Sunday, March 26, 2017

The Beast within the Beauty; CR93 Replica tales of woe

This was the first of two bikes that “came back” to me after ownership in the 1990s. It showed up at the recent Mecum auctions in Las Vegas, quite by surprise and completely unannounced. An impulse buy had it in my possession after spirited bidding, but I had no idea of what I was really buying until after the deal was done.
In the first go-round, the bike came to me from Utah, brought to my home in San Clemente in the back of a Honda Civic... really! We had to reassemble some of it, but it did turn out to be a pretty cool piece, but was somewhat unfinished in many ways. At that time the whole bike was CB160-based. While I owned it, a CL175 5-speed “sloper” engine was sourced from somewhere and eventually was shipped up to a new owner in San Francisco who saw it at an AMA-AHRMA event that I attended.
The bike slipped my mind for many years, but at some point I was messaged about it awhile back, asking if it had been mine at one time. Life goes on and the bike disappeared from my radar until 2017 at the auction. The bike showed up the night before the auction day and was not included in the catalog or any other preview information. It was just wheeled up to the auction block without fanfare, but I knew that it had to have been the bike I sold so long ago.

It had been announced at the auction as a CR93 replica with a CL175 5-speed engine, so obviously quite a lot of work had been done to it over the past 20-something years. After the auction, I returned to the holding area to look at what I had spent a large amount of dollars on and was pleasantly surprised to see the overall quality and workmanship of the end result. The early-version CL175 engine was mostly sand-cast and had been assembled pretty much with a rudimentary cleaning of the outer surfaces, but nothing that indicated an attempt at a 100 point restoration.

The chassis was powdercoated; nice polished AKRONT rims were mounted up on refurbished hubs and brake plates. Modified CB77 footplates carried some solid alloy footpegs in the most rearward position on the mounts. I discovered that I could barely manage the footpeg positions now, due to having both knees replaced which have more limited flexion than before. Otherwise, the overall fit and finish seemed to be well executed, but I couldn’t run the engine inside the facility so the engine condition was unknown. Based upon the overall look of things, I surmised that the bike had been fully finished, tuned and ready to run at the turn of a key.

While a stock CL175K0 bike engine was not originally fitted with an electric-starter, the CB160 starter parts will retrofit into the CL175 cases. So, thankfully, this bike was fully outfitted with an electric starter and even had an electronic regulator/rectifier installed. The little turn signals were replicas of Honda’s original CB72 lamp assemblies from the early 1960s, as used in domestic markets.

I hustled the bikes back to San Diego, off-loaded the CR93 and another auction purchase and then fired up the little Honda for the first time. It started easily and sounded throaty with its Thailand-made reverse cone mufflers. All seemed well until I took it out for an extended run around the neighborhood. As I throttled it up under full-power the engine note began sounding scattered and it felt like it was going lean in the mid-range. When I added some choke to the engine at speed, it picked up momentarily so I felt I was on the right track.

Back at home, I discovered a poorly repaired float on the right carburetor and a bit of an oil leak at the tach drive seal. A new float was fitted along with a new tach drive seal and the testing resumed. The more it was driven the worse it ran, choke or no choke. I fitted up some slightly larger main jets to the carburetors, which seemed to help in the first few minutes of riding, but then the bike went lame and wanted to run on just one cylinder.

Limping back to the house, the carbs were checked and the right side float bowl was nearly empty while the left side was at about the normal level. The petcock for the alloy tank only has one outlet, which tees into two fuel lines, one for each carburetor. It didn’t make sense that only one carb was getting fuel when the right side bowl was removed and the petcock turned ON, gasoline ran out freely when the float was in the down position. The float valve was a replacement and found to be slightly longer in the threaded section than the original. The float levels were carefully checked again, but that problem persisted somewhat until I refilled the tank again, then it seemed to have cured itself of that symptom.

What hadn’t improved was the poor running after a few minutes of operation. Chasing “fuel problems” for another hour lead to continuous dead-ends; nothing else of note was to be found in that system that could be causing the problems. After too much electric starting sessions, the battery finally gave up, dropping from 12+volts down to 7 or so. The battery was charged overnight and replaced back in the bike the following day, but even though it had full power, the engine continued to sputter and fail after a few minutes of riding.

Further checking of all the basics revealed that the right cylinder had fuel, spark, compression and air to the cylinder but it was not running on that side at all! A spark check revealed a somewhat thread and uneven spark, compared to the opposite side plug. As a last desperation effort, I swapped the plug wires from side to side and suddenly the opposite side cylinder went dead and the previously dead cylinder woke up!

