Why does overlap make a rough idle?

[Danglerb]

I understand the concept of reversion, exhaust gas flowing backward through the exhaust valve, combustion chamber, and out the intake valve.

I don't understand how a small overlap, with small valve opening, can create enough flow to be a problem. Isn't there always a much larger amount of reverse flow through the intake valve due to the exhaust gas left in the combustion chamber from the previous cycle?

I don't understand what exactly the problem is that is created by the reverse flow.

I don't understand the randomness of the effect.

[automotive breath]

I understand the concept of reversion, exhaust gas flowing backward through the exhaust valve, combustion chamber, and out the intake valve.

I don't understand how a small overlap, with small valve opening, can create enough flow to be a problem. Isn't there always a much larger amount of reverse flow through the intake valve due to the exhaust gas left in the combustion chamber from the previous cycle?

I don't understand what exactly the problem is that is created by the reverse flow.

I don't understand the randomness of the effect.

The longer the valve overlap the more exhaust gases flow into the intake port during overlap. The problem is more
pronounced at lower engine speeds when there is more time for exhaust gases to back up.

As the air fuel mixture is diluted with increasing amounts of exhaust gas, the burn time increases and the cycle
to cycle variations in the combustion process increases. Eventually a point is reached where engine becomes
rough and unstable.

With no or little reversion, combustion occurs before the exhaust valve opens cycle after cycle. At low engine speed
reversion increases and flame speed is so slow that combustion is completed just before to the exhaust valve opening.
As reversion increases further, in some cycles combustion is not complete prior to the exhaust valve opening and the
flame extinguishes before all the fuel is burned. The engine becomes rough and unstable. Finally misfire cycles start
to occur where incomplete combustion and misfires increase to a point where the engine no longer runs.

[engineguyBill]

Virtually all four cycle engines, even in stock form, have some valve overlap. Rough idle is the result of increased duration, which results in much longer overlap period. Intake valve is opened earlier and exhaust valve is closed later in the cycle, which results in exhaust gas diluting the intake mixture at low engine speeds. At higher engine speeds the engine smooths out as the exhaust gasses flow out much quicker, the result of some exhaust "blow down" when the exhaust valve opens while there is still some combustion pressure in the cylinder. The quickly exiting exhaust gas also results in a slight negative pressure which may even pull intake mixture into the exhaust system at very high RPM. Overlap can be the engine's friend, as well as an enemy.

[PackardV8]

Overlap can be the engine's friend, as well as an enemy.

To expand upon this, "At high RPMs overlap can be horsepower's friend, but overlap is alway an enemy of a smooth, efficient idle."

jack vines

[Rick360]

At idle the intake manifold vacuum is high so it will draw from the cylinder anything it can when the intake valve is opening.

Rick

[novadude]

OK... let's add to this question....

How MUCH overlap can an engine tolerate before the idle gets too rough. When I say "too rough", I mean rough to the point that it won't want to sit in traffic, work power accessories, etc.

Let's assume a common 9:1-10:1 355 Chevy street engine. Since the total overlap triangle area is what matters, what are some guidelines for maximum seat and 0.050 overlap for a true driver with power accessories.

Interesting discussion. Does the rough idle effect come more from what is happening in the single cylinder, or is it more the big "vacuum leak" created by other cylinders connected to the common plenum?

[paul s.]

Interesting discussion. Does the rough idle effect come more from what is happening in the single cylinder, or is it more the big "vacuum leak" created by other cylinders connected to the common plenum?

I believe it has MORE to do with the common plenum. Hence the OEM's fondness for dual planes.

There is A LOT of communication going on between cylinders and the carburetor.

[swatson454]

OK... let's add to this question....

How MUCH overlap can an engine tolerate before the idle gets too rough. When I say "too rough", I mean rough to the point that it won't want to sit in traffic, work power accessories, etc.

