cudak888 

Some steps forward, some steps back.

First off, I found the bit of dropout that was causing the issue. Filed it back a bit, problem is solved, to an extent.

Yes, one dropout was slightly off from the other - that was without question, However, the kinking is being caused because the pressed surface of the dropout is not perfectly flat on the outside edge (it is on the inside), causing the axle to walk downwards in the dropout when tightened. I'm hoping the fender eyelet - which, on this bike, is also the axle nut washer - is compliant enough to compensate for this. If not, I'm on the fence as to how I want to sort it out.

I really don't want to resurface the dropouts and muss up the finish. If anything, I'd rather adjust the outside edges parallel, and resurface the inside edge of the stamped drop.

Meanwhile, while servicing the GH8 Dynohub, I discovered that the axle threads had also take a beating. Wouldn't be surprised if this is because the axle nuts have no hope of being tightened square against the fender eyelet/dropout.





Did Sturmey even harden their axles? Because if not, I'd love to get some brass brazed to it as a quick fix.

Incidentally, the cotton belted rim strip is in excellent shape:



Dynohub also came out easily. The locknut and fixed bearing raceway didn't want to place nice, but I still think it's not a bad design. Better than the Bakelite version that came afterwards, that's for sure.







-Kurt

cudak888 

Time for another update. The handlebars visited an OA bath, along with the upper headset bits, which let me reinstall the fork and front fender.



For now, I have the spare 1970's front wheel from the '50 Superbe supporting it. Some beauty shots:





The Dynohub wiring is a complete mess. The 8 volt-era lamp is also wired different to the later 6-volt lamps, so that's another thing I have to research - once I get to this point.



Still haven't polished out the pie plate, or really given the chaincase that much attention. Still presents really well.



A few more hub pictures, and a look at that huge Wald kickstand:





More closeups:








More pics in next post.

-Kurt

cudak888 

Closeups, continued:



I'm missing one brake holder. If anyone has a black, wartime rod brake pad holder...I'm listening!







The Humber lamp bracket had some really fragile paint on the upper part of the bracket only - why black only on the tip, I have no idea. This was the only thing I felt obligated to touch up.





This also has me stumped - I think there's supposed to be a ferrule on the end of the cable, but it's missing - yet, the end of the cable is still soldered on. Based on a few photos I've seen, it should have one, and at least one of said photos indicates it may have a slot for removal/installation.



I don't have an answer - yet - for the V-shaped dropout shenanigans. I have a feeling I'm probably going to adjust them so the outer surface is square with the axle nuts, thus only the inner surface will have to be filed square.



-Kurt

Salubrious

The fact that its a rod brake machine means that the fender thing is moot. Prewar fenders on the non-rod-brake models were different from the post war fenders; they looked a bit more like what the rod brake machines had clear up to the end, but they were a bit more svelte and went lower over the rear of the front wheel.

cudak888 

Got it - now I know what you referring to. Didn't think of the Sports models when you mentioned it.

-Kurt

cudak888 

A lot came together in a short amount of time on this one. Mike added some brass to the axle threads today - I left the tap bottomed out on the axle to make sure it'd cut the threads square.


It's no Michaelangelo, but at least the socket will have a bit more to bite on.



The GH8 has a slightly different axle to the GH6. One side has a stepped edge which the "fixed" cone rests partially on. No idea why they simply didn't carry the threads all the way on axle and cone.





Back together. I'm surprised, but the GH8 goes together a lot easier and smoother than the pre-'51 Bakelite GH6 Dynohubs and the post-'51 metal version. The fixed cone and adjustment on the left side is a lot simpler than the later ones - where cone adjustment requires fighting the Dynohub and also cinching it at the same time. The GH8 uses a threaded lockring to secure the center armature after you've set the cones.

The open side of the armature and the threaded lugs also feel a lot sturdier than either of the GH6 variants.



The Continental front tube holds air, so I kept it. The continental EA3 tire, while decent, did have dryrot in it, so out came a new Michelin.



The finished wheel. Went for the bronze wool approach here and didn't touch the original spokes (which aren't stainless, but are Raleigh stamped).



