The stock handlebars on the CB350 are pretty comfortable, but they put your hands pretty high up so you look like you’re riding one of those gangsta bicycles from the 1950s.  The ergonomics of the stock bars are pretty good, though.

The mirrors are also mounted way high up, making the bike look like a goofy long-eared rabbit and screwing up the lines on what is otherwise a compact, pretty machine.  That aside, it seems like it’s working out pretty well for the guy in the CB350 ad from 1971.


I was looking for a set of handlebars that would drop the control position down a couple of inches, but nothing crazy.  I also wanted to find a way to install bar-end mirrors, and the stock bars had metal plugs welded in the ends to prevent that.  I didn’t want to change the way that the handlebars felt, but I did want a little more forward lean.  Clubmans or clip-ons would just be a little ridiculous (and soooooo 2014) on this little bike.

After some research, I decided on a set of reproduction handlebars made by Emgo for the CB400F. They’re “euro” style with a low rise and slightly less pull-back than the stock CB350 bars.  Once they arrived, I made some measurements and found that the pre-drilled holes were in the right spots for wiring and the mounting pins of the switch housings.  I was ready to install. 51WXCZH3oWL._SL1200_

It was one of the scarier operations that I’ve performed on this bike.  Because the switch wiring goes through the handlebars, it is necessary to disconnect all of these wires so that you can remove them from the old handlebars and install the switches on the new handlebars.  No less than 12 wires in the front harness must be disconnected.  For most of these terminal connections, it’s the first time they’ve seen the light of day in 40 years.  I pity the fool that doesn’t document their wiring before disconnecting it all.  I said a prayer and got started.

2015-08-23 10.19.33
Here’s everything disconnected inside the headlight bucket.

Next I disassembled the switch housings by removing the screws that help them together.  I disconnected the throttle cable from the throttle sleeve and removed the left grip.  I left the brake and clutch cables connected even though the service manual recommended removing them.  Honda can’t tell me what to do!

I taped up the disconnected end of each bundle of switch wiring in order to avoid putting stress on the terminals while pulling them through the bars.  I loosened the bar clamps to point the center hole upwards and began to slowly thread the wiring out of the old handlebars.  After about 20 minutes of gently pushing one end and pulling the other, I got the bundles out.  The handlebar was naked!  I removed it.  2015-08-23 11.14.53

I installed the new handlebars and reversed the process, slowly threading the wiring from the switch holes to the center hole.  I had some trouble getting the right switch housing mounted.  I also found that the throttle sleeve was rubbing against the inside of the housing, so I thought the sleeve was the problem.  I tried grinding down the outer edge of the sleeve, but this did not really help.  After about half an hour of cursing and moving the housing around without success, I figured out what was going on.   I had left too much wire on the outside of the handlebar inside the housing, and the bulk of the wiring was preventing the housing from closing on the bar cleanly. I pulled some of it through, and everything worked properly.

I hooked all the wiring back up and turned the key to see if there were any obvious electrical issues.  The starter begin turning over ON ITS OWN as soon as I turned the key, so I immediately turned the key back to the off position.  I almost had a freak-out, because there is nothing worse than an electrical issue.  I settled down and traced the starter switch wire.  I disconnected it and turned the key.  No starter turning.  Okay.

I opened the right switch housing back up and found that all of my fiddling earlier had worn the insulation off of the starter switch wire and it was grounding out against the handlebars.  I added some electrical tape and reassembled it.  I turned the key.  No starter turning!  I pressed the starter switch and the starter kicked right up.

The horn also seemed at first not to work, but then I discovered that I had just left its wire disconnected in the front harness by accident.

I installed the grips and discovered that the left grip actually was not long enough to cover the space between the left end and the housing.  After talking to the owner of my local motorcycle shop and comparing to some old stock he had in the back room, I believe that this is because the CB400F (which these bars were made for) has a larger left switch housing than the CB350 twin.  I’m planning to fill the gap with a bar clamp and to mount a mirror on that.  For now I’m just dealing with it and using a bar-end mirror.

Let’s do a side-by-side comparison.

Old handlebars.  Orange rabbit face.
Old handlebars. Orange rabbit face.
New handlebars. Eh? EH?


