Yet another "erratic at cruise"

Joined
Jun 15, 2020
Messages
118
Reaction score
102
Location
Køge, Denmark
I am experiencing erratic firing in my Imperial at cruise with a very light foot on the throttle. There are probably a million threads about this topic, but I could not find any with a precise match.

It only occurs when the engine is hot. I.e. after 10-15 min of driving.
After installing a wideband AFR gauge I realized that the misfires occurs the same time as "jolts" of lean condition. Cruising at AFR 14-15, and then jolts/spikes of AFR 16-17 lasting just long enough for the gauge to register it. It happens every few seconds.
The car accelerates fine if I step on it. I can not recreate it in neutral.

It is a stock '66, 440cui, automatic transmission, carter AFB 3856S (not exactly stock, but same CFM), new points, dwell set at 30, new coil. I have cleaned the carb and set float heights and float drop. It has a fuel pressure regulator and electric fuel pump. Last weekend I changed the gasket between carb and manifold. I have not been able to find other vacuum leaks.

A have a few more ideas of the cause, but I would really appreciate any inputs or links to old threads from this knowledgeable forum?

The suspicions I have left:
1. If the vacuum canister spring is weak and advancing too much causing a knock - but would that show as a lean spike?
2. If the step up springs are too weak - but I would think they should still be in the jets at light cruise? The carb has the 3 step - step up rods.
 
Might need a scope to watch the spark plug firings and see if something happens on the scope when the afr spikes. Are the afr spikes always on the same cyl or does it vary (according to the traces on the scope)?

What timing at hot base idle? At 2500rpm (total vac + centrifugal and also just the mechanical centrifugal advance by itself)? This is where a "dial-back" timing light can come in handy.

What on the carb is "not stock"? Primary jets? Metering rods? Power piston springs?

Just some thoughts,
CBODY67
 
You need to check the total advance to see if it is excessive, but unless the engine is also pinging, I doubt that this is a timing issue. The after market AFB carbs are set a good deal leaner that the factory units. I suspect that the metering rods generating a 14-15 AFR is too lean to start with and is causing the spike to 16-17 under load. At the top of the metering rod assembly on an AFB there is usually a small screw to set the height of the metering rods. You can take the screw loose about 1 turn and raise up the metering rods. This will fatten up the mixture to see if that cures the dead spot. Be careful not to drop the small screw down the carb throat. I usually take the carb off for this operation as it is easier to do the adjustment that way and if a screw falls out, at least it does no go into the engine. Usually moving the rod height about 1/32" will suffice. If the carb does not have adjustable metering rods, you will need to source an aftermarket jet package to increase the mixture settings. I would shoot for a base AFR of 13-14 as a starter.

Dave
 
It is fairly simple to check your vacuum canister rule and rule it out if you have a hand pump vacuum gauge and timing light. First put the hand pump on the canister, pump it up to around 20” of vacuum (you should also see your linkage move) and make sure it keeps a vacuum. If the vacuum drops your diaphragm is bad. If it holds you can then check to see if it is advancing close to spec.
Plug your vacuum advance line, Hook the hand vacuum pump to the canister. Start the car and turn on the timing light. Start pumping up the vacuum until you see your timing start to move. Take note of the vacuum it started to move at an check the specs in your service manual. if you have a dial back timing light you can be a little more precise to see the exact timing from the vacuum advance.
Mechanical check is similar with a tach and timing light. Again a dial back timing light helps. Plug or pinch off your vacuum advance. Start at idle and slowly increase your rpm until you see your timing advance. Check the rpm with the advance spec in the service manual. There are a couple of different rpm ranges. Remember the spec in the manual is distributer speed which is 1/2 your engine rpm.
 
Thanks for the replies. It sounds like I am on the same track as you would be.

