As I've been "upgrading" my car, I've been dynoing it to see what (if any) effect my crumpled scrilla infusion has had on the engine performance of the vehicle.

I've been getting the car dyno'd at the same place every time (MCE's indoor dyno @ THill), so this is about the most repeatable A/B testing as I can manage.



CF is SAE and Smoothing was set to 5 before exporting from Winpep to csv.

The engine is a low mileage BP4W that began totally stock. No internal work has been done, just an increasingly expensive set of bolt-ons.

The first set of dyno runs was with a DDMworks stage2 CAI and NB2 header.
I actually dyno'd it totally stock and these basic mods were good for a huge 10ft-lb+ increase in torque across the board - those dyno runs arent included here for clarity but if you havent done those two mods to your NB1 yet you are seriously missing out.

The second set of runs was with the MSM/BP5A intake cam installed.

The third set of runs is with a squaretop manifold and AC delete.



From these, it's pretty clear to see that the general impression that the squaretop decreases midrange torque is valid, the car is down by up to 5ft-lb until 5k rpm, but slightly more top end.

In order to see if the increased top end torque would result in overall faster acceleration, I also calculated the total area under the torque curve for all dyno runs from 4500-6800 rpm (general on-track rev range) and then averaged the three to produce the numbers in the graph's legend. In this rev range, the squaretop is ever so slightly better (.66%) than just a CAI and header.
I also did the same calculation for 4000-6800, and the results were less encouraging.

edit: made a mistake in importing - showing another BP5A run now.

Honestly, it looks like most of the midrange droop may be attributable to the MSM cam and not the squaretop?

 

Very cool. Propped

I think the squaretop woulda been a much bigger improvement had you started with a vtcs mani

 

Interested in seeing the AFR plot if you have no ECU. Probably a bit to gain, especially on MCE's dyno.

 

Here's the AF plot:


It's all over the place. Also, it's worth noting that this is from MCE's tailpipe sniffer, so post-cat.
This is stock BP4Y ECU.
I took some readings by pointing a GoPro at my cluster and WBO2 (pre-cat) and it went from about 13.2 to 11.6\

Overall my half-assed conclusion is that the MSM cam and squaretop probably arent a good value for a street car.

 

Why haven't you put an ECU in the car and gained all your midrange back yet?

 

Penny wise, pound foolish

Need to get Xidas for the MSM first, then I'll drive this car over to you for MS3 Basic.

Though its not like I make much use of the power I have now anyways

 

Oh, this is stock ecu! I was so confused.

With a proper ECU, you can tune the VICS manifold, and play with timing to make use of the cam, I hope.

I'm doing these same mods in my NA, with a MS2 soon. It'll be a bone stock '99 with VICs and an Ebay header. I do have a MSM cam, I wonder if I should dyno it with/without, see what gains there are to be had. The local dyno I use is 5 minutes away and only $60/hour.

Here's what I've done, just to add some fuel to the fire. Unfortunately I don't have the dyno files for the one '99 engine I've tuned. My car is getting a '99 soon, so I'll be able to show you that.

lower powered blue is a bone stock NB2 with 185k and a MSM axle back.
red is a 94 with NB2 pistons, ported NA8 head, and MS2.
green is a junk yard 100k VVT with gutted VTCS, stock header, short intake, decent amount of VVT tuning, and MSpro.
higher powered blue line is a rebuilt VVT with ebay header, K&N intake, not as much VVT tuning, and MS3 basic.
Brown is my fairly stock 1.6, with disgusting intake, leaking stock header, cat, 3" exhaust, and MS2.

 

The K&N does seem to help the NB2 dip, at least run speed density it does.

