Starting a 1971 IMSA 914 Restoration, A privateer run 914 with Sebring and Daytona history Options

[FourBlades]

People have been asking what the hold up is with me getting this car done.

It took me a while to realize but now I need to do something about it.

I don't like the way the front fenders line up with the front panel.

So I got some really nice used fenders from Bruce Stone so I can re-do them.

It would really help to have a NOS front panel as well, or one that is pristine from a car wrecked in the back half. The panel on there was sectioned horizontally from two different cars and is not dimensionally correct (I think, need to make a template to be sure).

I think I unconsciously did not want to go backwards to fix this but it needs to be done to be right.

John

[carr914]

I got this eMail from Sebring this Morning (look 1/2 Way Down) (IMG:style_emoticons/default/driving.gif)


Sebring Hall of Fame
Class of 2014 Announced

The Sebring International Raceway Hall of Fame has announced its 2014 inductees.

Vic Elford, Wayne Taylor, Bob Tullius, Peter Gregg and Brumos Racing join the prestigious Sebring Hall of Fame's class of 2014 to be inducted the day before the 62nd running of the Sebring 12-hour classic.

Vic Elford won the 1971 Sebring 12 Hours co-driving a Porsche 917 with Gerard Larrousse. He also finished second at Sebring in 1968, the same year he won the 24 Hours of Daytona. Elford won class twice at the 24 Hours of Le Mans (1967, 73) and has 13 F1 World Championship starts to his credit



Wayne Taylor won the 1996 Sebring 12 Hours co-driving with Eric van de Poele and Jim Pace. He also finished third in 1994. Taylor has won the Rolex 24 at Daytona twice and also won the inaugural Petit Le Mans at Road Atlanta in 1998. He currently campaigns a prototype in the new TUDOR Championship, driven by his sons.

Bob Tullius and his Group 44 team pioneered the concept of professional motorsports with immaculately prepared race cars and transporters. His first of 14 starts at Sebring came in 1963 at the wheel of a Triumph TR-4. Tullius won the GTO category in his Group 44 TR-8 and finished fourth overall in 1985 driving his Jaguar GTP prototype.

[URY914]

Looks like that 914 had just pasted the 935 in the background and was closing fast on that TR-8. (IMG:style_emoticons/default/biggrin.gif)

[FourBlades]

I got the same email too!

Hoping to make some progress this holiday.

John

[FourBlades]

I have been wondering about the straightness of the chassis. In the front, the
fenders do not look quite right and one rear control arm needs 1.5 cm more shims
than the other to achieve the same camber. I wanted to check this out so I did
some measuring using advice from Rick and others.

First I got the car up on stands and leveled it as best I could. I was not worried
about leveling it along the length so much as side to side. I wanted to detect if the
car was twisted along its length. Using different thickness of plywood I was able to
compensate for the slope of my garage floor.



I leveled it across the pinch weld on top of the longs. My thought is get the middle
of the car level and see what each end does.



Sorry about the poor picture quality. I feel bad for not providing better quality given
how much you all have to pay to see this. (IMG:style_emoticons/default/biggrin.gif)



I checked the floors from underneath using a level attached to a tram gauge. The
tram gauge has fingers on the ends that bridge the convolutions of the floor pan.



After checking multiple times and places I got it as level as my primitive equipment
would allow. Now I set up a laser level at the front of the car. Using shims of
different thickness paper I leveled the beam.



Now theoretically the car and the beam are level to each other. Using a square and
a ruler I checked the distance down from each front suspension bolt to the beam. I
picked this bogus ruler because the laser line showed up the best on it. It was really
hard to see on my steel rulers.



What I was really interested in was the difference side to side. I measured the two
front control arm bolts and the two bolts holding on the cross beam multiple times
to be sure I got a consistent reading. I found that the most any pair was off side to
side was 4 mm!!! Two pairs of bolts were within 1 mm!!

I think if my leveling efforts were really bad they would all be off left to right by a
consistent amount. Or if the car was really messed up then the measurements
would be off by more than 1 cm.

