Thom Fitzpatrick

Wow, interesting results!

For the sake of consistency, I think you should have done another pull with the stock airbox immediately after the 2nd pull with the drilled box. Since you were losing torque at every pull, it begs the question of whether you would have continued to lose it.

My 911+964 went into the shop yesterday for it's new head studs, after which I'll do another dyno run. I have the Zucs cone filter setup on mine, as well as a B&B exhaust and a cyntex chip, and I only pulled 208 RWHP. Since that test, I found two broken head studs, and a ~25% restriction in trottle travel due to cabling issues. I'm hoping to be in the 230RWHP range after the fixes.

jonfkaminsky

I'm not sure what that would show anything on the grounds of the following.

1. Sample size
2. The dyno measures torque, not horsepower. Because of the human input factor, there is some leeway (trying to start from 2500 RPM, flooring the pedal firmly), and of course instrument error, tire slippage, exhaust induction, etc, The difference in max power values are small enough to be insignificant.
3. The stock pull was the highest in max power, followed by the third pull, with the second bringing up the rear. Doesn't mean anything in this context (see #2).
4. The fact that max torque is made at or about the same factory value of 4800 RPM tells me that these data are good representations of the expected performance.
5.I am actually very pleased with how close the runs are to one another. I expected more experimental variation.

Thom Fitzpatrick

The next question is why the drilled airbox made less power. This flies in the face of the conventional wisdom that reduced induction restriction == more HP.

But I take it from your post that you're still running the stock chip? Maybe another set of tests is in order after a chip upgrade?

Sheesh! There's so many variables, you'd spend thousands of dollars and hours on the dyno!

Adrian

Dear Thom,
This is exactly what people like myself have been saying for years. Are Performance Enhancements on the 964 series of engines "Value for Money" or are there better ways of achieving similar results.
I always use the 964RS as my example. Porsche did heaps to that engine, blueprinted it, hand matched all the engine parts, lightweight flywheel, lightweight clutch, new DME, made it lighter. The engine capability gave up a whole 10HP increase. Yet the RS flies. Why, it is lighter."Remove Weight"
Ciao,
Adrian
911C4

Christer

Although I must say that a hot film air sensor conversion with remapping does usually cause a 25-30hp gain. A drilled airbox is viewed here in the UK as only for the extra sound when the engine is under load. The same goes for the various bypass pipes - again there will be some minimal gain in actual power but the main gain is the much reduced weight at the back of the car.

As I have already stated, I will be doing an AMD chip conversion initially, probably in the New Yearwhich is supposed to give you a gain of about 20hp (the catalytic converter is ripped out at the same time - not needed in the UK on cars made before 1995). I will make sure that I get before and after figures. The hot film conversion costs around £1700 so will not be tested until next summer at least.

jfkaminsky

Thom/Christer/Others-

This is why I am doing this. Some of these performance enhancements just didn't make sense to me.

For example, the Fabspeed Porsche cup airbox. You can buy it for $175. Or you can make your own for nothing by drilling the airbox. They claim 8 Hp. But read their ad copy carefully. Very carefully. It is actually Porsche Motorsport Germany that is quoted on the +8 Hp on 964/C2/C4/Carrera Cup/RS America. Who are they exactly (rhetorically speaking)? Where are the data? What did they base this on? Now read on and see what Fabspeed actually says:

"THE FABSPEED VERSION OF THE PORSCHE AG CARRERA CUP AIRBOX MAKES TRUE REAL-WORLD HORSEPOWER AND TORQUE. THIS IS THE PORSCHE FACTORY GERMAN DYNO TESTED APPROACH THAT ADDS BOTH POWER AND SPORTSCAR SOUND."

Sounds a little more dicey, doesn't it? How much of the that real world HP and torque are you going to see for $175? One 1/4 HP? Who knows. And where is their dyno data? Anyone can makes these claims. I asked myself, wouldn't this be a great story for a magazine? Why haven't seen an expose on this subject? Well who pays for the ad copy in the magazines?

In my opinion, the stock airbox is probably getting as much air into the induction system as the engine needs.

And Thom, please forget about the increase or decrease you think you see in the data I have posted. I have addressed all kinds of reasons for error in these measurements. The results show that there is NO SIGNIFICANT DIFFERENCE up or down between the stock or the drilled airbox.

We shall answer the Cup bypass question here in a few days. And then of course the chip, and the whole ball of wax together.

