Index Home About Blog
Date: Mon Dec 13 17:42:01 1993   
From: Luis Maclean <lmaclean@magnus.acs.ohio-state.edu>
Subject: Re: While we're on the subject of carbs....

>Hello all,
>     I have a '72 240 with a 280 engine in it. I recently acquired this car
> and am now in the process of trying to get it up to par. The engine appears
> to be mostly stock from conversations with the previous owner. The exception
> is the intake side of things. It currently has a Bob Sharp Racing manifold 
> and a Holley 4-barrel 390 cfm carb. (list # 8007)  It also has a stock  
> exhaust manifold that eventually connects to 2.5 inch exhaust pipe. (It used 
> to have headers I hear)
>     The problem is that when I try to accelerate from say 1500 RPM's, it bo
gs pretty hard.

(Stuff deleted)

> [I've meant to jump into this thread for a couple of days.  The problem with
> bogging on the 4 barrel conversions is the very poor design of the
> intake.  By the time the A/F makes a sharp turn coming out of the carb,
> a couple of more sharp turns getting to the interface with the old
> manifold and the turns inside the old manifold, there is more than
> ample opportunity for gas to drop out of suspension AND there is tons of
> surface area that must be wet on each throttle opening.
>
> Though I have never been able to get rid of all the bog and have as crisp
> a throttle response as with fuel injection, I have gotten it to be pretty
> good.
>
> The most important mod is to install Holly's 50 CC high capacity accel
> pump kit.  This costs about $30.  Next, install the most agressive
> accel pump cam.  I think that is the red cam.  Finally, Install larger
> shooters so that the ballast spring on the pump linkage does not
> compress as the throttle is opened.  That is, so the shooters squirt
> the full shot as fast as the throttle calls for it.
>
> These mods hammer around town fuel economy but there's really no other
> way to make a 4 barrel run with a jerry-rig manifold setup.
> JGD]


JGD,
     I really appreciate this info and believe that this will be my first plan 
of attack in solving this problem. The only thing I might mention is that the 
intake manifold is a one piece manifold that bolts up to the head directly. It 
does have the *wonderful* 90 degree turn as the A/F comes out of the carb, but
after that it is very smooth till it hits the head surface.

To:  Pete Paraska
     Bruce Haley
     Steve Sovaiko
     Peter Smith
     Frank McBath

   Thanks for all the info you gave me. I believe that I got a lot of good 
ideas and info from everybody.  The next step I intend to implement if the 
above dosn't solve it is to change the jetting on the carb and see what it 
does. After that... Maybe call Holley and see what they suggest. Maybe a
smaller carb is in order. But I have a lot of possibilities and ideas to look
into.

Thanks again,

Luis

[I have to stress again that the problem is a TRANSIENT mixture problem
and jetting does not address the problem, though running a too rich
mixture will somewhat plaster over the problem at the expense of 
economy only because an engine is a bit more tolerant of a rich mixture
than it is lean.  Transient mixture quality is dealt with almost
exclusively with the acceleration enrichment system.  

Let's look at the problem in more detail, ignoring for the moment
the effect of poor manifold design.  Let's consider ONLY the effect of
a lot of wetted manifold area.  Contemplate the amount of surface area
of the manifold between the carb and the intake valve.  Now consider
that some of the gasoline metered by the carburator is used to wet
these surfaces.  The amount of liquid gasoline on the manifold walls
is a complex function of the wall temperature, absolute pressure, 
air velocity and surface roughness.  We know that it is some proportional
function of pressure, and the lower the pressure, the faster the evaporation.

The engine's mixture requirements are fairly uniform.  Indeed, an engine
will run OK with a single mixture strength over most of its operating
range.  Consider an engine running at some throttle setting.  The layer
of gas on the manifold wall is at an equilibrium state with as much
gas being deposited on the wall being equal to what evaporates or blows
off.  It therefore does not enter into the mixture equation.  All the
gas metered from the carb enters the chamber.  Now the throttle is
opened some more.  The added airflow pulls more gas from the carb jets
and in a well designed carb, this added fuel will be more or les 
proportional to the added air.  The engine would theoretically continue
running well, only making more power.  The problem is the increased
pressure in the manifold supresses the evaporation of the gas on the 
walls.  Therefore the walls can hold more gasoline before conditions
again equilibrate.  Some of the gas metered by the carb sticks to
the walls and the result is the overall mixture entering the combustion
chamber becomes lean for a while and the engine stumbles.  Eventually
the walls absorb as much fuel as they can, an equilibrium is re-established
and the engine again runs properly.  In other words, the bog is a
transient function of how much manifold must be wet.