With the fuel tank and seat removed, the shiny black coil was accessible for testing. At first I blamed the condenser as the points were arcing more than normal. Changing the condenser did nothing for the problem, so onto the coil for the final test and therein lay the problem…

Honda 12v coils which are used with points and condenser ignitions have about 4.5 ohm readings on the coil primary windings. This shiny aftermarket piece read out at 1.5 ohms! Coils with these readings are generally used in conjunction with electronic ignition systems, not with points and condensers. The low impedance of the coil caused it to overheat after a few minutes of running, which then compromised the secondary output. These are double-ended coils, used on either 360 degree crankshaft twins or on two of the four cylinders of a Honda four of that era.

A trip to the local salvage yard eventually yielded a suitable 12v coil, with a 5 ohm readout on the primary side. With a few wiring modifications, the coil was installed and the engine fired up. It instantly sounded better, heated up both exhaust pipes immediately and took throttle with a crisp exhaust note exiting the little chromed exhaust mufflers. Success at last!

Looking back, you can see that this must have been one of the reasons why the bike was put on the market. It never could have run properly with the coil that was installed during the bike build. With all the good work that was done to the bike in the past, just this one poorly-chosen part pretty much ruined the CR93 Replica party until now. Now it runs as good as it looks!

Sunday, March 19, 2017

Miss Interpretation, the Type 2 CB77

Many boat owners apply some kind of name to their favorite watercraft like “Miss Behaving” etc. but I have seldom named a vintage Honda motorcycle before this one. “Miss Interpretation” seems appropriate in that I built the bike over 25 years ago, with the limited knowledge that I had then. Since then, I have acquired a lot of valuable bike/engine building experience and general work on the 250-305 twins. I’ve got to say that this one fooled me into doing way too much work, only to discover that the “problem” was external, not internal for the most part.

Back story:
I reacquired a 1962 Type 2 CB77 “cafĂ©” project bike that I had assembled from a rolling chassis many years ago. I was able to acquire a number of CYB72 “racing” parts for the bike build and then sold it, after completion, to the man whose widow recently created an opportunity for me to buy it back again.

In the revival process, I was reminded that there was an odd clicking/knocking noise in the engine when it was running and leaned over on the side stand. For years I thought that I had over-shimmed the gearbox with offset cotters and that the gear dogs were just touching when the gearbox slack allowed the gears to touch when the bike was leaned over.

I was reacquainted with the sound again and thought that it probably should be investigated sooner than later. The bike was a little hard to start (kickstarter only on this bike) and the performance seemed a bit “off” and idling was kind of random and uneven. After cleaning the carbs and rinsing out the fuel tank and petcock, I turned my attention to the ignition timing, which was still the typical points and condenser OEM system setup. After an initial static check, using 12v timing light, I checked the timing with the engine running, using a dynamic automotive timing light. What I noticed was that the timing marks were swinging back and forth about 8 to 10 degrees at idle. The “expert” in me jumped to the conclusion that the engine noises and the random ignition timing were somehow related. The first thought that came to me was that the camsprocket rivets had loosened up and that was causing both engine valve timing, as well as ignition timing problems.

I slipped the engine out of the chassis and left it on the bike lift while I popped the top cover off and exposed the camshaft and camsprocket.  After removing the camsprocket from the camshaft halves, I was surprised to find that the rivets were NOT loose and that the camsprocket seemed to be perfectly serviceable as-is. Undaunted, I dug through the old parts piles and found a Dream camsprocket that was in good shape, but did show signs of loosening rivets. I had the camsprocket tack welded in a few places by my good friend, Rob North, then put it back together once again and reinstalled the engine.

I kind of “forgot” about the lower end clicking noises until the engine was already back in the chassis, as I was so focused upon the “timing” issue to be solved. Well, of course, nothing really changed at all once it was started back up again. The same noises and same symptoms remained as before, so it became obvious that the engine needed to come back out and be torn down completely to ascertain what the noises were all about.

My dim memories of putting the bike together seemed to include installing a 337cc big bore kit in this bike, but as it was being disassembled it was clear that the pistons were stock OEM parts, but of two sizes ( the left side was .75mm and the right side was 1.00mm oversized). Apparently, I was short on spare pistons/rings and the machinist determined that the bores would clean up okay using this combination of parts. The actual weight difference is pretty tiny between a .75 and 1.00 piston and ring set, so the effect on engine balancing was negligible, at least in my mind at that time.

Everything inside looked pretty swell after mostly resting for the past 25 years, but I was shocked to discover that I had made one of those “rookie mistakes,” in that I had installed the crankshaft into the cases and failed to properly index the rotor side main bearing onto the little stepped locating pin. If the bearing isn’t indexed properly, the oil-feed hole, which comes from the crankcases, won’t properly lubricate the main bearing on that side.  

