Weber DCOE butterflies won't close while engine is running

[Konish]

Fellas,

As I suspected in a previous thread, my butterflies on my DCOE 40s won't close all the way when the engine is running unless I forcibly pull the linkage up by hand. I have 2 really stout helper springs already. Valves have been adjusted and the carbs can be synched within a hair of each other on the synchrometer.

 

Just to be sure it wasn't the linkage arm binding, I disconnected the linkages and if I open each throttle individually, they exhibit the same issue. The throttle just hangs there and I have to yank them shut. The helper springs don't help as they are too linear and don't "yank" the plates shut.

 

While the engine is running, I can induce the rpm to "hang"...sometimes quite high by slowly coming off the revs. If I pop the throttle open and let it snap shut, it'll come back to normal (although not all the time...say 80%). Also, if the rpm hangs, I can turn off the car, restart and it'll return to the normal 1000ish rpm.

 

When the engine is off, they work perfectly...open smoothly and pause just long enough for the jet pumps to bleed down, and snap shut all the way to the throttle stops.

 

What the heck is going on? I can see the throttle plates definitely covering the last progression hole fully through the inspection port. Oh, and I'm running a Cannon intake manifold with the balance tube cast into the entire runner and linkage.

 

I suspect I'm going to have to drill some little holes in the throttle plates to relieve some of the vacuum in the bores(?)...

 

Oh, and here's what I'm running:

Mains 130

Emms F-11

A/C 170

Drains 40

Pumps 45

Idles 55F11

Starters 100F5

Chokes 30 (for my type of driving I was thinking about going to 28s if it'll help the dipping idle issue)

Needle valves 175

 

R/

Dustin

Edited March 22, 2010 by Konish

[Konish]

Okay, just thinking about this during dinner. Could it be that as I close my throttle, and the progression ports come back online (throttle plate nearly closing over the holes), the fuel flowing from either the second or last port is creating a hydrostatic wedge? I suspect the car is running rich, but man, that'd be a lot of fuel to create such a condition...?

 

I'm pretty sure the seal between the throttle plate and last p-port is air tight because I have been able to witness a little vapor coming out of each them when I was inspecting the position via the inspection cover...completely covered, no vapor...slightly open and I'll get a wisp.

 

Finally, in a moment of desperation, I completely removed the throttle stop screws and actually got the smoothest idle I have yet to achieve and the throttle was snappy and the engine returned to idle immediately. I thought I had licked the problem but it didn't take long for them to hang at a pretty high idle.

 

However, I did notice that even when the throttle was as closed as they could possibly be, I was still getting enough air flow to measure it on the synchrometer (say just under 2) and visually noticed a little light leaking into the bore prior to installation. I have read that a *comletely* closed butterfly (i.e. without throttle stop screws limiting the throttle position) would cause the air flow to stop completely. At any rate, the engine would barely run or die when the plates were closed without the limiting screws.

 

Man, this is just so frustrating!

Edited March 22, 2010 by Konish

[cygnusx1]

 Do you know for a fact that the butterflies are closing all the way.  The little tabs that rest against the stop screw can bend pretty easily.  If they are bent, they could still be preventing the throttles from closing.  Even with the screws backed all the way out.  Also, there could be some carbon build up in the bores preventing free movement of the plates.  Also, sticky accel pump pistons can cause the plates to close slowly or hang. 

Edited March 22, 2010 by cygnusx1

[Konish]

 Do you know for a fact that the butterflies are closing all the way.  The little tabs that rest against the stop screw can bend pretty easily.  If they are bent, they could still be preventing the throttles from closing.  Even with the screws backed all the way out.  Also, there could be some carbon build up in the bores preventing free movement of the plates.  Also, sticky accel pump pistons can cause the plates to close slowly or hang. 

 

Well, I guess I don't know for sure if they close all the way, but I can see the throttle plates completely closing the rear-most progression port. Actually, I can get the plates to pass the last progression port but the motor won't run with them in that position (or *just* barely). Actually, with the stop screws removed I can't pull them any more closed and the stop screw tabs are not touching the carb bodies, so I don't think there is any interference. The bores are clean as a whistle.

 

Pumps are clean, smooth and are operating correctly...I can actually hear them squirting fuel when the engine is off and I pump the throttle shafts. Obviously, when I first open and close the throttle, you can feel the gas in the pump chamber dampen the throttle a bit, but it bleeds off quickly and they close without any of the hesitation I feel when they are running. I've already had them off the manifold trying to diagnose this problem last month, and I was also visually able to verify that each squirter was working as well.

 

Even though I know the pumps are operating correctly, I'll check their operation tomorrow. Still can't imagine why they would stick when the car was running and not while it's off? So strange...

