Car Shakes at 50 mph, but not always

[expat]

There are 4 speeds at which I can 50mph, engine speed makes no difference.
I have also run the car in gear while up on a lift.
Nada.

The drive-shaft was changed because the Stock unit used a CV joint at one end, that did show some wear.
A new shaft was made up by a reputable shop that specializes in Drive-shafts.
No change.

The engine runs perfectly. This is a fully balanced (to within 1/10 of a Gram) engine, Carburetors (Twin SU) and ignition (Nippondenso electronic) are fully sorted.

 

[Cujet]

Frame resonant frequency would be my guess. FYI, this is a typical dynamic of convertible cars. (and Bronco II's)

This will happen when unsprung weight is too great, or wheel/tire inertia can not be effectively damped by the chassis.

The first thing I'd try is lowering the tire pressure to a very low number (guessing 15PSI) and trying again. The next thing I'd try is borrowing a lighter set of wheels/tires.

Miata's are very prone to something similar. They typically shake at 65MPH. In the case of the Miata, lighter tires with softer sidewalls will help. I'm not saying the car performs better with softer tires, just that the shake is reduced or effectively eliminated with light/softer tires/lighter wheels.

Even my S2000, which is considerably stiffer than my Miata has a bit of "cowl shake" resonance.

 

[MaxZX]

Originally Posted By: Cujet

Miata's are very prone to something similar. They typically shake at 65MPH. In the case of the Miata, lighter tires with softer sidewalls will help. I'm not saying the car performs better with softer tires, just that the shake is reduced or effectively eliminated with light/softer tires/lighter wheels.

Even my S2000, which is considerably stiffer than my Miata has a bit of "cowl shake" resonance.


Our 99 miata has the same shake around 65/70. Best cure for it we have found it to make sure the tire PSI is set to factory spec when cold, which I believe is 26 psi. That virtually eliminates that annoying shimmy. In fact, whenever it sppears, wife asks me to check the tire pressure and sure enough, it needs minor adjusting. OP, I'm sure you keep a good handle on the tire psi, but just in case it'd be worth checking and setting to the factory cold spec, measured when tire is cold (like first thing in the morning before driving).

OP, another thought, have you checked the ball joints? I know nothing about the car you have (other than I've always wanted one, such cool cars). But I've read bad ball joints can cause shakes at certain speeds in some cars. Just a thought.

 

[Silverado12]

Check the ball joints.

 

[default]

Originally Posted By: expat

I have also run the car in gear while up on a lift.
Nada.


On a lift the engine/powertrain is unloaded, a failing coil for instance will often free rev just fine, but run terribly under a load. I'm not saying its your problem, but you seem a bit quick to rule it out.

 

[yvon_la]

Lift car and put something then turn the wheel do each corner ,make sure to also check on the brake side (might need to be 2 )probably a disk or drum brake issue (i assume you did a good job on the wheel)might be either too many part in some place and not enough in other place .you ll neeed to find factory spec picture to get the idea of where the issue might lie (espacially if you modified.

 

[yvon_la]

Unless you didnt align the car in x?so you dont end up with a flat rectangle or weird shape?measure from a frame fixed point and from there measure to each corner (try to find the one from factory (the frame measuring point ,on one truck i drove all axle were proper distance with each other same side but in x one side was 1 inch more forward .(error when the shop calibrated the computer machine that measure.

 

[expat]

Originally Posted By: Cujet
Frame resonant frequency would be my guess. FYI, this is a typical dynamic of convertible cars. (and Bronco II's)

This will happen when unsprung weight is too great, or wheel/tire inertia can not be effectively damped by the chassis.

The first thing I'd try is lowering the tire pressure to a very low number (guessing 15PSI) and trying again. The next thing I'd try is borrowing a lighter set of wheels/tires.

Miata's are very prone to something similar. They typically shake at 65MPH. In the case of the Miata, lighter tires with softer sidewalls will help. I'm not saying the car performs better with softer tires, just that the shake is reduced or effectively eliminated with light/softer tires/lighter wheels.

Even my S2000, which is considerably stiffer than my Miata has a bit of "cowl shake" resonance.


I think this could be it.
During restoration I did attempt to stiffen the Cowl in an attempt to stop the steering wheel a Dash dancing around in front of me. But now the Frame in front of the Cowl cold be flexing.
I would still point my finger at the Tires as the cause, because, as I said, it only became noticeable almost immediately the new tires were installed.
Perhaps a small imbalance or Hard spot causing a Harmonic oscillation (as was suggested)

By the way, these are not Huge tires, 175-70-13's

Tomorrow I will take the car out and run the same stretch of road repeatedly, to try to get the car to shake in a predictable fashion.
If I can do this, I will adjust the tire pressures to see if this can effect the degree of shake.