I liked how the bike felt before, but I really like the way the bike feels now.  In the new riding position, I feel that I’m closer to the machine.  The position is still upright but definitely more aggressive.  If you’re thinking about doing this modification, I would recommend that you go for it.  Drop me a line if you have questions about this.

Spark timing

I had originally thought that if my motorcycle was running on both cylinders, then I shouldn’t need to manipulate the ignition timing.  If both cylinders were firing, what could possibly need to change?

The points and advancer adjustments are under the chrome cover pictured here.

I could not have been more wrong.  After bringing my ignition timing and points clearance into spec, the CB350 runs much much better.  It purrs through first and second smoothly instead of rattling anemically.

Here’s what I’ve learned:  A motorcycle engine works by tightly controlling a combustion process.  The precise timing of delivery of fuel, air, and spark determines how the engine will run.  The electricity for the spark is provided via two electrically-conductive strips of metal that control the opening and closing of the ignition circuit.  These strips are placed within a few hundredths of a millimeter of a slightly-oblong metal ring that sits on the end of the camshaft.  This ring is called the points cam.  The strips are called the points.  As the engine turns, a chain on the crankshaft attaches to the camshaft, turning it and the points cam on the end of it.

As the oblong end of the points cam spins, it will briefly brush into each of the points.  This (or more specifically, the moment when it STOPS touching each of the points) causes a spark to jump across the gap at the other end of the circuit connected to that point.  You guessed it, that’s the spark plug.

There’s also a mechanical device called the spark advancer that works only at higher engine speeds, causing the spark arc to take place a little earlier in each rotation of the engine.  The spark advancer spins with camshaft, and has weights attached to it that move outward in proportion to centrifugal force and are returned to their original position by springs when engine speed falls .  The weights put pressure on the points and cause timing to be altered.  This component is necessary because there is only a limited amount of time to deliver and mix air and fuel in the cylinder before the exhaust stroke begins, and higher engine speeds necessarily shorten that period of time so the spark must be delivered earlier to provide better combustion, or in other words, the spark timing must be advanced.

Adjusting ignition timing should really be done at the same time that you adjust valve clearance.  Both adjustments require you to drain the oil from the crankcase, remove the stator cover, and manually turn the crankshaft with a wrench.  Both adjustments involve changes to components underneath the points cover.  Also, you should really do valve and spark adjustments before messing with the carburetion and exhaust of the engine,because if the engine doesn’t breathe and spark properly, then no amount of other tuning will make it run properly.

ANYWAY.  the actual operation took about two hours total.  Measurement of the points clearance was simple.  The factory service manual says to measure at the point where the gap is greatest, but it doesn’t say at what point on the crank that should be.  I just left my 0.30mm feeler gauge at the gap and turned the crank until it went in, and then kept re-inserting it while turning the crank to determine when the gap interval was over.  Looks like the greatest gap is about 90 degrees after firing.  Both sides were within 0.3-0.4mm.

I then figured out how to connect my timing light to the points, and used it to determine that each  spark plug has a complete circuit until about 60 degrees in advance of the “F” mark on its cylinder, and then the circuit is broken.  The light comes back on (spark!) around the F mark.

2015-08-02 20.45.00
Sorry for picture quality, sun was going down as I was finishing up.

I found that my left cylinder was about advanced by about the same amount of crank distance as between spark fire and top-dead-center.  This is probably about 15 degrees.  The right cylinder was retarded by the same amount, which makes sense.  I adjusted the breaker plate using a screwdriver while holding the crank at F, and then checked again at LF.  All in line.

2015-08-02 20.45.16


I put the stator cover back on, filled the case with oil, and took the bike around the block.  Man, what a difference.  So smooth, so much more power.

I realize that I forgot to re-check my points clearance again after adjusting the spark timing.  Because I put some Marvel Mystery Oil in this fill-up to see if it helps with the oil burning issue on the left cylinder, I’ll be changing the oil again soon anyway, and I’ll just re-check them again at that time.

Flasher relay and LED indicators

The CB350 had this weird issue when using the turn signals. When you switched on either side, the signals on that side would stay lit for a few seconds, and then verrrry slowwwly flash on and off. You wouldn’t think this would be such a big deal, but when you’re trying to change lanes or signal when approaching an intersection, a turn signal lit without flashing just looking like a broken parking light.

