632 ci Big Block Chevy Build for Project Wild & Willys | Engine Power – Full Episode


(Narrator)>>Today
on Engine Power we’re building a bullet that’ll
make more horsepower than any naturally aspirated
engine we’ve ever had in the shop. When we’re done with it
this big block Chevy will power X-O-R’s over
the top Willys Wagon. (Pat)>>Welcome
to Engine Power. Today we are going to
embark on an engine build that is a true race piece,
and it will make more naturally aspirated
horsepower than we have every made before. The more power you
make the stronger the engine you need. So it’s important to start
with a solid foundation. This short block is the
creation of Blueprint Engines, and it’s made
right here in the U-S-A. It’s a big block Chevy
that’s available in both standard and tall
deck heights. Since we’re looking for
maximum displacement we opted for the tall deck. Another thing that will
increase the displacement is the finished bore size
of four-600, drastically larger than a stock
big block Chevy. The included forged steel
crank shaft has a stroke of four-750, which gives
us 632 cubic inches. You have seen us run
Blueprint engines in the past and it’s
for good reason. They’re well designed,
reliable, and they’re perfect for someone who
wants to save the time that it takes to spec
out and build their own. Last year we helped build
Factory Five’s new ’35 hot rod pickup, which was
powered by the Blueprint 306 cubic inches
small block Ford. It made 390 horsepower and
370 pound feet of torque. In keeping with the
hot rod theme, triple Strombergs
delivered the fuel. We fired it up, synced
the carbs, and it ran like a top. ♪ ♪ Recently we beefed up a
’74 Glencoe jet boat with a Blueprint 540 cubic
inch big block Chevy. It pulled down 680
horsepower and 669 pound feet of torque
in the dyno cell. [ engine revving ] (Pat)>>Last week
we spent the day at the lake and it was
a face flapping, rooster tailing heck
of a good time. We can’t wait to get this
boat back in the water, and we’ll be bringing
you more of that soon. This 632 cubic inch bullet
is coming together for our friends down in X-O-R. Eliza and Jeremy are
working on a classic piece of American off
roading and it needs some stout horsepower. Let’s go down
and check it out. ♪ ♪ Hey what’s going
on down here? (Eliza)>>Hey,
what’s going on? (Pat)>>Well i came
down to see what this engine is going in. Tell me a little
bit about it. (Eliza)>>Well it’s a 1963
Willys Wagon, and we want high horsepower to
get through the mud. So we’re going to custom
four link, probably like 16 inches up. 46’s, maybe 54’s? Two and a half ton axles. So we’re looking to get
about 120 miles per hour wheel speed. (Pat)>>Now I’m not
familiar with these type of vehicles. Why does it need that
kind of power for this application? (Eliza)>>Well usually
most of our vehicles have a lot of gearing. So they don’t need a whole
lot of horsepower, but since we want to have 120
miles per wheel speed to turn the mass of the tires
we need high horsepower. (Pat)>>Now this thing
actually needed some power when it came in because it
sounded, it sounded okay but I didn’t see
what was in it. (Eliza)>>So it
had a 305 in it. They had already swapped
it out because originally these came with a four
cylinder Hurricane engine. So maybe 100 horsepower? (Pat)>>Maybe, it
didn’t sound bad. It definitely had the
v-eight exhaust but it sounded a little
on the anemic side. (Eliza)>>A little bit,
for what we’re doing a little bit yeah. (Pat)>>So say it made
100, 150 horsepower, it’ll roughly have
10 times that now. (Eliza)>>That
would be wonderful. (Pat)>>That’s what
we’re shooting for. The engine’s
gonna be good. We’re gonna crest
1,000 with this. This is gonna be the
highest horsepower naturally aspirated
engine we’ve done. And congratulations,
you guys get to have it. (Eliza)>>Thank you! We’re excited that you’re
building us the engine. (Pat)>>Now I might be
stating the obvious but you don’t see a lot of
power adders like a blower or a turbo on these
type of engines right? (Eliza)>>No you don’t. It’s a little too
sloppy out for that kind of setup. We don’t want any kind
of water or mud getting into the system. So that’s why you don’t
see a lot of power adders. So that’s why we need
big cubic inches and a lot of horsepower. (Pat)>>Well I think we
can help you out because the engine is going to
make a lot of power, and if you don’t mind would
you come down and help a little bit? Maybe torque
some stuff on? (Eliza)>>Oh yeah,
I’d love that. (Pat)>>And you wouldn’t
want to run the dyno on something like
that would you? (Eliza)>>No I wouldn’t! (Pat)>>I didn’t think so. So when it’s ready we’ll
come down, let you throw the lever on it. (Eliza)>>Awesome,
I would love that! (Pat)>>All right,
well get back to work. I’m taking too
