Snaking

[xtrailman]

No, every snake is different, mine for example snaked over the lane i was traveling in.
surfers was across three lanes.
A simulation just isn't the same.
If it was F1 tunnel wind testing for example, would give the same result on the track, too many variables.

I had no warning of my snake, no time to reduce speed, no instability, before or after the snake.

 

[kelvin3]

I’ve been pondering my navel over this?

Here’s what I’ve come up with, what do you think?

The van swings on the fixed pivot and this equates to 1st harmonics, so as long as the ‘trigger/affect’ is removed it settles back down, with no action being taken.

The pivot starts moving (side to side) because the ‘trigger/affect’ remains. Dependant upon the magnitude of the trigger/affect is proportional to the pivot movement i.e. back end of the car travel side to side or 2nd harmonics. This can only grow proportional to the magnitude of the trigger/affect. As long as the trigger/affect is removed by what ever means and you remain in a straight line then it will drop back down to 1st harmonics and the settle. To me this is not a snake because there is no forward movement of the sway.

If a snake is third harmonics (an example being flicking a still rope and the movement travels along it). Then: While in the 2nd harmonic stage another trigger is applied i.e. movement from the straight line by either the car/van or trigger/affect then it will move into 3rd harmonics where the back end movement of the van is moved along the rig to the front end of the car (flicking the rope).

So! Lutz is correct about learning not to panic and remaining straight.

I think I’m correct with my labelling of the harmonics but I am sure someone will correct me if I’m not.

 

[ProfJohnL]

Thank you Lutz for the link to the Chris Killer study.

At last we have some scientifically produced data on which proper debate can ensue.

Clearly a considerable amount of work went into the study and some useful but limited findings were produced. Interestingly there is a lot congruity with my own views which are derived from empirical thought and experience.

The study requires further careful reading, but based on a speed read through a few points arise.

The study (as is common with many final pieces of graduate work) approaches the issues from an academic perspective, but is ham strung by lack of funding, facilities and time to have explored all aspects, and certain aspects of the topic were openly excluded some variables, but the justification for exclusion is not clear.

It tries to produce a a method of establishing a towing factor for a given set of basic characteristics. This is fine as long as the factor is seen as the ideal or topline, and it is understood that virtually all other factors will detract from that level. Chris mentions some of the factors such as height of the CofG and tyre compliance etc.

Consequently real road surfaces with humps and potholes are likely to detract from the topline

It seems there are still opportunities for further investigation into those excluded areas.

Of particular note is the non linear effectiveness of friction based stabilisers. Whilst they may raise the speed at which inherent snaking may start, it is the more violent way the snaking may start which will narrow the band in which the driver has time to correct the problem, and importantly, given a speed at which the outfit may be stable on a good surface, if the outfit has been disturbed and caused a major sway, the scale of the sway may overcome the damping effect of the stabiliser and be maintained until the speed of the outfit is reduced.

Of similar interest was the reference to the compliance of both the car and caravan tyre walls. I have previously suggested that the degree of lateral stiffness afforded by the cars and especially 4x4s rear suspension and tyres may be a factor in RTI's

I feel these two areas may provide some answerers as to why large and heavy vehicles can sometimes seem to be involved in a disproportionate level of rti's

 

[Lutz]

No, every snake is different, mine for example snaked over the lane i was traveling in.
surfers was across three lanes.
A simulation just isn't the same.
If it was F1 tunnel wind testing for example, would give the same result on the track, too many variables.

I had no warning of my snake, no time to reduce speed, no instability, before or after the snake.

If Surfer's snake was over three lanes and your's only over one lane then all this means is that it took him longer to notice it and react than you. Your's could have developed as seriously as his if conditions hadn't changed but, in principle, there was no difference between the two snakes.
Maximum amplitude of a snake is never achieved when the caravan sways for the first time. Because of the interia of the caravan anything else is physically impossible. It takes at least two full cycles to build up to the magnitude that Surfer experienced.

 

[kelvin3]

Typical final year project, which is biased towards learning rather than discovery. I can see a major flaw in that it is self induced rather than external induced and the initial focus changed i.e. from the toppling over of a van to stability at varying speeds. Therefore I would have liked to have read a definition of snaking?

My thoughts are that snaking is the transfer of rear end movement to front end movement as I have already said previously and it is this front end movement that moves you across the lanes or into the ditch. Toppling may or may not be a resultant issue.