Let's assume a common 9:1-10:1 355 Chevy street engine. Since the total overlap triangle area is what matters, what are some guidelines for maximum seat and 0.050 overlap for a true driver with power accessories.

Ok, I'll shoot.

This is totally subjective and "too rough" will vary from guy to guy... probably a lot. In your example engine and my humble opinion, I'm in the 55 degree range for overlap in a healthy idle, reasonable for the street convertor kind of deal. I think 60 degrees is starting to get a little nasty for traffic. I'm kind of a vag with street cams though.

But then you have the question of LSA. 55 degrees of overlap on a 112 cam is a different horse than a 108. Although the overlap period is the same, how much ill effect will the later closing intake on the wider cam have over the tighter one?

We did a Magnum 280 h/f cam in a 9 to 1 350 last year. I wanted a single-pattern 268 Voodoo on a 108 but when the customer saw that the performance shop had the Comp cam on the shelf, I was voted out of office

The Magnum had 60 degrees of overlap and although it isn't as bad as I thought it would be, it's more than I'm personally ok with. Of course the guy has a 3k convertor too, which I also don't like on the street.

I've also done Isky 270 cams on a 108 in otherwise stock 350s and the guys were really pleased. One even ran Howell F/I and it apparently gave him no tuning fits. That one works out to about 55 degrees.

None of these engines are how I would liked to have done them but it seems like a reasonable comparison. Again, I'm kind of a puss with street cams.

[novadude]

Again, I'm kind of a puss with street cams.

I am too. That's why I am asking these questions. Looking at a 270 deg 108 LSA cam for a 9.5:1 350 I'm planning, and I am worried that it might be too rowdy. I suspect that I worry too much, and that it will be easy to live with in a 4-spd street car.

[Danglerb]

Thanks, this is good I feel like I am learning. I've been poking around idle questions for a few years, and what many have told me is that the LS7 cam is the text book to study on what you can get away with on a street/smog cam, its the practical edge for a mass produced motor. Details on why never went past LSA and overlap, and its quickly confusing to me.

I found the grind info sheet for my cams, which are going into a 5.0L V8, around 10:1 comp. My notes say people get "good" vacuum despite the 13 deg overlap, but old notes etc. It took me awhile to find, so stuck it in this post so I can find it again.

Valve Timing Information Engine: Set (2) Porsche 928 SOHC 16v - Hydraulic Grind # 552/553
... Intake ... Exhaust
Valve Lash 0000 0000
Valve Lift .506 .462
Duration 266º 240º
Duration @ .050 244º 222º
Lobe Center 110º 110º
Intake Opens 12° Before TDC
Exhaust Opens 41° Before BDC
Intake Closes 52° After BDC
Exhaust Closes 1° After TDC

[540 RAT]

The following may help you get a better idea of how this all shakes out.

Valve opening overlap is of course determined by the Lobe Separation Angle (LSA) and the amount of duration being used. If you hold the LSA fixed, and vary the amount of duration, you will end up with different amounts of overlap. For example, in the chart below, the cams could all have say a 110* LSA, but increasing the duration, increases the performance, while at the same time increasing the overlap. And if you want to get an APPROXIMATE idea of how a given cam will stack up in the overall scheme of things, you can take a look at this reference chart. The numbers in this chart are not engraved in stone, but they can be used as a general guide to see how SOLID LIFTER ADVERTISED OVERLAP amounts typically compare:

APPROXIMATE SOLID LIFTER “ADVERTISED OVERLAP” PERFORMANCE REFERENCE CHART

300ci…………………………400ci………………………..500+ci
10*………………………………25*………………. ……….40* towing
30*………………………………45*……………….. ………60* ordinary street
50*………………………………62.5*…………….. ………75* street performance
70*………………………………80*…………………………..90* street/strip
85*………………………………92.5*……………………….100* race
95*………………………………105*………………………..115* Pro race

Here's how to calculate your SOLID LIFTER “ADVERTISED OVERLAP”, to use in this chart above:

• Add your intake and exhaust ADVERTISED duration (duration at .050 will not give you the correct overlap here)
• Divide that answer by 4
• Subtract the lobe separation angle (LSA) from that answer
• Multiply that answer by 2, and you have it!