I eliminated an engine misfire as the cause, when I discovered the shake would continue even with the ignition turned off and coasting

 

[Gokhan]

In order to not have oscillations, you want,

n >= 2*[square root(k*m)],

where n is the viscous damping of the shocks, k is the spring constant, and m is the mass of the car.

So, in order to suppress oscillations, you want stronger shocks, and/or weaker springs, and/or lighter car.

 

[CapriRacer]

Originally Posted By: Trav
PM CapriRacer he is the man for tire issues, if anyone would have some idea if its tire related he would.


Well, I just stumbled on this thread - and thanks for the complement.

First, wheel end vibrations occur in the 50 to 70 mph speed range and would be there pretty much all the time at the specific speed they occur. They would not come and go. (Please note, "wheel end" means tires, wheels, brake rotors, and other rotating components.)

Rotating tires (and wheels) should have done something if the problem was tires and/or wheels.

Tire (and wheel) related vibrations would come strongly through the steering wheel, not so much visible on the hood or cowl. You have the opposite.

So I don't think it's the tires or the wheels.

So what is it? You need to do some more diagnostic testing.

I think the first step would be to find a stretch of road where you can conduct tests. The problem should not be there at 40 mph, and should appear at 50 mph, and disappear at 60 mph. Do each of those at steady speed. What you want is to assure yourself that the road isn't causing the vibration.

Please note, that I have seen roads where certain lanes cause vibrations and the lane next to it doesn't. Same for direction of travel. Be very careful as what you want is to be able to consistently get the vibration, but also to consistently NOT get it.

So once you can get the problem to appear regularly, the real testing begins:

I see where you have driven in different gears and eliminated engine - Although, coasting out of gear with engine at idle might also be worthwhile to verify it isn't engine.

Try leaning on the steering wheel - bearing slightly in one direction. That will stress one side of the suspension components. Do both left and right. A change would indicate worn suspension components.

I also see where you said that the driveshaft has only a single CV joint. Mmmmmmmm. So what's on the other end?

Try crowding the throttle - that is press gently on the accelerator pedal so torque is being applied. If that aggravates the problem, it might be a trans and axle bearing or even the rear axle angle.

Try accelerating hard. Trans and rear axle bearings and rear axle angle.

Let us know what you find.

 

[expat]

Thank you, I was going to contact you (on Trav's advice) but wanted to do some Real testing first.

As I said, I will be doing some runs today.

I'll keep you posted.

 

[expat]

Quote] "I also see where you said that the driveshaft has only a single CV joint. Mmmmmmmm. So what's on the other end?"


Originally the car's Drive shaft had a UJ at each end and a Strap Drive to accommodate any small movement in length.

(Note: the car has independent rear suspension, so the Diff is mounted to the vehicle Frame)

When the car was re-built I fitted a Laycock Overdrive. This required a shorter Drive shaft.

It so happened, that Later model Spitfires (while under British Leyland management) had shorter Drive shafts, that happened to be the correct length for my application.

However, BL thought fitting a UJ at one end of the shaft and a CV joint at the other would be a Good Idea

This arrangement is known to be problematic, But as I had such a drive shaft at hand, and in good order, I used it. and it worked just fine.

However, when this Shake problem manifested, I DID suspect it could be this 'problematic' shaft, so I had a custom DS shaft with a splined slider and UJ's at each end made at a shop that specializes in drive shafts.

Incidentally, the car can be driven with the Transmission cover removed.
It is possible to see the DS running smoothly, but a shake in the transmission and all things forward.

 

[expat]

OK.

There is a stretch of Highway near here that gets little traffic, has log straight sections, gentle curves and has recently been re-paved (it's consistently smooth and in good order)

I have been running a 5 mile section back and forth.
I started with all four tires at 30 psi, then 25 psi then 20psi

Findings:

Tire pressure did not seem to make much difference.

The car could not be made to shake consistently (no one section of the road could be said to cause the car to shake)

The Shake or oscillation, could start at 40 mph and increase frequency as the speed increases. Peak Amplitude would be 50-55mph and would fade completely by 70mph.

A slow curve in the road (left or Right) could provoke the shake, OR cause it to stop.

At a steady speed, for no determinable reason, the Shake could start, and CAN BE FELT TO BUILD in amplitude only to fade and disappear in much the same manor.

Left foot braking (while maintaining speed) Had no effect.

Coasting (even with the ignition off) had no effect.

The shake is not felt through the steering nor the Dashboard or Windscreen. the shake is most visible at the gear lever and by observing the Car hood (it may be necessary to understand the Spitfires construction to make sense of this)
http://files.conceptcarz.com/img/Triumph/72-Triumph-Spitfire-DV-09_PVGP-001.jpg

File photo.

 

[Gokhan]

Nice car.

It appears that the driving force of the oscillations is the wind, as the wind speed is equal to the vehicle speed in calm weather.

You're probably familiar with the Tacoma Bridge shaking and collapse. Similar destructive oscillations happened with other bridges in windy conditions. Read that and also about similar cases.