What this project does however is show factors that effect ‘this threshold point” and are IMHO good reading for the I’ll informed. BUT! this threshold point my or may not move dependant upon the external affect, Variables that may or may not effect are the type, magnitude and duration of the affect. An example of this may be: I am doing 40mph, with a threshold point of 58mph. I am overtaken by a 44tonne artic at 50mph – according to this study; because I am well below the threshold point I cannot go into a snake! Obviously they have never been in this situation.

 

[Lutz]

I have corresponded with Prof Jos Darling on the report and he conceded that it is obviously not all-encompassing and there is room for further investigation. Financial and time constraints put a limit on what it could or could not cover. Perhaps the caravan industry would consider funding studies that carry on where Chris Killer left off.

 

[ProfJohnL]

Hello Kelvin,
I think your last conclusion is a little unfair. The study as you and I seem to agree does provide some intersting and usefull information but it is fettered by it purpose - which is to gain a degree, rather than to answer a question.

The study was looking at speed or self induced instability, it did not look look at external effects such as bow waves.

Regardless of the reason for the inception of the study, I think we should be greatful to Chris Killer for the work,and if nothing else it is a basis on which to start further investigations.

 

[kelvin3]

Lutz it is a project not a report and should be considered as such. Which direction do you think it should take; for better caravan design or how external influences effect the stability. And is there enough evidence to warrant the expenditure? Not that it’s not important but if rules are followed what is the proportion of accidents that come from none user induced snaking?

Is a gun dangerous or is it the person holding the gun?

 

[Lutz]

What I mean is that the effect of some of the variables needs to be looked into in more detail. For example, the study reflected the behaviour of only one towcar (1998 Mondeo Estate) with obviously only one suspension setup. Varying tyre pressures, both on the car and the caravan, were not taken into account, neither was a true comparison of the effect of changing noseweight included. The degree of improvement when fitting shock absorbers on the caravan was also left out. I could go on .......

 

[Dustydog]

The most important factor that I noted from Chris Killer's Thesis is the relationship between speed and the damping ratio. Thus at 60 mph he found the ability of the stabiliser to damp out a snake had all been depleted.
I've taken the time to read this document and at least it is better than nothing. His major factors which he feels affect the handling are well documented and in fact all have been covered on this forum in some depth previously.
I feel Chris's statement below will be the basics that any new caravanner should know.

11.1 Practical advice for caravan owners & designers
Nothing fundamentally new was discovered in this study, but theoretical factors
suggested in the past have been substantiated by road-tested evidence. The
theoretical advice that is now known to be significant in increasing stability includes
the following points.
For owners:
•
Make sure car is suitable for the size of caravan
•
Load heaviest items in the car if possible
•
Locate other heavy items in the centre of the caravan, just forward of its axle
•
Make sure tow ball load is adequately high
•
Avoid sharp steer inputs at high speed •
Never exceed 60mph
As our Prof says this was only a PHD theseis but imo opinion better than anyone else has done! It was never meant to be the authority but it does demonstrate what happens as speed rises.

 

[xtrailman]

The most important factor that I noted from Chris Killer's Thesis is the relationship between speed and the damping ratio. Thus at 60 mph he found the ability of the stabilizer to damp out a snake had all been depleted.
I've taken the time to read this document and at least it is better than nothing. His major factors which he feels affect the handling are well documented and in fact all have been covered on this forum in some depth previously.
I feel Chris's statement below will be the basics that any new caravaner should know.

11.1 Practical advice for caravan owners & designers
Nothing fundamentally new was discovered in this study, but theoretical factors
suggested in the past have been substantiated by road-tested evidence. The
theoretical advice that is now known to be significant in increasing stability includes
the following points.
For owners:
•
Make sure car is suitable for the size of caravan No mention of ratio's
•
Load heaviest items in the car if possible What if it isn't possible?
•
Locate other heavy items in the center of the caravan, just forward of its axle What if the nose weight limit is exceeded.?
•
Make sure tow ball load is adequately high Define adequate?
•
Avoid sharp steer inputs at high speed • What if an emergency situation makes this unavoidable?
Never exceed 60mph Well at lest thats something definite!!
As our Prof says this was only a PHD theseis but imo opinion better than anyone else has done! It was never meant to be the authority but it does demonstrate what happens as speed rises.

This chap could go on to be a politician!

 

[Royston]

I have read with great interest, the report, Every caravanner should apply with all the recomendations.
Even complying with the said recomendations, you can still get unstable caravans.
I had the misfortune to purchase a small fixed bed caravan, which when towed over 45 mph became unstable.
My thoughts were why should this van behave like this, when the previous van same external weights and dimensions, was very stable, On reflection the fixed bed model had all the internal heavy weight units on one side , This was also mentioned in the report as a cause for unstable towing.
It was also interesting in the report that a longer caravan complying with the recomendations to be a safer tow, I have a 5 berth max length for a single axle, tows like a dream.