much of your time. (Eliza)>>Have a good one. (Narrator)>>Up next, we
crank up the compression on the big block. (Pat)>>We’re continuing
on the buildup of our 632 big block Chevy, and for
the application down in X-O-R we’re gonna have to
make a bunch more power than this engine is
currently setup to make, which means we’re
gonna have to change out a few parts. Right now it has a piston
that will make 10.8 to one compression, which is
friendly for pump gas but for what we’re doing we’re
going to need a bunch more compression than
that and here’s how we’re gonna get it. We’ll be putting in a
new set of Mahle Elite Sportsman drag
series pistons. Now the big
difference with these. These have a 42cc dome,
which will increase our static compression ratio
from 10.8 all the way up to 15.36 to one. Now these have hyper
accurate ring lands and lateral gas porting to
help improve ring seal, an zero-43, zero-43, three
millimeter ring pack, wider pin bosses to
help manage the increase inertial loads from these
long stroke, high piston speed applications, and
also come with an upgrade wrist pin that
will withstand heavy nitrous use. All of Mahle’s pistons
come ready to run right out of the box with a
properly sized wrist pin bore, a dry phosphate
coating over the entire piston, and Mahle’s
patented Grafal anti-wear coating on the skirt. We’re also changing up
the connecting rod in this setup and I’ll tell
you why in a second. We’ll be using a set
of Eagle Specialties Forged H-beam rods. They’re made out of 43-40
steel, have and H-Beam design, and our setup
for a nine-90 wrist pin. They are six-700 in
length, and have been upgraded with an A-R-P
L-19 rod bolt for high horsepower, high
r-p-m applications. We needed the piston
that we’ve chosen for the higher dome volume to get
the compression ratio that we want for this
application, but it won’t work with the rod that
came in this engine. That’s why we’re
switching to the Eagle. Here is a quick look
how we came up with the calculation of what
we needed to use. What we are referring
to is block height. The cylinder line of the
crank shaft to where the cylinder head bolts on
is called block height. In this case it
is 10.200 inches. The piston’s compression
distance, plus the center to center length of the
connecting rod, plus half the stroke, which is
called crank arm, should equal the block height. We tailor the compression
distance and rod length with our crank
shaft to get that. So theoretically our
piston is flat with the deck at t-d-c. In our case we have a
stroke of 4.750 inches. Divide that by two and
we have our crank arm, which is 2.375 inches. Add the connecting rod
center to center distance, which is 6.700 inches as
well as the 1.120 inches piston compression
distance, and we have a total length
of 10.195 inches. If our engine’s deck
height is truly 10.200 inches and we subtract the
10.195, we can determine that our piston is
five thousandths of an inch in the hole. Of course we have to
physically measure the deck height to see
if that’s correct. The first step is to put
a rod and piston assembly together for mock up. ♪ After removing the current
piston and rod combination we’ll drop our mock
up setup into place. ♪ ♪ [ drill spinning ] (Pat)>>Then we’ll
run the piston up to top dead center, checking it
with a dial indicator. Then we rock the piston
back and forth finding the highest and lowest point
in the bore at t-d-c. The average of those two
measurements tells us how far the piston
is in the hole. In our case it’s eight
and a half thousandths. This is one of the
measurements that we do use to calculate final
compression ratio for the combination, and this
number also effects other things in the engine, but
for right now we have this number and we can get
the rest of the rods and pistons put together. When putting together our
piston and rod assemblies we like to use Permatex
Ultra Slick Engine Assembly Lube. This specially formulated
lubricant protects components from excessive
friction and wear during initial engine start up. ♪ ♪ Because the wrist pin’s
bore intersects the ring land on the oil ring
we’re gonna have to run a support rail, and it is
held in place by this mark that keeps it
from rotating. It supports the oil
ring to keep it from burning excessive oil. ♪ ♪ We’re using the Goodson
power ring filer to gap the rings. This second compression
ring is being gapped to 28 thousandths. A fine stone is used
to deburr the edge. With the ring installed,
a Summit adjustable ring squaring tool gets it
perfectly into position, and a feeler gauge is
used to check the gap. We’re gonna be using an
upgraded ring pack in this operation because of the
amount of horsepower it’s gonna make. It still has a standard
tension oil ring to control the oil and a
