Data Loggers - What they are and why you'd use them.

[Maxx]

I started doing a MUCH briefer answer to this question "What exactly is a Data logger - does it just measure lap times?" posed by capt Slow in this thread http://openpitlane.co.uk/forum/showt...p?t=747&page=5

Finger trouble ended up with me throwing it away so decided, whilst in a few brief sober moments whilst on holiday, I would do an even fuller answer to the question as I do get asked this question a lot, and what to do with them when you've got one (upcoming post on that also). I typed this bit by bit in notepad and having found a free wi-fi hotspot decided to post it now rather than on my return.

Formatting is probably cr@p so, if it's too bad anyone with access is welcome to format it better.

Probably won't be able to answer questions/comments until Wed when I get back home.

Maxx

[Maxx]

Data logging is much more than a product to provide lap times, there are much cheaper and
easier ways to do this. In realtime this could be (i.e. stopwatch/mobile phone etc.) or post
event (in-car video). Also of course for realtime there is the beacon/laptimer option or products like
the Alfano which uses the magnetic strips beneath many top circuits to provide lap and sector times.

From a trackday point of view however no timing is allowed, which I interpret as no "realtime" timing
as video is allowed which can be used for post event timing. On test days of course, which are
totally different any form of timing is allowed.

Anyway, back to data-logging. There are various types but the most popular ones for trackday use
are GPS enabled data-loggers. These can be further divided into those that include "accelerometers"
and those that don't. An accelerometer measures G-Forces and most units have a minimum of 2
(longitudinal: to measure braking and acceleration forces) and (lateral: to measure cornering forces).
Some (e.g. the DL1 from Race Technology) has a 3rd which measures up/down g-forces which in itself
can be useful but is mainly provided to allow side mounting of the unit (so up/down accelerometer
actually measures side-to-side G-forces).

An examples of a GPS only data-logger would be Virtual Cup which is software that runs on a GPS
enabled PDA or some recent GPS enabled smartphones. There are a number of similar products which
are generally inexpensive (£25) or even free.

The GPS element of data-loggers can provide positional and speed data which therefore can produce
a "map" of your driving line around a circuit and your speed at any point. You can then enter a
lap marker and optionally sector markers to give you lap and sector times for all the laps you have
driven. This allows you to compare lap and sector times for multiple laps and (if supported by the
software) give you a fastest theoretical laptime based on your best sectors.

Interesting stuff and useful "verified" willy waving fodder for the post trackday drinks session.
But, how can it actually help you improve your driving/achieve quicker laptimes.

Well, as it tracks your driving line on each lap you MAY be able to spot things that you did on
your best sector that POSSIBLY made it your best sector and, by doing this regularly from then
on your ACTUAL best laptime MAY improve but general speaking, without doing something else, other
than line variations you are never going to BEAT your fastest theoretical time by information
gained from this type of data-logger.

Now there are lots of MAY's and POSSIBLY's in there, the reason is that the driving line provided
by GPS is not particularly accurate and can vary quite substantially session to session and even
lap to lap.

Even at it's best GPS accuracy is only accurate to about 1.5 mtrs, remember this is AT BEST and
often is much less accurate. Lower end products tend to have less accuracy, mainly as they take
less samples of positional data than the higher end (dedicated) products. Most PDAs etc. have
fairly low scanning frequencies hence are less accurate overall, not just for positional data but
can also give oddities in speed and are much more prone to dropouts where you effectively lose all
GPS data for a period of time.

As any track driver will know (I hope) if you drive round a circuit turning in 1.5 mtrs from the
side of the track, missing the track by 1.5 mtrs at the apex and don't use all the space (say by
1.5 mtrs) at the exit your laptime will be much slower than had you not missed all these points.
So is the racing line as produced by these data logging products of any use?

Well, yes but perhaps not as much as most people will think. lap to lap they can show variations
in line. Say for instance you takes a particular corner, lets say Druids at Brands. There are two
quite different lines you can take there. There is a) the traditional hairpin line where you will turn
in very late on a tight radius thus allowing you to unwind substantially from the apex. Then there
is b) the shallow entry where you get into the curb well before the apex, hug the curb all the way
round and unwind more gradually from the apex.

You should be able to spot the difference with any GPS data logger, even if done in different
sessions. By isolating that section you can see which worked better for YOU and YOUR CAR. This may
not be the best way for another car or even another driver in your car and of course may not stay
static for you as you gain experience.

Perhaps more importantly though and the reason GPS only data-loggers are very limited with what
they can do for you, you have no idea WHY one line worked better or even if it worked better
because of line or another factor. Now there can be lots of other factors and, as you generally
don't do (as you don't get time) to do the data analysis straight after a driving session you can't
really remember if any of those factors occurred. Examples : track grippier, traffic, you didn't
drive it so well (i.e. didn't use all grip available, didn't use all available track, didn't exploit
the weight shift properly), car was lighter on fuel, was heavier as you had a passenger, the air was
cooler, engine was hotter etc. etc.

Also, don't forget to analyse the whole relevant section, option b) would be the quickest through
Druids but a) would potentially give you better exit speed which you can carry to Graham Hill
bend, so you need to analyse the section from just before braking for Druids until braking for
GHB.

More subtle changes in line, which can still make a big difference are much harder to spot and
even more so between sessions. One reason is simply variations in weather conditions/cloud cover.
At one time you could be "connected" to 6-7 GPS satellites which can change to 3-4, possibly
even totally different ones dependent on conditions. That s another reason why the higher the
scanning frequency is the better.

Another reason there is this issue between sessions is a phenomenon called GPS drift. As time progresses
the various GPS satellites that your logger is "connected$to" will move and at different rates/directions.
Often there is very little difference in the line that is produced between sessions but I have seen
variations of around 3 mtrs, add this to the 1.5 mtrs "general" inaccuracy and analysing line between
sessions can be difficult and not particularly constructive.

The higher end DL2 product from Race Technology offers an additional service to "post-correct" for GPS
Drift. A day or so after a logging session you log onto a particular site on the internet and download
GPS drift data based on the location of your data capture. This is then used to correct the readings
taken to essentially remove the effect of GPS drift (although you still have the 1.5mtr general
inaccuracies).

It's worth mentioning that some of the older and even some current "integrated" systems do not use GPS
at all and can draw reasonably accurate track maps using accelerometers alone, speeds data is collected
via wheel speed sensors. Most of the GPS+Accelerometer systems have options within software to either
use GPS data alone or combine GPS data with accelerometer data to [generally] produce more accurate
track maps.

So, we've seen that GPS only data loggers can be somewhat useful in at least logging and comparting
laps & sectors, corner speeds and to a degree, the driven line. They can tell us which sector is faster
but only give very limited information as to why.

GPS + Accelerometer Data-loggers like the DL1 and the Performance box are able to give us additional
information, specifically g-forces. This (in software) can show us when we brake, how hard we brake,
whether we trail-brake, how much of the total available "grip" [in g-force terms) we use throughout
the entire braking/turn-in/apex/exit sequence, whether it is consistent. Accelerometer data can also
show where gearchanges are made, if we use any curbs etc. These are all the things that tend to
differ greatly between drivers, most experienced trackday drivers will drive a similar enough line
to make little difference.

This then is the really useful stuff we can get out of data-logging. If you understand the dynamics
of vehicle handling, track driving and things such as the traction-circle then you can improve your
driving quite a bit by using this type of analysis. It's by means certain though. You may be able to
see from your traction circle graphs that you are not using all the available grip throughout the
corner but HOW you get there is another thing. That s where people like myself and the other driver
coaches/instructors on this forum come into play. In fact, we don't NEED data-logging to determine
things like this and of course are on hand to not only identify the problem but also help to correct
it.

I do use data-logging myself though as an additional tool to actually show the things I often talk
about. Also, we generally can't operate at the accuracy of a data-logger and the more proficient a
driver you are coaching, the harder it is to identify the weak areas, plus what works well for
a low-powered FWD car may not work as well for a high-powered RWD car. I also use it myself as a
driver, as it's often much more difficult picking up on things from he drivers seat as it is the
passenger seat.

