Dyno Numbers

[alphasaur]

Originally Posted by wheelspeed

To someone ignorant on boxer engines (me), this is interesting. It seems that a dip may be inherent to boxers that isn't there in inline 4s. You don't see this on the Miata engine and I don't think there was one on my old Mazda 3 2.3L (though I didn't scrutinize any dyno charts in 2009). Anyway, I'm thinking maybe trade-off to the boxer layout to have a torque dip somewhere, but you get the lower CoG of the boxer layout?

Anyway, I'm trying to save judgement until I drive one. I was perfectly happy with my Mazda 3's 150hp/150tq engine because, despite low numbers, it felt "willing and happy" and was fun to row through the gears. I was just sick of it being FWD. As long as the new 2.4 boxer engine feels willing and happy, I'll be happy with it.

It's emissions-related.

[alphasaur]

Originally Posted by wheelspeed

To someone ignorant on boxer engines (me), this is interesting. It seems that a dip may be inherent to boxers that isn't there in inline 4s. You don't see this on the Miata engine and I don't think there was one on my old Mazda 3 2.3L (though I didn't scrutinize any dyno charts in 2009). Anyway, I'm thinking maybe trade-off to the boxer layout to have a torque dip somewhere, but you get the lower CoG of the boxer layout?

Anyway, I'm trying to save judgement until I drive one. I was perfectly happy with my Mazda 3's 150hp/150tq engine because, despite low numbers, it felt "willing and happy" and was fun to row through the gears. I was just sick of it being FWD. As long as the new 2.4 boxer engine feels willing and happy, I'll be happy with it.

It's emissions-related, it can be tuned out.

[itsovr9k]

My stock AP1 S2000 had a similar "torque dip". It was a drop of about 5-8 ft/lbs from 3500 - 5500 rpm.

The drop is more significant / notable on the BRZ because the torque output prior to the dip is good considering the displacement. The 2.0 BRZ is up about 20 ft-lbs on the s2000 from 3-3.5k, and still up 5-10ft-lbs during the lowest portion of the dip.

[stilesg57]

As mentioned it’s not inherent to the boxer configuration. My ‘95 Impreza 1.8 dynoed all of like 78whp (and that was corrected for altitude hahaha)…but there wasn’t any dip.

Fun story about how that econobox got on a dyno: got a two-for-one at the first AWD dyno in Denver almost 20yrs ago and the other car — my stage 2 bugeye WRX — almost 3x’d it

[Dzmitry]

Originally Posted by wheelspeed

To someone ignorant on boxer engines (me), this is interesting. It seems that a dip may be inherent to boxers that isn't there in inline 4s. You don't see this on the Miata engine and I don't think there was one on my old Mazda 3 2.3L (though I didn't scrutinize any dyno charts in 2009). Anyway, I'm thinking maybe trade-off to the boxer layout to have a torque dip somewhere, but you get the lower CoG of the boxer layout?

Anyway, I'm trying to save judgement until I drive one. I was perfectly happy with my Mazda 3's 150hp/150tq engine because, despite low numbers, it felt "willing and happy" and was fun to row through the gears. I was just sick of it being FWD. As long as the new 2.4 boxer engine feels willing and happy, I'll be happy with it.

There are definitely plenty of cars out there with a torque dip. I have done dyno searches before for various sports cars and big power naturally aspirated beasts. In the big powered ones, it is often quite small looking, as 20-40 lb-ft of a dip in a 400 lb-ft car doesn't look like much.

IMO, I believe the reason is more related to shooting for as much power as you can get out of an engine. In order to do that, the engine design has to be quite good and the tune has to be extremely on point. This is where emissions comes in to bust it. Because it is the tune that can't be brought to a maximum these days due to emissions. Just as removing the front cat and getting a proper tune removes the torque dip completely.

Even the MX-5 has a small dip of about 5-10 lb-ft. But I don't think this engine was built and tuned up to the same extreme the FA20 was. Again, this is just my theory, but if they were to tune the power up a bit more all around, I bet that torque dip would remain more or less in the same place, making it larger.

My guess would be that most cars in general (that aren't sports cars or very high powered monsters) are not tuned up to maximize power, obviously, and therefore avoid having a torque dip. This is not always the case of course, as even ordinary cars have a dip sometimes, and again I believe this is thanks to modern emissions. I am sure companies that heavily support better emissions (like Subaru with their PZEV) experience this more with their vehicles.

[Frost]

Originally Posted by itsovr9k

My stock AP1 S2000 had a similar "torque dip". It was a drop of about 5-8 ft/lbs from 3500 - 5500 rpm.

The drop is more significant / notable on the BRZ because the torque output prior to the dip is good considering the displacement. The 2.0 BRZ is up about 20 ft-lbs on the s2000 from 3-3.5k, and still up 5-10ft-lbs during the lowest portion of the dip.

Don't forget that S2000s also only gained peak torque for SUCH a narrow band as well. The new twins are within 70-100% of max torque from 3500 to near redline.

S2000s? 125ish till VTEC kicks in yo and then you get a spike of torque before falling off.

[OkieSnuffBox]

Originally Posted by wheelspeed

To someone ignorant on boxer engines (me), this is interesting. It seems that a dip may be inherent to boxers that isn't there in inline 4s. You don't see this on the Miata engine and I don't think there was one on my old Mazda 3 2.3L (though I didn't scrutinize any dyno charts in 2009). Anyway, I'm thinking maybe trade-off to the boxer layout to have a torque dip somewhere, but you get the lower CoG of the boxer layout?

Anyway, I'm trying to save judgement until I drive one. I was perfectly happy with my Mazda 3's 150hp/150tq engine because, despite low numbers, it felt "willing and happy" and was fun to row through the gears. I was just sick of it being FWD. As long as the new 2.4 boxer engine feels willing and happy, I'll be happy with it.

https://grassrootsmotorsports.com/ne...iata-nd2-dyno/

It's not as pronounced, but it's still there. I suspect it's something to do with emissions. You can see on lots of new vehicles, the torque flattens out/stops rising in that RPM range before picking back up again.

[stilesg57]

Could be worse I guess: we could have TWO dips



BMW N51 3.0L with 3 stage (valved) intake manifold. You can see the two DISA switchover points pretty clearly. Red is stock Blue is tuned btw (i.e., the tune can’t fix it). Been doing a lot of old Bimmer research lately and they pretty much all look like this on a dyno. It’s hilariously consistent.

[Grif]

No one likes it when a person double dips...

[ermax]

Originally Posted by stilesg57

Could be worse I guess: we could have TWO dips



BMW N51 3.0L with 3 stage (valved) intake manifold. You can see the two DISA switchover points pretty clearly. Red is stock Blue is tuned btw (i.e., the tune can’t fix it). Been doing a lot of old Bimmer research lately and they pretty much all look like this on a dyno. It’s hilariously consistent.

The valved intake may create a few dips but the curve would be a lot worse without the valves.

[OwlDance]

Think I've said this earlier in the thread but given what we know about 421 header performance and based on what I read on some small technical article on the development of the FA20 vs the FB20 -

The reason for the dip likely has to do with the catalyst and OEM header. The technical document stated that the FA20 was able to move the header cat a little further back than the FB20 while maintaining cold start emissions. Longer headers with 421 configurations also create a mid torque peak where the dip used to be.

It's my belief that because of the difficulty of creating longer and more optimized runners for performance without sacrificing emissions is the reason we have the torque dip (lessened however, in 2nd gen, likely made up with the larger displacement) and we'll likely see very similar kinds of gains and improvements in the aftermarket.

Nothing to do with tune (STG1 doesn't eliminate the dip whatsoever and yields next to no gains, only swapping/changing the headers do), everything to do with exhaust flow, even if the two aren't mutually exclusive.

[wheelspeed]

Hi everyone,

When I said "inherent to boxer engines", I meant, including everything related with selling a modern car, so including emissions related needs. If an engine layout throws out dirty emissions at a certain rpm range, then certain pipe designs can't be used, or certain fueling changes are needed for those 'dirty' rpms, or other requirements to deal with that. So, those become inherent traits of an engine layout.

Referencing old cars that didn't have such emissions requirements isn't very fair, but I can also see that referencing my '09 was going back a ways also. I'm in a state with emissions inspection each year, so can't necessarily just tune it out (without some risk in my mind).

But that was all just a conversational interest. A fairly narrow range of a small dip is certainly acceptable, so I just can't wait to drive one.

[nikitopo]

Originally Posted by OwlDance

I still think it's a very overblown issue that people only find as a problem because of number theory/oggling dynos over real world stuff. Yes, the dip can be annoying depending on how you drive but it's far from making the car "gutless" IMHO. IIRC our dip isn't any worse than what the S2000 makes until VTEC kicks in lol.

People are usually confused, because the engine is small and NA. You have to rev such kind of engines. Newer engine still has a dip but feels better, because it is bigger and has more torque.

[Jianlun]

Originally Posted by Jianlun

Can't help but notice that the AVO results are uncannily siniliar to the phoenix power results. With the 1st peak at 3300rpm and the 2nd at 4600rpm. And the hole at 4300rpm. Both are dynapacks.

And till date NO ONE got peak torque at 3700rpm. All in fact got it at between 3200 and 3400rpm. I love the consistency.

Screenshot from the recent HKS SpecL2 video.

Again a 1st peak at 3200 - 3300rpm (not 3700rpm in press materials) and a 2nd peak at 4600rpm. And max dip at 4300 - 4400rpm thereabouts.