So I went looking for solutions and found that I could use a modern flasher relay to replace the ancient mechanical one. I decided to replace the flasher entirely because I was thinking of replacing the indicator bulbs in the gauges with LEDs, and the turn signal indicator required an LED-compatible relay anyway. I ended up with the LF1-S-PIN flasher from superbrightLEDs.com.  I’ve used superbrightLEDs.com for replacing bulbs in my last two cars, and their stuff is always high-quality (even if on the expensive side).

A picture of the flasher is below.  It’s got two prongs, just like the stock mechanical relay.



I already had the tank off, so I went ahead and removed the old flasher from the left side of the bike.

2015-07-19 17.17.38

I removed the metal bracket that provided the mount for the old rubber relay holder, and threaded the new plastic relay onto the bolt that supported the old mount.

2015-07-19 17.29.38

Clean and stable.  And turn signal indicators now flash properly as well!

Oil filter cleaning

The main reason that I decided to restore a CB350 instead of any other bike is that it is a simple, uncluttered machine that has very little weirdness to its upkeep.  One exception to this is the oil filter cleaning.

The oil filter on the Honda CB350 is not of the spin-on/cartridge type that you can find in pretty much every modern car/motorcycle application these days.  It’s actually a steel cylinder whose centrifugal motion (driven directly by the crankshaft) pulls heavier particles to its outer walls while allowing oil through its central outlet.  This is the first weird thing about the oil filter.

You can gain access to the oil filter by first draining the oil from the crankcase and then removing the oil filter cover on the right side of the crankcase.  Do not lose the o-ring on this cover.  My cover was actually pretty tight in its bore, and I had to wiggle it out using a long flathead screwdriver.  Although the cover is cylindrical and unmarked, it can only be re-attached to the crank case in one of the three possible positions (three because of the three equidistant screws supporting it).  That’s the second weird thing about the oil filter. Make a mark on the crank case and cover so you know how it will re-attach.  The three screws are the same.  2015-07-05 09.33.02

In order to remove this filter, Honda recommends that you use special oil filter removal tool.  People on hondatwins.net recommend using a M6x1.0 bolt instead. This seemed to work fine.  Just get the longest M6x1.0 machine screw that you can find.  The fact that Honda didn’t just recommend that in the first place is the third weird thing about the oil filter.

First you’ve got to carefully remove the circlip holding the outer piece onto the filter barrel (not shown) by squeezing the two holes towards each other with a needlenose pliers.  Honda recommends replacing this circlip with each oil change, but you can re-use it if you’re careful.  Also, it’s not a cheap part.  (I was not careful and broke one of the two circle tabs — still usable but about ten times more annoying to use.)

Once you’ve got the circlip off, simply thread the M6 bolt into the threads at the back of the filter barrel as shown, and then when it bottoms out, pull the barrel out using pliers.

2015-07-05 09.33.45

All that’s left is to clean out all the muck in the filter barrel.  It may look like there’s nothing in there, but believe me, once you start scraping a towel around with some WD-40 on it, there will be layers of gunk coming off.

2015-07-05 09.33.59


Mine wasn’t too bad, but I’m planning to run some Marvel Mystery Oil in the next change of oil so I plan to see some more muck at that time.  Luckily there were no metal shavings or rust.

Anyway, assemble in reverse order and re-fill the oil, and you’re good to go.


Left carburetor rich, unaffected by idle air screw

If you’ve been keeping up with my progress, you’ll remember that I recently rebuilt my carburetors and re-installed them on the CB350, but that I was having some trouble tuning them, and that some lightish smoke (with black condensate) was coming out of the left muffler.

I did some further investigation today, using the old pull-one-spark-plug-lead-off method to run the bike on one cylinder. While running on the right cylinder alone, the bike’s idle was in fact controlled by the idle speed screw on the right carb. However when I ran the bike on the left cylinder alone, the idle speed was entirely unaffected when I manipulated the screw. I could screw the idle screw all the way in or remove it entirely and the bike ran either way.
This would seem to indicate that there’s some sort of condition on this carb that is causing fuel to make it through even when the idle air jet is pulling nothing but air. There’s likely a rich condition.  That COULD also explain the smoke from the left muffler.
Because I’m not seeing anything coming out of the overflow tube for this carb, I have a hunch that the floats are set too high in that carb. When I have some free time I’m going to try adjusting it downward to see it that helps.

