On January 2nd of this year, Jeff asked us all what our project car resolutions were for 2019. I said, “Finish the ’74 Honda CL125S by fall.” I’m happy to report that I did keep that resolution. About a month ago, I finished the bike. As in, for-real finished it. Running, riding, streetable, not-leaking-anything, no-loose-wires finished it.

You might already be familiar with this bike, since I have written about it for the Hooniverse Project Car SOTU five years in a row. Yes, it’s true: I had been working on this bike for half the history of this website. The irony of it is that this was never intended to be a full-on project vehicle. When I bought it, I intended to replace the regular wear items, perhaps a couple of gaskets, and have a fun neighborhood runner while I worked on my pre-existing, glacially progressing, real project, Bultakenstein.

I bought this particular bike in a fit of romantic nostalgia, because it is the same model as my first bike in high school. It’s a CL125S—a “street scrambler” (that is, faux off-road) derivative of Honda’s ubiquitous CB125 street bike. Unlike the street version, which was imported 1973–85, the CL was only sold two years, and had a fair number of unique components.

This animation shows the differences in the tank, bars, seat, front fender, and muffler between the CB and CL variants.

Unfortunately, my pre-purchase inspection of the bike failed to expose the full extent of its decrepitude. I bought the bike in Lawrence, Kansas, home of Kansas University. It had been owned and flogged by a long progression of students who, being college students, wrecked it, neglected it, ignored scheduled maintenance, and left it out in the elements. The engine was in remarkably good running order, but once I got it home and started tearing into it, many of the chassis parts turned out to be practically unsalvageable. Usable replacements for the CL-specific items are rare, expensive things. So, I made a custom exhaust (one of the most surprisingly easy and straightforward parts of this project). The inside of the rusty tank was coated with sealant. The carb was rebuilt. New tires, chain, sprockets, grips…

And told myself I had a runner. But I didn’t.

My one attempt to commute to work on the little Honda resulted in a flatbed ride home with a mysterious fueling or ignition issue. On the one excursion I took accompanied by my wife on her bike, she reported that the cheap, Chinese turn signals I’d installed were barely visible in bright sunlight. It leaked from the transmission. It leaked from the forks. The carb refused to idle consistently, the supposedly “sealed” tank was still clogging the fuel filter, and the electrical system was a frayed, cobbled-up disaster.

Meanwhile, I had spent a great deal more on this bike than I ever intended. The old Honda parts I bought off Ebay were usually only marginally better than the discarded ones they were intended to replace. One good example is the speedometer, which was broken. The CL125S had a speedo unique to the bike, so I searched and found one “in good condition” for more than I wanted to spend. It lasted 2 miles before its innards flew apart. So I bought a used CB125S speedometer. It was cosmetically different, but basically the same. It failed before I even got out of the driveway. So I got smart and bought a cheap, no-name, knockoff Honda speedo, which ended up not fitting the stock bracket at all. I then bought a cheap, digital speedo-tach unit, but it was wildly inaccurate, the LCD display washed out in sunlight, and looked positively horrid on the bike. I eventually bought a generic “chopper style” mini speedo unit and a CB350 idiot light console to house the indicators that had previously lived inside the speedo housing.

Did you notice a theme there? —Cheap. Cheap is the goal, but cheap is bad. Used components off a 45-year-old bike are still, unavoidably, 45 years old. Universal aftermarket components are readily available, but whether they’ll adapt properly and work right is always a crap shoot. For example, I spent over $200 on speedometer replacements. I ended up with a $50 speedometer. If I had guessed right the first time, I could have saved a bunch of money. But there’s the rub: you often don’t know what will be the best option until you try it. And trying to conserve money often leads to poor initial choices. I spent $157 and a whole bunch of time trying to restore the original fuel tank. It eventually ended up in the trash. I then spent $125 for a replacement tank designed for a Honda CD70. I spent another $147 for fittings, fuel valve, an a fuel filter that would allow it to work on my bike. And the only reason I was able to make that work is that I have a lathe and was able to custom-make a special threaded brass fuel outlet that would clear the cylinder head. Overall, I ended up spending nearly $900 on parts that ended up broken, couldn’t be adapted to my bike, or were simply too crappy and lame to use. I bought a universal carburetor. It required a different air filter, a different mounting flange, and a custom-fabricated adapter to bolt it to the head. To solve the poor lighting issue, I upgraded all the lights with LED units. That required upgrading the 6-volt system to 12 volts and a custom wiring diagram. The larger carb and a larger AGM 12-volt battery required me to fabricate a new battery box. The new tank fit properly, but required me to fabricate brackets to relocate the seat 2 inches further rearward.

Eventually, I did get the bike running. Running really well, in fact, with a larger carb, free-flowing muffler, clean fuel system, fully functioning instruments, and bright, LED lighting equipment. I even added some tank badges and a spiffy tank stripe, so it looked good. I took it out and rode it, and it did bring back lots of vivid memories of the summer of 1980. And once I had enjoyed about 50 miles of reminiscence, I decided that I had accomplished all that I’d intended for the bike; I couldn’t bear to have something else go wrong and descend into the rabbit-hole of unintended, unavoidable repairs all over again. I listed it on Craigslist for $1450 — its realistic market value. A day later, I helped load it into its new owner’s pickup. The guy is a car racer and accomplished wrench who has is own CL350 scrambler of similar vintage. It went to a good home.

I maintained careful records of my spending, money-wise if not time wise. The total cost, including the bike, was $3673.23. Of that, only $2315.78 worth of stuff actually ended up being a part of the bike when I sold it.

Here’s the final tally:

Purchase price $ 750.00
Sales tax, property tax, title fees, inspection, licensing, scheduled maint. $ 267.74
Exhaust $ 134.01
Speedometer $ 50.79
General chassis repair parts $ 490.21
Money spent restoring old tank that was eventually trashed $ 156.79
Parts purchased that ended up broken, unsuitable, or not used $ 735.93
Subtotal $ 2585.47
12V/LED/AGM electrical upgrade $ 380.50
Fuel tank & fittings $ 271.91
Carburetor $ 238.36
Subtotal $ 890.77
Parts for aborted front brake and stator upgrades $ 196.99
Total Expenditure $ 3673.23
Sale price of bike $ (1450.00)
Estimated market value of serviceable parts on hand $ (350.00)
Net loss $ 1873.23
Cost per day over 1,977 days $ 0.95


Did I lose my shirt? You can clearly see that I did. But look at that last number. For less than a buck a day, I had some great fabrication practice, lots of mental exercise thinking up solutions to unforeseen issues, and even though I didn’t really intend or desire for this to be a 5-year-plus project, a lot of rewarding moments along the way. When it was all over, the cruises I took around town, with the bike starting on the first kick and purring perfectly, were totally, absolutely, 100% worth the investment.