Dynamic / Static compression ratio, Cams -my head hurts

tmichaud

New Member
I have been reading through a lot of this stuff, and have to admit, I still do not quite have it. So I am going to post up a question and maybe work backwards from the answer....

(there are other posts about what I am doing, I am just really slow at getting going)

I have a stock 86L, auto, 3.07 gears. My goal is to just bump it up a little, with close to the same rpm range (well - up to 5500ish)

I have a FIRST intake
1.6 comp rollers
going to clean up the stock heads (aluminum) - bowl blend etc.
Headers
Crane Cams 114142

That cam is on the high end of what is recommended (by Grumpy) for the stock auto and stall / gears. (I believe the stall is 2200 for an 86)

So my question is this - I would like to get it up to 10, or 10.5:1 I was thinking of using a thinner than factory gasket of 0.29 to get there. After reading several compression posts - I am confused..

looking for your feedback on this, which will hopefully help me understand......
 
ok lets go thru the basics
static compression is a concept where the volume of the cylinder is measured from the lowest point the piston reaches in the cylinder (BDC +BOTTOM DEAD CENTER) compared to the highest point in its rotation (TDC=TOP DEAD CENTER) then you take that volume and compress it at TDC into the combustion chamber volume to figure the effective STATIC COMPRESSION RATIO


EXAMPLE

if we take a 383 with its 4.030 bore and 3.75" stroke and 5.7" rods
with a 67cc combustion chamber and a .023 deck(factory deck height)and use a .032 head gasket, we get a QUENCH of .054(not ideal)and a piston with 6.6cc of valve notch clearance
and we get a static compression ratio of about 10.2:1
use the calculators below

viewtopic.php?f=99&t=4458&p=12155#p12155

http://www.wallaceracing.com/cr_test2.php

http://kb-silvolite.com/calc.php?action=comp



the idea with dynamic compression is the piston can,t possibly compress anything in the cylinder until all the valves are closed, and they don,t close untill the pistons well up the cylinder on the compression stroke

http://www.wallaceracing.com/dynamic-cr.php

http://kb-silvolite.com/calc.php?action=comp2

http://www.projectpontiac.com/ppsite/co ... iew/16/30/

use all the same data in the calculator below but input the valve closes at 60 degrees past BDC or 60 degrees ABDC
and we get a dynamic compression ratio of about 8.5:1.

lets say we want a max of 8.2:1 cpr with that 383,because we have crappy gas and the engine has a dynamic cpr of 8.5:1.and we find it pings under heavy loads, what can we do?
well most guys will suggest a slightly thicker head gasket like a .040 , that gives you 8.35:1, you now RETARD the cam 4 degrees to close the intake valve ,later at 64 degrees ABDC and you find your new dynamic cpr is 8.1:1


play with the calculators, it will eventually become clear


btw heres a cam timing chart, notice that a cam on a different LSA can vary the intake valve close point a great deal comparing identical DURATION CAMS built on DIFFERENT LSA(LOBE SEPERATION ANGLES
EXAMPLE

an INTAKE VALVE ON A 230 duration cam on a 105 LSA has this timing at .050 lift..open 14-closes 36
an INTAKE VALVE ON A 230 duration cam on a 114 LSA has this timing at .050 lift open 05 closes 45


http://www.crower.com/valve-timing-chart

piston location related to degrees of rotation and stroke

http://victorylibrary.com/mopar/piston_position-c.htm

http://www.wfu.edu/~rollins/piston/animation/graph_1/

I had a couple guys ask individual e-mail questions on this thread, so I realized I need to drop back to some basic concepts that seem to be over looked
duration on a cam lobe is the factor that determines how many degrees the valve is off its seat and potentially flows gasses in or out of a combustion chamber over the piston.
lift on the cam lobe obviously dictates how far the valve opens ,
but its duration that determines how long its open, and once a valve lift exceeds about 1/4 its diameter its flow rates close to maximized, so the longer its kept over that lift ratio the more effectively it will flow . but remember the longer the duration used on the intake valve the further the piston moves in its rotation.

USE THE CALCULATORS

http://www.rbracing-rsr.com/runnertorquecalc.html
http://www.wallaceracing.com/chokepoint.php
http://www.wallaceracing.com/header_length.php
four_stroke_diagram.jpg

but keep in mind the port flow rate is restricted by both the intake cross sectional area and the duration , or time the valves off its seat WHILE theres a significant pressure differential between the intake port and the cylinder, the intake runner can be designed to induce inertial pressure in a set rpm band, but its the exhaust flow that's far more influential at helping to drag in the next intake charge into the cylinder as its fast moving gasses exit the exhaust, thus the overlap period is vitally important to maximizing cylinder fill efficiency, once you reach the rpm range where there's a significant inertial energy level in both the intake and exhaust to draw in and cause the exiting gases to be drawn out of the cylinder.
if you do some research you quickly find that the engines power curve very closely relates to its cylinder fill efficiency,
volumetric.gif

you also find that the intake port cross sectional area and valve size plus the cam duration ,lift and overlap have a huge effect on the engines ability to operate efficiently art any rpm range, because of either flow restriction if the port cross sectional area or cam timing are restricting flow or low port gas velocity levels, if the compression ratio,exhaust scavenging ,engine displacement or cam timing are not maximizing the flow rates in the ports
camcomp.jpg

the chart above can be used as a rough guide to match cam duration at .050 lift and static compression in engines obviously other factors come into play so its only a rough guide
lsadig.jpg


LSAChart01.jpg

tighter LSA tend to increase the OVERLAP but thats not always the case
pistonposition2a.jpg

EXFLOWZ4.jpg



p174971_image_large.jpg

DURATION AND LSA are NOT necessarily related
ramprate.jpg


portfl3.jpg


FlatVsRollerChart.gif


Duration_v_RPM-Range_wIntakeManifold01.jpg




related info you might want to read thru carefully
http://www.epi-eng.com/piston_engine_te ... torque.htm

viewtopic.php?f=52&t=5154

http://www.veryuseful.com/mustang/tech/ ... enging.pdf

viewtopic.php?f=52&t=148&p=34936&hilit=curtain#p34936

viewtopic.php?f=52&t=8460

viewtopic.php?f=52&t=727

viewtopic.php?f=52&t=3729&p=9689&hilit=ramp+acceleration#p9689

viewtopic.php?f=52&t=333

viewtopic.php?f=52&t=2627&p=6780&hilit=ramp+acceleration#p6780

viewtopic.php?f=52&t=181&p=5342&hilit=ramp+acceleration#p5342

viewtopic.php?f=52&t=1489&p=3360&hilit=ramp+acceleration#p3360
 
So.. based on that, using stock numbers of
Cylinder head Volume (CC) = 58
Piston Top Volume (CC) = 1.56
Cylinder Bore (Inches) = 4.0
Cylinder Stroke (Inches) = 3.48
Deck Height (Inches) = 0.070
Head Gasket Thickness (Inches) = 0.051
rod length 5.7

I get a compression ration of 9.48. moving to a .028 gasket gets me to 9.96

Running some numbers through the calculators seem to indicate with this cam, according to the table closes 36degrees Intake Closing Point (Degrees)ABDC @ 0.050 lift.

That would mean static compression with the stock thickness gasket would be 8.9, while with the 0.028 it is 9.3.

Hopefully I am on the right track? To me this looks to be an ok choice.
 
tmichaud said:
So.. based on that, using stock numbers of
Cylinder head Volume (CC) = 58
Piston Top Volume (CC) = 1.56
Cylinder Bore (Inches) = 4.0
Cylinder Stroke (Inches) = 3.48
Deck Height (Inches) = 0.070
Head Gasket Thickness (Inches) = 0.051
rod length 5.7

I get a compression ration of 9.48. moving to a .028 gasket gets me to 9.96

Running some numbers through the calculators seem to indicate with this cam, according to the table closes 36degrees Intake Closing Point (Degrees)ABDC @ 0.050 lift.

That would mean static compression with the stock thickness gasket would be 8.9, while with the 0.028 it is 9.3.

Hopefully I am on the right track? To me this looks to be an ok choice.

When calculating Dynamic Compression Ratio (DCR), then you can't use the timing number @ .050" lift. You have to use the actual closing of the intake valve.

Getting the Crane numbers off their cam specs, you can calculate the Intake Valve Closing (IVC) angle, which turns out to be 63 ABDC. This is 5 degrees advanced.




Retarding the camshaft a 2-4 degrees might help keep you away from possible detonation, but ask Grumpy he can speak from experience. If you wanted to deck the block, then you could get your Quench distance down to around .040". This would help greatly with stopping detonation.



If you want to do your own calculations you can download this file if you have the Microsoft Excel program or want to download a compatible version called Open Office. It the same file that I used above. Anyway all you need to know is in this thread.

viewtopic.php?f=99&t=4458


 

Attachments

  • CR_Calcs_With_Crane114142.JPG
    CR_Calcs_With_Crane114142.JPG
    51.5 KB · Views: 143
  • Crane_114142_454_480.JPG
    Crane_114142_454_480.JPG
    56.3 KB · Views: 143
  • IVC-Crane_114142.jpg
    IVC-Crane_114142.jpg
    54.5 KB · Views: 143
Back
Top