There are two basic solutions. One is to cover up the underside of the bridge so that the wind can't get into gap areas. In your case, you would put a good, uniform chassis undercover or perhaps some spoilers behind the front bumper to cut and redirect the wind.

The other solution, which could be implemented together with the previous solution, is to use stronger shocks that have more viscous drag force (more damping). Perhaps you could even look into some kind of adjustable shocks. Note that, with stronger springs, you need stronger shocks to prevent shaking. With weaker springs, you could use weaker shocks without shaking problem.

 

[Silverado12]

I don't see any weights on the wheels from your pic. You sure the tires are balanced?

 

[CapriRacer]

I'm going to truncate the posts to make it easier to read.

Originally Posted By: expat
Quote].......Incidentally, the car can be driven with the Transmission cover removed.
It is possible to see the DS running smoothly, but a shake in the transmission and all things forward.


Are you saying you say see the shake in the trans with the cover removed?

Originally Posted By: expat
...... a 5 mile section back and forth.
I started with all four tires at 30 psi, then 25 psi then 20psi

Findings:

Tire pressure did not seem to make much difference.....


That should have done something, so I think we can eliminate tires from the equation.

Originally Posted By: expat
...... The car could not be made to shake consistently (no one section of the road could be said to cause the car to shake)...


Again, I think anything rotating on the wheel end can be eliminated.

Originally Posted By: expat
...... The Shake or oscillation, could start at 40 mph and increase frequency as the speed increases. Peak Amplitude would be 50-55mph and would fade completely by 70mph.

A slow curve in the road (left or Right) could provoke the shake, OR cause it to stop.......


There's a clue. That should point to something worn in the suspension.

Originally Posted By: expat
...... At a steady speed, for no determinable reason, the Shake could start, and CAN BE FELT TO BUILD in amplitude only to fade and disappear in much the same manor.

Left foot braking (while maintaining speed) Had no effect.

Coasting (even with the ignition off) had no effect.

The shake is not felt through the steering nor the Dashboard or Windscreen.....


So 3 times, we can eliminate tires, wheels and other wheel end components.

Originally Posted By: expat
...... the shake is most visible at the gear lever and by observing the Car hood ......


That would tend to point to something in the drive train. With the driveshaft not pivoting as the suspension moves up and down, the gear lever should not shake at all if that portion of the car was fine.

 

[expat]

Originally Posted By: Silverado12
I don't see any weights on the wheels from your pic. You sure the tires are balanced?


The car picture Is not my car. It was just a pic off the web to illustrate the arrangement of the hood for those people how may not be familiar with the designe of the Spitfire.

From what I see there, I would not trade that for my car.

 

[expat]

Thank you Capriracer, and others here.

I will have the car up on stands this weekend, check suspension/steering components (again) rotate the wheels (again) and see what happens.

I did do this while the wheels were being re-balanced last year, But the only slightly worn component was a trunnions on the rear suspension.
It was replaced, but made no difference.

I am also thinking of fitting a cross brace between the front suspension turrets.
(that could be a little difficult to configure, and I have never seen one used before in this application)

The engine's front plate does act as a brace here. But of course, the engine sits on rubber mounts, so there IS some Give

 

[expat]

Are you saying you say see the shake in the trans with the cover removed?


Yes!

In fact, I did have a torque arrestor fitted to limit rotation of the engine/transmission during slalom launches and over-run.
But this was detached when this problem started, because it just transmitted 'more' of the shake into the frame.

 

[Gokhan]

It's not anything in the drivetrain because then changing the tire pressure would have some effect on it.

It's not a loose or broken part in the suspension because then you would see it at all speeds and it would greatly depend on the pavement conditions.

The fact that it's a nice Lorentzian curve with the amplitude starting around 40 MPH, maxing out at 55 MPH, and going away at 70 MPH is showing that it's a resonance caused by a driving force. Since the variable here is the vehicle speed, which is equal to the wind speed when there is no head or tail wind, the driving force is more than likely the turbulence caused by the wind, just like in the old Tacoma Bridge and several other bridges that experienced similar problems with designs generating air turbulence and the damping and flexibility parameters being in the wrong zone -- too rigid (strong springs in your case) and too little damping (weak shocks in your case). You can test if this is the case by driving the car against the wind and in the direction of the wind on a windy day, and it should make a big difference with the tail wind vs. with the head wind.

As some others said, it's a heavy car, and chances are that when the car was restored, they put springs that are too strong and/or shocks that are too weak. Heavy car + strong springs + weak shocks = oscillations.

The solution, as I said, is to put some chassis undercover and/or bumper spoilers to eliminate the wind turbulence and/or stronger or adjustable shocks to dampen the oscillations. Eliminating the wind turbulence and/or putting stronger or adjustable shocks that increase the damping should solve the problem.