 

[Dustydog]

This chap could go on to be a politician!

Not bad for a sunday afternoon Ray

 

[Lutz]

My comments in red:

11.1 Practical advice for caravan owners & designers
Nothing fundamentally new was discovered in this study, but theoretical factors
suggested in the past have been substantiated by road-tested evidence. The
theoretical advice that is now known to be significant in increasing stability includes
the following points.
For owners:
Make sure car is suitable for the size of caravan No mention of ratio's Ratios are of no help here because they do not reflect actual conditions and even then they cannot be seen in isolation from the respective moments of inertia ot the caravan and the towing vehicle.

Load heaviest items in the car if possible What if it isn't possible? All the author was trying to convey was that one should do the best that is achievable. A degree of weight transfer from caravan to car is always possible, however small, and every little bit helps.

Locate other heavy items in the center of the caravan, just forward of its axle What if the nose weight limit is exceeded.? It should be understood that the noseweight limit must be adhered to no matter how the load is distributed within the caravan

Make sure tow ball load is adequately high Define adequate? Enough to ensure a stable outfit and within the legal constraints is the obvious answer

Avoid sharp steer inputs at high speed • What if an emergency situation makes this unavoidable? It is for exactly this reason that technical aids such as stabilisers, ABS, ESP, TSP, etc. are fitted to caravans and cars, respectively. They are all there for the exceptional circumstance where the driver is unable to cope.

Never exceed 60mph Well at lest thats something definite!!
As our Prof says this was only a PHD theseis but imo opinion better than anyone else has done! It was never meant to be the authority but it does demonstrate what happens as speed rises.

This chap could go on to be a politician!
[/quote]

 

[Surfer]

No, every snake is different, mine for example snaked over the lane i was traveling in.
surfers was across three lanes.
A simulation just isn't the same.
If it was F1 tunnel wind testing for example, would give the same result on the track, too many variables.

I had no warning of my snake, no time to reduce speed, no instability, before or after the snake.

If Surfer's snake was over three lanes and your's only over one lane then all this means is that it took him longer to notice it and react than you. Your's could have developed as seriously as his if conditions hadn't changed but, in principle, there was no difference between the two snakes.
Maximum amplitude of a snake is never achieved when the caravan sways for the first time. Because of the interia of the caravan anything else is physically impossible. It takes at least two full cycles to build up to the magnitude that Surfer experienced.

Lutz you must have an amazing crystal ball that allows you to view my reaction time to the snake we had but either way I don't think any of your posts offer anything of value on snaking.

 

[Lord Braykewynde]

I can't comment on this subject because I have a phobia about snakes. When I mentioned this at a lecture by a psychiatrist whilst on an ambulance refresher course he told me not to worry because all it means is that I'll never become a homosexual

 

[graham_somersham]

Much has been mentioned about speed and fore and aft weight disposition but an important factor is also the centre of gravity and the effects of free surface from liquids ( and loose objects that start to move in a snake)and hanging items raising the vertical position of the C of G. I believe that righting moments (to use a nautical expression)play a big part in snaking as the suspension loading on each side of the caravan is changing rapidly and the c of g is moving dynamically until the righting moment is insufficient and the unit rolls over. Keeping weights low and over the turning point of the caravan as well as removing any possible free surface effect would contribute greatly to stability. I might have to dig out my old stability books from 40 years ago regarding metacentric heights and other parameters but I am sure the rolling effect and to a lesser degree pitching can be the trigger in a situation where the stability threshold is critical.

Has anyone seen research regarding snaking and the safety effect of shock absorbers of different ratings?

Safety matters such as this is what the NCC should be encouraging research into, but maybe I have expectations of the NCC that are too high for the structure of the organisation.

 

[Lutz]

Lutz you must have an amazing crystal ball that allows you to view my reaction time to the snake we had but either way I don't think any of your posts offer anything of value on snaking.

No crystal ball. As I said before, it takes 2 full cycles for the snake to build up to maximum amplitude. If you have a look at Chris Killer's paper 2 cycles take about 3 seconds. At 60mph that's 88 feet or more than twice the length of an average outfit. Anyone unable to react within that time shouldn't be driving.