Thom Fitzpatrick

Are you documenting this anywhere? It would be a great place to point someone to in the future, because this subject will come up again.

Of course, if you document it too well, you may wake up with a horse's head in your bed, courtesy of the Aftermarket Mafia (fabspeed, ghl, b&b, et-al). Or maybe a pair of carbon-kevlar shooza.

Bill Gregory

There's a great 964/993 engine upgrade article in Pano (August 1999) by Joel Reiser of Reiser-Callas Rennsport Racing, who said that many who installed mass airflow devices found little gain, in fact, nick naming them cash flow devices. However, he goes on to say that as you modify the internals of the engine, that some of the upgrades, including the mass airflow intake, take on a real impact, where you might not see any on a street engine. Another that he mentioned, for example, was putting 993 injectors on the 964 engine. He wouldn't recommend it on a stock engine, but on a modified engine, it makes a measureable difference. Also, he starts off the article talking about the cup airbox cover as an easy mod. Logic was the factory (and some racing teams) were using it and claimed some benefit, so why not.

Jon, thanks for your efforts dyno'ing various combinations. It'll benefit all of us.

jonfkaminsky

Thom-I think these threads get saved off somewhere, or are downloaded in digest form, so I don't think we are in danger losing anything. Also, I believe Adrian will probably address the issue in his book and I expect he would provide ample supporting evidence.

By the way, read Herb E.'s post called "Mass Air Flow Performance" in the 911 forum. Interesting.

Thom Fitzpatrick

I'm not sure - remember the disk crash recently? If they were on the web somewhere, they could be more easily referenced than in the archives or in Adrian's book. Isn't there a technical archive somewhere, besides the Pelican one? Heck, I'd host the page on my site if you wanted.

jfkaminsky

This afternoon I was able to get the "Cup" by-pass pipe put back in after removing it for the earlier dyno runs. I could not get in the dyno today as they were booked.

However, I decided to perform a dry-run chip change to make sure I was going to be able to do the swap in a timely manner while the car is on the dyno (paying by the hour while the car is strapped).

My initial reactions are thus-
The idle appears to be detectably cleaned up. I had an occasional "hunting" at idle. That is gone. Acceleration seems smoother through out the RPM range, especially at launch. I cannot tell nor would I want to cloud the discussion by offering an opinion of whether the car feels more "powerful" or quick. I'll leave that to the dyno.

Tonight I'll change back to the stock chip (an Intel, my employer by the way ) and hopefully get on the dyno tomorrow if it isn't raining and they can get me in.

jfkaminsky

Hi all-

I performed some more work today, albeit under some change of plan. But before I get into the results, I want to discuss where I deviated and why. I had recalled that after re-setting DME, there is a learning period where the system adapts to "best found" conditions to operate the vehicle. I had no data on this, and as such, I contacted Adrian who proceeded to confirm my suspicion and provide some background on the issue. An excerpt from that discussion is as follows:

"These maps cover every function including fuel schedules, timing, dwell angle, engine knock etc etc. Now obviously it is not possible to use all these maps at once. Bosch designed into the system the adaptive mode. You take your 964 for a drive and it learns which maps which are going to need and use. It then pulls these and installs them into the adaptive volatile memory. In other words the DME learns how you drive the 964. 10 minute of operation after a reset is the minimum recommended by Porsche"

That being said, I had installed the chip last night in order to dry-run the procedure so I would be able to perform it quickly at the shop. I went out drove the car around town and on the freeway last night. My perceptions are captured in the previous post. I also drove the car to work today, and then on to the shop this afternoon. My original plan was to arrive with the stock chip, take a few pulls and then perform the chip change, and proceed to pull a few more to test the new chip. What become apparent was that I would have to take the car off the dyno, change the chip, go drive it around, return, re-strap and re-run with the new chip. Because I did not have time last night to return to the original chip, I had planned to do it today before I arrived at the shop. Unfortunately, the shop could only fit me in between 4:15 and 5:00 pm so the plan had to be altered unless I was willing to show up on yet another day to fit this all in.

My solution was to not change the chip back to stock, and arrive with all the "upgrades" in place. Then if significant changes were observed, I could return to the stock chip and come back on another day to investigate how much change was due to the "Cup" pipe and how much to the new chip. The theory being was that if there was little or no change with all the performance enhancements in place, there would be no reason to investigate further.