The acceleration enrichment system is designed to fill in this temporary
need for extra fuel.  Ideally it would deliver just that amount of fuel
and over just the time span necessary to supply the wetting needs.
Implementations only approximate the ideal but it works pretty good 
nontheless.  Methods of enrichment vary significantly.  The SU 
accomplishes the deed by using an oil or air dashpot on its slide
to momentarily restrict its opening.  This does two things.  It 
makes the mixture rich and it controls the rate at which air is admitted
to the manifold when the throttle is suddenly opened.  It therefore attacks
the problem from two angle - supplying additional fuel AND controlling
the demand for that fuel by controlling the rate of change of the 
manifold pressure.  A very clever and simple design.

The Holly system is a somewhat Rube Goldbergish mechanism that works in
reverse.  Instead of trying to control the rate of change of manifold 
pressure, it tries to implement an enrichment system capable of pouring
in whatever fuel is needed for the conditions.  It only approximates
that function.  

There are several control parameters.  First there is the relationship
between throttle movement and required enrichment.  It is intuitive
that a throttle valve is highly non-linear.  The first motion has the 
most effect on flow and manifold pressure so that motion requires 
the most enrichment.  The throttle shaft cam determines this profile.
The second parameter is the total volume injected.  This is a function of
the pump stroke capacity and the cam lift.  

The third parameter is the time profile of injection.  If the positive
displacement pump were rigidly linked to the throttle shaft and the
driver stomped the throttle wide open, the entire shot of enrichment fuel
would by definition have to be injected during the time it took the
throttle to open.  This huge blorf of fuel would momentarily flood the
engine and then go lean because it was delivered so fast little of it
had an opportunity to distribute itself onto the walls of the manifold.
This doesn't work, of course, so the pump is connected to the throttle
through a buffer spring.  This spring absorbs the energy and meters
it out at more or less a constant rate to the pump.  The flow profile
is therefore controlled by the strength of this spring and the 
flow capability of the enrichment shooters.  This combination spreads
the shot out over sufficient time that the walls can be properly wetted
without momentarily flooding the engine.  The buffer spring is usually
preloaded so that the pump follows the cam and therefore the throttle
motion up to a certain throttle velocity.  Preload is another tuning 
parameter.

So you have the followng tuning parameters:

*	Pump capacity - consider this a rough ranging parameter that gets the 
	pump within tuning range.
*	Throttle cam - sets the rate and total amount of fuel delivered 
	up to the point the buffer spring lifts.
*	Buffer spring - preload sets the threshold of proportionality.
	Spring rate determines the maximum pump pressure.
*	Shooter flow rate - sets the maximum flow rate once the buffer
	spring has been activated.  

Tuning this mess is complicated because of the interaction between the 
parameters.  In general, if the engine immediately bogs with applied
throttle, there isn't enough fuel which means either the cam is 
not agressive enough, the shooters are too small or the buffer
spring has too little preload.  this assumes the pump capacity is adequate,
of course.  One can put a bit of brittle cement, something such as 
body putty, on the spring to detect if it is being activated.  Motion
will break the putty away.  If the engine initially pulls but then
bogs, then the shot is too short.  If the buffer spring is not being
compressed, the shooter is either too large or the cam lacks lift.
Putting in a smaller shooter is an easy test.  If the engine then 
pulls smoothly, you know the shooter was too big.  If it now bogs immediately,
you know the old shot was too fast and that you need to install the old
shooter and increase the cam lift.  

One diagnostic technique I've found extremely useful is to observe
the shot.  Now I have been known to lay on the engine peering down the 
carb while someone else drove the car but there is now an easier 
technique.  Simply mount s small video camera - a cheap surveillance 
B&W camera will do - and either watch a monitor while someone else
drives or record it on your camcorder.  Obviously not a good idea to
put your camcorder out there.  If you position the camera so that it
can see some of the engine compartment, one can judge the torque
output by how much the engine moves relative to the chassis.

Of course the best way to tune a holly is to toss it in the scrap bin
and install a port EFI system :-)

Hope this helps.  JGD (btw, that's me, jgd@dixie.com, yer list keeper)]

Date: Tue Dec 14 19:10:29 1993   
From: Luis Maclean <lmaclean@magnus.acs.ohio-state.edu>
Subject: Re: While we're on the subject of carbs....