Carburetor rebuild

2015-07-02 23.02.44

Rebuilding carburetors a carburetor for the first time is a lot like having sex for the first time.  You pull off the cover and find a whole bunch of parts, tubes, and pieces that you’ve heard about before and maybe even seen some illustrations in textbooks, but the moment you start prodding around you instantly feel like you missed a step and you’re in way over your head, and that carburetor knows it.  All your friends are going to laugh at you when they find out you tried to stuff a secondary jet into the idle jet hole.  Compounding this difficulty is the fact that Honda used at least three different carburetors during the 4 production years of the CB/CL350, so online information is often incomplete and unclear.


2015-07-02 17.19.41


And oh yeah, it’s filthy down there.  Note how the old float bowl gasket is actually flattened and bonded to the groove of the float bowl.  This is the reason that gas was seeping out of the sides of the carburetors before I took them off of the bike.  2015-07-02 17.19.34

I used a set of K&L carburetor rebuild kits to rebuild my carburetors.  They contained the right replacement parts for the primary and secondary main jets (#68 and #105), needles, float bowl gaskets, and float valve assemblies.   I was afraid that I’d crack the ancient brass floats if I attempted to pop out the float hinge pivot, so I never got down to the float valves to replace them.  I really wanted to replace the slow jet from the left carb, but the kit had the wrong size jet included.

When you remove the three jets, you should be able to see straight through the primary/secondary holes to the throat of the carb (two larger holes on the left), but not through the slow jet hole.  This is because the slow jet feeds fuel through a passage that makes a turn towards the outlet side of the carb where it lets out just on the engine side of the throttle plate.  2015-07-02 23.02.57


So I didn’t get a chance to replace the slow jet or float valves, and because I didn’t detach the floats, I didn’t soak the carb body in carb cleaner.  I did set the float levels by eye.   I also pulled the slide and replace the needle.  In hindsight, it probably would have been a good idea to lubricate the slide with some WD40 or something as well.   I just didn’t think of it.


2015-07-02 23.12.44

There are three passages that connect to the area under the diaphragm.  I am not sure what these are for, and I did not clean them either.  I will likely have to do that in a future rebuild.

I painted the top covers because they were rusty and disgusting, and then reassembled the carbs.  I installed them on the bike and sync’ed the throttle cables.  The bike started right up.

I am finding that the bike has lost the problem it had before of holding revs after the throttle is closed (probably due to cable synch issues), and that power is greatly improved.  Tuning was still not possible however because the idle screw on the left carb doesn’t seem to have any impact on the idle speed of the bike.  I will probably have to tear that carb apart again and thoroughly clean the idle jet and/or idle screw.  Hmm.

Much like my first time having sex, I did pretty much what the books told me to do and conquered some of my fears in the process, but I ultimately left the job less than satisfactory and will need to do another job to really clean things up.




Valve clearance adjustment

Honda’s factory service manual (FSM) lists as 150-170psi as within specification for cylinder compression.  I did some measurements when I first got the bike and found the right side firing at 170psi and left side at 119psi when warm.  I let the bike sit for a week with Seafoam Deep Creep in the cylinders, topping them up every couple of days.  I got the bike up to running temperature and did another measurement.

Right: 165.  Left: 119psi.

Per the FSM, I added a tablespoon of motor oil to the cylinder with low compression and re-tested.  I got the same result: 119psi.  This indicates that the compression issues were caused by something in the head of the engine and not in the piston rings.  Possible causes include the head gasket, valve timing, clearance, or seals.  I didn’t see any leaks around the head gasket, so I decided to try my luck with adjusting the valve clearance.  I didn’t see any reason to check the right cylinder because it was running well already.  Don’t fix it if it ain’t broke.

2015-07-03 15.06.39


There’s three things you have to do in this operation: Hold the crank at top dead center (TDC) for the cylinder you’re examining (“LT” mark on the generator rotor) with one hand, and with the other hand alternate between slightly adjusting the tappet clearance adjuster screws and checking the rocker arm clearance with feeler gauges.