The result? Very little change at the wheels was observed. Using uncorrected values for comparison, my best stock run was 229.5 max HP and 216.4 ft-lbs max torque. The best run today with the drilled airbox, "Cup" by-pass pipe, and the performance chip was 232.7 max HP and 220.5 ft-lbs max torque. The gain at the wheels were 3.2 HP and 4.1 ft-lbs of torque.

[Note: I went back to this, and added some more explanation below, as I don't think I was telling the whole story earlier]

One way to look at this might be to consider that with 15% drivetrain loss, the 247 HP flywheel spec should translate tb about 210 HP at the wheels, which is about what was measured on the dyno, figuring in the 1.086 fourth gear ratio, i.e., 1.086*211.3 = 229.5 HP (observed in best stock run). So, what value at the flywheel would translate to the observed max power value of 232.7 HP with the "Cup" pipe and new chip? A quick calculation shows the following:

If 229.5/1.086 = 211.3 HP (wheels, measured), and
247 HP (flywheel)* 0.85 = 210 HP (wheels, assuming 15% loss),

Then,

232.7 HP/1.086 = 214.3 HP (wheels, measured) and therefore, 214.3/0.85 = 252 HP (flywheel, assuming 15% loss)

Therefore, at the flywheel, we might argue that max power increased by a total of 5 HP.

The claim on the chip was to add 6-7% power. At the flywheel, this claim should have increased power from 247 HP to 261.8-264.3 HP. This is clearly not the case (if the above is correct).

Results are shown below.

Porsche specifications:
SAE max power: 247 HP @6100 RPM
SAE max Torque: 228 ft-lbs @4800 RPM

The collected data are summarized as follows:

1991 964 Carrera 2 (July 1990 build date)
Ambient temp: 50 deg F
Barometric pressure: 30.21 in. rising
Humidity: 62%

Run 4: Drilled airbox, "Cup" pipe, new chip
SAE max power: 232.7 HP @6100 RPM (est.)
SAE max torque: 220.5 ft-lbs @ 4900 RPM (est.)


Note that we did not capture RPM on the above plot (by mistake) but I was able to estimate RPM by using the plot of run 1 overlain on run 4.


Run 5: Drilled airbox, "Cup" pipe, new chip
SAE max power: 230.1 HP @6000 RPM
SAE max torque: 214.9 ft-lbs @ 4900 RPM



Run 6: Drilled airbox, "Cup" pipe, new chip
SAE max power: 231.0 HP @6200 RPM
SAE max torque: 216.4 ft-lbs @ 4850 RPM


Comparisons...
Run 1 compared Run 4 (stock vs. all upgrades)


Again, we did not capture RPM on Run 4 so it cannot be shown on the plot above. What is interesting on this plot is where the "gains" were made. The pull is taken in 4th gear, and the data suggests that gains are not realized until about 75 MPH, about 3800 RPM. The sag in torque and HP at about 4000 RPM is cleaned up, but everything else is about the same in terms of torque and power curve morphology.


Run 1 compared Run 6 (stock vs. all upgrades)


The plot above is interesting from the standpoint that is actually shows a decrease in performance until about 4800 RPM, which is about the peak in torque. Thus most of the gains are outside the useable range.

That's it. I will be contacting the tuner and relate the findings. I will be quite interested in what the explanation will be. Until then, let the discussion continue!

Jon

Bill Gregory

Jon,

Did you say or did I miss what brand chip you tried as the alternate to stock?

FlyYellow

Jon,

OK you motivated me to go in and try a dyno pull as well. I've got the following configuration two configuration to try:

Modifications performed:
- fabspeed primary muffler bypass pipe
- cat bypass pipe
- powerhaus ram air intake

and now here comes the trick - I have two chips. The first is a custom made fabspeed chip. The second is an upgraded Authothority Chip. I recently learned that there are several Autothority chips to be used on the 964. Here is what I've learned and the last 3 digits of the chip part number determine the chip to be used in your specific application.

.001 Standard Autothority Chip
.002 unrestricted exhaust & air intake mods
.003 lightweight fly + .002

So I'm proposing to do two configs. The first one with the custom fabspeed chip and the second with the upgraded autothority chip. I'll have to address this adaptive/learned dme problem, but I'll figure something out.

Cheers,

Boris

jfkaminsky

Bill-
As I said in the beginning (in the original thread that was lost), I will not be naming the chip vendor until they have a chance to respond. I could damage the credibility of this effort by going off half-cocked. I'll wait until an explanation is given, then I shall reveal all.

Jon