18″ high shouldered (H-type, H-shape, flanged or valanced) rims are essential point of both, CBX550 spoked wheels project and whole KZ650 project. I bought first pair of such rims on Italian ebay. They were descendants of Borrani Record rims and they were definitely made in Italy, not somewhere in nowhere. They looked great and I was happy with them up to the point when I started spokes tightening and began my work on achieving satisfactory lateral (radial, out-of-round) and axial (side-to-side) runouts. My standards are 0.7mm for lateral and 0.5mm for axial run-out. I use two indicators (dial gauges) simultaneously, as usually.
When it was looking like I was close to good result I levelled up indicators and spun the wheel. And Holy Molly! The opposite to “lateral” indicator side of rim was wobbling like a drunken sailor…
After first shock faded I used third dial gauge to clarify the picture.
It turned out that rim was narrower for more than 2.5 (two and half) millimetres on quite long area around the welding seam. It’s unimaginable, but it’s a fact. I checked once more and found out that only outer walls of rim were narrower in that area, while distance between inner walls was more or less constant. However, it was clear for me that I had to find better rims…
The search took some time, but luckily for me I found that Takasago Excel makes 18″ x 2.5 shouldered rims. And luckily for me I decided to order them from
It is said on Central Wheel website:
“Here at Central Wheel Components, we pride ourselves on being Europe’s largest stockist of classic rims. We have been collecting data for over 120 years, and drill the rims in house. So if you can’t find what you’re looking for, get in touch!”
And it’s true. I wrote email to Central Wheel with a question if they have Excel flanged rims in stock and if yes, is it possible to drill them to fit my conversion of CBX550 hubs. Sure, I attached some photos and hub drawings. I was instantly answered by Richard Hoyland and he confirmed that they could do the job. I ordered first rim and, thanks to Central Wheel Components, order was processed very fast and soon parcel with rim was on its way to us.
However, right when I was ordering rim, Ukrainian government lowered limit for non-taxable parcel down to EUR 100. Thus when parcel with rim arrived to Ukraine it was handed to customs. It wasn’t a big deal to pay another 10 EUR of custom duty; the problem was that custom held parcel up to the last term described in law, which is 7 working days, that automatically means 9 calendar days. Only after the deadline passed I’ve got SMS with a sum of tax and requisites to pay. I made payment in ten minutes and waited for another few days while parcel was released by customs and delivered to me. It’s ridiculous: it took only 5 days to deliver parcel from England to Ukraine and mostly two weeks delivering it from incoming centre (Lviv, Western Ukraine) to me.
As soon as I could lay my hands on parcel I unpacked it. Content was looking like a ring wrapped in gift ribbon.
And where is a gift ribbons and ring, you for sure may find some girl name…
Central Wheel Components made a great work. I think, they might have supply of undrilled polished Excel shouldered rims, thus they could drill them according to specific hub. But hubs of CBX550 have quite a large diameter, larger than most, eve dram brake hubs. So I guess in this case Central Wheel firstly deepened spoke dimples and then drilled them at angles corresponding to spokes holes in hub flanges.
I put some extra polishing on rim and proceeded to front wheel spoking. With such a rim sit was an easy game. I managed to get 0.4mm for both lateral and radial runouts (including the zone of welding seam).
Spoked polished wheel looked great:
My next task was putting tyre on rim.
As all components are highly polished I didn’t dared to entrust this work to tyre service. I have quite an experience with tyre fitting and simple tools to do the job, so I managed to do the work without any scratches. With tyre and some components installed, front wheel looks magnificent.
Yep, I know about it. To be honest I know every suitable head on German ebay. This head has a broken cooling fin on right side, right under front oil channels plugs. In my opinion it's too close to plugs therefore welding may cause plugs loosing and thus oil leaking. I chose and ordered another lot from German ebay. Waiting for it with my fingers crossed...
It’s well known, that clutch baskets of Kawasaki KZ650/KZ750, as well as that of Zephyr 750 are equipped with rubber dampers. These rubbers shrink and even crumble after a long time of exploitation, therefore cease to perform damping functions. Clutch basket joins secondary shaft and gearbox, so it’s essential to keep those rubbers working. Clutch basket’ outer, where dampers sit is made in “non-separable” style, and Kawasaki does not provide replacement dampers. However, aftermarket dampers made of Viton rubber are available, and methods of clutch basket disassembling are well known. Thus I am not the first person who did such replacement, and not the first person who wrote tutorial for it, but may be you could find something interesting in my approach to the issue, as I am used to make the work as I deemed necessary.
So what do we have at start? As you may see from photos below aluminium clutch basket outer is riveted by six 5mm rivets to steel gear.