I think this means the front end is ok. Now on to check the back end.

John

[FourBlades]

I set up the laser level again at the rear of the car.



Then I checked the levelness of the floor of the car in the engine compartment.



I did this by comparing the measurements from side to side. They were almost
exactly the same given the crude method I was using. The beam is about 1 mm
wide and not sharply delineated making it hard to read exactly. I used a tape
measure and hung it down like a plumb bob.



Then I measured the inner and outer attachment points for the trailing arms.



The angle of the photos make the tape look really crooked. When I actually made
the measurements I took care to keep it vertical.



The outer attachments were off by 1/4 inch and the inners are off by 1/8 inch. I
am sure Hubert De La Metric is spinning in his grave that I mixed Metric and English
measurements in the same project, but I don't have a metric tape measure. (IMG:style_emoticons/default/huh.gif)

Being off 1/4 inch seems pretty good. (IMG:style_emoticons/default/beerchug.gif)

A member suggested I check the trailing arms to see if maybe one was bent.

I made this simple jig to check them.



The left and right differ by 5 mm at the hub. I need to calculate what this translates
into in terms of camber differences. I measured them repeatedly and always got the
same answer.

Now I am not sure if 5 mm at the hub equals 1.5 cm of shims?

What do you all think?

John

[trojanhorsepower]

You never cease to amaze. Very impressive use of what you have on hand.

[FourBlades]

Thanks! I wonder how accurate all this cobbled together stuff is though.

What makes me somewhat confident things are good is that it seems unlikely that
the inaccuracies in my approach would all come together in pretty close to exact
agreement. If the laser level was off or the car level was off, I would think the
difference side to side would be big and not almost nothing. (IMG:style_emoticons/default/idea.gif)

I can only guess that Vince and Manuel started with a pretty straight car and the
cage they installed kept it basically straight through at least two accidents and all
the welding I have done to it.

John

[rick 918-S]

Nice work. There is no reason a guy can't check your chassis without spending a fortune. All it takes is time and effort. You don't need a Celette for every little thing.

For an old race car that's not shabby. Your 1/4" = just over 6mm. The factory tolerance is 3mm or less. (three dimes thick) Add that to the 5mm from the hub and carry it out over the distance of the wheel center line, add some camber and I think that is what you are seeing when you say the car looks crooked.

I'm less concerned with the individual bolts in the front. In a cluster like that you can take an average and your ok. If your still wondering about the front check the torsion bar tubes to your laser line just behind the forward torsion bar cover. Centerline here is a consideration. If one torsion bar is pulled in your alignment will be wonky.

One final check you may want to make is the center line forward to back. You can take some #9 tie wire and hang it down from the suspension mounts, take a measurement at the top of the wire. take a level or something you are sure is straight and hang it from the wire. You will want three points down the car. mark center of the levels or what ever your using. Shoot your laser down the center line and see what you get.

The normal part of the process is to establish a datum line. We don't have datum in the 914 info link on the site. But if your upper body measurements are good you can just set the cross bars evenly and establish your own for this exercise.

They make old school centering gauges for this function but they are not necessary. With the cross bars set an equal distance from the floor pan shoot your laser down the outer end (front to rear) of the cross bars. This will show chassis twist. Then it's a matter of determining which corner is high. I doubt you'll see much difference. It looks like you found the issue already. My post is to bring to light the distance you can go with simple tools. Now repairing chassis issues is a completely different subject.

[FourBlades]

Thanks Rick!

Your idea for using a laser level was the key to doing something useful. (IMG:style_emoticons/default/first.gif)

I will check the torsion bar tubes next. The idea here is that they need to be parallel
to each other and to the center line of the car, right? I think I can find the center
line using my tram gauge pretty easily. Maybe I will run a wire or chalk line down
the bottom of the car on the center line. It should be easy to measure the torsion
bar tubes to the center line.