napier style second ring, which has a small hook
on the edge that will actively scrape oil
off the cylinder. The top is the
special one. It is an uncoated steel
nitrite ring, which will hold its tension
better than a standard ductile ring. (Narrator)>>Our quest for
power continues with an aggressive cam and a
heavy duty belt drive. (Pat)>>Welcome back
to Engine Power. We’re building a 632
cubic inch Chevy big block that’s gonna go into a
’63 Willys Wagon down in the X-O-R shop. Once the pistons are in
we can torque up the rods. The most accurate way to
do this is using a rod bolt stretch gauge. This tool measures how
much the bolt stretches under a specific
torque load. The spec for this L-19
rod bolt is 73 to 77 ten thousandths. The goal is to tighten the
fastener until it reaches its maximum clamping force
without overtightening the bolt until it becomes
distorted or compromised. In fact, whenever you’re
tightening a fastener to a torque spec what you’re
really doing is stretching the fastener to the ideal
length under load, which on most fasteners is 75
percent of its yield. For the ultimate in
accuracy and adjustment we’re installing a
Jesel belt drive system. It comes with a billet
aluminum cover that’s o-ringed for a
leak proof seal. Next the cam shaft gently
slides into the block. This is an off the shelf
solid roller offering from Comp Cams. The specs are impressive
by any standards. Duration at 50 thousandths
lift on the intake is 285 degrees and 300
degrees on the exhaust. Lobe separation
angle is 114 degrees. Gross valve lift on the
intake side with a one point eight rocker is
873 thousandths, and 848 thousandths
on the exhaust. The thrust washer gets
installed on the cam shaft adapter provided by Jesel. Then it’s wiped completely
dry with lacquer thinner to remove any
trace of oil. Next a thin but thorough
layer of silicone covers the thrust shim before
it’s put in place. The cam shaft thrust plate
slides in and is torqued to 96 pound inches,
or eight pound feet. ♪ ♪ The crank gear is
seated into place. Then the belt and the cam
gear are installed and torqued to 70 pound feet. As always, we degree
the cam shaft. This allows the engine
to make the most power where we want it. The intake center line
comes in at 113.5, which is a half degree advanced,
and that is within a half degree of straight up,
meaning the cam shaft’s intake centerline is
the same as its lobe separation angle. Now we don’t plan on
moving the cam around a lot because we want the
engine to make power in a specific place in
the r-p-m range. So we are gonna check our
piston to valve clearance with the cam shaft
in this position. Our solid roller lifters
are from Comp Cam’s premium sportsman series. Even though this is only
mock up, we’ll be certain the correct thickness
head gasket. New gaskets will go
on for final assembly. Once the head is set into
place we’ll drop in a few A-R-P studs
to keep it there. ♪ ♪ We need to check
for push rod length. So to do this we’ll
drop in a couple of adjustable push rods. We’ll set the adjustor
on our Jesel rocker arm to one full turn
of adjustment. Then we’ll lengthen the
adjustable push rod while the cam is on base circle
until we have the correct lash, which is 24
thousandths on the intake and 26 thousandths
on the exhaust. ♪ ♪ We’ll set up our dial
indicator to check piston to valve clearance. The intake is checked at
10 degrees after t-d-c on split overlap. We zero out the indicator
and press down until the valve contacts the piston. We have a roomy 100
thousandths clearance. We’ll do the same for the
exhaust valve but it’s checked at 10
degrees before t-d-c on split overlap. The result, a whopping
135 thousandths. In other words this thing
has a ton of clearance. (Narrator)>>Up next,
high flowing heads for our bullet. (Pat)>>Now studs will
help so you can align it. (Pat)>>Welcome back and
we’re continuing on our 632 buildup, and next to
go on is an A-T-I super damper we picked up
from Summit Racing. Since everything is
torqued up and ready to go, we’ll install our
Melling high volume oil pump and our new
seven quart pan. To make the power we
want to make we need a lot of air flow. So we called up Summit
Racing and got a set of A-F-R 385cc rectangle
Magnum 24 degree big block Chevy heads. Now there’s nothing
small about these. They have a two-350 intake
valve, one-eight-eighty exhaust on a
121cc chamber. So they take a minimum
of a four-250 bore. Now the 385 refers to
the intake port size, which is 385cc. Now that will flow
a massive 452 c-f-m at 800 lift. It also has a 135cc
exhaust port, which flows at 344 c-f-m at
that same lift. They are set up with a
one-625 o-d spring, which is good for 850 lift,
which is right in the range of our cam. These heads will make big
power on a big engine. Because this bullet
is going in the Willys project in X-O-R I thought
it’d be a great idea to have Eliza and
Jeremy down to help torque up the heads. ♪ ♪ All right, that
one’s for you. Now studs will help
so you can align it. This on the bottom
right there. (Eliza)>>There it goes. (Jeremy)>>Ooh,
those are pretty! (Pat)>>All right,
studs are next. The reason we do studs
first down here just for guide, and then later
because we don’t want to drag a head over a bunch
of installed studs cause it’ll drag some
aluminum in. (Eliza)>>Oh, and you’re
still letting us run the dyno on this right? (Pat)>>Yes. If you get a little
crazy on the dyno you can wound it. We’ll keep you
out of the woods. (Jeremy)>>Just tell me
what buttons to push. (Pat)>>Tell me what
buttons to push. (Jeremy)>>That big
lever, I’m just gonna push that forward. (Pat)>>First click. (Eliza)>>Okay! [ click click ] (Jeremy)>>Makes you
a little scarred sometimes when you’ve
got to pull on them. (Pat)>>Now once you’ve
torqued several thousand of these you’re
gonna get used to it. I know it seems kind of
complicated but it’s not. You’re just going
in a circle. (Jeremy)>>Do you
always turn your torque wrenches back out? (Pat)>>Absolutely, on
a click type I always return it to zeroes. (Jeremy)>>Perfect, well
I’m gonna turn this one back out, and that gives
us a point to also get out of here and let you finish
putting this back in. (Pat)>>I’ll take
that from you. (Eliza)>>Take
care of our baby! (Pat)>>Nice job and I’ll
have you back down for the next part, which
is probably gonna be on the dyno. (Jeremy)>>Sounds great! (Pat)>>Good job. For more information on
anything you’ve see on today’s show visit
Powernation T-V dot com. ♪ ♪ When building any engine
one of the most important factors to consider
is compression ratio. Now the parts that you
choose for your build are designed to achieve a
specific compression ratio but that is totally
dependent on what you’re doing. Now simply put,
compression ratio is the maximum volume of the
cylinder divided by the minimum volume
of the cylinder. Or another way of putting
it, the volume of the cylinder at b-d-c divided
by the volume of the cylinder at t-d-c. And what makes compression
ratio so important? It directly effects the
power level and operating range of a given
engine application. Generally speaking the
higher the compression ratio of the engine the
higher the potential power output. Higher octane fuels
are required for most high compression
engines to achieve optimum power output. Strictly speaking, static
compression ratio measures a volume only,
not pressure. There are high
compression, low cylinder pressure engines that are
designed primarily for fuel efficiency and not
just out right power. In our world we want to
generate as much cylinder pressure as possible
to increase power. Higher static
compression makes this easier to achieve. On the other end there
are actually low static compression but high
cylinder pressure engines that also require high
octane fuel, but that involves something called
dynamic compression ratio, and that’s a topic
we’ll get into later on in depth. These engines are
typically purpose built race bullets that have to
conform to very specific compression rules but
still make big power. There are five variables
that effect compression ratio, and changing anyone
of these factors will raise or lower it,
and we’re talking about volume. They are, number
one, the swept volume of the cylinder. If you change the bore or
stroke of a cylinder you will change the volume
of that cylinder, and therefore it’s
compression ratio. Number two, combustion
chamber size. Increasing it will lower
the compression ratio and decreasing it will raise
the compression ratio assuming that no other
changes have been made. Number three, head gasket
volume, meaning the gasket’s bore size and
the gasket’s thickness. You might not realize it
but the variations in head gasket volume can increase
or decrease compression ratio by over
a half a point. Number four, the volume of
the cylinder at t-d-c with no other components. In most engines the piston
stops just shy of being level with the deck
of the engine block. This small volume of
additional space lowers the static compression
of the engine slightly. Number five,
piston volume. A dome shaped piston takes
away volume, increasing static compression. A disc shape piston
adds volume, therefore decrease compression. Now you know the five
factors that determine engine compression ratio. You can use this in order
to more carefully plan your future engine builds. Next time we’ll show
you how to measure and calculate the static
compression ratio in your project vehicle.

Posts Tagged with…

Write a Comment

Your email address will not be published. Required fields are marked *