Our first task is to remove rivet caps from the side of the gear. I drilled them with 3mm drill to some depth.
And then remove caps with countersink bit. I recommend to use countersink bit instead of large drill, as it works much more predictable and precise.
After rivets’ caps removed we could remove steel sear cover and see what came from original rubber dampers. In case of my KZ650 basket they shrunk and all of them have cracks.
Our next task is most important. We have to remove rivets without even slightest damage to aluminium basket, as later we have to cut the thread in holes where rivet sits. Do not try to punch out rivets as they are, and do not try to drill them out with 5mm drill bit. Instead use 3.5-4mm drill bit and drill rivets’ bodies MOSTLY through.
This will remove the tension between the rivet body and aluminium housing so you could punch out rivets from basket without application of excessive force.
If you made everything correctly, you have six undamaged holes each of them by 5.1mm in diameter. We have to cut M6x1mm thread. According to specs, such thread needs 5mm hole. However I cut the thread and found result being very sufficient: thread looked fine/
But appearance is not only parameter of the thread. The maximum torque for M6 thread made in aluminium is 12-14 Nm (camshaft caps). Thus, I tightened 6mm bolt with 14Nm to check if thread holds firmly. Test went successfully. That’s why undamaged holes were such essential: be they a little bit larger and thread might be loosed.
I bought two sets of Viton dampers from Turbosteve, one for KZ and one for Zephyr. When I installed one of them on KZ650 clutch basket I found that I wasn’t satisfied with backlash between gear and basket. Thus I took second set of rubbers, and combined dampers until backlash became minimally possible. It was alike the scene from movie “The Good, the Bad and the Ugly”, where Tuco assembles a gun in gun shop.
Once I was satisfied with backlash I proceeded to assembling.
There is a variety of methods how to fix gear cover back. Here is my version. I chose flanged screws with grade 12.9 and 2701 Loctite (which is temperature improved version of 2700). And if you are a control freak, here is my little invention that will allow you to sleep without bad dreams about wrecked crankcase. In fact it not exactly invention. Everyone who have to replace driving chain before next day ride and found out that his chain riveting tool was lent, but not returned, or not functioning properly. could of invented this method. In fact it’s old school chain riveting method that may be called “bearing ball and hummer”.
To check if it works on screws that I’ve bought for clutch basket, I drilled the end of one of them with 3mm drill bit, screwed a standard nut on it, and then with help of suitable bearing ball and hummer widened drilled opening. After this I tried to unscrew nut and here it is, it didn’t come off.
So I modified all six screws in same manner:
Torque them with 13Nm and then riveted screw ends. Here is how the same screw looked before and after such operation:
Screw heads were protruding for about 3mm about gear cover, while suitable protruding is not more than 1.5mm. I protected clutch basket from abrasive penetration and then removed the extra heights with file tool. I equalized heights of heads and then made final grinding using as base thick mirror with sheet of sandpaper on it.
Result of my work looks fine. Yet one unit of project is ready.
I’ll check clearance between screw heads and crankcase before assembling and I also have a thought to soak gear in oil before install it.
Very impressive work and attention to detail. Make me wonder what lies beneath my clutch basket!
KZ RIDER RELAY FLAG CARRIER #55 1977 KZ650B Ready & Fit to Roll Anywhere! 1974 F7 175 -first bike and still own 2002 ZRX1200R -Red the faster color
2011 Concours 14 -Now a Snarling Wolf in Sheep's Clothing!
TexasKZ: Sure, I could do that, but it'll be up to moderators to stick the thread.
It was a busy week: I was assembling rear wheel and working on carburetors. Not counting that yet one KZ750 cylinder head arrived. I found it acceptable so now I am working on it. Hope to finish channels cleaning on weekend.
Saturday was completely f*cked up. I ordered cylinder to be honed for Wiseco K700 kit (64.00mm for barrel). Yesterday I took them ready from specialized workshop. Cylinders looked nice but I decided to measure them. Even presuming that different bore gauges may give slightly different results there is no way to be mistaken with different measurements of same gauge. I clearly see 0.02 difference between measurement which is unacceptable.
Seems I've been slightly carried away when working on cylinder head. I started it like cleaning but then it turned... no, couldn't call it porting. Shaving is more close to what I did to intake and exhaust ports. Work on head is mostly finished but still is in progress.
By the way, turned out that black and silver KZ750 cylinder heads have different part numbers not only due to color. They were cast in different molds. About differences I'll write later.