NOTE: Because of the differences resulting from not having any lash, and the way hydraulic cam advertised duration is rated, if you want to figure the “advertised overlap” for a HYDRAULIC LIFTER cam, so that you can use this chart above, REDUCE the Hydraulic cam’s listed ADVERTISED duration by 8*, for both intake and exhaust, then follow the calculation procedure as shown above.


And if you also want to get an APPROXIMATE idea of what a given cam will sound like, you can take a look at this chart below. The sound of a lumpy idle, which most gearheads like, is very subjective, meaning that people won’t all agree on what a lumpy idle should really sound like. For some, it doesn’t take much, while for others, it takes quite a bit. But, here are some common engine sizes, and the approximate minimum, solid lifter “advertised overlap” needed, for that bad mofo lumpy idle (at around 1,000 rpm) that will turn heads at the local cruise-in.

APPROXIMATE MINIMUM, SOLID LIFTER “ADVERTISED OVERLAP” LUMPY IDLE REFERENCE CHART

350ci………..77*
383ci………..78.5*
400ci………..80*
454ci………..84*
496ci………..87*
540ci………..90*
572ci………..92*
632ci………..96.5*

And LARGER advertised overlap numbers than shown here, will sound even lumpier.

NOTE AGAIN: Because of the differences resulting from not having any lash, and the way hydraulic cam advertised duration is rated, if you want to figure the “advertised overlap” for a HYDRAULIC LIFTER cam, so that you can use this chart above, REDUCE the Hydraulic cam’s listed ADVERTISED duration by 8*, for both intake and exhaust, then follow the calculation procedure as shown above.

ALSO NOTE: Lowering the idle speed below 1,000 rpm to say 750 to 800 rpm (if your particular combo can go that low) will also make the engine sound lumpier. So, that can be a bit of a band aid, in the event that your combo falls below the desired MINIMUM ADVERTISED OVERLAP shown in this chart above. Lowering the idle like that can even make a stock 502 with its mild hydraulic roller cam, sound pretty good.

As a real world point of reference, my 540ci BBC Street/Strip engine has a solid roller cam with 303*/309* advertised duration, 266*/272* at .050 duration, 108* LSA, and has 90* of advertised overlap. And it “just reaches” the threshold of a nice lumpy idle at about 1,000 rpm. It takes a lot of overlap to truly reach that bad mofo lumpy idle that will turn heads at the local cruise-in.

[85gt]

what would my advertised overlap be if its 49.49 degrees at.050?

[Danglerb]

Here's how to calculate your SOLID LIFTER “ADVERTISED OVERLAP”, to use in this chart above:

• Add your intake and exhaust ADVERTISED duration (duration at .050 will not give you the correct overlap here)
• Divide that answer by 4
• Subtract the lobe separation angle (LSA) from that answer
• Multiply that answer by 2, and you have it!

for a HYDRAULIC LIFTER cam, REDUCE the Hydraulic cam’s listed ADVERTISED duration by 8*, for both intake and exhaust, then follow the calculation procedure as shown above.

Ok, my cam data says, Duration 266º 240º, take off 8º on each for hydraulic and its Duration 258º 232º, add that together is 490º divide by 4 and its 122.5º and now I need to figure out LSA.

[PackardV8]

The other thing to keep in mind is 99% of the info here is for carbureted SBC single plane intake and distributor ignition

IIRC, You are messing with a Porsche 928 SOHC 2V with a semi-IR EFI intake manifold and a crank-triggered ignition. A given duration and overlap will behave completely differently in your application than it would in a larger displacement SBC or BBC with antique induction and ignition.

The Ford ModMotor SOHC guys would be your closest cousins and best source of info on cam duration idle characteristics.

jack vines