When I got this bike it would start fairly easily although the electric start was a bit hit and miss. However, although it would fire up easily enough, it wouldn't run without virtually full choke. The plug was very sooty indicating rich running. The PO had said that it took a long time to warm up and he didn't know why... he didn't add that in fact it never warmed up and simply wouldn't run without choke. This obviously pointed to a carburation problem but to my surprise the carb had actually been replaced already, it was clearly newer than the rest of the manifold and inspection showed it was a Chinese Keihin copy by Kaiwan, a PZ19.
In the past I've found this sort of problem to be due to a blockage somewhere in the carb so it seems like a strip, clean and a check are in order.
Remove the legshields, 4 bolts and a lower catch clip on this bike to reveal the carb. Honda fitted Keihin carbs, this bike seems to have been fitted with a cheaper Chinese substitute. In this case a Kaiwan PZ19 but I think this is a direct copy of the Keihin item.
First clamp the fuel hose then disconnect the choke cable.
Loosen the screw
Free the cable from the clip at the top of the bracket
Wind it round to release the nipple
Unscrew the carburettor throttle cap.
... and pull out the throttle slide, spring and needle.
I left this hanging on the bike whilst I removed the carb
Pull off the two overflow hoses
Noting their routing down through the frame. Both pass through retaining clips, one drains to the left whilst the other drains through the frame.
![]() |
| Right hand drain tube passes through clip and enters frame |
![]() |
| Left hand drain tube passes through guide below the block |
![]() |
| withdraw the carb and pull it off the air hose |
![]() |
| Float chamber detached |
Check inside the bowl and clean out any sediment.
Undo the large 17mm nut cap under the tap chamber with a spannerLift off the cover, remove the filter cage and clean out any dirt, make sure the lower O ring on the filter cage is present and in good condition. Replace the filter- making sure it pops fully home and refit the cover plug-checking the O ring beneath.
Looking at the top of the carb, the float and needle valve are held by a free-sliding pin
The pin just pushes out to release the float
So that the float and needle valve can be lifted out. Note that the needle just hangs from a metal bracket on the float by a wire clip. In other carbs the needle has a lug on the top which slips into a groove in the float, these are non-adjustable but the type fitted to my bike type can be altered.
This exposes the two jets which on this carb are removable. The larger central is the main jet which screws into its holder and the smaller is the slow running jet.
![]() |
| Central jets in the carb. The slow running jet has been unscrewed and is being removed. |
In this carb the main jet is in two pieces which unscrew, in others they are made as one. In any event if the jet is clear its not necessary to separate the two parts. Both jets unscrew and can be blown through with carb cleaner. I found that the slow jet was blocked which would certainly account for the problems in running. I realise that these are of critical size but when they are blocked like this no amount of blowing will clear them so either replace or poke them through. I used a high "E" electric guitar string which crunched through removing the crystallised white blockage. I then blasted carb cleaner through all the passages moulded into the body and verified all were open.
Reassembly is the reverse of dismantling and all went well. However when reassembled the motor would now run off the choke, but throttle response at low openings was poor. Opening the throttle just a little led to a misfire which picked up again as the throttle was opened further. There are probably a number of explanations for this but in the past I've found this can be caused by too low a fuel level in the float chamber and this can be addressed by adjusting float
I understand that the original Keihin carb may not have been adjustable, but the PZ19 clearly is. Checking Haynes manual float height is specified as 10.7 mm for this bike. This seems a little optimistic as I doubt anyone can set it within 0.1mm anyway. This value is for a genuine Keihin, but think it should be the same in this copy-cat carb.
I stripped the carb again to check float height. Carb height is defined as the distance between the carb body flange base and the base of the float when the float is in contact with but not pressing hard on the needle. You can arrange this by holding the carb just off vertical so the float hangs against the needle.
However in this carb the float is pivotted at one side and clearly doesn't hang straight so it makes a difference where you measure its height. By convention you check across the main jet. I have a float height gauge which you set to bridge the float making contact with the mating flange on each side and then depress the slide until it just contacts the float. However, it seemed to me that it was set much higher at 15mm. I readjusted the float height by bending the metal needle tab- in this case upwards,away from the float, so that the float would sit lower when the needle was seated.
.
![]() |
| Resetting the float height to 10mm |
Many people recommend setting fuel level rather than float height. This is easily done using a clear tube on the drain spigot that can act as a U tube and indicate fuel level in the float chamber. This sort of adjustment can be awkward on the bike but the level should be checked at least in position because the carb isn't horizontal in use. Trouble is I can find no info on what level to aim at. There is a moulded mark on the lower section of the float bowl to the right of the fuel tap. Experimenting I found that setting the float height to 10.1mm gives a level above this mark when horizontal and higher again when in position on the bike. I'm not able to find the "real" height expected so I'll just show the height my fuel rises to in position with the float height set to 10.1mm.
![]() |
| Moulded mark on the carb bowl rim |
![]() |
| Level indication mark highlighted in blue |
The level observed in situ with float height 10.1mm is shown below.
![]() |
| Choke cable clamped onto packer behind the ferrule |
Still not sorted
Well I took the thing out for a run and it runs like a dog. It starts and moves off easily but there is a decided lack of power, a real hesitation and even pushing the motor doesn't give me much past half throttle.
Reading around I discovered that many of these cheap carbs come jetted for something other than the C90. Maybe a chainsaw or something!!
The jets in a Keihin are simply specified by the orifice diameter and for the C90 these ought to be pilot-40 and main 80-85. Using a set of jewellers hand drill bits (and checking them on a micrometer) I found my carb is jetted at pilot 30 and main 70. You can use these drills to carefully enlarge the main jet at least to 80 although this is a rough and ready approach that could destroy any venturi effect. The pilot was too small for this approach to work without breaking the drill bit! Having enlarged the main there was some improvement but still no real response to full throttle and the overlap between pilot and main as the throttle is opened seems rough.
I ordered a Keyster carb set 0323 which comes with both correctly sized jets and a new needle. Sadly, and to my mind oddly, it doesn't come with the needle jet female counterpart so I'm not sure whether the Keyster needle will work in the Chinese carb hole but I'll try fitting this lot and report back.
I fitted the pilot and main jets plus the needle from the Keyster kit, setting the needle clip to the middle position which should be right now. I readjusted the carb. Surprisingly the bunsen blue is visible over a wide range of air screw adjustments so I had to set it for best running at idle (eliminate misfires) and good response to throttle blip. Road test shows the bike is now running the best I've ever had it. The new pilot jet gives much smoother response at low to mid throttle but really full throttle is still unresponsive so the main hasn't improved matters.
I'm going to explore non carburation faults now I am beginning to wonder if this is an ignition fault where the coil breaks down at higher revs or the cdi can't respond quickly enough. More serious issues might be a burnt or leaky valve and badly adjusted tappets. I did check the compression on this motor which was 160-165 psi, so pretty much perfect and this eliminates any leaky or burned out valves. Checking the tappets did throw up one problem in that the inlet tappet should be 0.08mm and the exhaust 0.1mm. The inlet was too big and accepted a 0.13mm feeler so I reduced that to 0.08. Exhaust was fine. Sadly none of this helped.
I am now reduced to electrical and ignition faults so I will swap out the coil and cdi unit. Whilst I'm in there I will swap the flasher because this one's a bit fast. I'll start a new post for this electrical stuff.




