> Luis
>
> [I have to stress again that the problem is a TRANSIENT mixture problem
> and jetting does not address the problem, though running a too rich
> mixture will somewhat plaster over the problem at the expense of
> economy only because an engine is a bit more tolerant of a rich mixture
> than it is lean.  Transient mixture quality is dealt with almost
> exclusively with the acceleration enrichment system.

Okay. So changing the jetting will affect the overall A/F mixture. And I don't
want to change that because it is not the source of my problem.

> ( More stuff deleted)                          Now the throttle is
> opened some more.  The added airflow pulls more gas from the carb jets
> and in a well designed carb, this added fuel will be more or les
> proportional to the added air.  The engine would theoretically continue
> running well, only making more power.  The problem is the increased
> pressure in the manifold supresses the evaporation of the gas on the
> walls.  Therefore the walls can hold more gasoline before conditions
> again equilibrate.  Some of the gas metered by the carb sticks to
> the walls and the result is the overall mixture entering the combustion
> chamber becomes lean for a while and the engine stumbles.  Eventually
> the walls absorb as much fuel as they can, an equilibrium is re-established
> and the engine again runs properly.  In other words, the bog is a
> transient function of how much manifold must be wet.
>
> The acceleration enrichment system is designed to fill in this temporary
> need for extra fuel.  Ideally it would deliver just that amount of fuel
> and over just the time span necessary to supply the wetting needs.
> Implementations only approximate the ideal but it works pretty good
> nontheless.

(more stuff deleted)

Excellent! So, my goal, given the large (cold) surface area, is to maximize the
pump shot volume and then install a cam to take advantage of this. Then
install a stiffer spring to get an immediate response from the pump shot. You
also suggested larger shooters to allow the extra volume of fuel now available
to flow into the carb at the higher rate (then adjust as necessary). Yes?

[Well, don't interpret my discussion as a cookbook solution.  I know you'll
need the large pump but what cam and/or shooter you need, you have to figure
out.  Something I forgot to mention in my previous note (fingers ran
out of gas :-)  Heating the manifold does wonders for throttle 
response.  My CarTech manifold had water jacket heat which helped.
I added more heat by epoxying some copper tubing to the bottom of each
runner.  If you heat some tubing dull red and plunge it into water,
it will be annealed dead soft and can easily be bent to conform to the 
runner shape.  Paint the manifold and it becomes un-noticeable.  Plumb
it into the same water supply that goes to the heater.  This is counter-
productive to best power but to me it is a decent tradeoff in order to
get some throttle response.  JGD]

From: John De Armond
X-Source: The Hotrod Mailing list
Date: Feb 1992
Subject: Re: Holleys, mostly, and a couple of Qs

> You could be bogging due to several things (less than comprehensive list).
>Especially at low RPM, being over-carb'ed is a common problem. Insufficient
>flow gives lousy metering. The accelerator pump is supposed to correct this
>but may take a fair bit of messing to get right. There is an "experimenter's
>fun kit" of accelerator pump cams from Holley that contains about 10 different
>profiles. Trouble is (and therefore I haven't even opened the one I bought)
>there seems to be no helpful info on how to determine beforehand which one
>might be the best for a particular behavior. It looks like you just have to
>run through them and decide on the best using the ole whiplash gauge.

Or you can be a bit more scientific :-)  A little background.  Transient
throttle bogging is caused by instantaneous leanness.  This leanness has
three causes.  A) lack of vacuum signal at the metering jets, B) inertia
in the fuel column, and C) fuel being diverted from atomization to 
rewetting the manifold walls after high vacuum dries them.  Funny how
port fuel injection addresses all these problems in one fell swoop :-)

The accelerator pump is a tag-on device that is designed to cover all
these deficiencies.    It does a passing job when properly tuned.  Ideally
the pump would supply a very large shot of fuel that would wet the 
walls and cover the burst of air that fills the manifold vacuum and then
supply a smaller sustaining flow until flow is established on the main 
jets.  Unfortunately, the holly pump can only approximate  this behavior.
If you're getting the idea I don't like carbs, you'd be close.