Meanwhile, lets talk about intake. I am about to use Mikuni VM24 and I need not only rebuild them but also re-tune them to fit Wiseco K700 kit and KZ750 cylinder head. So I decided to try Keyster repair/tuning kits for Kawasaki KZ650 carburetors:
I like the look of K&N dual pod filters and once I found out that K&N make dual pod filters with flanges center-to-center distance compatible with Kawasaki KZ650 round slide carburetors (75mm it is) I placed the order. filters look great, and of high quality , traditional for KN products.
But anyone who might of check RC-2382 on K&N website find out that inside diameters of its flanges is 54mm (52mm according to measuring made with calipers), while KZ650 Mikuni VM carburetors have 40mm intakes. Difference is huge and quite obvious, that RC-2382 couldn’t be put strait onto carburetors. The solution for this task is applying adapters in form of velocity stacks. I am not inventor of it, however I made drawings of my own version of adapters. I started with measurements and good old pencil and ruler…
I used to make blueprints in different drawing programs (CAD), but this time I had to master a bit new to me 3D drawing program from which drawings could be transferred to CNC lathe. Lucky the task was quite simple…
I modified blueprints for better fitment and ordered parts to be made on CNC lathe. Photos below I shot just after receiving, so to say fresh from under the lathe cutter. At that ерфе I also moved a bit further into 3D modeling program mastering so here are rendered 3D pictures VS reality:
Very cool! are you keeping them shinny?
Those carb kits look top notch, which needle is stock?
Yep, I'll polish stacks. I decided to compare these needles and original needle when I'll be assembling carburetors. What I noticed from first glance these sets have no pilot screw for carbs with accelerator pump (it regulated air but has different configuration than pilot screw in carburetors without pump). Also starter plunger from set has no needle on its end.
Nice. Do stacks hold on to carbs with o-rings only?
The question of KZ650 carburetors is quite extensive. One may catch a serious case of headache just counting modifications of Mikuni VM24 that were installed on 650 during years of its manufacturing. My knowledge in this is quite superficial, however I counted two modification of VM24 without accelerator pump and three modification of VM24 with accelerator pump. Engine on which I based my project had carburetors without accelerator pump and with late type of pilot screw that adjusts air. However, as soon as I learned about VM24 with accelerator pump, I became curious about it. One may find information that those accelerator pump implicit fuel mix only on idle, so to say produce single jet of fuel in every carburetor when you jerk throttle to start moving. However, only one modification of VM24 with accelerator pump has closing fuel valve (triangle thing on float bowl of second carb), and vacuum actuating circuit that causes valve closing on higher rpm. Second modification of “pumpers” still has valve, but it has no actuating vacuum channel. And the last, third modification has no closing valve at all. Thus, in these two types of VM24 there is no reason for accelerator pump not to work on other than low rpm. However, its only my suggestion and I couldn’t confirm it until I finish with “pumpers” I bought for the project. And even if experiment with accelerator pump will lead me nowhere, I could deactivate all pump system just by replacing float bowls with bowls form carburetors without pump.
So I bought modification of “accelerated” carburetors with closing valve but without actuating channel. They turned out to be in very good condition:
Sure, forty-years-old carburetors need to be completely rebuilt and cleaned through and out. So I put them apart. Counting age, insides also looked rather nice.
Some soot, dirt and depositions in my opinion are manageable.
To my knowledge, there is no vapor blasting here in Ukraine . The closest thing we have is a workshop that specializes on glass beads blasting, which has a range of glass beads of different size and thus coarseness. But even if there was vapor blasting available, I am not sure that I would dare to use it on carburetors, as it is also media blasting. I also don’t like how blasted aluminum looks, so my way is simple: carburetor cleaner, brush and those synthetic detergent I used on crankcases. It works nice and removes all visible depositions, so there is no reason to consider it might leave invisible ones untouched. I like how carburetors look after cleaning:
Thus carburetors’ bodies were cleaned and so were all other components, like heat shield.
Float bowls’ gaskets gave me a bit of work. as they were glued to bowls from factory and after mostly forty years glue and gaskets hardened. To cut them off with a knife might look good idea, but I didn’t wanted to damage float bowls. So instead I put bowls into hot water for an hour, or so, and after some soaking gaskets became softer and thus more removable.
Rough rug and carburetor cleaner helped me to remove last traces of glue. I was glad that I preserved contact surface of float bowls completely undamaged.
I also polished all four top covers.
I prepared and gave most of steel components in workshop that specializes on zinc plating. Refreshed parts look very nice.
Sure, it wasn’t hard to notice that I grinded outer parts of carburetors’ intakes.
I made six cuts in every velocity stack to fix them on carburetors with help of clamps.
These works I made to warrant the tight fit of velocity stack to carburetor. And as you may see from photos below inner transition from velocity stack to carburetor is also very smooth. Thus, there is no need in any sealing gaskets or O-rings.