Calculating new targeting solution.

Since we've hit a snag with the delivery of some essential items for the Sportster's overhaul, I decided to put that bike back in storage while we work on other projects.

To begin with, we have a fun and interesting task to complete, as we attempt to build two Gilera 150 5V Autostrada, out of what is basically one complete bike and two halves in various crates.

This exercise has already yielded some surprises: we are learning all kinds of details about the MkI and MkII versions of these small lightweight bikes. By the time we're done, we should have a complete MkI (in crates) and a MkII "special" of some kind, ready to ride. We have to donate some components from the "red" bike to the MkI pile, and use other MkII parts that are currently in the crates to replace what the "red" bike will lose. Sounds completely naff, but there is some logic behind all this. 

As we plot our moves and try to figure out what will go where, the weather quickly takes a turn for the worse so I run outside to look for myself.

Thunder cracks violently overhead, amidst the chaos of swirling lead-colored clouds, a flurry of dead leaves blown across the scene by howling winds, and squawking pitch-black ravens. The dark outline of the cypress trees in the background makes the Monolith appear as a black cathedral. More thunder, lightning, and an ominous rumble from deep underground; the limestone slabs glide open and the Rising Star emerges.

I guess it's time I took care of this too...

1200S: service time, part 9

This is where I'm going to be labelled a nerd, no doubt, but I'll stick to my guns because I think there is a small but important difference. The service manual specifies that the clearance between the shift pawl and the shift drum pins should be set by using the shank of a #32 drill bit. Simple enough, right? It's a convenient, and real-world way of finding something cheap and easily accessible through any hardware store, that has the correct dimension to set an important clearance in the shifting mechanism, without resorting to a specially made tool.

Except that a #32 bit is unheard of in Italy, and I would guess in Europe, where we use the metric system, including for drill bits. So, aside from the fact that I had a friend bring me a couple of drill bits from the US a few years ago, the fact is that nobody in this country sets the clearance with a #32 bit. If anyone bothers to do it at all, they probably do it guessing and eyeballing what the clearance should be, and the very very few who bother to search for a #32 drill bit might find some charts with the decimal equivalent dimension and determine that "it's close enough to 3mm". Yes, that is true, it is pretty close to 3mm but it isn't 3mm, so setting the clearance using a 3mm drill bit is still wrong. It's pretty close to being right, but it ain't right.

Incidentally, I did check the shift pawl clearance "as found" and it was just a little bit loose, nothing crazy though possibly enough to at least contribute to the problem. Obviously this is a great opportunity to inspect everything I can reasonably look at while I'm in here: it always pays to spend a little extra time to look at and feel as many moving parts as possible, you'll either spot a problem that you can fix, or have peace of mind that all is well down there.

In the next photo, you can see the end of the shift drum minus the detent plate: this makes it easier to see the shift drum pins and the shifter pawl, to the left. 

The clearance to be set with the #32 drill bit is between the pins and the pawl. You have to shift into third gear, which aligns a hole in the detent plate with the spot between pins (yellow arrow) and pawl (green arrow), it's a little hard to see in the next photo but it's shown by the double-ended red arrow:

After cleaning the shift mechanism and applying fresh transmission oil, I installed it (as well as the detent plate and a new retaining clip for that) and tightened the new nyloc nuts, by hand only at first. The gearbox must be shifted to third gear. At this point you insert the #32 drill bit's blunt end into the hole and press the pawl down to eliminate the gap (but not so hard that you shift into another gear).

The absolutely important part of the process is to maintain pressure while you tighten the nylocs to the same torque spec of 11N⋅m. I am confident that this is now set correctly on my motorcycle. Of course I also installed a new retaining clip when performing the adjustment:

Well whaddya know, we've reached another natural stopping point. Let's get ready to refit the primary transmission, shall we?

TO BE CONTINUED...

1200S: service time, part 8

With a parts list now ready, I headed over to the nearest Harley-Davidson store (not the one where I bought the bike in 2015) to ask for a quote: they said they would get back to me but, disappointingly, never did. I asked around a couple of dealers in Europe and the US but none seemed too interested in my business. In the end it was Mr White who came through, went out of his way to help me source all needed parts and was an overall ace about it. W&W in Germany was the only outfit where I could reasonably purchase OEM gears, and - crucially - the updated parts that were developed to remedy the 1st gear problem. As you can see in the side-by-side below, the update consists in the lobes being machined with virtually no ramps when compared with the original 2001 component. Everything is straighter and this makes for a more positive engagement.

Above: countershaft 1st gear; original/damaged one on the left, updated part on the right.

A close-up view reveals the main difference between the two:

Above: countershaft 3rd gear; original/damaged one on the left, updated part on the right.

Below: the updated gear is in place on the countershaft:

All these parts were not only hard to find, but very expensive too so, thinking back about the time I originally bought the bike and asked the shitty dealer to fix the gearbox, it's obvious that they knew they couldn't fix it without losing serious money on the sale. Absolutely terrible people.

Reassembly begins with cleaning every part thoroughly and laying everything on the bench. I won't go through each gear, thrust washer, needle bearing and retaining ring - and if you're considering doing this type of maintenance on your Sportster, you shouldn't use this as a reference: get a manual! But all those parts go back on the two shafts in the reverse order they were taken off. The important thing to note here is that the needle bearings (one is shown below), thrust washers and, perhaps most crucial of all, the retaining rings all get replaced with new parts. Everything was lubricated before assembly and tested for correct operation.

Once the gear cluster was ready, I added the shift forks and slid the shift drum back in, then secured it to the trapdoor with its retaining double plate (together with the detent positive stop mechanism on the outside). The forks' guide pins go back, and each fork gets a brand new cotter pin.

