Blueprinted engines
Veeduber
Chris asked the Newsgroup:
>
>I have seen many Internet ads for aircooled VWs which tout the engine
>(or engine rebuild) as being "balanced and blueprinted".
>
>Please, could someone tell me the meaning of that phrase? And why it
>would be significant to a potential buyer?
>
------------------------------------
Chris,
'Blueprinting' as a generic terms refers to the comparison of a finished part
against its original specifications, typically by reading the specs off the
engineering drawing. (Real 'blueprints' -- white-on-blue Ozalid prints -- went
out with high button shoes, or when I was in high school... one of those.)
In modern-day automotive terms however, 'blueprinting' has a more specific
meaning. But to understand this specialized usage of the term you first need to
understand about tolerances, and here's an example that may help.
Let's say you have a 1" diameter shaft that must be supported in a bearing. In
non-machinist terms you would think the hole in the bearing would also be one
inch in diameter but in the real world, things don't work like that. First
off, you would spec the diameter of the shaft as 1.000". The three zeros tell
the machinist you want the part fabricated to an accuracy of a thousandth of an
inch. But if you spec'd the part at EXACTLY 1.000", to achieve that degree of
precision the shaft would have to be ground (rather than turned) and the
wastage rate -- the percentage of shafts that came out a little bit too large
or too small -- and the production cost, would be fairly high. So you define a
TOLERANCE -- a range of sizes that will be acceptable. For the sake of this
example, lets say you spec the tolerance at a thousandth of an inch. Now your
spec would read 1.000 +/- .001. This tells the machinist she has TWO
thousandths of an inch to work with, since the acceptable range is from .999"
to 1.001". Using standard production techniques, the distribution of the parts
across this tolerance would follow a classic plateaukuric curve -- about 66% of
the parts would be dead-on 1.000", about 15% would be smaller (ie, .999") and
15% would be larger (ie, 1.001"). (The other 4%, you hide under someone else's
lathe when they aren't looking :-)
Now you've got your shafts. The next step is to spec your bearing.
Since the shaft has to rotate, the bearing must be larger than the shaft,
otherwise there would be no room for the lubricant. So you spec the bearing as
1.0015" (Alright, all you rowdies in the back there who already see the punch
line, just stick a sock in it -- it's an example, fer crysakes!)
And again, the bearing must have a certain tolerance. So you go crazy and spec
plus or minus five-ten-thousandths of an inch -- 1.0015, +/- .0005.
(Can you see where this is heading? :-)
Your bearings are going to range in size from 1.001 to 1.002 Which is good.
But your shafts range in size ffrom .999 to.... oh-oh 1.001... which is bad.
So your shafts go into one bin and the bearings go into another and a robot
puts them together... oops. The robot just tried to put a 1.001" diameter
shaft into a 1.001" bearing and discovered they didn't fit. Or mebbe they did
fit (imagine it's a STRONG robot!) but once fitted, the thing would not rotate.
Or if it did rotate, it quickly overheated and failed because the fit was too
tight -- there was no room for lubricant.
Then too, there were the 15% of your shafts that measured .999". Mate one of
these with the 15% of your bearings that measured 1.002" and you have
three-thousandths of an inch of slop instead of the desired .0015"
So whatz to do? In the real world, you make the tolerances pretty wide -- you
design a lot of slop into any device meant to be mass-produced. Because if you
don't, the price of the thing goes right through the roof -- like a Rolls-Royce
-- or a hand-built, blueprinted engine. (Fact is, modern-day automated
production facilities hold to tighter tolerances than in the past. But a
hand-built engine can still show some improvements over a 'factory job'.)
The tolerance for most VW engine parts is from three to five thousandths of an
inch. On average, chances are you'll get a pretty good engine out of that --
the errors tend to balance out -- but in reality, the quality and fit of the
final product reflects the same statistical curve as the tolerance of its
individual parts -- most are pretty good and a few are exceptional. But a few
are real dogs. That's the reality of mass-production, be it automobiles or
washing machines.
You can't win a race if your engine is a dog. Even 'pretty good' will only get
you to the finish line. To get to the finish line ahead of everybody else, you
need an exceptional engine. And that's where 'blueprinting' comes in. When
you 'blueprint' an engine you determine its PRECISE dimensions. And find a
matching part that PRECISELY matches those dimensions, with exactly the right
amount of clearance and a perfect fit. Often times, that means going through
two dozen connecting rods to find four 'good' ones or re-machining one part...
or a dozen parts... to achieve the most perfect fit.