The holly pump's delivery is governed mainly by two variables - the
nozzle orfice size and the pump spring. Contrary to popular belief, the pump 
cam has little effect on injection rate.  Consider what happens when you
stab the throttle.  The primaries open immediately, forcing the cam to push
the actuation lever fully down.  Gas, being incompressible, resists 
compression and so there is a spring in the pump actuation train.  This 
spring is compressed and then exerts a relatively constant pressure on
the pump diaphram during the injection stroke.  The flow is regulated by
the orfice size.  The injection duration varies according to the spring 
pressure, the pump capacity and the orifce size.  What the cam does do is
regulate what proportion of the pump capacity is injected for small throttle
openings.  In practice, I've seen little difference between all those cams
in the kit.  Perhaps if I were UNDER carburatored, finer regulation would
be needed.  As it is, I generally just leave the stock cam on and tune on
other areas.

Spring pressure is not easily tunable since Holly does not generally offer
a range of springs.  I have made my own springs but that's not a common
capability.  The two tunable parameters are pump capacity and discharge
orfice size.  In pump capacity, you have a choice of 10 cc and 50 CC for 
common aftermarket carbs. Generally if you're overcarburated, you'll
need the 50 cc pump.

From there, the orfice size is the tuning parameter.  There are two 
conflicting requirements due to the fixed pump volume.  One requirement is
to supply enough fuel to cover the initial air surge and the wetting 
requirement.  The other requirement is to supply fuel long enough to allow
the mainjets to begin to flow fully.  This can be a surprisingly long
interval.  

Tuning consists of installing a large enough orfice to cover the initial
bog.  Then if there is a flat spot a second or so after punching the 
throttle, either the pump capacity must be increased or a smaller orfice 
must be used in order to stretch the injection interval.  It may be 
necessary to go just a bit rich on the main jets in order to help
cover the leanness caused by injecting at a slower rate than needed.

You can try to do this by the seat of the pants but a much better way is
to actually observe the operation of the carb while underway.  Now I
have strapped myself on top of an engine and had a friend drive the car
while I peered down the carb throat.  Not recommended!  A much more 
practical technique is to mount a small video camera over the carb
inlet and video the operation.  Much easier on the eye when the damn
thing spits back.  What you want to observe is the duration of the 
squirt vs when the main jets start discharging.  There should be some
overlap.  If the squirt ends before the mains start flowing, you'll
get a delayed bog.

> I have been working against the same sort of behavior on my car, which is
>probably a worst-case example of sorts - a 231 V-6 with a 750CFM Holley.

About as bad as you can get. :-)  That's about twice the carburator that
engine needs.  Look at the math.  At 6000 rpm, this engine does 3000 
intake strokes per minute.  3000 X 231 cu in = 693,000 cu in/min or
401 cu ft per min.  That, of course, assumes 100 volumetric efficiency
which is not the case.   A small 390 CFM carb would make that engine
run very good.

>in the picture when it's opened up at low speed. A vacuum gauge is nice; 
>I put one in the dash and at half pedal at low speed it drops right down to 
>0; that there is one indication that you are overcarbed. 

Yup

> Another thought - if you have solid secondaries try disabling them. Double
>pumpers are popular but a bit impractical off the track; they require more
>attention to the pedal because they only do what they are told. If it's a
>vacuum secondary, also try disabling the secondaries just for yucks; if the
>bog goes away then you might want to try out different springs in the
>actuator biscuit to hold off opening. The enabling linkage can also be
>bent to change the opening allowed.

Mechanical secondaries are another scourge that needs to be stomped out (kinda
like the coolant rumor.)  I can't imagine any situation short of Daytona,
where the throttle is bolted open for many minutes at a time, where 
mechanical secondaries are appropriate.  Even in drag racing, all 
too much throttle gives you is a big bog.  Vacuum secondaries are
so smooth and work so well, they are probably too complicated for the
average hotrod mag writer to understand :-)


> Anybody familiar with nitrous systems? I know this guy who used to be into
>the street racing scene but has given it up. He sez he has a bunch of stuff
>left over in his garage, among which is a NOS nitrous system. He says he'll
>take $150 for it 

Probably a pretty good price.  About 20 cents  on the dollar.

>Second, the nitrous systems seem to have been evolving a bit
>and I'm not sure the how the more primitive ( ca. 1980 ) setups stack up to
>the current state of the art; 

There's not been much in the way of advancement for simple single stage
units often seen on the street.  You can get electronically controlled
proportional systems but they involve $$$.  Some improvement had been
seen in injection nozzle design but nozzles are relatively cheap.
Besides, you'll probably be using a small, under the carburator
injection system.  