With a little work, any driver can learn to use a data-logger effectively to analyse their own
driving and to a degree to perhaps improve it (and SEE IT improve via the data) but it does take
a bit of work which anyone spending many hundreds of pounds on a data-logger should be prepared
for.

[Maxx]

It's worth in the section on functionality talking about the sorts of thing that make data-logging
EVEN more useful and is often the differentiator between different products in the mid-priced
bracket.

Some data-loggers also has the possibility of working with additional sensors which can measure pretty
much anything you want it to measure. The difference between the two most popular products, the DL1
and the Performance box is that the later has very limited capability to attach additional sensors
(not sure you can connect any in fact). Racelogic produce a more expensive product which does have this
capability, the VBOX range.

The DL1 comes with 8 Voltage based inputs and 4 Frequency based inputs which essentially account
in part for the extra cost.

So, why would you need extra inputs. Well, you'd almost certainly would like to see the engine
revs, from this and g-force data you can very accurately measure the power the car is producing
at any given point/speed range.

Another common one is throttle position, from this you can actually see where the power is being
applied and by how much. It's not uncommon to see a driver who is earlier on the power actually
getting less accelerative force out of the corner (scrubbing speed). This is often even easier
to spot when you also use a [steering] wheel position sensor.

Driver A seems much better under braking than driver B, what would perhaps give you the biggest
clue is if you had a brake pressure sensor. You can also have many engine related sensors, oil
pressure, fuel pressure, water temperature, turbo pressure etc. etc. etc.

There are then a host of chassis and handling related sensors you can add. You can have sensors
on individual dampers, spring compression sensors, wheel speed, slip angle etc. etc.

Most of this is way beyond the needs of the trackday driver but if you were really serious about
data-logging for driver improvement you would perhaps want the first 4 mentioned, revs, throttle
position, brake pressure and steering wheel position.

QUALITY OF SOFTWARE

I can't close this without making what should be an obvious point and that is the quality and ease
of use of the software is all important. No point having the best, most accurate data-logger in the
world if the software is not able to give you the information you want in the way that you want it.

The difficult thing is when you start out looking at data-loggers you don't really have a very
good idea what you want to see and how you want to see it and, as they all use there own proprietary
data structure you are tied into that companies software.

They should all give you the ability to do the main things that you, as a driver, would want. From
using them however I can say that none of them are "easy to use" or intuitive and I think this is
the reason that many who buy a data-logger never use it to it's full potential, many simply using
it as a fancy stopwatch to give lap and sector times.

As this has already turned out to be quite a hefty post I will leave discussion of what you should
perhaps be looking at in your data to another post.

[m3Chris]

Maxx - thanks for that, possibly the most informative post on here to date.

[Mick]

very informative, might be worth copying that and put it in the track Q&A bit for future read.

[m3Chris]

I have moved it into the Track Q&A section.

[Captain Slow]

Thanks Maxx - tons of information there. At what point do you think they become important to have. I can see just the times being important as a beginner just to see improvement, but I guess at some point you have to do analysis like this to find the parts of your driving you need to improve.

[Mick]

For now James, an on-board cam and the nite-Pace software should be good enough for lines and lap/sector time. That's what I'll be using for now until I'm good enough to start using the data from a data logger in order to modify braking/acceleration point, by which time I'd prob use both together.

[Maxx]

Guys,

Thanks for the comments but one thing I've found out since writing that was that the Performance Box does not have any accellerometers in it, so analysis of g-forces or the combining of accelerometer data with GPS for more accurate track maps/speeds is not supported.

Whilst I was on the wi-fi connection I downloaded the PerformanceBox manual.

There are a number of other units that do have GPS + Accelerometer data such as Motec, Traqmate etc. but these are all pretty expensive products which would put them firmly in the professional category.

Maxx

[daz]

Max,

Thank you soo much for such an informative post, I will be reading and re reading several times to get my head round it, would you mind when I am stuck if I pm you with questions about my box, or open forum ok