The carb is not standard to the car, but from what I can find online, it is original to a '65 383cui. I.e. it is not an aftermarket carb, but not right for this car.
It has "16-160" metering rods (forgot to note the jet #:BangHead:), which has one more step than the metering rod kits I can see online. Will those work anyway?
The step up rods are unfortunately not adjustable.
If I can find original step up rods, what would be a good alternative to fatten it up a little. 16-164? It is 0.002" smaller in all 3 steps.

step-up-sizes.JPG


Because the distributor is a CAP version, the base timing is set to 5°ATDC (!) as per the FSM. The vacuum canister is not leaking and it is not adjustable (round top).
I actually have another distributor on the way (Not a CAP version), but I still chose to start a thread - in case my issue was easily identified as something else. It can take a long time to get parts over here (unless you are willing to pay 3X), and I don't want to miss half of the driving season waiting for parts.

I will make curve of the vacuum advance and centrifugal advance as suggested.

The lean "spike" I see is not a lean "spot" as people normally describes. Meaning: I will be driving down the road at AFR 14.5, and then without moving my right foot it will have a lean "spike" for a half second, then return to normal for a few seconds, then spike and cough again... and so on.

Wheather it is pinging or not - I am probably not very good at judging - as I have been driving diesels for 20years :rolleyes:
 
AFBs were all 2-step rods, AVS were 3-step rods. I put AVS rods in my '67 AFB and could tell no difference in overall performance. One of the main differences in the 383 and 440 AFBs and AVS carbs were the main jetting, primary and secondary. It takes both rods and jets to make things right, usually. IF the rods are correct for the current carb number, then changing the pri and sec main jets might put it pretty close to what it needs to be?

A metering rod carb can probably be more finely-tuned by using a basic jet and then using rods to make the fuel calibration to what it needs to be at various operational situations. Then, all of that can be staged with the power piston springs.

For example, use the "Eco" size to meter cruise and the "Pwr" size to meter WOT, with a spring that lets the Pwr section be where it needs to be at about 5"Hg. The bad thing about the FSM is that they do not decode the rod sizes. But a dial caliper can do that, too.

As for a temporary fix for the jets, a small drill bit can usually take care of that. Seems like 440 carbs usually had a .101 jet diameter? FWIW

Enjoy!
CBODY67
 
Let's first check on how loose the timing chain is. After that is checked and very likely replaced try 10-15 degrees of initial timing with a 195 degree T Stat. 34-36 total.

At part throttle cruise and at full advance with vacuum [50-56 degrees total timing] 16-18:1 afr should be fine if not higher depending how goldilocks everything else is.
 
At part throttle cruise and at full advance with vacuum [50-56 degrees total timing] 16-18:1 afr should be fine if not higher depending how goldilocks everything else is.
16-18:1 afr is where the Lean Burn system is, which requires lots of spark advance to get the overly-lean mixture burned in time for the piston to be at TDC. Which is much leaner than the 14.7:1 ideal afr for normal engine calibrations. FWIW

Enjoy!
CBODY67
 
I got a little further in my carburetor research.

I read on one of the carb rebuilders' website, that AFBs had 3-step rods up until '68, and 2 step from then on.

If my car would have been a non-CAP car it would have had an AFB 4130S, with 16-404 step up rods.
The 3856S carb it has now is original to a '65 newport or 300, and would have had 16-165 step up rods.
These two carbs have same size venturi and throttle bore diameters. Original primary jets are 0.089" in both carbs.
The 16-160 step up rods in the carb now are "one size lean" according to the '65 factory service manual. This also corresponds to the Carter table posted above.

Upon rechecking the vacuum advance curve, I found the canister leaks between 0 and 5" Hg. More than 5" Hg; it holds vacuum (?). This would mean, the vacuum advance falls off to soon.
I will post the numbers/curve later, but it did not advance as much as it should (and I am aware of the distributor vs. crank degrees :))

It does indeed run with AFR 16-18, at light cruise, so I am happy to be reassured that this would not be problem in it self. This weekend after discovering the vacuum advance did not advance as per the FSM, I tried advancing initial timing 5°, and it seemed to have lessened the erratic moments.