Blueprinting an engine is an expensive, time consuming process requiring a host
of tools most mechanics don't even own. Assembling such an engine is even more
expensive -- I've gone through as many as five sets of main bearings to come up
with the 'perfect' four needed for a high-output engine.
And that brings us to 'balancing'.
Just as all parts have a dimensional tolerance, parts which rotate or
reciprocate also have a mass tolerance. And like Volkswagen's dimensional
tolerances, the mass tolerances of your VW engine are woefully wide -- another
whiff of low cost, mass-production reality.
For example, Volkswagen sez its okay to use a connecting rod that is four grams
-- a quarter of an ounce -- heavier or lighter than its mates (ie, 580 to 588
grams). By comparison, the spec for modern American-built automobiles is a
gram or less.
Guess what happens if all of your connecting rods weigh within one-tenth of a
gram of each other? And if they have the same center of mass as well? The
answer is really no surprise -- the engine runs better. A balanced engine
doesn't have to waste energy overcoming those imbalances. Same amount of fuel
but more torque delivered to the flywheel -- free horsepower! (Well... not
really. Balancing costs money. But once you've paid the price, you get to use
those 'free' horsepower for as long as you own the engine. The bottom line is
that a balanced engine pays for itself.)
Volkswagen's biggest failing in the balancing department is that they don't
balance the rotating components AS AN ASSEMBLY. The rule is, everything that
rotates in a given plain should be balanced together, which means your pulley
and crankshaft and flywheel and pressure plate are first balanced individually
then assembled into an intergal unit and balanced all together. Anything less
and you can't really say the thing is 'balanced'... because it's not. That's
why it's a bit of a joke to buy a short-block [ie, sans flywheel]. Or to
change flywheels [or pressure plates] without having them match-balanced.
(Saying things like this always causes a lot of kickin' & screaming from the
shade-tree types, most of whom haven't a clue as to what I'm talking about.
Sure, it'll run... just throw that thing together and drive on... for a while.
The truth is, most veedub owners have never driven a properly assembled VW
engine. And never will. Too expensive, unless they build it themselves.)
But all of that speaks only to mass-balancing. You also have to consider
volumetric balancing -- making sure each cylinder has exactly (or as close as
you can get it) to the same displacement as every other cylinder. If each
cylinder has exactly the same volume it will draw in exactly the same volume of
fuel/air and, on ignition, provide exactly the same amount of power. (Okay, so
I'm waving 'exactly' around like a flag when the truth is, nothing is every
'exactly' this or that -- the message is the same. You don't have to be a
rocket scientist with a shop full of tools to cc your heads to within 1cc
across all four, nor get your deck-height to within a thou. Takes time and
calls for good attention to detail but there's no mystery to doing it.) The
result is very similar to the benefits of mass-balancing -- more power for the
same amount of fuel -- the engine runs more EFFICIENTLY. It produces more
power for the same fuel. It also lasts longer and runs one hell of a lot
smoother while doing so. And that's always nice.
So on the whole, 'balancing' and 'blueprinting' are good -- all VW race engines
are balanced and blueprinted. At a cost of about five grand per copy.
As to buying a 'balanced & blueprinted' engine for your bug... Why don't you
start with one of those factory-built VW engines from the Puebla plant in
Mexico? Cost you about a thouand bucks for a brand-new, spec-built Volkswagen
engine... from Volkswagen. Over the twenty years or so it'll take you to wear
it out, you can come up to speed on building your own balanced, blueprinted
engines. But here comes the kicker: When it comes to really sweet-running
engines, balancing and blueprinting is just the tip of the iceberg.
-Bob Hoover
Dreamer
--
Dreamer - '69 Autostick - pbu...@fastlane.net
___
/___\
(o\ | /o)
u-----u
Veeduber <veed...@aol.com> wrote in message
19981029061200...@ng137.aol.com...
PumaRacing
something to do with building to factory tolerances. It is building to the
tolerances set by the man who designed the engine before the limitations of
production engineering reared its ugly head.
As an engine builder who does build to tolerances that the engine wants rather
than what the factory can achieve I would like to throw my tupence in.
99% of the things that people believe make a difference to the way an engine
performs and nearly everything that has been mentioned in the thread actually
make almost no measurable difference at all.