While the other three main bearings have larger 7mm locating pins, the small pin on the right side is easily overlooked while assembling the cases and can be pushed into the engine case bearing boss, cracking a chunk of aluminum loose as the pin is pressed beyond the bore depth of the hole in the case. Yep! That’s what I did back then….
So, the pin was extracted, reset with Locktite and then the broken out piece was replaced with JB-Weld and let to harden overnight. I have had to do this to quite a number of engine cases, always with success, but the trick is to MAKE SURE that the pin end engages with the bearing locating hole properly.

 Before the cases were reassembled, the transmission gears/shafts were all inspected for excess play or signs of unusual contact. Re-examining the transmission gears, it was clear that the 2nd gear had been replaced with a new one when the engine was last re-assembled; however it wasn’t quite the right part for that application. Early transmission gear dogs have straight sides where they engage with other gear dogs that have straight sides. Despite the early model engine, the transmission gears were mostly the later style with back-cut gear dogs, which engage each other at a slight diagonal. This angular engagement pulls the adjoining gear dog into the first one, reducing the tendency to disengage under hard power requirements.

From a large pile of transmission parts, I found a good used 2nd gear that had matching gear dog shapes and installed that one on the shaft. I rechecked the gear cotters to see just what I had installed so long ago and discovered that one of the offset cotters was broken in half! In the end, each and every gear was inspected and the engagement checked for proper overlap. Once I was satisfied that the transmission was in good shape and ready to go, I slipped an endless camchain over the crankshaft, lined up ALL the main bearings properly and reassembled the crankcase halves with a little liquid sealer.

During the teardown, it was noted that the oil filter drive chain was quite stretched as was the primary chain. Having just made up a new oil filter chain from #25 chain and a master link for a Dream engine, I replicated the process and made a new one for the CB77 engine. A primary chain, picked from a handful of used ones, was selected to take on clutch power duty, until some new aftermarket ones arrive from the UK.

When the bike was first fired up, the clutch was “stuck,” so the cover was pulled and the plates peeled apart and cleaned. The clutch springs seemed to be short and stiff, but I reinstalled them at the time. When the bike was operated, the clutch pull was kind of excessive and the transmission wouldn’t select neutral very easily. I somewhat expected that as the inner clutch hub was machined for a 6-plate clutch pack and the plates were from a later 5-plate setup. There are a half-dozen different clutch hubs for all of the various 250-305 engines, but the last one is the best. A decent candidate was sourced from an eBay seller and the clutch pack was assembled with the proper A plate, secured by a thin wire retainer ring, then one set of plates installed, once again held in place with a retainer wire, as designed. I selected some lighter clutch springs and the whole assembly was reassembled with the upgraded parts, which improved the clutch function greatly.

I had noticed some wrist pin play noises when the engine was running under part-throttle and sure enough that was what was found when the pistons were removed from the rods. I did have a spare .004” oversized wrist pin (Honda option), but after a lengthy process of reaming and honing the rod end, the fit was only slightly better than before. The pin holes tend to wear out of round, so it is difficult to make a nice clean oversized hole with anything other than a precision drill bit or mill end to bore the hole perfectly round at the slight oversize diameter.

With the engine back in the chassis again there were other problems related to fuel supply. Some of the old tank liner had melted into the gasoline that was used after the first tank flush to get out the big pieces and even following an acid bath to improve tank cleanliness. The contaminated fuel caused the carburetor slides to get gummy and stick, which required more cleaning and examination of the whole fuel system.

After all this work to rebuild the engine, scour the fuel system and assemble the bike once more, the engine was still difficult to start and the ignition timing was still pulsing back and forth…. After installing a new set of OEM ND points, the ignition timing was checked statically with a 12v test light to confirm that the point cam lobe was kicking the points open at the F mark alignment. When I turned the engine over ONE turn, to check the timing on the opposite point cam lobe, the points kicked open about 8 degrees earlier than the previous cycle! The whole ignition timing issue was that the point cam lobes were not a perfect 180 degrees apart on the opening ramps! One turn of the crankshaft had the ignition timing at F (5 degrees before TDC) and the next rotation had the points opening well before the F mark, out there around 8-10 degrees further.

Armed with that new-found, over-looked knowledge, the two repair options are either to pull the engine and right side camshaft to replace the point cam with a better one and reinstall OR gently take a Dremel tool with a grinding drum and work the point cam profile to where the point opening moments are exactly 180 degrees apart. Option 2 was employed and the difference now is about 1-2 degrees.

With all the testing and fixing that has followed the engine and carburetor work, the vibration levels of the Type 2 engine have come to the forefront. Honda made 180 degree firing engines for a good reason on the majority of the Super Hawks and this odd-duck 360 crankshaft model reminds me more of riding a Scrambler then a Super Hawk now. Like revisiting an old girlfriend after 25 years, things change between the two of you and often the memories of what went wrong come back to mind in a very strong fashion.

Maybe it is time for the bike to get a 180 crankshaft rebuild instead….