When I install my new distributor, I will try to measure the timing chain slack. As far as I can tell, the engine has 54000miles on it.
If it was easy to obtain a set of 16-404 step up rods, I would do that, but for now I will wait on the new distributor, and see how the car behaves.
 
Thanks for the update. Per the '65 FSM, or the '66 and '67 FSM, the AFBs all had two-step rods. The three-step rods started to happen on the AVS carbs. As I recall, most vac advance curves need about 9"Hg to start the vac advance.

Remember, too, that everything on the Internet, even on some websites, might not always be accurate. Which is where places like www.mymopar.com, www.jholst.net, and others come in handy with their official Chrysler-produced information. FWIW.

Additionally, the as-produced calibrations are for "sea level" elevations. Leaner calibrations are needed at higher altitudes to keep the carb mixtures near to what is needed at those thinner atmospheres. IF the spark plugs' ceramic starts to look chalky, that's an indication that you need a richer calibration. Lean is good, compared to rich, but too lean can cause durability issues with spark plugs and pistons, by observation.

Enjoy!
CBODY67
 
Thanks for the update. Per the '65 FSM, or the '66 and '67 FSM, the AFBs all had two-step rods. The three-step rods started to happen on the AVS carbs. As I recall, most vac advance curves need about 9"Hg to start the vac advance.

Remember, too, that everything on the Internet, even on some websites, might not always be accurate. Which is where places like www.mymopar.com, www.jholst.net, and others come in handy with their official Chrysler-produced information. FWIW.

Additionally, the as-produced calibrations are for "sea level" elevations. Leaner calibrations are needed at higher altitudes to keep the carb mixtures near to what is needed at those thinner atmospheres. IF the spark plugs' ceramic starts to look chalky, that's an indication that you need a richer calibration. Lean is good, compared to rich, but too lean can cause durability issues with spark plugs and pistons, by observation.

Enjoy!
CBODY67

I agree; one should be very careful about random statement made by random persons on the internet. Including statement made by myself :). However in this case the FSM supports, that step up rods are indeed 3 step rods and at the same time 2 stage rods. Actually I think it is very confusing, how they mix up the nomenclature in the table of specs vs. figure texts.
I believe the 2 stages refer to the step up pistons having a short build in spring, hence: 2 stages: The long step up spring also found in the Edelbrocks, and the short spring within the piston assembly. So 2 stages (springs) enables the piston to switch between the 3 steps.
I had to skim the AFB chapter twice to realize what they meant. Pictures from the '66 FSM below.

Good point about the elevation. I tend to forget that this country is completely flat and close to sea and the rest of the world is not. I live less than 2 meters above sea level. The last place this car was used, was in Lake Havazu, Arizona, which is 700 feet above sea level. Could this be the explanation why it runs "one size lean"?

This figure from the FSM gives the explanation - the difference between steps and stages.
Capture_2stage_3steps_figure.JPG


In the table of specs they confuse us, because they list the number of stages and not the number of steps....:
Capture_2stage_table.JPG
 
Thanks for your supporting evidence. Here's another orientation to what I mentioned. I'm only concerned about the number of different machined diameters on the metering rods, not the piston that runs them. A two-diameter AFB rod would be a "2-step" rod to me, that goes with the "old style" short metering jet. A three-diameteLet's first check on how loose the timing chain is. After that is checked and very likely replaced try 10-15 degrees of initial timing with a 195 degree T Stat. 34-36 total.r AVS rod would be a "3-step" rod to me, that goes with the "new style" tall metering jet.

I do know that my '67 Newport's AFB (383, federal emissions, factory a/c, automatic trans) had the correct (per the FSM) 2-step metering rods and short primary main jets in it when I bought it used in 1981. At that time, I also had a 1970 Monaco still had the orig AVS with the correct (per the FSM) 3-step metering rods and tall primary main jets in it when I bought it in 1975. I bought one of the old Carter Strip Kits (a selection of jets, rods, springs) for the AVS, to see if what I'd read was really true (about fuel calibrations and such). After I "upgraded" to a 9801 ThermoQuad, I got a Strip Kit for it too.

