My last notes, before the exhaust addition, said that I had left off with 30 pilots in the VM28s on this bike. However, when I pulled the pilot jets they were 32.5s. I had trouble getting a good idle and keeping it running enough to even mess with adjusting the mixture screws so I thought that the 32.5s might be too rich. I had no real proof of that and probably should have left them as they were but going down this road did lead me to find something else later. For reference I have 30 pilots and 220 mains in the VM30s on my 76 KZ400. Looking back I think the issues with not running properly are two-fold, one the bike has not run in nearly 2 years, and second I found something else later that may have been the main culprit.
I went down to 27.5. I adjusted the air screw to best idle, about 3/4 to one turn out. The bike was tough to start but idled and revved pretty good while in neutral. But it wanted to die as soon as I tried to put it under any load. I re-watched videos about tuning Mikuni round slides (by Mikuni Oz) and set about figuring things out the right way. The 27.5s were quite too lean so I put in 30s.
But I could not get the bike to start. I thought that there was no way that simply going up a small pilot jet size was the cause. Perhaps I had fouled the plugs. I changed those, no difference. I checked the plugs again and noticed they were oddly pristine like they had not tried to fire at all. Then I pulled a plug and cranked it over and saw I was getting no spark. So why no spark? It was a brand new Dyna green coil. I really hoped that was not bad, it was brand new but I bought it 3 years ago or more. Perhaps it was a component of the electronic ignition that I am running from an 81+ KZ440? I have a spare of everything for that except the ignitor, though I do have one on my 76. I set about swapping in good parts from the 76 when I found the cause - the ground wire to my ignitor had become undone. I plugged it back in and cranked, now I had spark - good spark. I believe that I had been seeing symptoms of this for a while. A loose connection may explain the hard starting and trouble keeping it running with the 32.5. I also had the bike just completely die on me a few times while idling when adjusting the carbs - not sputter and die, just die. Additionally when test riding it around the yard I had some trouble with revving past about 3-4K when under load - part of that may be that the carbs' mid-range still needs adjusted but it could also be because of weak spark.
So once I had that corrected I started testing the upsized pilot jets. The 30s are better, it definitely started right up with no choke (it is 85 degrees out today) and I was able to get a good solid idle and adjust the idle down finally. I couldn't really do that with the 27.5 jets. Also wIth the 27.5s it wanted to surge a bit up and down while idling and it would take a second or two to return to base idle after revving. Not sure now how much of that is due to the jets and how much could be a possible loose ignition connection. But the 30s seem to be quite better all the way around so far.
The tuning directions say to adjust the idle pretty low and turn the air mixture screws in 1/2 turn, waiting 10 seconds between each turn, until the bike starts to stumble. Then turn them out following the same procedure until the bike starts to stumble. The final setting should be midway between the 2 points and should be somewhere between 1 to 3 turns out. I found that with the 30s I get best idle around 1 turn out but it never really stumbles even when turning the air mixture screw all the way in, the idle definitely gets quite a bit lower but no stumble. According to the video that I watched that means that the 30 pilot jets are still a bit to lean. So I just switched back to the 32.5 pilots and buttoned everything back up during my lunch break. Hopefully I'll get a chance to test those tonight. Worst case these are too high and I have to go back to the 30s but I feel good that I am getting it figured out now.
BTW, here is the video I mentioned about tuning Mikuni VM round slides. I really like this method, much simpler, than any of the others I have read online: