Pumping loss & Friction Calculations 3

[willeng]

If we spin an engine at 2000 rpm-without ignition what are the easiest methods of getting figures for pumping losses & friction. Preferably without electronic gizmo's.
What are the calculations for above.

 

[enginesrus]

What are you spinning it with????

 

[ElwoodGreen]

Assuming you spin the engine with an electric motor, you could calculate Watts from the measured voltage and amperage. To get a more accurate value for the engine losses, you will need to know the efficiency of the electric motor. I think that should be all you need. Good luck.

 

[MaxRaceSoftware]

There are 2 different SAE papers on FMEP
by Honda and GM

some of data results were used by SuperFlow
in their software as part of Weather Correction Factor

if you don't need exact results , you could use similiar
code to estimate these losses ??

Larry Meaux (maxracesoftware@yahoo.com)
Meaux Racing Heads - MaxRace Software
ET_Analyst for DragRacers
Support Israel - Genesis 12:3

 

[MaxRaceSoftware]

My 1st algorithm was this ;

FHP = ((RPM^2)* Stroke * CID) / ((13.7/((Stroke^(1/3))*(RPM^(1/4))))*350000000



later in 1990
this QuickBASIC source code

'=================================< FTQ.BAS >
COLOR 15, 1: CLS
start:
LOCATE 5, 5: PRINT USING &quot;Stroke = ##.#### &quot;; stroke
LOCATE 7, 5: PRINT USING &quot;Rpm = ##### &quot;; rpm
LOCATE 9, 5: PRINT USING &quot;CID = ###.### &quot;; cid

LOCATE 5, 23: INPUT stroke$: IF LEN(stroke$) THEN stroke = VAL(stroke$)
LOCATE 5, 5: PRINT USING &quot;Stroke = ##.#### &quot;; stroke
LOCATE 7, 23: INPUT rpm$: IF LEN(rpm$) THEN rpm = VAL(rpm$)
LOCATE 7, 5: PRINT USING &quot;Rpm = ##### &quot;; rpm
LOCATE 9, 23: INPUT cid$: IF LEN(cid$) THEN cid = VAL(cid$)
LOCATE 9, 5: PRINT USING &quot;CID = ###.### &quot;; cid

pspeed = rpm * stroke * .16666667#

factor = 1.4105813# + 1.3602189# * COS(.0001257675# * pspeed + 3.2909413#)

ftq = factor * cid

fhp = (ftq * rpm) / 5252

LOCATE 15, 5
PRINT USING &quot;Piston Speed=#####.# Factor=#.#### FTQ=###.### FHP=###.### &quot;; pspeed; factor; ftq; fhp
GOTO start


those were my 1st 2 attempts an emperical algorithm for Friction HorsePower


since then i've developed an exact fit algorithm to SuperFlow and GM data

its been 10+ years since i've researched this stuff at College Libraries

Some SAE papers you might find interesting are
# 870088
# 840179
# 830165
# 880268 <= Honda R & D
# 910347 <= Honda R & D but last time i looked at this paper there were math errors.
but still probably best paper on subject ??? maybe not ??
# 900223 <= Honda R & D

also
# 890836
# 890838
# 870610

and much more info from GM that i can't find tonight

you could just code a lookup table in software
based off this data


Larry Meaux (maxracesoftware@yahoo.com)
Meaux Racing Heads - MaxRace Software
ET_Analyst for DragRacers
Support Israel - Genesis 12:3

 

[MaxRaceSoftware]

FHP = ((RPM^2)* Stroke * CID) / ((13.7/((Stroke^(1/3))*(RPM^(1/4))))*350000000
===============================================

should read ;

COLOR 15, 1: CLS

RPM = 8500
CID = 355
Stroke = 3.480

FHP = ((RPM ^ 2) * Stroke * CID) / ((13.7 / ((Stroke ^ .3333) * (RPM ^ .25))) * 350000000)

LOCATE 5, 5

PRINT FHP
------------------------------------------------

Should give you 270.8 FHP

plugging into FTQ.BAS
RPM = 8500
CID = 355
Stroke = 3.480
should give you 249.1 FHP

the SuperFlow and GM code should give you 243.3 FHP
based on data


Larry Meaux (maxracesoftware@yahoo.com)
Meaux Racing Heads - MaxRace Software
ET_Analyst for DragRacers
Support Israel - Genesis 12:3

 

[willeng]

xr7755

kpuddarby

Larry

Thanks for the replies & the algorithms.
I would still like to set something up to measure friction &
pumping losses directly from the test engine. This is of major importance in this particular class--very restrictive rules.
I would like to separate the friction values from pumping losses. Any & all ideas on this would be appreciated---
E-mail if necessary.
Could you have a look at this simplified formula for Thermal efficiency. is it correct?.

TE = 1- (Te-Ti)/(Tp-Tc)- PL - F
Te = EGT
Ti = inlet Temp
Tp = temp @ PCP
Tc = comp temp @ 2000rpm without ign
PL = pumping losses
F = friction.
Your thoughts please.

 

[MaxRaceSoftware]

WILLENG ,

since you are still trying to solve problem with code
the &quot;formula&quot; you have
and not by actual measurements

why not use Honda's tremendous research and their
developement of formula for FMEP ..its going to be very hard to come up with something thats better than their
effort in R & D in that area, as far as a formula/code

if you want to actually measure this==
you could call or email SuperFlow for answers as to
what you might need to actually measure this,
they also offer an ECA Engine Cycle Analyzer software
along with equipment to measure in cylinder pressures



Larry Meaux (maxracesoftware@yahoo.com)
Meaux Racing Heads - MaxRace Software
ET_Analyst for DragRacers
Support Israel - Genesis 12:3

 

[MaxRaceSoftware]

WILLENG ,

you could calculate the Thermal Efficiency directly
from your live dyno data

TE = 2545 / ( BSFC * Fuel_BTU )

BSFC = dyno brake specific fuel consumption
Fuel_BTU = energy content of Fuel used in BTU's per Lbs.

you wouldn't be guessing that much if you knew BTU of fuel from manufacturer ?
----------------------------------------------

i don't know how you are going to calculate ;
Te = EGT
Ti = inlet Temp
(Te-Ti) without ignition ??
its a &quot;circle&quot; ..both dependent on each other
one will influence the other
if the engine isn't actually running under full load, then
data will be wrong compared to motoring the engine without
ignition..no accounting for wave-effects influences during overlap period (especially at low RPM and low effective cylinder pressures will be dramatically influenced by ignition/running loaded engine) or during any of the strokes without engine fully loaded and running with ignition

what units ? Rankine ? it looks like it couldn't be dimensionless ?

and
Tp = temp @ PCP
Tc = comp temp @ 2000rpm without ign

Tp = ??? what is ? and what units in all cases

Tc = compression temperature






Larry Meaux (maxracesoftware@yahoo.com)
Meaux Racing Heads - MaxRace Software
ET_Analyst for DragRacers
Support Israel - Genesis 12:3

 

[MaxRaceSoftware]

heres some info from deducing formulas from Honda's SAE papers ..could be errors , its just from old notes i have,

the actual code i have is in my computer program that i can't reveal

FMEP = (((RPM * Bearing_ID)^3)* Bearing_Length * Number_of_Bearings)/ ((Bore^2)* Stroke * Number_of_Cylinders)

FHP = (((RPM^2) * Honda_Factor) + .225) * .66339 * Engine_CID * RPM) / 5252

where .225 = 176 deg F oil temperature

PMF = (((RPM^2) *.000000006) + .225 ) * FlowFactor * ((Stroke * Crank_Journal_Dia)/ Bore)^.5

where
FlowFactor= Cyl_Vol_Cubic_Inches/ (((IntakeValveDia^2)*.7854) * IntFlowDischarge_Coefficient)^2

Larry Meaux (maxracesoftware@yahoo.com)
Meaux Racing Heads - MaxRace Software
ET_Analyst for DragRacers
Support Israel - Genesis 12:3

 

[MaxRaceSoftware]

and one more method i just thought of

try at your own risk !!!!

Example=>
Firing Order SBC 1-8-4-3-6-5-7-2

ground plug wires for 8,3,5,2 cylinders
make another fully-loaded , acceleration dyno test

note data

then ;
switch cylinder grounds to 1,4,6,7
repeat dyno testing

note Torque/HP/BSFC/Fuel Lbs/ etc
and should indicate pumping losses
still not the best accuracy because
of loss of wave effects/charge robbing , etc
but probably good enough

you would be using the actual engine as its own
motoring agent


Larry Meaux (maxracesoftware@yahoo.com)
Meaux Racing Heads - MaxRace Software
ET_Analyst for DragRacers
Support Israel - Genesis 12:3

 

[patprimmer]

If you use Larry's last method, you could minimise charge robbing effects by useing stack injection and zoomies, taking care that the intake trumpets were shielded from each other

Regards
pat

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips Fora.