These are my experiences. Yours might vary.

Enjoy!
CBODY67
 
Changing the metering rod would probably be the easiest. Find something thats just a little different in the cruise section. Maybe you could try to richen up the idle screws temporarily by half a turn and see if that changes it?
 
Let's first check on how loose the timing chain is. After that is checked and very likely replaced try 10-15 degrees of initial timing with a 195 degree T Stat. 34-36 total.

At part throttle cruise and at full advance with vacuum [50-56 degrees total timing] 16-18:1 afr should be fine if not higher depending how goldilocks everything else is.

Checked the timing chain slack today by setting the timing mark on the 0° mark going clockwise. Then rotated crank counterclockwise until the rotor started to move. I got between 8° and 10° of slack measured on the crack in 5 tries.
The service manual does not have a spec for this measured on the crank. Only max. 3/16" measured directly on the chain and with a preload from a torque wrench. From what I can find in old forum posts 10° is not considered at the point of failure, but in the high end?
 
Timing chain "slack" is more of a diagnostic thing to me. My late machine shop operative had a theory on timing chain "jumping". He said it was more prevalent in engines where the timing cover had a good bit of room between the cover and the chain. If it had too much slack (stretched chain or worn timing sproket), then the harmonics (flopping) would let it jump enough to get past the sproket's teeth and "jump time". Engines, as in the orig small block Chevy, had the cover and chain with less distance between them. On those engines, the chain would hit the cover and wear a hole in it, resulting in an unexplained oil leak from an unusual place. But still not jump time per se.

Then came the cam sprokets with the nylon teeth, for noise control, that did good to last 70-80K miles OEM, back in the 1970s and prior. When a tooth broke off from age or whatever, then the chain would jump teeth.

Probably a better indication of timing chain stretch would be having to re-set the base timing as the dwell (on a points distributor) is the same. But on a vehicle you haven't ownerd since it was newer, something that you can't really follow over time.

Default mode . . . to me . . . is that if there are any questions about the timing chain, even on a sub-80K vehicle, put a complete roller chain set in it and be done with it. THEN you know what's there and won't have to worry about it. Cheap insurance and piece of mind. BUT, not that things are going to fail tomorrow, so being attentive to IF anything changes can allow for more time to get the chain done, as other things are done first.

Just some thoughts,
CBODY67
 
Last edited:
I got a little further.
To recap; my original issue was sudden jolts of lean condition at light cruise.
I suddenly got the idea, that it could be related to the PCV valve (i.e. an intended vacuum leak). The valve is brand new, but I think it might have the wrong characteristics. I clamped the hose to the valve and drove a few miles. The AFR dropped to 13.8 at cruise, the extra lean jolts were gone and the car gained low end power. So I think the PCV jumps between it's extreme positions when I see the sudden lean spikes and I think it flows too much in general.
It is Standard motor products V180. It has an "F" stamped on the "plunger". The one I took out (because it did not fit the rubber grommet) before I even started driving the car has an "A" stamped.
I have a NORS one on the way in the mail. But I am curious: Does anybody know what the letters mean? According to google it is related to flow and pressure characteristics, but no precise explanation is available. Is it problem seen before; having a PCV with wrong charateristics for the engine?
 
When the valve is under normal cruise vacuum, the plunger is pulled against the body of the valve and the calibrated restriction in the valve is "the flow". As the vacuum drops, the internal spring will allow for more flow, as the plunger becomes more in the middle of its travel. When the vacuum drops enough (as at WOT), the valve closes all of the way and there is "no flow".

Let us know how the NORS valve works.

