Engine technology thread.

[Dimman]

Originally Posted by arghx7

To make it look pretty the smoothing factor is high and the window has been stretched. It's a common trick used by tuners. Go to the Dynojet website and install their free WinPEP7 software. Play around with it and you'll see what I mean. I have a side business tuning cars. Aftermarket performance tuning is on a day-to-day level a completely unregulated industry... there's a lot of Wizard of Oz/smoke and mirrors stuff that goes on.

Hmmm, I'm familiar with the smoothing, but hadn't heard about the stretching part.

I read a Grassroots Motorsports article where they had a pedal bike on a Dyno-dynamics just for fun, and it was sensitive enough to spike with each pedal push when the smoothing was low.

Smoothing works both ways with shady dyno sheets. Some parts installs claim gains from the highest spike on a sheet with low smoothing, even if it's anomalous.

[arghx7]

Originally Posted by Dimman

Hmmm, I'm familiar with the smoothing, but hadn't heard about the stretching part.

The scaling of each axis can cause different visual effects. It gets even more confusing when you have a bunch of data plots on each graph, and each plot has its own scaling apart from the scaling of the chart.

I read a Grassroots Motorsports article where they had a pedal bike on a Dyno-dynamics just for fun, and it was sensitive enough to spike with each pedal push when the smoothing was low.Smoothing works both ways with shady dyno sheets. Some parts installs claim gains from the highest spike on a sheet with low smoothing, even if it's anomalous.

absolutely. Some chassis dynos (as opposed to research-lab engine dynos) are more noisy than others so that contributes to it. The most blatantly inflated dyno sheets are those on a Dynojet with smoothing set to "0" and correction factor set to "STD"

[blur]

So what you are suggesting is that these dyno sheets can't be trusted?

Regardless, the flat torque curve is the most appealing aspect of the dyno result. Is there a reason this engine nailed that characteristic?

[Dimman]

Originally Posted by blur

So what you are suggesting is that these dyno sheets can't be trusted?

Regardless, the flat torque curve is the most appealing aspect of the dyno result. Is there a reason this engine nailed that characteristic?

What he is suggesting is that by using a lot of 'smoothing' (averages more of the data points is a way to describe it, sort of...) and stretching the graph horizontally, it makes the curve look way flatter than it really is.

There are a lot of games played by tuners and dyno shops. The stuff I'm familiar with involves rather questionable claims of what constitutes a 'stock' bottom end for certain Supra claims. Fiddling with dyno charts is another.

[arghx7]

Originally Posted by blur

So what you are suggesting is that these dyno sheets can't be trusted?

Regardless, the flat torque curve is the most appealing aspect of the dyno result. Is there a reason this engine nailed that characteristic?

I don't want to say too much about a blurry cellphone camera shot of a dyno sheet. If I had the actual run file I could load it into the software and look through it more to have a better idea of that specific run.

There is an SAE standard, J1349, which shows how many manufacturer dyno tests are done (GM & Ford follow this standard more than most other manufacturers). You can PM me about this if you would like more information. Those tests are done under very precisely controlled conditions and must be certified by an independent monitor. A chassis dyno pull at Billy Bob's performance is a very different animal. There's just more room for error and manipulation.

Performance shops have every incentive to manipulate the numbers. Manufacturer dyno tests still have the potential to be manipulated but overall they are done under much more controlled conditions (even though those conditions may not always represent the real world).

[blur]

Thanks for the insight.

[Gardus@Supersprint]

I totally agree, we have a very precise dyno at work (it's a Maha 4x4) and we often have people coming in with tuned cars. The charts the tuner gives to them is most of the time manipulated or plain wrong.
The most evident case was a C 63 AMG where the tuner just modified the speed limiter and claimed the car had 50 more hp.
Reality was the map was bog standard and the 50 hp more of crank power came from a different loss of power (in one test there was 70 hp of power loss, in the next one 120, the whp was the same but as the cranks is the sum of the two obviously it was showing 50 hp of crank power more).

This is one of our dyno charts (it's my car actually).
100 nm are 73.7 lbft
We do 3-4 runs in second from the last gear (this for example was in 4th).

Imagine if the C&D number of ~170 (I'm going to pick 168) lb-ft @ 4000 rpm is correct. Now imagine that the brochure number of 151 lb-ft (which would be 90% of the peak using 168 lb-ft) @ 6600 rpm is also correct. And the peak power of 197 bhp @ 7000 rpm and the 7400 rpm redline are correct, too.

That would be an outstanding 197 bhp NA performance motor. It would pull from everywhere.

For comparison, this has 140 lbft at 4000 rpm, peak is 152 at 5465, and at 7k is down to 125.
170 lbft (equal to 230 nm) at 4000 is A LOT of torque from a 2l NA engine.
Good NA engine struggles to get over 100 nm/liter.

[nogaroblues4]

Does anyone know if this engine will have any type of variable valve timing?

[serialk11r]

It'll have cam phasing for sure, and almost certainly no lift/duration adjustment. Aka one cam profile. Not ideal...I am still a long ways off from needing a car, but when I get one, it won't be a first gen FR-S. I smell a better one coming out later.

[Sully]

I plan on waiting until at least year 2 for the same reason. Maybe the mid-model refresh. We'll see how my current vehicle holds up though.

I'm still kind of baffled why we don't have full VVT though. Unless they're saving it to one-up a competitor later.

[serialk11r]

Toyota and Subaru have stopped using variable lift in basically all their engines for whatever reason lately. Toyota promised Valvematic would appear in many models in the near future back in 2007, but so far it's only made it to European/Japanese Corollas and similar cars. Most companies have just a cam phasing system in place nowadays, of course with the exception of Honda which has 2 stage lift on basically everything, and BMW/Nissan which have the most advanced continuously variable duration/lift systems on most engines.

My guess is Toyota, for whatever reason, feels like it is not ready to release these new valve control mechanisms everywhere yet. It could be that they don't want to retool factories too soon, although it's not REALLY that big of an add-on. The FT86 on the other hand had to be released sometime around now, as they can't just keep telling people to wait, and so they kinda half-assed the engine a little and got the car to production first.

Eventually, we'll see Valvematic everywhere for fuel economy regulation reasons, but they are really delaying it. We can only hope that a mid-cycle refresh will have a better engine, it's not a guarantee as Toyota seems to be pretty good at squeezing cheap mpgs out of their cars to meet CAFE standards.

[arghx7]

Expecting them, in the 2012 marketplace, to use both continuously variable valve lift and an expensive direct injection system on a supposedly affordable sports car is just too much. They have to control costs, and this car is not based on currently mass-produced models.

Be realistic.

[serialk11r]

If Japanese Corollas get continuously variable lift, I don't think it's a really big deal. If Hyundai can direct inject everything...

If Honda can have stepped variable lift (which probably costs as much as continuously variable, having its sliding pins and whatever which are more difficult to ensure reliable operation) on Civics, and Toyota is giving us just a simple cam actuating valves directly (okay okay through a follower maybe), I'd say there's something missing.

[Gardus@Supersprint]

The most advanced and flexible system is the one used by Fiat, the "Multiair" system, that allows valve opening independent from the cam profile:
http://en.wikipedia.org/wiki/Multiair
It's used on cheap cars like the Grande Punto / Alfa MiTo.
BMW uses the Valvetronic on the Mini Cooper and S.
I think Toyota could afford to use a similar system on the FT but there is the problem of the two heads, so every part has to be mirrored...
I don't think they were able to extract 200 hp from a NA 2.0 without any kind of valve lift or timing control.
Renault on the Clio RS engine has it, even if it's a basic system, since the 1st '98 model.