Engine run on with a new engine

[silasmoon]

Howdy, so I recently had my whole engine rebuilt. 292 cam, new 38/38 DGES, E12 Head, new intake manifold, 9.5:1 compression, 123+ Tune distributor. I put about 120 miles on it before we realized the cylinder head had a crack in it, weeping engine oil. That's being replaced and installed for free to my mechanics immense credit and reputation. But while I lick my wounds I need to keep myself distracted. It did have some serious run-on / dieseling issues and I have been doing a lot of reading. Here are the symptoms:

 

• Engine runs on intermittently after being driven around a bit
• Smells rich
• Pulled the plugs (BP6ES) and they had carbon build up on the rim, with a dark wetness in the threads, smelling of gas. 
• Car runs great otherwise, pulled strong, etc.
• Experienced a slight stumble from time to time when I was really on it and then slowed in 3rd gear.

 

How I tried to fix it:

• I went ahead and set the timing by physically moving the 123+ while a friend kept her at 3,000 rpms and I timed it to the bouncing ball. 
• Got the idle speed screw set to 900 rpms and she idled well enough
• The mix screws had little effect, but I didn't want to futz with jets so early on into the break in period. 
• The manifold's vacuum ports were plugged. 
• The 38/38's manifold vacuum port was plugged. 
• Octane boost - I live in California and run 91 octane (which I honestly thought would be alright)

 

Here are my suspicions thus far:

• Vacuum leak somewhere - possibly at the carb to manifold mount?
• Float bowls weren't set properly, as they weren't checked out of the box - possibly explaining the rich mix?
• The rich mixture had left carbon deposits on the plugs which glowed hot.
• The machine work left something to be desired, and perhaps there was a shaving acting as a hot point?
• The nature of the crack had let engine oil into a cylinder? 
   

I know this all seems moot since it's being rebuilt, but take pity on me and let me at least talk about 2002s. 

 

Edited February 25, 2017 by silasmoon

[TobyB]

It's not moot-  the problems you had before will still be there when it's back together- something to look forward to!

 

I like numbers.

#2- yes, start with the float level,

and with fuel pressure

#c- the idle mix screws not doing anything usually means the throttle's too far open,

and you're running on the transition circuits  That's not trivial to fix-

timing, air leaks, fuel level all play into that together.  The throttles being too far open

will certainly contribute to run- on

#g- did the octane boost make a difference?

#3- yes, carbon build up will certainly make a run- on condition worse, especially when the tthrottles are a bit too far open.

#4- no, that would burn off pretty soon and it's hard to leave odd shavings in the head... unless you're Metric Mechanic, and even then, it doesn't seem to bother.

#5- probably not, as that would only affect one cylinder.

 

Other things:  It's really odd to have a crack let oil leak into a cylinder.  It only passes #4, and that, not so closely.  Water's by far the more usual problem, as it surrounds

the combustion chamber AND is corrosive AND expands when some doofus doesn't use coolant.

 

These engines do tend to run- on when the timing and throttle closure aren't correct.  Advanced timing at idle seemed to do it with my car on

a standard dizzy- I wasn't getting as much advance at 4500+ as I wanted, so I advanced the whole curve, and run- on got a lot worse.

But Idle suffered, and if you're idliing smoothly at 900, that wouldn't be the first thing I'd look at.  That said, you should be able to dial in your 123

to retard idle a bunch under 1000 and see if that helps- or hurts.

 

hth,

 

t

[silasmoon]

Thanks Toby for the thoughts. I forgot to mention I have a mechanical fuel pump now with a pressure regular, that I believe is set to 3.5 PSI, but I would need to double check that.  That octane boost didn't make a noticeable difference, poured half a bottle into half a tank and it still ran on. I didn't try it again, but I still have half a bottle left. I checked the curve that my tune is set to and it shows:

• RPM / Degree
• 500 / 0 
• 1000 / 0

I am confused as to whether I should be setting idle RPM by hand or by changing the curve in the actual distributor app. 

Let me go into some more details of things I have done. 

• Initially when I noticed the dieseling, I suspected advanced timing, so I retarded it a smidge at idle by ear and got a very low idle - 650 rpms or so and it seemed to help a touch. Maybe 50% of the time it stopped dieseling. I then upped the idle speed screw to 950 rpms. 
• The next day I set the timing with a gun - bouncing ball in the hole at 3,000 RPMs. I did this by physically twisting the distributor. 
• I also adjusted the electric choke to have weaker tension on the choke plates, as it had been causing my car to idle at 2,500 - 3,000 RPMs after I had let it sit for 10 minutes. The electric choke now has my car idle at about 2,000 rpms for a few minutes. 