Thanks,
CBODY67
 
That is a great Idea. There is actually an adjustable PCV valve out there so you can adjust it to your engine conditions, drawback is its billet machined and cost $150. When I had that carter afb and some lean issues, I explored that option too but never could bring myself to spending that kind if money. Its an easy test....plug the hose for the valve and see if it still does it!?
 
Years ago, I found an old Exxon service manual with pcv valve flow specs in it. ONLY place I'd seen that at the time (mid-1970s). Almost all of the OEM valves flowed the same, no matter what brand of vehicle.

In the world of OEM Chevrolet small blocks, there were TWO valves. One for the generic small block V-8 and one for the L82-style V-8, with the latter one flowing just a hair more air, to balance the greater air flow from the bigger-cam engine. It had a dark purple dye on the metal portion for ID purposes. Shaking it, it sounded a bit different. Different spring, maybe.

I tried to use it as a tuning device after I upgraded the cam in my '77 Camaro 305, but could not tell any difference in off-idle performance of idle adjustment idle speed or mixture. Kind of an inconclusive experiment of sorts, to me, on that engine combination.

There was an old SAE paper which GM did in the earlier 1960s on pcv flows and engine condensate in cold weather. One valve was better than the old draft tube, two valves were bretter at keeping the engine cleaner inside, but not much more to justify the additional costs and plumbing.

Enjoy!
CBODY67
 
I guess I owe an update to all of you who took time to reply:thankyou:.
I made a lot of small updates, and I believe most of them made a small improvement to the way the car drives - or at least were good preventive measures:
After my first electronic ignition kit stranded in Amsterdam, when the customs broker went bankrupt, I waited for another one. I opted for the kit from Rick Ehrenberg. After getting this installed, I am sure it was pinging before! And it starts a lot better now. The CAP distributor needed base timing of 5 deg. ATDC in order not to ping at cruise. This made car hard to start.
The NORS PCV valve also arrived. It looks more correct, and hisses less than the one it replaced, but did not cure the lean spikes. My original test drive with blocked PCV was too short.
pcv.jpg

Then upon re-checking, I found I had misadjusted the float heights. It helped a little.
Then I started looking for vacuum leaks and found the AFR dropped, when I sprayed brake clean around the intake manifold. So I changed the valley pan gasket (I almost can't lift the manifold into the wide Imperial). No harm done, but it turned out the dropping AFR from the brake clean spraying, was only the heat cross over leaking and the brake clean reaching the sensor that way. So it did not help on anything but peace of mind. Quite sure it was the original pan/gasket.
new pan gask0.jpg
new pan gaskt.jpg


Since it was "easy" I wanted to try some different metering rods. I found a set of 3 step rods (see discussion above) a few sizes fatter for $10, but long story short they ended up costing $70 to get to Denmark. Ouch. They made the primary circuit fatter, but not the transition area between low speed and primary.
nos_step_ups.jpg


Since the lean condition was only present with a hot engine, I started wondering if it could be the fuel boiling. So I taped a K-type thermocouple on a float bowl on the hot idling engine. It showed 52°C/127°F. On the data sheet from a local fuel company I found the most volatile parts would boil at 40°C. So I thought it would be worth the effort to add a phenolic spacer. I could not find one to fit an early AFB with the small bore pattern, so I made my own from a piece of 5/16" (8mm) G11 (epoxy impregnated fibre glass board). If anyone wants the CAD file, let me know. I also had to make a new choke rod about 5/16" higher.
It dropped the temperature of the float bowl to 42°C/107°F - heat cross over still active.
carb temp53.jpg

spacer1.jpg
spacer 2.jpg
new choke link.jpg
carbtemp_42.jpg


Almost about to give up, I found Knebel's thread on 2nd air bleed mod, and that is what I am experimenting with now. I put my pictures in his original thread. I think this mod have made the biggest impact of them all. It is almost too the fat side now, but still has a small lean spot. However I am not done experimenting.
Carter AFB Air bleed Mod
 
Back
Top