The rim isn't perfect, but it's better than it was:



The brass stamped ring on the GH8 cleaned up really nice:



More to come...

cudak888 

Now that the wheel was built, it was time to sort out the dropout issue. To recap, this is how much engagement there is, but the dropouts are wallowed out so much that any attempt to tighten them results in the wheel kinking.



I decided on the drastic route. I took a Park Tools disc brake facer and lined it up with a quick release axle to square up the outside edge of the dropouts.

Yes, it takes paint and metal off. Yes, it's a significant modification on a bike that's survived this many years unmodified.

No, I don't feel bad about it, because the wheel would have never sit straight otherwise and the axle would have been destroyed again from loading the nut off from square (which damaged the threads in the first place).





This made perfect room for the fender eyelets (which double as washers) and everything else.



I'm happy to say this means the front wheel fits the way it should - which has probably plagued it since new.













-Kurt

Salubrious

NIce! I imagine that fork mod would solve a lot of Raleigh front end problems.

pastorbobnlnh 

Just WOW! Fantastic Kurt. Don't forget my offer on the saddle when you find a replacement for the original.

cudak888 

It's a workable solution, but I only see it working on the pressed-end dropouts used on these roadsters or the pre-1940's Sports forks which were of similar construction.

Always found it interesting that it took a lot longer for the rod brake bikes to get proper dropouts/fork ends.

Would you like to tinker with the one off the '56 DL-5? That bike is so rough that I wouldn't be concerned if a repair doesn't look absolutely original.

-Kurt

pastorbobnlnh 

Kurt, absolutely. Send some pictures to my email so I can see what is needed. Sides, top, and underside would be best.

cudak888 

If you can, shoot me an email when you get a chance; it's a busy week and I might forget!

-Kurt

mitchito

I just picked up a Humber and the rear hub says "patent applied" . It has a more conventional front wheel. Trying to figure out what year range it might be from. Any idea?

mitchito

OK, did some research and it turns out that my bike is a 1948. Very cool.

cudak888 

Do fill us in on the details.

-Kurt

pastorbobnlnh 

The saddle on the ladies Humber certainly looks to be a in nice shape! Great score.

Salubrious

The serial number is not a reliable method since the documentation was lost in a fire IIRC. Usually the date stamped on the rear hub is the best indication. The shifter is clearly 1960s but it appears to have been replaced, at least the shift cable has.

Yes. One indication of an older Humber will be the way the shift cable pulley is mounted. What is the pulley made of?

Hammalot

Just lovely!

cudak888 

Went through the usual hoops required get the rear wheel out tonight. Raising the stays made it easier to get the fender out without fuss.



Like any slack-angle, bolt-on stay Raleigh, this case opens up from the top. The current Royal Roadsters from Raleigh of Denmark finally revised this so the fender doesn't have to come off to allow the cog to clear the chaincase.

Why it took so long for Raleigh to re-engineer this, I'll never know. Chain has to be split as well; I find spare spokes ideal for keeping the chain from running away into the case.



Going to give the wheel the same cleaning as the front. Hub will also get a cleaning and rebuild.

Meanwhile, a look at the Humber maypole chainring:



New Michelin tire and fully-threaded Specialized tube are lined up for installation already. The Continental tube has no less than four snakebite punctures in it, even though the rim strip has no sharp points at all in the areas in question.

-Kurt

SirMike1983 

I have never been able to nail down an exact date for "no date code" hubs. You are correct that the "Patent Applied For" predates the "Patent" marking. They all have an old style stamping otherwise. My experience is they were probably made between 1939 and 1949, with the majority being within a year or two of WWII on either side. Some hubs from that era have date codes though. Why these have no dates, I am not sure.

The embossed shifter with external spring is rarely seen today. They are also from the era right before and right after WWII. It became apparent fairly quickly that the post-war shifters with smaller springs were a lot better, so the long-arm shifters with external springs weren't really around for very long, when you remove the WWII years. Sturmey touted these early bar shifters when they first came out before WWII. It is easier today to find a good 1930s quadrant shifter than a nice and complete external spring bar shifter.

I think your estimate of pre-1948 is a good one. It looks like it's a really nice, unique project. It's not the run-of-the-mill 1970s Raleigh Sports.

The axles on the dynohubs up front are a chronic problem. Bending, thread bashing, and thread loss all were issues going back to early dynohubs. At least that is my experience with them. I don't think they ever fixed it. Even the much later GH6 hubs I've worked with pretty regularly turn up with bashed threads or bends. I heard someone (RHM maybe?) mention there was some kind of modern replacement for the axles, but I never found what those would be. I wish I had. I have a couple GH6 hubs that need axles.

cudak888 

Different production lines, perhaps? Shift managers with different ideas of how to mark the hubs? Issues with the tooling stamping out the date? Could be any one of these - or simple re-use of pre-war hub shells sitting about.

The GC3 shifter does have one benefit over the later ones, and that's that you can gradually work the spring out if you need to bend it for a bit more tension. Can't do that with the later GC2 3-or-4-speed versions.

These axles are soft steel - not hardened. It's not surprising their threads are weak, though in this case, it might work in the favor of the nuts surviving - consider how there isn't a single left-hand Sturmey rear hub nut out there that has survived mechanics tightening them to the hardened steel axles.

As for replacements, any 3/8" x 26tpi axle could be used as a replacement, but axle centering will probably be a pain without the unthreaded area in the middle to bottom out the first cone upon. Loctite 603 could potentially perma-attach the cone onto the axle. The GH8 is slightly different than the pre- or post-'51 GH6, so I couldn't look to a later hub for spares. @rhm might have found a modern axle with just enough threading to be mostly compatible; still, probably easier to go with the Loctite 603 method.

Side note:

It came to my attention tonight that the hub has the conventional Sturmey lockwasher, but no locknut on the drive side. It has it on the non-drive side. Both sides had the typical eyelet-style tensioners for rear-facing ends. Question is, did Sturmey delete the nut from factory, or is this the oversight ofa mechanic from BITD?






-Kurt

cudak888 

Making progress. Dunked and cleaned the AW bits (not that they even needed cleaning - everything looked perfect), gave everything a good re-lube, and put it back together:





Hub isn't so clean, but that's about to be rectified:



...and polished up and back together:








There's only so much I can clean off the rim, but as you can see, the brake blocks have worn the chrome away down to either the steel or the nickel anyway. They'll probably clean a bit more off when this thing is finally rolling.



I also discovered that the locknut on the left side of the Sturmey AW had cracked, which led me to reconsider the apparent omission of a locknut - only the lock washer - on the drive side of this hub. Unlike a Sports, this bike has axle tensioners (after all, they're rear-facing), so I was considering whether this was an intentional difference at the factory for the rod-brakers - but the cracked LH locknut suggests to me that the right hand locknut probably failed in a similar manner, and might have been removed by whoever installed the Continental tires at some point in the bike's life.

Still, I sent a message to @SirMike1983 who seems to have seen more than a few of these early hubs. Tagging @nlerner just in case too.

-Kurt

cudak888 

A few steps forward...and all of them right back again. Let me start from where I left off:

Kickstand joined the oxalic acid bath a few days ago. I also picked up some Newbaum's cloth tape to line the inside, padding the contact points against the frame.





Kickstand installed. Believe it or not, the wartime left hand nut didn't strip.





Pedals out of the OA bath:



Looks good, right?











But that's not where this ends.

cudak888 

So, even though the AW isn't hooked up, I decided to give it a quick test. Besides the pedals being wonky (no surprise there - either the shafts are bent or I didn't clock the outside plates spot-on), I realized it was doglegging bad. Bad enough that you could see it under you.

That's when I actually bothered to eyeball the back end. Apparently, I'd completely overlooked the one thing I've done on every Raleigh product to come through the shop as of recent...check and cold-set the rear triangle. How I managed to miss something this obvious, I don't know.

Mind, the rear fender makes it look a lot worse than it is, but it's still bad, nonetheless.



I took some basic (but not really accurate) measurements with the frame alignment gauge, and the rear is offset to the right by no less than a half an inch (!) to the right.

Put simply, I've got to take everything off again - this time including the chaincase and the bottom bracket. The case also has a loose rivnut in it, so that ought to be fun to sort out as well (not). I'm just mad at myself for not immediately checking this in the first place.

The shop is also an absolute mess at the moment due to some plumbing work, so this isn't getting torn apart until next week, at the earliest.

-Kurt

pastorbobnlnh