To make sure I understand about your second suggestion, you are saying to
suspend 3 levels cross ways between three sets of suspension points lengthwise
down the car and see if they all line up horizontally. If I use equal length wires then
a straight car would show them all lined up in a flat plane under the car.

It the car is twisted lengthwise then I won't be able to use equal length wires and
get the levels to be level?

I might need to draw a picture of this procedure to get sure I get it.

John

[ChrisFoley]

The idea here is that they need to be parallel
to each other and to the center line of the car, right?

Nice work John.

Parallel to each other - yes.
Parallel to the chassis c/l - no.
The torsion bars are higher at the front.

A measurement suggestion at the rear pickups - attach a bare trailing arm pivot shaft to both sides using no shims at the outer pickups. Then measure their inclination with a digital level which has been carefully zeroed. The factory built about one degree of negative camber into the pivot shafts.
Also - use the digital level to set up your trailing arm measurements. When the pivot shaft is level and hub c/l is at the same elevation, there should be approximately 1.5 degrees of camber at the hub flange.

[FourBlades]

OK, thanks Chris.

When you say digital level, do you mean like those used in digital camber gauges?

I need to pick one of those up.

John

[rhodyguy]

can you find 4 points to measure and check for squareness on the diagonal? having never traveled this path, its just a suggestion. the transformation you've done is incredible with this sad old racer. love the interior light.

[ChrisFoley]

OK, thanks Chris.

When you say digital level, do you mean like those used in digital camber gauges?

I need to pick one of those up.

John

I bought a Smartcamber gauge when they first came out. I also bought a 4 foot long frame with a pocket for the same level. They make a perfect combo for leveling the work platform and making wheel camber measurements. With the level removed from the frame it has numerous other uses around the shop.

There are other digital inclinometers available too.
Amazon has a number of inexpensive options.

[FourBlades]

can you find 4 points to measure and check for squareness on the diagonal? having never traveled this path, its just a suggestion. the transformation you've done is incredible with this sad old racer. love the interior light.

I had the same thought while doing the major rebuilding of the car.

When I had the car upside down I did a lot of side to side and diagonal
measurements of all the suspension and other major points. Those all checked
out to within a few mm. Before welding the new front part on I checked this
several times.

I have done a bunch of diagonals across the front trunk and most check out ok
with one exception that I will post soon.

This project has taken a long time but I am still stoked about it.

I only work on it when I want to so as not to make it into a drag.

PS. I just ordered the SmartCamber gauge. Not sure why I waited this long to get
one. I need to get their string gauge next when I get further along.

John

[rick 918-S]

(IMG:style_emoticons/default/agree.gif) with Chris. The front of the torsion bars are higher than the center section of the pan. So looking at the car from the side the distance from the datum line would be higher to the front than the center. You can still use the levels set at the same distance from the bottom of the car and sight down the plane to pick up visible twist in the chassis or a deflection off center. I wish I still had my old chassis manuals. I would have the drop down measurements. (IMG:style_emoticons/default/confused24.gif) For those that don't know and are following along the datum is an imaginary line to use as a reference to determine various points on the chassis are true.

[FourBlades]

Rick, the datum is a horizontal line or plane under the car, right?

I found this reference with lots of auto measuring info:

http://books.google.com/books?id=zh94U0W_h...air&f=false

This is going to take a while to study... (IMG:style_emoticons/default/drooley.gif)

John

[maddin]

chassis4712912215
what a great read!
found some pics of the car...and results here
world sports racing prototypes

great reference sites.
cheers marty

[rick 918-S]

Rick, the datum is a horizontal line or plane under the car, right?

I found this reference with lots of auto measuring info:

http://books.google.com/books?id=zh94U0W_h...air&f=false

This is going to take a while to study... (IMG:style_emoticons/default/drooley.gif)

John

Yes, Nice find. That is very clear information.

[FourBlades]

Ordered the book from biblio for $17.56 plus shipping:

http://www.biblio.com/details.php?dcx=5521...gn=feed-details

Thanks for the links, Marty!

John