Other than the float-bowl adjustment, I think you're on to something with the throttle closure issue. Weber says the idle screw shouldn't be screwed in more than 1/2 a turn. I can't remember how far in it was, but I bet it was more than 1/2 a turn. This would note that the carb is too lean, which according to Weber, can show as being too rich which would explain my plugs. Seems to be next steps are set the float-level, check all is tight and plugged then calibrate the carb with a jetting kit at the ready. 

Edited February 26, 2017 by silasmoon

[avoirdupois]

My memory tells me the starting point for Weber idle mixture screws is generally 1 1/2 turns from closed.  Regarding jet sizes, I wouldn't hesitate to change them - if I suspected they were the wrong sizes - even if the engine is not fully broken in yet.  Running too rich or lean is not good for any engine - including something new.  The notion that your fast idle is any more than 1,500 rpm for minutes suggests your choke and/or fast idle cam, could benefit from some adjustment or something may be sticking.  Otherwise, this may be an issue related to throttle plates that are not fully closing at idle - and as Toby mentioned - an engine idling off of the part throttle circuit rather than the idle circuit.  (The 1/2 turn you mentioned would also seem to indicate not running off of the idle circuit - or the idle jet/s is/are too large.)

 

SInce you mentioned a cylinder head crack, was this preexisting or something that occurred during your first few miles?  Was overheating involved?

 

The spark plug pictured does not appear particularly abnormal considering the limited amount of driving time.  It suggests possibly a rich condition at idle and a leaner condition at higher revs (that burns off much of the carbon created at idle or off idle).  If the plugs continue to have the same or similar appearance, you might consider stepping up to a hotter plug, i.e., BP5ES - although that may be contraindicated by a run on condition.  Finally, the one thing you did not mention is whether your carburetor is equipped with idle solenoid.  A properly functioning I. solenoid should prevent run-on.  If, however, an engine is idling off of part throttle, it is defeating the solenoid's purpose.

 

HTH.

[silasmoon]

Interesting, really wish I had the carb in front of me now :0 - The cylinder head crack occurred in the first 100 miles I believe. The engine never seemed to overheat according to the temp gauge. I didn't drive it very hard and its been cold. It doesn't have idle cut-off solenoids, and I would consider them, but fear they're just hiding the root cause. 

[Simeon]

A couple of points here. Timing doesn't have much effect on dieselling as the ignition is switched off so by definition the ignition won't be firing by the normal process. What you do find is that incorrectly set ignition means that the throttle plates are too far open at idle to set a reasonable idle speed. The exception to this is if you have a 'box-o-transistors' ignition set up which may continue to be powered after the ignition is switched off through small backfeeds into the ignition.

 

If you have hotter than usual plugs this can cause the ignition. Temps can be raised for a lot of reasons, including timing but wouldn't show itself without the additional air and fuel available from the open throttle plate(s). 

 

The car needs a good tune plus the jetting being looked at. You talk about timing to the ball at 3,000 rpm. You don't explain what curves you have in your 123 but that sounds really retarded. The ball is 25 degrees BTDC. Typical timing would see you shooting for 36 degrees BTDC at 3,000rpm. This is likely to see your initial advance really low at idle hence needing to crank open the throttle plates and get into the transition for idle. 

 

Give us a screen shot from your 123 app and we will see if we can get this running. 

[silasmoon]

Hey Simeon, thanks for the advice!

I have been reading up more on the 123 static timing setting, and some people never touch it and simply set static in the app, and some actually move it to establish idle / static timing. I realize that if I set say 6 degrees at static, I would need to add that to each degree of advance on the curve. So it seems that 1,400 rpms / 25 degree advance is where I should be seeing the ball? 

Here are the current settings:

 

[silasmoon]

Looking at that curve it seems like 3,000 RPM is 22.5 degrees, so I must be running about 2.5 degrees of static timing to be seeing the ball. Is that correct? I guess this curve is the default and is pretty retarded. 

Edited February 26, 2017 by silasmoon

[Simeon]

That curve is all kinds of wrong. When entering new values in the table, click out of fields that you change to make sure that the text change is recorded and then press save. 