 

[willeng]

Larry:

Thanks once again for the suggestions but what we are trying to establish is real world accurate figures on the particular engine we are testing.
I know all the scientific equations--physics etc to calculate the friction H/P
What i lack is an easy & quick method to get real values for friction & pumping losses for this engine.
As you are well aware i'm sure that in some cases every single H/P can be an advantage & it is in this case.
You mentioned the MORSE test for determining the values.
The MORSE test is usually performed by disabling 1 cylinder at a time & from this you can work out the indicated power using only brake-power measurements. From this you can calculate FP also mechanical efficiency etc etc.
Unfortunately the MORSE test as ingenious as it is underestimates the friction power & overestimates the indicated power because the assumption of constant friction in the disabled cylinder is not really valid. When a cylinder is disabled, friction is less because piston forces are lower.

The formula TE = 1- (Te-Ti)/(Tp-Tc)-PL-F was given to me several years ago by an old aeronautics engineer & who also worked for a fuel research company as the dyno operater, apart from being the best unrecognised engine man i have seen he was a good bloke, unfortunately before i could query the formula he passed away. I was told at the time it is very usefull in calculating TE without having to measure the energy content of different batches of fuel that can vary enormously.
AS far as i know Te,Ti & Tp are taken from the running engine. Tc is taken @ 2000rpm no ign.
The things that are missing are PL pumping loss & F friction. How do i measure these seperately quickly & accurately for our given test engine.
Thanks again.

 

[GregLocock]

Well that's interesting. Given that most of the sideload on the piston is usually inertial I'm a bit surprised that a no-spark test will underestimate the losses... in fact recent threads might suggest the opposite.

I can't see how to get round this unless you have true IMEP mapping, even then, the errors will be big.

Hey Larry, good to see you back.

Cheers

Greg Locock

 

[MaxRaceSoftware]

Good idea PatPrimmer !

Thanks Greg !!!

and congratulations on being this Forum's MVP !

old English/Austrailian terms =>
&quot;Bob's your Uncle&quot; means = &quot;Spot On&quot;
= means number One = means Forum's MVP


Larry Meaux (maxracesoftware@yahoo.com)
Meaux Racing Heads - MaxRace Software
ET_Analyst for DragRacers
Support Israel - Genesis 12:3

 

[patprimmer]

For a practical test on an engine, could you mount the driven engine and the drive on seperate frames that can rotate relative to each other, but only around the crank axis and concentric with it.

If there was a device to resist this rotation between the frames, with a load cell mounted to measure this resistance, the torque to drive the motor could be very easily calculated.

Regards
pat

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips Fora.

 

[MaxRaceSoftware]

TE = 1- (Te-Ti)/(Tp-Tc)- PL - F
Te = EGT
Ti = inlet Temp
Tp = temp @ PCP
Tc = comp temp @ 2000rpm without ign
PL = pumping losses
F = friction.
=======================================

Looks better as ;

Thermal_Efficiency = 1- (((EGT-Ti)/(Tp-Tc))- (PL + FL))

Te = EGT (Units ?)
Ti = intake Temperature (Units ?)
Tp = Temperature at Peak Cylinder Pressure (Both Units ?)
Tc = compression temperature @ 2000rpm without ignition
PL = pumping losses (Units ?)
FL = friction losses (Units ?)

6 variables to work with = 720 min permutations

EGT = the exact placement and depths of EGT probes
will definetly effect accuracy of above equation
this i know for certain makes a difference from my dyno experiences with placement ..can be easily 150 to 300
degrees off if not very careful

Ti = again, where will the reading be taken and exactly when ?

Tp = again, extremely hard to measure because research has
shown Peak Pressure and Temperature to vary on EACH cycle..and whats printed/recorded in dyno data is really an AVERAGE of that single RPM point as it goes thru sweep/acceleration Example=> 2000 RPM =16.67 firings/strokes per second

PL = will not be exact

FL = will not be exact
--------------------------------------------------
then compare to this simpler equation;
Thermal_Efficiency = 2545 / ( BSFC * Fuel_BTU )
BSFC = dyno brake specific fuel consumption
Fuel_BTU = energy content of Fuel used in BTU's per Lbs.

where 2 variables = 2 permutations min
a tremendous accuracy advantage over other previous equation

its easy to accurately measure and control the BSFC readings ..so that only leaves Fuel_BTU which from all my
race gas dyno testing leads me to believe this is also
very accurate once BTU is actually known

Larry Meaux (maxracesoftware@yahoo.com)
Meaux Racing Heads - MaxRace Software
ET_Analyst for DragRacers
Support Israel - Genesis 12:3

 

[willeng]

Larry:

I couldn't agree more, well done.
Thanks to everyone it's always helpfull to hear other opinions on things. Keep up the good work.
Thanks: