, more of CLBs are on my shopping list.

In the mean time, the levers sport the drilled-out look - kinda cool. So I'm thinking of removing the paint from the calipers and levers, and polishing the aluminum to a nice luster. This should enhance the overall look of the Path Racer.

Tuesday, September 11, 2007

More Path Racer (w pix)

Ok ok, I promised some pictures. So we'll start with a couple of shots of the Path Racer from when I built it up to test clearances. Most of the build is temporary, but the bars, brakes and tires are part of the final product. The tires are large, and even though the brakes have arms of decent length, it's important to see that everything works together. I'm fairly pleased with how everything is coming together, so take a peek:

Here's and image showing the clearance for the tire and easily a fender: If you enlarge the image and look closely, you will see into the underside of the steerer tube.

What you can't see is that this crown has between 4
and 6 hours of filing smooth it out and give it its shape. Originally it was quite asymmetrical and stuck out considerably around the fork tubes.

Because this frame is made from older materials and with older methods, everything about it takes time. One of the nice things, however, is that the older materials are much easier to work. Lugs can be adjusted more easily, shapes can be changed more easily, and tubes can be mitered more easily.

Last night I spent several hours building a chain stay bridge (the last bit of brazing for this frame). It would have been possible to use a pre-formed bridge, but there were two reasons not to: a) With the fastback stays, so much would be chopped off of a pre-formed bridge as to seem wasteful; b) The NOS brakes on this bike expect to fit on a simple tube. There are special washers on the brake nut designed to clamp around the tube.

This shot shows how the washers fit and several other things that we'll get to soon.
Below you can see that the bridge is mitered on each end to fit against the seat stays.

And here you can see that it is bored
through to take the brake bolt. Also that a water bottle boss has been brazed into a hole on the bottom side (or in this picture the top side).

And below is a view of the boss head-on:

The boss does create a problem for us.
It is long enough to interfere with the brake bolt passing through the bridge. Here is a picture of showing the filler holding the boss from the inside of the bridge and you can see how the end of the boss has been bored off to accommodate the brake bolt.

Finally, here is a picture of the test fit of the brake to the bridge. It may look nice here, but going back to the first picture, you can see that the fender boss interferes with the brake washers. Hence the washers will need a little 'adjustment' to make sure that all goes together smoothly.
That's if for tonight, hopefully we can meet again soon!