 

Your max advance should be about 36 degrees BTDC at 3000rpm. Try an initial advance of 14 degrees BTDC between 0 and about 1,050-1,100 rpm and then just let it rise up in a linear manner to 36 degrees BTDC at 3,000 rpm. If you add an additional point at 25 degrees at a speed not too far off the curve (maybe 2,200?) then you have a figure to use when setting the timing to the ball. In theory you don't need to do this but it does allow any play in the ignition and timing chain to be dialled out plus any variation in setting the static timing at TDC using the 123 light. 

 

The vac curve is OK for a start but to get it working for you you need to work out the appropriate points on the x axis for the vacuum signal from your engine. 

[silasmoon]

If the engine was heavily retarded wouldn't that cause it to run very hot? Hot enough to crack a head? 

 

When I picked it up I was told the timing was set to the Tii curve. What's the Ti curve? I feel like that would be a good baseline. 

[Simeon]

The TII and the TI shared the same distributor but the TI had (in some markets and configurations) slightly more static advance.  The TII distributor is not necessarily a good curve for a carbed car. It has less total advance than you may use with carbs (28 degrees max) and comes in quickly. The performance prescription for this distributor from the likes of Jeff Ireland is to add more static advance than you would when running on a TII. You do this by timing it at 36 degrees at 3,000 rpm max advance (which is where I got the figure above from). The initial advance (I.e. Advance at idle) then 'is what it is' as we are less concerned about idle on a performance engine. Of course, with a 123 you can set this to be whatever you want. 

 

Have you read this thread? Examples of curves are included. 

 

 

[silasmoon]

Ah I had never seen that thread before, despite hunting around for 123 threads. Thanks for the tip! 

[AustrianVespaGuy]

Is the 'L' light on the dashboard lit when it's dieseling?  On my EDIS setup I had this problem because the current from the charging circuit (through that light) was enough to keep the EDIS system alive.  Same thing might be happening to you with the 123 setup.  It's also enough current to keep a relay energized, in case you put in a relay somewhere in your ignition circuit. My main point is double check how your ignition is wired, as you may still be getting spark as opposed to true compression-ignition.

[silasmoon]

So I got my engine back finally and had put about 20 miles on it before the L light came on and the battery died. I jumped it, worked fine, drove on the highway for 20 minutes and it died again after getting into stop and go. I got it to the house, put it on a tender for 2 hours and came back. The alternator wasn't charging the battery. Wires on the alternator were touching. Odd. Moved them apart, the L went off, and the alternator was charging again. Drove it around the neighborhood. Turned it off to adjust the idle speed screw, and when we turned the key to fire it up we heard a large clunk. Then nothing. No lights, no dash lights, nada. So we tried to jump it with a battery pack, no dice. Then we tried to jump it with another car and the cables started to smoke. We triple checked the orientation of the jumpers and they were correct. We tried bump starting it as a last ditch effort but nothing. Turns out the starter (randomly?) died and almost burned down the whole car. Had it towed back to the shop and was given a new starter, grounding wires to the alternator and starter. Works fine and I have since put 300 mi on the engine. I find the whole electrical debacle rather confusing, and have no idea what was to blame. 

With that said the run on is gone after installing a slightly thicker head gasket during the (re)rebuild which dropped the compression down to 9.3:1 or so from 9.5:1, and stole probably about 5 HP.

I have put in a pretty mild timing curve based off of Simeon's suggestions. I have a MAP curve as well, but didn't hook the carb up to the distributor vacuum port yet. I see the bouncing BB at around 2,200 RPMS. I do notice a bit of a vibrating sound coming on at 4,000 RPMs if I go WOT. I don't exceed 5,000rpms since its a new engine. I can't tell if it's pinging, it sounds like a light chattering noise. It concerned me a lot so I retarded the engine to the attached curve and drive like I'm with my mother. 

 

Final side note - the 123+ Android app is pretty garbage. I am a software engineer by trade, and I was appalled at how poorly it works. Constantly crashes, buttons don't work, I can save an edit about 10% of the time. Really - REALLY - hoping they fix it. I had to use my girlfriends iPhone just to set a timing curve.

Edited March 30, 2017 by silasmoon

['76mintgrün'02]

I am just curious.  

You said "bouncing BB".  Is the BB steady under the light?  

 

A bouncing BB generally means erratic timing.

(I would think it would be nice and steady, with the new 123).