I've been on both sides of the equation. I've used nitrous with my
turbocharged Z engines and I've sold it from my welding gas supply warehouse.
Nitrous is wonderful but habit forming and will really punish indiscretion.
If you turn up the horsepower screws, you can go up to the breaking point
without even trying.   Fastest way I know to acquiring a piston kit (you
know, sweep all the pieces up off the track, glue 'em back together and 
you have pistons.) The second problem is the kick in the pants is so
much fun that you'll get to be on a first name basis with your nitrous
supplier.  I know, I had several junkies tethered on my string :-)

If you're really interested in Nitrous, get the book "Nitrous Oxide 
Injection" by David Vizard.  ISBN 0-931472-16-4.  Publisher SA Designs.
I got mine from the local book emporium a few years ago.  Lots of good
poop.

John

From: John De Armond
X-Source: The Hotrod Mailing list
Date: Feb 1992
Subject: Re: wetting manifold walls

>> in the fuel column, and C) fuel being diverted from atomization to 
>> rewetting the manifold walls after high vacuum dries them.  
>
>Now, I am curious as to this wetting phenomenon. I have long
>since conquered it myself, but how is a highly volatile liquid
>like gasoline supposed to wet the several hundred degree walls 
>of the manifold, and to what end? 
>
>Does this serve as a 'fuel capacitor' of sorts, storing small amounts 
>of fuel until they are whisked away by air?

Precisely.  And the capacitor discharges and recharges at precisely
the wrong time.  The walls evaporate their load of fuel when the 
throttle is closed, contributing to emissions, and reload during
power operation, causing transient lean conditions.  This is one
of the major reasons OEMs  have flocked to port fuel injection in
recent years even for non-performance vehicles.  Remember the emission
qualification process involves collecting all the exhaust emitted during
a programmed dyno run and then analyzing samples of this mix.  
The OEM has X grams per mile of each pollutant.  If he uses up a 
significant fraction of X during coastdown, he has less margin for
acceleration, cold starting and the like.  This process is also why
you see soot from most new cars under WOT acceleration.  The engine
can be dirty during this phase if idle and slow speed operation is clean.

The wetting occurs via two mechanisms.  First is the direct squirt of 
heavy droplets/stream of gas from the mainjets and accelerator pump.
This gets the area under the carb and the bottom of the runners for
a distance.  The other mechanism is aerosol deposition.  That is,
atomized droplets of fuel impinge and stick to the walls of the manifold
just like an aerosol spray does.  A secular equilibrium establishes 
itself at a given throttle setting whereby evaporation and deposition 
are balanced.  You can observe this mechanism by heating a piece of 
metal to just below the boiling point of water and then spraying it with
a fine mist of water.  As the surface heat is drawn out by the evaporation,
the diameter of the spot of wetness expands until a balance is reached.
(assumes a constant input of heat to the metal.)

An interesting experiment is to mount a carb on a manifold and then
blow air through it in the normal direction and observe the mist/spray
emitting from the manifold.  I once made a smelting furnace to recover
lead from batteries that was fired by gasoline.  It used a small
hair-dryer type blower, a 30 mm motorcycle carburator and a metal tube
that extended to the bottom of a firebrick chimney.  The idea was to
pile the batteries in along with charcoal and lime, fire the puppy up
and use the hot gasoline/air flame to melt the whole mix.  The first
attempt was pretty sloppy.  Great globs of unevaporated fuel slathering
out the tube and making large red flames.  I finally welded up a "U" tube
such that the flame impinged on it.  This "u" served as the vaporizer.
This kluge then burned with a nice blue flame and I smelted a large 
number of batteries.

John

From: emory!OAS.PSU.EDU!FBS3     (SZYMKOWSKI.FRANK)
X-Source: The Hotrod Mailing list
Date: Mar 1993
Subject: carb cfm question
X-Sequence: 4386

All the information that I have read says that the formula to caluclate
cfm needed for a particular engine size is correct.
Then how come my sb chevy (331cuin), which calculates somewhere around
600-625 cfm runs so much better with a 750 cfm Edlebrock then the old
600 cfm Holley? I'm running 11.5-1 compression, Isky 280 hyd. cam, minor
head work (1.94" s), 1-1/2" headers,and a 1963 340 hp 327 alum. intake.

Frank fbs3@oas.psu.edu

[Because the "cfm" figures Holly et al quotes are at some specified but
according to the various pieces of literature, ever changing pressure
drop across the carb.  Typically 1" of mercury.  Your engine makes the
best power, of course, with NO pressure drop in the intake.  That
"750" carb is only flowing whatever your engine needs but it is doing
it at a smaller pressure drop than the 600.  JGD]


Index Home About Blog