Getting at the fuel screws without burning my hands is going to be an interesting feat.

You just need more tools.

Sitwon, laugh, ain't that the truth?  

Thank you rshaug for coming over and solving that problem for us. It would have taken me ages to figure that out, if ever. 

I'm not still 100% sure I understand what the issue was.

Yermoyou are most welcome. I think I can boil down the issue (or what I believe it to have been) thus... It was a throttle position problem. The behavior with the choke was a Red Herring, making it seem like a potential choke or fuel deliver problem. What it seems to have been instead is an overall misalignment of the various throttle adjustment parameters. The more I thought about it the more it made sense, as there are several different adjustment points which control throttle position either in whole (all carbs) or in part (single carbs). Watching the vacuum gauges really solidified it for me as I could tell that the "high idle" was in fact not an idle at all but was happening from an open throttle (watching the transition from the idle jets and idle bypass to the mains and airflow through the throttle bodies). From there it was a question of determining where in the throttle travel that transition occurs, then adjusting everything back to the point where releasing the throttle closes the butterflies and drops vacuum. Once getting the carbs and throttle cables and actions synced up such that closing the throttle truly closed the throttle and brought the system reliably into idle (only using the idle bypass for air and idle jets for fuel), we then could simply adjust the idle speed back down to the desired level - which we set at about 1150-1200 rpm. 

The summary is that what was happening is that the throttle was actually partly open, potentially in one or more carbs, and the condition was exasperated by having the throttle pull and return cables improperly adjusted, which created weird tensions on the cams and likely was inducing the behavior in the center two carbs (probably num3 actually, based on what I saw with the vacuum gauges as we cycled them in operation from idle, to off idle, to part throttle, to WOT).

See if it still behaves tomorrow, but from what I saw it should be fine. Once everything is back together you should be able to adjust idle speed with only the idle adjuster. It should be obvious if it transitions from high idle to on throttle, just back it back off if it does. Remember with the idle and throttle cable changes...small changes, then wait for it to settle before assessing.
 

It occurred to me later... we adjusted everything with the airbox and air filter removed. Should we have done the carb adjustments with the air filter in place? does it make a difference?

Thanks for the explanation rshaug. I think I get it but until I have another set of carbs to play with to explore my understanding I'm not going to know for sure.

Sitwonyou should be fine, that shouldn't impact what was done.

Bike is all back together idling perfectly. Throttle response is correct. I am now departing on a test ride.

I declare the Bandit S fixed. It runs perfectly. I put it through it's paces tonight and could find nothing wrong.



Thank you again, rshaug. 

The bandit passed inspection with flying colors.

Specs of the Bandit by type and year:

 

Not on a bandit but some good information for a related carburator:

 

A problem has developed, which was probably there earlier but I did not notice, such that it is impossible to get the bike to hold 1750RPMs with a slightly open throttle without it immediately shooting up to 2750RPMs. 

This makes being in stop and go traffic very challenging not to mention doing tight slow turn exercises in parking lots. 

Ben Mendis and I carefully rebalanced the carbs and adjusted the idle speed again. On a lark, we used my Twin Max gadget to check the accuracy of the mercury sticks and found that the mercury sticks were not reading accurately for some reason. I suspect the plastic in the lines has gotten so old that it's not making a good seal around the vacuum nipples.

So we rebalanced two carbs at a time over and and over again until we go them all to within the accuracy limit of the Twin Max. The throttle jumping problem improved but did not disappear. 

My suspicion, after doing a bunch of reading, is that now that we have the throttle linkage and balance set correctly we've exposed a lean pilot circuit issue.

The Bandit low RPM throttle response issues were bugging me so I went out to the garage with the intention of experimenting.

Using starting fluid this time I tried to find any vacuum leaks once again. To the best of my knowledge there are no vacuum leaks. I had read that sometimes the little cap on the airbox breather line which is routed under the bike can cause vacuum problems, so I checked that as well and it's present and sealed.

I decided to experiment with the pilot screws. The center ones are quite challenging to get to. I need a better screw driver for the task.