This is it as far as reassembly goes, then comes testing to see that all gears engage as they should, so it's over to the power flow diagram once again. I ran through it a few times and found that everything shifts and engages smoothly going both up and down the sequence (1-N-2-3-4-5). 

The complete gear assembly can now go back where it belongs, you have to be careful that all three shafts engage their respective needle bearings, and under no circumstance should you hammer it into place: if it doesn't want to go, it means you've missed something: go back and check again, easy does it.

The trapdoor is bolted back up, each fastener should have a sparing amount of blue Loctite and the correct torque setting (18-23N⋅m) applied.

I tested once again afterwards to see if all gears would still engage, and everything seems to work as expected.

I can show you one more interesting detail here, looking towards the back of the transmission case, with the gearbox in neutral, you can see the pin on the end of the shift drum pressing against the ball bearing to activate the instrument light:

And with a gear engaged to show the difference:

I had received warnings of impending catastrophe, that I was doomed to fail were I to attempt such a complicated, difficult and risky job.

I was told only truly experienced mechanics could grasp all the finer points and have a chance at succeeding here. I have a sneaking suspicion it was a sly attempt at charging me for labor, with zero guarantee the job would actually get done. Sorry guys, I'd rather take years to get around it, than let someone else do it for me.

There is something very important to know in case you're considering buying a Sportster: 2004 was a year of big changes for the model, and one thing many people don't know, is that H-D removed the gearbox trapdoor altogether. This means that if for whatever reason you need to get at the gearbox, the entire motor has to be disassembled, the cases split, and then you're looking at at full rebuild after you're done. Stay away from anything other than a 1995 to 2003 Sportster, those are the good ones.

Now, this is a good stopping point, we'll pick up next time with a critical adjustment in the shift mechanism.

TO BE CONTINUED...

1200S: service time, part 7

With the gear cluster safely on the bench, we can now proceed with inspection and disassembly.

In every photo I've seen, whether it's someone else documenting their transmission maintenance, or even the Clymer manual, the cotter pins are installed like this:

But I actually found them like this:

which looks much more intentional, much more secure and I'm almost certain this is how it was done at the factory in 2001, so the new ones will be going back in exactly like this.

With the help of the transmission power flow diagram from the factory manual, I checked operation of all gears going up and down the sequence, and only really found engagement of second gear to be less than optimal. Everything else meshes and couples fully.

The fork pins come out using a little bit of compressed air over each opening, the pressure differential effectively pulls the pins out, no dark magic required.

The detent arm pivots on a bolt that, cleverly, also does double duty as the mounting point for the two lockplates that secure the shift drum to the trap door. Notice also the three fork pins and respective cotter pins at the top-right of the image. The cotter pins must always be replaced.

Once removed, the shift drum can slide out and be set aside, then each fork can be lifted out.

It's over to the mainshaft and countershaft, removal of all the components is pretty straightforward and consists of removing a series of retaining snap rings; easy and fast, provided you have the right tool:

I only reached the damaged gears on the countershaft (1st and 3rd), there is no need to remove anything else.

Here's the deal, I was expecting, and to some degree hoping, to find something clearly and badly damaged in here. And while there are plenty of indications of wear on various parts (circled in red in the photos above), none of it seems to be enough to cause such a big problem. So yes, the ramps on the countershaft's first gear are a little bit beat up, definitely, as are the corresponding dogs on third gear, and I found a few very little burrs on practically all the dogs throughout the gearbox. The forks show very minor signs of wear but truly nothing concerning. Although it doesn't really make sense to me, I must accept that it would all add up to cause first gear to "pop out".

At least part of the issue in all of this seems to come from the fact that H-D does not make shims in different thicknesses for the gears, relying instead on a one-size-fits-all spacer that obviously can't always be right. Checking the transmission on the bench, I can see how the forks might be unable to push gears all the way, making full engagement difficult. Obviously this just invites incorrect wear issues, and a less than sympathetic rider will quickly cause damage as was the case here.

Another thing to check are the tracks in the shift drum itself, as these guide the forks and can severely impact proper function of the whole transmission if damaged. I checked all three pins in their respective tracks to see if they could run smoothly and they do precisely that. 

TO BE CONTINUED...

1200S: service time, part 6

Off comes the shift-shaft with the pawl still mounted to it (no need to disassemble here). The inboard end of the shaft rides in a floating brass bush (second photo below, top-left) that seems to me like a very generous clearance fit into the crankcase: no signs of wear. The two 9/16″ nyloc nuts really ought to be replaced during an overhaul like this. Re-using these because they look fine would be a false economy here.

Old ones on the left, new replacements on the right:

There is a thin washer in here (shown by the arrow) and it's important not to lose it:

The five ½″ retaining bolts come out without trouble, the whole gearbox now comes out as a unit. Beautiful.

Pay attention so you don't lose the two locating dowels for the trapdoor:

Above, you can also see mainshaft 5th gear, which is captive with the front belt pulley on the other side.  

I noticed this pin at the end of the shift drum (see next photo) and at first didn't recognize it, then a quick look inside the casing, which is immaculate by the way, and the penny dropped: this pushes against the ball bearing situated right above the shift drum needle bearing, and when the transmission is in neutral, activates the switch for the neutral light in the instruments.

Above: the pristine inside of the gearbox case, with the ball bearing for the neutral switch shown by the green arrow. Notice also the needle bearings for the shift drum, the countershaft and the mainshaft (inside fifth gear), and a breather tube at the top-right of the image.

TO BE CONTINUED...