The engine that I build most of is the Ford 1600 CVH engine and most of these
over the last 5 years have been for a race series which calls for engines to
remain at standard Ford spec. Any horsepower gains to be had therefore call on
very careful engine work because no porting, camshaft, carb work etc is
allowed. This has given me the opportunity to study some 40 engines over the
years and to try different things and to get feedback from dynos on what makes
a difference to power.
The only thing that makes any measurable difference is the difference in
cylinder head airflow between different castings.
You can mess about with piston/bore clearances, bearing clearances, balancing
things to 0.1 gramme, boring exactly at 90 degrees to crank centreline and see
no more than 1 or 2 % in power improvement.
Over that range of engines I have hundreds of flow tests and the difference in
flow between the best and worst standard cylinder head is more than 10 bhp.
When you buy a standard production car they range between lemons, good standard
and the occasional flyer. Nearly all of those differences are related to
airflow not internal engine tolerances IMHO.
To build an exceptional engine for customers with deep pockets I got them to
search scrapyards and send me as many cylinder heads as they could find. I
flowed them all and put the highest flowing head on the engine. Over the years
I developed computer programmes that analysed flow data and related that to
power figures. I can now predict engine power on a range of engines to as close
as 1% (with certain limitations which I won't go into here). Power is directly
and linearly related to inlet valve airflow. Every engine that I tested gave
the power that the airflow predicted despite the fact that thousands of other
things like exhaust system, oil viscosity, internal tolerances etc should have
been affecting things. I believe that all these other factors made minimal
difference and also just tended to average out of the equation.
The only other "blueprinting" type exercise that I would say makes any major
difference is the exact way that the bore is honed so as to seal to the piston
rings as well as possible. The grade of stone, cross hatch angle, depth of cut,
type of plateau finish etc can greatly affect power and oil consumption. One of
my engines which produced 86 bhp at the wheels went to another engine builder
when the car owner sold the car to another driver who didn't know who had
originally built the engine. This "engine builder" knocked the power down to 78
bhp at the wheels on the same dyno by what I can only describe as completely
fucking up the bores with a cheap flapwheel type glazebuster tool rather than
using the correct honing equipment.
I found this out when the engine came back to me a year later after the new
owner found out who had done the original build and wanted to regain the
performance it once had. For those specialists out there who will understand
this - the bare bottom end, no cylinder head, had a turning torque of 30 ft lbs
even after a seasons running with this awful bore finish. The pistons had to be
drifted out of the bores with a club hammer because there was so much friction
against the rings.
As to all the other stuff, if you stay within factory specs on bore & bearing
clearance and all thrust tolerances the engine will be fine for most
applications. For very highly stressed uses it can become more important to
ensure that bearing clearances (especially big ends) are spot on if oil
pressure is to remain good.
As to the cost of blueprinting I fail to see why it should be expensive as
people suggest. To have cranks and bores machined to an exact size and to
balance all parts properly costs next to nothing over an average "chuck it
together" engine build. For an average 4 cylinder engine I charge about £300 to
do all this work to race tolerance rather than road tolerance so where on earth
the figures of thousands of dollars come from as have been suggested I have no
idea.
Dave Baker at Puma Race Engines (London - England) - specialist flow
development and engine work.
nosp...@oceanstreetvideo.com
>
>Guess what happens if all of your connecting rods weigh within one-tenth of a
>gram of each other? And if they have the same center of mass as well? The
>answer is really no surprise -- the engine runs better. A balanced engine
>doesn't have to waste energy overcoming those imbalances. Same amount of fuel
>but more torque delivered to the flywheel -- free horsepower! (Well... not
>really. Balancing costs money. But once you've paid the price, you get to use
>those 'free' horsepower for as long as you own the engine. The bottom line is
>that a balanced engine pays for itself.)
>
balancing pays for itself in added longevity, with little
additional horsepower gain. that is per rodger crawford of heads-up
performance, in response to a discussion i had with him on this very
subject.
rodger has a computerized dyno that he has made thousands of
pulls on. this guy knows how to build flat four performance engines,
his car still holds the pra pro stock 1/4 mile record.
my personal uneducated take on this subject is that energy is
never wasted, merely redirected where you don't want it to go. if you
had any kind of significant horsepower loss because of poorly balanced
internal engine components it would have to show up somewhere. i
believe it would show up as very noticeable vibration, the degree of
which would be directly proportional to the amount of the imbalance.
but then, i'm no mechanical engineer... :-)
dan