 

[Lutz]

Much has been mentioned about speed and fore and aft weight disposition but an important factor is also the centre of gravity and the effects of free surface from liquids ( and loose objects that start to move in a snake)and hanging items raising the vertical position of the C of G. I believe that righting moments (to use a nautical expression)play a big part in snaking as the suspension loading on each side of the caravan is changing rapidly and the c of g is moving dynamically until the righting moment is insufficient and the unit rolls over. Keeping weights low and over the turning point of the caravan as well as removing any possible free surface effect would contribute greatly to stability. I might have to dig out my old stability books from 40 years ago regarding metacentric heights and other parameters but I am sure the rolling effect and to a lesser degree pitching can be the trigger in a situation where the stability threshold is critical. Has anyone seen research regarding snaking and the safety effect of shock absorbers of different ratings?
Safety matters such as this is what the NCC should be encouraging research into, but maybe I have expectations of the NCC that are too high for the structure of the organisation.

No doubt you are right and the variables that you mention also effect stability. However, this is adding even more variables into the equation and we are already in a position that we don't know fully by how much the few that have already been looked into really make a difference.
It just goes to show how complex to subject is and we will therefore probably never get the full picture, or if we do, it'll be so overwhelming that no-one will be able to understand it all.

 

[ProfJohnL]

Hello Graham,

You rightly point out there are more factors that must have some influence on instability in a snake,and probably at other times also.

As Lutz implies we will probably never have a universal model that will accurately predict the stability performance of all outfits, but I do think more work and investigation is necessary so some practical steps can be taken to understand and hopefully reduce instability issues.

 

[xtrailman]

No, every snake is different, mine for example snaked over the lane i was traveling in.
surfers was across three lanes.
A simulation just isn't the same.
If it was F1 tunnel wind testing for example, would give the same result on the track, too many variables.

I had no warning of my snake, no time to reduce speed, no instability, before or after the snake.

If Surfer's snake was over three lanes and your's only over one lane then all this means is that it took him longer to notice it and react than you. Your's could have developed as seriously as his if conditions hadn't changed but, in principle, there was no difference between the two snakes.
Maximum amplitude of a snake is never achieved when the caravan sways for the first time. Because of the interia of the caravan anything else is physically impossible. It takes at least two full cycles to build up to the magnitude that Surfer experienced.

Lutz you must have an amazing crystal ball that allows you to view my reaction time to the snake we had but either way I don't think any of your posts offer anything of value on snaking.

Thats right Surfer.
I value any feedback from some one who has experienced a snake and actually recovered from one, but frankly take with a pinch of salt, any comments from some one who hasn't.
Anyone who has made a real life emergency stop, knows its completely different to driving test simulation, actually theres no comparison.

 

[Lutz]

I value any feedback from some one who has experienced a snake and actually recovered from one, but frankly take with a pinch of salt, any comments from some one who hasn't.
Anyone who has made a real life emergency stop, knows its completely different to driving test simulation, actually theres no comparison.

I'm glad you value my feedback then, Ray. I've experienced both deliberate and unforeseen snakes and can't say that there's any difference between them other than the initial shock of an unforeseen one. The way both types develop are the same and so is the action required. The only difference is roughly 0.3 second of extra time that you need to appreciate what is going on if taken by surprise rather than when already able to anticipate a deliberate hazardous situation. In view of the 3 seconds required for the snake to develop fully, that still leaves 2.7 seconds to act.

 

[Dustydog]

2.7 secs is nothing when your heart is in your mouth.
So in the final analysis why do some survive and others end in absolute catastrophe?

Dealing with the latter which will be of more interest to the general populus.
Is it speed?
Poor loading?
Tyre blow out?
External / extraneous factors?

Or whatever.? I'd like to know please ; if only so I can do my best to avoid such things?? A rhetorical question but still an interesting one for most.

 

[Lutz]

It could be any of those things that you mention, Dustydog, and more. Experience and due care can reduce the risk, but it can never eliminate it altogether. Therefore, unless your outfit is blessed with technical aids that reduce the risk of disaster even further, the only alternative is practice, much the same as one should learn how to handle other hazardous situations, such as aquaplaning, driving on patchy ice, avoiding a child running out into the road, etc.

 

[kelvin3]

Dusty

I was only doing about 40 and was passed by a lorry. I ended up jumping a lane and following him? I remember there being a split second (or 3) when I thought I was going to go into the side of the lorry. The pull of the lorry and my acceleration pulled it out and the fact that the affect had gone.

Is it speed – Not for me
Poor loading – Maybe (young and inexperienced)
Tyre blow out – No
External/extraneous factor - yes (if I knew what extraneous meant)

Remember what I said earlier it was light van! So it got pulled away very easily. So the chances are I countered the pull on the van and this put me into the snake.