Thanks for your input and experience.
Can dyno figures be exaggerated easily to serve a tuners marketing agenda?
Thanks
Barbz
No probs
Yep certainly can. It's easy to get exaggerated figures out of a dyno. For example: the dyno software needs an rpm reading in order to calculate horsepower **bhp = (torque x rpm) / 5252** The tuner/dyno operator can take a direct reading via an inductive pickup connected to an HT lead. But more often they will use a reading calculated from the wheel speed; The car is held at 3000 rpm indicated on the dashboard tacho, then the dyno software stores the wheel speed at that point and uses it to calculate the rpm with reference to wheel speed. This is absolutely fine; it saves the dyno operator time getting to HT leads to connect to. Not to mention, some cars (like 20VTs) don't have HT leads.
However this does leave room for potential 'naughtiness'. An tuner could underhandedly (or just plain accidentally) set the 3k point too low. So the dyno thinks the engine is revving higher than it really is. So, say, your car's making 400 lbsft @ 5500 rpm, that works out as 419 bhp.
However if the dyno thinks you're making 400 lbsft @ 6000rpm: ta-da, 457 bhp. Instant +38 bhp.
Another thing is being "selective" of which dyno results a tuner prints. Throughout a tuning session you'll usually do many power runs. Some of those will come out with wildly inaccurate results. For example due to wheelspin on the rollers, or heatsoak, etc, etc. It can be very tempting (I know!) to print the result with the lowest reading as your "before" reading and print the highest reading result as your "after" result, despite the fact they may not bare any relation to the actual trend of results throughout the tuning session.
A tuner can also be sneaky with the results graph itself, if they wish. If you've ever looked at two results graphs and wondered why one looks all jagged like a rocky mountain but the other is a perfectly smooth line, it's because the graphs have been "smoothed". Dyno software usually allows you different levels of smoothing, from light to heavy. The idea is that it makes the final printout easier for the customer to read.
This isn't necessarily a bad thing, but it is possible to hide a multitude of tuning sins by heavily smoothing the graph, eg. hiding flat-spots, or jagged torque lines due to (for example) poor boost control.
That's not to say that all smoothed graphs are hiding things, its just important to be aware of what you're looking at and to know that ALL dyno graphs are bumpy. The one's that are perfectly undulating have been smoothed to a lesser or greater extent - not because the tuner responsible for the graph is stupendously awesome!
I personally never use the smoothing option on a dyno printout. Mainly because it's no use to me as a tuner. The lumps and bumps can show up important issues, not only to do with the engine but in the drivetrain too. For example a flat-spotted or under-inflated tyre can sometimes show up on a graph. For me unedited graphs are tuning tools, whereas smoothed graphs are only useful as publicity tools.
Next time I'm on a dyno I'll have to see if I can print off some examples to illustrate what I mean.