'83 Slow return to idle when the throttle is released from 3K-5K RPM

[pullshocks]

My 83 starts great, idles great, makes great power when you get on it.  It does require careful throttle technique off idle to avoid stumble off idle, but I do not find that a major issue.  My problem is this:

When I am taking it easy, shifting at 2500-3000 RPM, the throttle drops back to idle quite nicely.  When I gas it up much above 3000 and into the 3-6 K range and release the throttle to shift, the RPM is very slow to drop.  It drops slowly  to about 3500 and stays there for several seconds, and then finally drops.  This is with my hand completely off the gas.

It is kind of nerve wracking and I can't really use engine braking to regulate my speed.  Rideable, but not optimum, and could end up being unsafe in some situations.

It has been this way for quite a while.  Trying to anticipate your questions, here is some more info.

The throttle cable is properly routed per factory service manual.  The throttle snaps shut, there is no sign that it is cable-related.

The effect is not as pronounced with the bike on the center stand, revving it in neutral.

Right now the YICS is off and the ports capped. 

I have checked extensively for vacuum leaks, and recently replaced the throttle shaft seals.

The carbs are synched at idle.  As the RPMs are increased, the fluid in my U-tube gauge moves toward the the front cylinder, indicating it has stronger vacuum at higher RPM.  Valve clearance was set a few hundred miles ago.  A cold compression test last summer indicated the front has lower compression than the rear.

The flapper moves as suction is applied.  I do not know what other tests can be done.

The flapper vac line is correctly connected.  The little vent filter is correctly connected.

I tried adjusting the floats to lower the fuel level a bit.  The engine seems slightly more of a stumble character accelerating gently at low RPM, but the slow RPM drop is as bad as ever

I am going to try synching at a higher RPM and see if that has any effect. 

Any other ideas would be appreciated.  I really do not understand how the various carb circuits and the air box system work.

[cvincer]

So,   you are at 5000rpm & go to change gear,  clutch in & back off on the throttle but the revs remain high   .... at that point in

time (before you have let the clutch out after changing gear), can you move the throttle backwards & forwards  1/4 of an inch

or so without the revs altering ??

[pullshocks]

I will try that and let you know.

[Rikugun]

A little history might help the diagnosis... How many miles and when did this start?  Did the problem develop over time or develop suddenly after some routine maintenance or upgrade? Have you checked compression? What does the engine idle at when warm?

[pullshocks]

Just turned over 20,000.  I bought it 6 or 7 years ago with 12,000 non running and after installation of a Keyster carb rebuild kit and many carb cleanings got it going.  It has had numerous episodes of carbs gunking up and needing to be cleaned (primary jet). 

I do not remember if it was ever free of this issue, probably not.  Since performance is otherwise excellent, I don't think I recognized it as an issue in the first few years of riding.  But I can remember being aware of it on a memorable ride in the fall of 2011

The change of sync with increasing rpm has been present since the first time I synched the carbs.

Compression is 125 front 138 rear (cold).  This was taken last year after adjusting the valves http://ridersofvision.net/rovforum/index.php?topic=14397.msg130728#msg130728

It will idle a bit below 1K.  I have it set it for about 1250.
With the bike pointed up hill, the idle goes up 1 or 2 hundred, pointed downhill it goes down by 1 or 200.

Recently I corrected a vac leak through the petcock.  Slow RPM drop not affected

I wondered if a too-high fuel level in the float valve could cause it.  Adjusted the floats.  Slow RPM drop not affected.

I wondered if there was an air leak through the "accelerator control valve"  I capped that line.  Slow RPM drop not affected.

[pullshocks]

Today I confirmed that the throttle plates are closed fully to the idle position while this is happening. 

So I would have to think air/fuel mixture is getting into the engine by some other route.  Noting that the primary circuit has a small hole below the throttle plate and the cold start circuit has a large hole below the throttle plate, these seem like possible pathways.

I checked the rubber boots on the cold start circuit plungers.  They were slightly loose.  I replaced them with some spares that fit snugger.