I experimented with increments of 1 turn from 1 turn out to 4 turns out.

At 1 turn out, the bike shows symptoms of running lean. The "throttle jump" from 1700 is most pronounced. 



At 2 turns out the bike runs as it has been.
At 3 turns out the throttle jump lessens a bit.

At 4 turns out the bike seems to be running slightly rich but not overly so but the throttle jump (surge?) from 1700 is the least pronounced but still present. I haven't taken it on a test ride yet for a more involved test.

My assumption is that the carbs are bone stock. The jets we pulled out seemed to be the stock sizes.

It seems to me that it's likely changing the idle mixture screws this drastically to get an improvement is masking some other problem in the system.

The stock setting, as far as I can find online from various sources, is supposed to be 1 3/5 to 1 5/8 turns out. But apparently many people turn them out as far as 3.

I must admit that this is quite the head scratcher, particularly after yesterday with the shim experiment. A couple of things come to mind. Firs Fast Larry's Bandit FAQ: 

http://www.portablepct.com/fastlarry/banditfaq.htm

If you look in the carb section there is a listing of potential mods, and a reference to shims that can be bough from Radio Shack (of all places!?!)

The reason for buying them there, is that apparently they are very uniform at .05 width.  Here's another interesting page from a Valkyre site:

http://www.rattlebars.com/mtz/shims.html

Notice what this pages says about adjustable needles in STEP 9. In addition, if you change the needle jet position with shims, you need to adjust the pilot circuit mixture.

So after refreshing my memory on this, I was reminded that the pilot or idle circuit controls the mixture of the carb to 1/4 throttle, after which the the needle jet takes over. IIRC, this is exactly the throttle position where this problem is occurring, right?

So, perhaps a set of radio shack shims (Radio Shack shim part 64-3022 or 3022A) and set of calipers (to insure that they are uniform in thickness) might be worth picking up. I don't know what thickness the shims we got yesterday are, but this is really a very fine adjustment so it probably matters.

So in summary, I'd say pick up several shim packs, then try adjusting the carb mix back down on the idle circuit to stock. Add shims and check the 1/4 throttle transition. If the bike is bogging with 3 turns out and one set of shims (of unknown thickness and uniformity), then I'm guessing it's too fat right now. Or, just order a set of adjustable jet needles so we can try different needle heights:

http://www.jetsrus.com/a_jets_by_carburetor_type/needle....

now this may not be the *right* needle for this exact carb, you'd have to verify the number, or find one that's identical in dimesions. OK, now for one more ref from the bandit faq:

http://www.jetsrus.com/a_jets_by_carburetor_type/needle....

and finally, a link on the general tuning of CV carbs:

http://www.factorypro.com//tech/carbtun.html

As stupid as it may seem to have to go through all this, some times they DO get it wrong from the factory. But remember carbs *can* work just fine. Several forum members here ride Suzuki DRs, which are carbed

Per the msg I sent on Facebook, after test riding it last night, I've noticed that the symptoms have changed.

Despite the balanced carbs and low idle speed setting, it still occasionally sticks at 2000 rpms on the way back down.  We had thought that shimming the needles had an effect on that symptom but it turns out it was a fluke. It seems that shimming has actually had no positive effect.

The bike now will often stall at traffic lights on the application of light throttle. (i.e. running too rich.)

The jumping from 1750 to 3000 rpm problem is still present and makes the bike very challenging to ride in parking lot situations. Once under way it runs fine. So I believe, as you mentioned above, that it's something in the pilot to main circuit transition. 

My feeling is that this carb rail might be bad. There is some indication that it may have gotten tweaked. The throttle lever where the cable goes was a bit bent as was the idle speed screw. The throttle linkage was sticking ever so very slightly, but some lubrication took care of that immediately while the carbs were off. My thought is maybe the whole rail has been tweaked such that it binds slightly when it's torqued onto the boots.


We can spend endless hours tweaking and adjusting to work around some problem, but it seems to me that finding a hopefully known good set of carbs to put on there is probably the most efficient use of everyone's time. They are not nearly as expensive used as I had thought.