[RB30X]

Do yourself a favour a completely rule out a vacuum leak on your inlet manifold. Go over it with a piece of fuel hose and listen for a leak or spray some 'ether' or 'start ya bastard' around where manifold meets the head and listen for rpm increase. I had exactly the same problem you have described with my webers and played around with linkages and fuel mixtures for months and still had high idle until I pulled them shut. Then I removed them and used a new manifold/head flange gasket with sealer on both sides and OMG, my webers worked perfectly. I tried what I recommended above and didn't hear any leaks but there still was.

[Konish]

Should have updated this. I sealed up any suspected vacuum leaks, readjusted the throttle linkage from the firewall (there was a bit of binding at the extreme end of the throw and now it's as smooth as butter), and bought some really nice heim and turnbuckle linkages.

 

The binding contributed to about 85% of my high idle and the other 15% (which was the hardest to track) was for the slop in the worn out link arms that came with my carbs. I'm only talking hundredths of an inch (or less maybe) but it makes a HUGE difference in the car's air volume through the carbs and the idle speed.

 

Everything is working 100% now but I still believe that the butterflies on DCOEs won't close all the way on their own under vacuum even with no linkage attached and the throttle arm is manipulated by hand. It's like a storm door in an a house that won't close all the way while the A/C is running. Turn the A/C off and the door shuts under it's own spring power easily.

 

Even with everything running as perfectly as they are, I can tell there is a difference in the amount of force necessary to close them running vs stopped...it's not much but it's *just* enough to hold open the butterflies at a higher idle if there is slop in the linkage.

 

With everything set, I can now get the car to idle sub 1000 rpm...say 800 or so, run smoothly (i.e. my tractor-like idle is gone) and the revs return to idle almost instantly.

 

R/

Dustin

[Tony D]

Webers WILL shut with air flowing. As a matter of fact, look at the cross-section of the carburetors and you will see that vacuum/flow induces them to close due to the angle of the leading edge of the throttle plates.

 

Whatever you do DON'T drill a hole in the throttle plate. That is to fix another issue with them! Usually when a throttle shaft is bent a transition port is left uncovered, you can start with a VERY small drill to introduce more idle airflow to match the fuel coming from an uncovered transition port. This will get you running with a decent sized idle jet and give you adjustability with the idle mix needles. Generally BOTH barrels will ave the same angle, meaning both ports are opened in both barrels and both plates will need the same size orifice. Then you will need to balance airflow to that carb as it will have the highest flow of the three. In the opposite case, where the throttle plate is too far closed and tip in popping is occurring, it may be necessary to taper the top of the throttle plate near the transition port to let it 'come in' earlier to prevent transient tip-in backfiring.

 

Have you checked synch between barrels on the same carb, or just carb-to-carb? A bent throttle plate will give one plate to close fully, and the other plate will idle up the engine. If the idle speed stop is adjusted correctly you should be able to let the carbs close and the plates will close to almost watertight. If you have spooge in the barrel from poor air flitration it can hold the throttle plates open.

 

Also, in cases of high humidity and small throttle openeing it's possible the bodies are icing, causing the throttles to be held open. This goes away almost immediately on an L once you stop moving forward and the manifold heat starts warming the carb bodies, but it is possible.

 

In every case of high idle (not returning) it's been binding / incorrect linkage setup, or a bent throttle shaft. Usually the bent shafts happen on the Mikuinis as they have a tendency to have plates that stick to the bores and bend on quick opening.

[Konish]

Webers WILL shut with air flowing. As a matter of fact, look at the cross-section of the carburetors and you will see that vacuum/flow induces them to close due to the angle of the leading edge of the throttle plates.

 

Whatever you do DON'T drill a hole in the throttle plate. That is to fix another issue with them! Usually when a throttle shaft is bent a transition port is left uncovered, you can start with a VERY small drill to introduce more idle airflow to match the fuel coming from an uncovered transition port. This will get you running with a decent sized idle jet and give you adjustability with the idle mix needles. Generally BOTH barrels will ave the same angle, meaning both ports are opened in both barrels and both plates will need the same size orifice. Then you will need to balance airflow to that carb as it will have the highest flow of the three. In the opposite case, where the throttle plate is too far closed and tip in popping is occurring, it may be necessary to taper the top of the throttle plate near the transition port to let it 'come in' earlier to prevent transient tip-in backfiring.

 

Have you checked synch between barrels on the same carb, or just carb-to-carb? A bent throttle plate will give one plate to close fully, and the other plate will idle up the engine. If the idle speed stop is adjusted correctly you should be able to let the carbs close and the plates will close to almost watertight. If you have spooge in the barrel from poor air flitration it can hold the throttle plates open.

 

Also, in cases of high humidity and small throttle openeing it's possible the bodies are icing, causing the throttles to be held open. This goes away almost immediately on an L once you stop moving forward and the manifold heat starts warming the carb bodies, but it is possible.

 

In every case of high idle (not returning) it's been binding / incorrect linkage setup, or a bent throttle shaft. Usually the bent shafts happen on the Mikuinis as they have a tendency to have plates that stick to the bores and bend on quick opening.

 

Tony,

Thanks. I know the holes in the plates is to balance air between bores on the same carb (prior to the intro of the external bleed screw on carbs like the 152s), but since I had a slight synch issue, I thought it might help eliminate the hanging throttle as well. However, after seeing 3 sets (1 brand new, 1 rebuilt and 1 refreshed...2 of those were mine) and about 100 videos of them in operation, I still think Webers have a harder time returning to full closed throttle when running.

 

Both my sets close extremely positively when the car is off or the carbs are off the manifold. Linkages are all high quality pieces and adjusted correctly (no binding at all). I just observed my latest set closing relatively slower near the end of the arc with the helper spring removed. Now they did close all the way by themselves, but it for sure wasn't the positive "snap" they have when the motor is off. Obviously with the helper, they close quickly and positively.

 

On my older set, with just the throttle arms, the carbs tend to want to "fall" open while running but still need a nudge to close. I had one incident where one of the linkage arms was still connected to the throttle lever while I was adjusting the something on the upper linkage. The link arm fell and with just the weight of that one link arm hanging off the throttle lever, it pulled the throttle open quite a bit...enough to make the motor race about 1000 rpm. No way could the weight of one linkage arm do that with the motor off or the carb off the manifold. I was shocked that it was so easy to open the throttles just with the car idling...but it was and is.

 

I know the leading edge of the butterflies angles towards the direction of flow, but it also has the trailing edge that angles away from the direction of flow with the pivot point splitting the leading/trailing edge equally. I won't even try to argue fluid dynamics, but I've also seen a few videos with the same exact issue (car off, they close great, car running, the don't), so I can only go off observations.

[cygnusx1]

 Here is another approach.  What is your idle timing?  You can reduce your idle timing to get the RPM's down a little at idle so that you don't have to have your carbs darn near shut at idle.  Timing too far advanced at idle will cause a host of issues with the DCOEs 

Edited April 18, 2010 by cygnusx1

[AZhitman]

Bringing back an old thread...

 

I'm running into the same issue with my DCOE 42's on a built L20b.  When I pulled the carbs to check the "gaskets" (I'm running the anti-vibration isolators), I held the carb up and looked down the bores - There was a sliver of light coming past the throttle plates on one side. 

 

Seems the throttle shafts were adjusted too far to one side or the other, which leads me to believe I may be missing something in the tightening of the outer nuts on the throttle shaft, causing the whole internal assembly to pull slightly to one side of the bore or the other.

 

Once I got them adjusted, it ran fine for a while - However, after a few sessions of hard throttle open/close, the problem has returned.  No vac leaks anywhere - I've checked over and over and over.

 

Anyone else run into problems with this, and how can I fix it permanently?

[Leon]

A dead-reliable linkage and stiff internal return springs seem to solve most of the issues. There are some good threads on classiczcars.com about it.

Edited July 16, 2013 by Leon

[Tony D]

Agree with  Leon, many of these carbs are old, and have worn out return springs. I was amazed at how definitively a new set of 44IDF's snapped shut compared to some I bought out of the JY in Japan years ago and which sat closed for 15 years... The internal springs are more than enough to close the individual carbs.

 

The problem after that (weak internal springs) is from the linkage being all synched together and something binding. A small square is your friend to set up your linkages to make sure they push STRAIGHT down and allow return STRAIGHT up without side-loading anything. The original linkage off the firewall was pretty much a weighted over-center design and which would snap closed reliably. But people 'fix' it and it starts binding. Keep it straight, the ball at the firewall CLEAN and properly lubricated, same for the pivots for the pedal bellcrank. 

 

The choices for spring location are limited, and in many cases putting a spring onto the linkage at an angle COMPOUNDS binding. I had a DREADFULLY heavy pedal and binding issues. After getting into someone's car with NO return springs, and FLAWLESS return to idle I had to rethink the 'put a spring assist on it' thinking line. I went back and cleaned everything, made sure it was all STRAIGHT and there was no angular side-loading... and damned if it didn't start working lightly like originally designed!

 

Since then, I've gone to throttle cable which was more of an 'interchange quickly' decision, and have no return springs on the carbs whatsoever other than their internals, and the single Mikuni Return Spring on the crossbar. That spring was NOT used when I had a mechanical linkage, the overcenter of the original design seemed to be more than enough to close the throttles on their internal springs. I had to make my cable come STRAIGHT off the quadrant because having it at a 5deg angle (looked straight enough...what could it do?) from dead straight and it started having return to idle issues. That got the Mikuni spring reinstalled...but to unsatisfactory final effect. That's when I put the new fabricated bracket on to get it 'straight-straight'... In retrospect it probably would have been easier to slip the quadrant 2-3mm to line it up... Everybody has a Homer Moment! D'OH!