This is harder than it seems because the appropriate clearance is measured in thousandths of an inch, and even a couple thousands off will cause problems.  The adjustment screws only need to be turned a thousandth of an inch in order to translate that change to the rocker arms.  It took about an 30 minutes for me to adjust each of the two screws.  That’s half an hour sitting on the ground, resisting the inertia of the cylinder falling from TDC, trying to turn a flathead screw a thousandth of an inch and accidentally going a few hundredths in the wrong direction each time.  Ugh.

Once I got the valve clearances into spec, I did a quick compression test on the cylinder and read 155psi, up from 119psi!  That’s with the carburetors off the engine, so the engine was cold.  With the engine warm, it’s likely that that number would be a bit higher.  Either way, the cylinder is now in spec for compression.

Carburetor removal

My carb rebuild kits came in earlier this week, and I figured that it might be a good time to actually take the carbs off the bike to clean them. It was pretty scary. The carbs were so gummed up that even when I unscrewed the drain screws at the bottom, gas came out from the drain screws themselves (and not the drain holes!). When I turned the left carburetor sideways to remove it, gas started leaking out through its gasket. These things need a serious overhaul.

Here’s a picture of the bike with the air filter removed.  The fuel petcock was turned off, but I think that the fuel lines were still flowing a little bit of fuel.  I’ll have to check that out sometime.  I want to get some nice chrome filters, anyway. You might also notice that the right carburetor is missing its choke plate.  I know, pretty lame.  I’ve got the plate on the left carb, but I might just run with that if I can get the bike to agree.

2015-07-02 15.42.58

Next a look down the intake manifold form where the carburetor was attached.  I sprayed some deep creep into each manifold for good measure.  Can’t hurt.

2015-07-02 15.43.08

This is a picture of the bike with the carbs and air filters removed.  Yeesh.

2015-07-02 15.50.14

And the stars of the show, the carburetors.  Ready for a good cleaning.
2015-07-02 15.50.19

Tachometer oil seal

When I first took the bike out for a 5 mile ride at about 45mph, I noticed a new feature: hot oil gently spritzing my right pant leg with the fury of a tiny sun. When I stopped to take a look at it, I found that oil was seeping slowly from the port where the tachometer drive cable entered its housing on the right side of the engine head.  The cable was secure in the housing, but there was a couple millimeters of play in its seating against the oil seal.  This is probably why there was a leak.  I think that over the 40 years since the bike was assembled, the seal had hardened and shrunk.  honda-cb350-super-sport-350-k4-1972-usa-cylinder-head-side-cover-breaker-advancer_bigma000148e02_d4ae

I ordered new Honda part 91211-286-003 to replace the oil seal that was apparently failing.  That’s part 19 in the schematic above.

2015-07-02 12.56.39


At first, I had a lot of trouble getting that oil seal out of there.  It was apparently fused to the aluminum of the housing.  I tried using combinations of penetrating oil (Seafoam Deep Creep), dental picks, wood screws, and tweezers, but nothing worked.  I looked to www.hondatwins.net for guidance and found a recommendation to just pull out the whole tachometer drive element (#5 in schematic), and the oil seal would come with it.


After some wiggling and gentle but firm yanking, the whole assembly dislodged.  I had pretty much destroyed the old seal while trying to remove it earlier, as you can see.

2015-07-02 15.01.54

I replaced the seal and re-assembled the connection.  The play in the seating of the cable is now almost gone, and the leaks appear to have almost completely gone away.

Fuel door gasket

It wasn’t until I gave the CB350 its first fresh fill-up at a BP a couple miles away from my house that I discovered a serious issue.  When the fuel tank was full, and the bike was accelerated or leaned slightly, gas would start pouring out of the fuel door on top of the tank.  Any time the bike was not standing perfectly straight, there would be a leak going down the side of the tank.


Naturally I was worried that the gas would flash-ignite when it struck the cooling fins of the motor, so I pulled over and put one of my socks over the fuel door to help keep the leak manageable.  I  kept my acceleration and turning to a minimum on the ride home

2015-06-26 10.23.542015-06-26 10.24.13.

Luckily the fix for this was simple.  I just needed a new gasket to replace the seal where the fuel door meets the tank.  This was Honda part #17534-323-300.  I ordered it from ebay and it came in the mail today.2015-06-26 10.24.21  2015-06-26 10.33.21

The old gasket was missing a chunk, rotted, and brittle.  The new one was …the opposite of those things.   No more leaks.