 

buffalo the faq seems to only suggest shimming for the 600 if you have changed the exhaust (which makes sense). In a completely stock configuration (stage 0 on the 600 chart) they mention no changes.

I started at the back of this thread and worked my way to the front and these comments above have me thinking A/F ratio issue. Essentially some sort of lean spot that could be a clogged pilot jet (or whatever they call it in this type of carb), damaged needle...etc.  I know you say the carbs were rebuilt, but it's not always easy to get all the dirt out of those small orifices or notice nicks in slides, needles, etc.  

I say this because if it was some sort of mechanical hang such as cable adjustment or throttle shaft issues I wouldn't think they would be affected by pulling the clutch in causing engine drag....at least in my limited mind.

Here's what it's currently doing:

 

Summary to date.
It idles perfectly.

It runs perfectly once off the line all the way up to redline.

Obviously the first thought was a vacuum leak but for the life of me I cannot find one. 

The petcock had failed and the crankcase had quite a bit of gasoline in it so we changed the oil twice. 

The bike is completely stock as far as I can tell matching numbers with what I find the manual.

Here's what we've done:

. oil/filter change.
. new petcock.
. fresh fuel.
. now using a aux tank with a fuel filter just to eliminate the tank.
. new air filter.
. carbs separated, thoroughly cleaned (chemical dip and then heated ultra sound cleaner), reassembled with new gaskets, inlet needles, floats and carefully balanced using a calibrated carb balancer. 
(same result with second carb rail ordered from ebay, also cleaned, etc)
. floats set.
. clean jets. 
. we even tried a brand new set of oversized idle jets no joy thinking that it's a lean running condition.
. clean idle circuit bypass ports (as far as we can tell by pouring fluid through them)
. new floats and needles.
. carbs carefully balanced using an adjusted carb balancing tool.
. no vacuum leaks that we can find anywhere (tested using starting fluid sprayed over carbs, boots, etc)
. new plugs correctly gapped.
. valves adjusted.
. firing on all cylinders as best as we can tell using IR meter on the exhausts.
. compression of 150PSI on all cylinders warm.
. leakdown test done. First time ever so I'm not sure about these numbers. left gauge on leakdown tester set to 90 on a 100PSI source per instructions that came with it. Right gauge when cylinder at TDC read: #1 84. #2 86, #3 88. #4 85 (BUT we made the error of not eyeballing TDC instead of pulling the signal generator cover so we will repeat this)
. manual calls for loosening and retorquing the head bolts every 4000 miles. We tried but I'm not comfortable loosening the head bolts, so we just checked them for 28 ft lbs of torque per the manual. One bolt on cylinder #2 turned a slight bit, otherwise the others we tight.)
. resistance check on both coils within spec according to manual.
. battery voltage at 12.7V when the bike is off. 14.72 @ 5000 RPM (too high?)
. problem occurs at about 1/16th throttle opening so the slides and mains aren't even in play yet.

We haven't yet done a signal generator resistance test. And I can't for the life of me figure out how I'm supposed to verify the timing and advance on this. No inspection port that I can find or do you just do it with the signal generator cover off spewing oil? We haven't taken the ignition unit to have it tested since apparently there's no way to test it, but everything I've read says they are either good or they aren't.


Since we are relatively inexperienced it's always possible that we are doing one of the above steps incorrectly and that's masking the source of the problem. This whole time I've made the assumption it's going to be something stupid simple that I've overlooked. 

I would add that throughout all the tinkering and experimentation we have done the surge always occurs at precisely 1600 RPM and jumps to about 2600 RPM. Nothing we have done to adjust the fuel-air mixture has caused the jump to happen sooner or later in the RPM range. Similarly, it does not seem to be affected by engine temperature. It happens when the bike is first started as well as after it has been running for a while.

I posted a question on the Suzuki Bandit Forum:


Feedback from Facebook is that we're just missing a vacuum leak somewhere or possibly a clogged crankcase breather.  

Can you see the butterflies on the carbs while performing the test in the video?  Do you they open more once 1600 rpm's are reached and it spikes?  I know the movement would be small and difficult to see.