The BP powerband shape is very sensitive to duration @.003. The few degree increase from a 4W to a 5A shifts peak torque a few hundred rpm. We found that an Integral stage 1 NB2 cam with a bit more lift but 10° more duration at .003 was no faster on track than the stock cams even though it made a bit more peak. If it is a 5 speed or driven on street with a stock longblock, we prefer stock cams for best driveability. Breathe on the longblock or do a full I/H/E +ECU and you can afford a little more duration as the low end improvements from the other stuff over up the losses from cams. For instance, the Tomei's we offer on a stock engine with just an ECU, are awful.
Add I/H/E + ECU with the Tomei's, particularly with a 6 speed and you have a great balance between powerband and peak power.

Gotta love that VVT torque over the NA8.

OP's build is crying for an ECU. I'd venture to guess another 10whp at least.

 

^^^ Hey Emilio, when you say with a 6 speed transmission, do you mean with the 3.9 ratio rear end?

 

5 spd gaps vs 6 spd gaps. 2-3 upshift on a 5 spd happens all the time on the street (not so much on track), is around 2300rpm drop. Most BP builds lose 15-20% power during that upshift if done at redline. 20-30% if shifted say, 1000rpm below redline. Broaden the powerband and/or run a trans with closer steps.

For NASA PTE, we ran a low rear end ratio (4.77 or 4.875) to allow the driver to use 3-4-5 instead of 2-3-4, broadened the power so it made about 95% of peak over 3000rpm, all to be save 3 points by running a 5 speed.

 

Curly, any way to repost the dyno graph with more noticeable colored lines? At least using the mobile theme to the forum website, the lines just show up as different shades of gray. However, the color does show up pretty well in the legend listing each car. The colored circles where you have the intersecting line aren't great, but there is enough color to tell the apart. However, I can't quite follow which line is which, when they run very close together and even pass over each other. Even labeling each line in the middle of the graph would help tremendously. Thanks for posting up the dyno data in any case.

Regarding the famous Flat Top Mazda intake manifold, is there any concensus on how it impacts FI (turbo'd) engines? I would guess that a turbo would help fill in the mid range, but would be interested to know if the Flat Top slows down turbo spool significantly. (Most of the intake manifold comparison dyno graphs which I have seen have been from NA (non-FI) engines, besides a few from FM.)

 

Yeah it's whatever. Color winpep7 displays them in. I'll try to get brighter colors. I'm sure there's a line thickness option.

 

Okay, here's another question then:
Should I install my JDM/EUDM 4-2-1 NB2 header to replace the USDM 4-1 NB2 header?

Everything I've read (mostly from Emilio) says it flows better.

 

Definitely

 

Bump of this old thread to provide some updates.

I installed the EUDM/JDM 4-2-1 exhaust manifold but never got it dyno'd on the same dyno, so that's a missing datapoint. Butt dyno didn't notice any difference, though.

I installed an MS3 Mini during shelter-in-place, got it running well on the base tune, and had it dyno tuned by a local shop this past week.
These runs are on the same dyno, so can be compared to each other but not against the previous dyno charts.

In short, I gained about 12hp peak (122hp vs 134hp) at the high end of the rev range and increased the area under the torque curve by ~10% compared to the same bolt-ons with the stock ECU.



This is with the stock CA-spec midpipe. A better flowing midpipe might be the next modification but I can't imagine it will add much appreciable power.
Thoughts?

 

Thanks for updating this thread. You're making a lot more power with the ms3 looks really good, pulls all the way to redline very nice. I bet a midpipe would pick up another 5hp/tq it's probably the biggest bottleneck left in the system

 

Sub'd, looking forward to how much you'll gain from a midpipe as I'm at a similar point in my build as well.

 

It probably won't happen any time soon. I just spent $500+ on a new OEM midpipe last year. Now I wish I had bought the Roadster Sport midpipe for only a bit more :(

 

How long did you run w/ the square top and OEM ECU? This thread is the only place I read that goes against everyone saying you gonna blow it up cuz of bad AFR. I got a square top waiting on my shelf for my NB1 w/ similar bolt-one.

 

About half a dozen track days + 5k miles. I had an AFR gauge installed with the stock ECU, and even with all of the bolt-ons it runs rich enough at WOT to not have any issues..