This did not eliminate the slow RPM drop but it did change warm up behavior drastically.  I did get some very eratic behavior during warmup with the "choke" on—engine speed surging to 3K+ but but when I tried moving the cold start lever very slightly, the idle would suddenly drop to below 1K.  Very notchy, all or nothing response, very different than what I normally experience.

I tweaked the sync.  On the stand it seemed to be slightly better with the sync biased to the rear.  Out on the road, behavior was the same as yesterday.

Last thing to report is I checked the float bowl fuel level.  Front was right on 20 mm,  but I guess I overdid it on the rear as it was way low—24 or 25.  Correcting that will be at the top of the list next week.

[pullshocks]

Yeeee-Haaaahhhhhh!!!!!  Got it!!!!

I knew the fuel level in the rear carb was low, and after hours of staring at the carb diagram, I realized that this was probably the source of my problem.  The other vacuum leaks werent helping matters, but the float level was the issue.

Made that adjustment, and a test ride confirms RPM now drops right back down

Here's why (I think):

The fuel level was below the inlet of the pilot jet, so instead of picking up fuel, it was allowing fuel laden air  from the float bowl head space to be pulled in to the pilot circuit by the strong vacuum below the closed throttle plate.  Sort of an inside-the-carburetor vacuum leak, with air coming in through the vent tube, which is there to keep the float chamber at atmospheric pressure.

I need to spend some more time on the mixture screws and synch, but wow.  This is great.

[fret not]

Congratulations on your triumph!  It always feels good to overcome an obstacle.

[The Prophet of Doom]

Well done.  There's nothing quite as satisfying as properly diagnosing a problem properly - a task made harder by the absence of a diagnostics checklist.

[pullshocks]

Here is my fuel level check rig.  While the carbs were out of the bike I carefully laid out the reference line 20 mm below the top of the carb body

[Hartless]

glad to see that you got it figured out. I was going to suggest that if the fuel screws aren't adjusted right it can cause similar symptoms.. too rich and it can bog down, too lean and it will slow to idle...slower...lol if that makes sense. I guess for future reference :-)

[devotee]

Nice troubleshooting! What's not to love about a manometer?

[Rikugun]

Nice troubleshooting! What's not to love about a manometer?

Yes, they are nice but I'm not sure if this technically qualifies as one? The tubes aren't interconnected but rather go to the carb bowls. Apparently he lacks X-ray Vision  8) so has employed a visual aid leveraging the fuels need to seek it's own level thusly providing an external representation of where the fuel level is in the two carb bowls.  :D

Well done.  There's nothing quite as satisfying as properly diagnosing a problem properly - a task made harder by the absence of a diagnostics checklist.

Is this just a comment or is there a challenge in there too!? I guess when pinholenz is done with his carb videos he needs to also create diagnostic flow charts??  :laugh:

[pullshocks]

No I do not have xray vision.  Far from it--until I did this, I have been running blind, thinking my fuel level was correct when it wasn't.  2 years ago I set the fuel levels  with the carbs off the bike, which apparently did not translate to correct fuel levels with the carbs on the bike under use conditions.  Checking on the bike is the better way.

Rikugun is correct, the tubes are separate.  This is an adaptation of the one-carb-at-a-time fuel level check in the factory manual.  The factory method is difficult to impossible to do with the fuel tank in place, since the petcock and tank hide the rear carb.  My little gizmo makes it possible to do it under use conditions.

Once I got the fuel level correct, the slow RPM drop was drastically improved but not eliminated.  After readjusting the pilot screws, it is gone altogether.

Interestingly, the pilot screw adjustment sort of worked with the low fuel level, since gasoline vapor was coming up the pilot jet.  But now that the fuel level is correct and the pilot jet is drawing liquid, the pilot circuit is working more consistently and the adjustment is more defined.

As a result, idle is much smoother and more consistent.

Now, to re-check sync.  Lets see, shall I do it at idle, or 2500?

[Rikugun]

Now, to re-check sync.  Lets see, shall I do it at idle, or 2500?

Try it both ways and use the method that works best for you.  :)