Hello guys, so glad to read a thread like this. I have wondered about this issue with MINIs having different power outputs aswell. I believe it has to do with flow, intake and exhaust.
I've started messing around with mine already, starting at the TB. Once I removed the Throttle Body and got to see down the intake runner that feeds the supercharger, I knew this could be a good place to start judging by the extra composite material that is left inside the tube after it's fused together. I broke out the Dremmel and went to town on mine. I don't have the means to test flow or anything, but do believe it's been improved.

Wutcha guys think about that?

 

How about the ECU adaptation?
If the fuel trim between two cars are very different, then they will pull very different numbers on the dyno.....

How about the break-in procedure? Drive like a grandma way vs. Drive it like you stole it out of the dealer's lot way?

I do think that car-to-car variation is probably the biggest factor here, but my feeling is that ECU adaption also plays a role. I have no idea how big of a factor break-in procedure plays.....

Thoughts?

 

When you're comparing measured HP between vehicles that were measured on different dynos, I see your point. My point is that on the same dyno (Dynojet) in Houston, TX in January of this year, there were several MCSes dyno'd with great differences in measured horsepower. The variance from the dyno would be consistent throughout all vehicles, so the measured variation must be from the vehicles themselves - assuming each MINI were stock, or had the same exact mods.

I'm talking numbers ranging from 155HP to 165HP on the same dyno, with atmospheric conditions that were pretty constant throughout the dyno-ing period (84 degrees F), relatively humid - this is Houston. External conditions aren't supposed to matter on a Dynojet, which is supposed to compensate for them. Those are big differences, not explained by test equipment variance.

 

.... also assuming all the MINIs were dyno'ed in the exact same manner, with similar IAT and ECT at the start of each test.

 

All of the MINIs had been sitting together all day, inside the garage. [It was a Randy Webb pulley-party.] Randy and Joel were handling the dynos - mostly Joel, with Randy looking on, as it wasn't his shop. There were variances in HP between each pull, since there was heat soak, but typically about 3HP.



The delta when comparing HP between like cars was much greater than the heat soak, environmental variations (what little there could be in such a short time), and equipment could account for. I only have my dyno, which is labelled incorrectly. The runs are from bottom to top, with a longer wait between the 2nd and 3rd pulls.

 

dotBob,

The SAE correction that a dynojet provides is only an estimation at corrections and is not even close to being a true correction value. It is a broad correction value, that applies to a large variety of cars and conditions. It doesn't take into account affects on cars systems. For example humidity is going to have a greater affect on a supercharged car with an intercooler than a naturally aspirated car. The are a myriad number of other factors that play into the car that the SAE corrections don't compensate for. It is much better to significantly control the variables (which requires a lot of knowledge), and have a control that you can run against. Without knowing the variability that the measuring device or the conditions introduce you cannot even argue about why the numbers are variable. Even your graph shows a difference of almost 7hp with the exact same car and if you claim the exact same conditions. That right there shows you variability of the dyno. The manufacturing process is not exact. I have seen studies that show that a 5% variability in power is within production tolerances. That would mean another 8 or 9hp for the Mini. Add these together and you have easily differences of 16hp. A couple extra hp of not exact conditions and you can see how you can easily get a difference of 20hp with like cars.

I think the ECU adaption is a big factor as well, and has been well discussed on this board. Depending on how the car is driven plays a big role and can significantly change the values. How the car was run in will change values. If the engine is nice and tight with it's tolerances, if the supercharger blades have set in well. Minis even have differences in their boost levels of a psi or so. That changes it as well. The version of their ECU will have differences, how tight bolts are screwed down at the factory to prevent small leaks, bugs or bent blades on the intercooler, worn parts......... This list goes on and on.

The way that dyno's are being used and people trying to argue about why the numbers don't equal between cars, is trying to make the dyno a tool that provides absolute measurements. Which I don't think is something you can attribute to a dyno, nor is it anything an experienced tuner will tell you. A dyno is truly a measurement of power for a single vehicle under a single set of conditions and runs should be used on the same car for tuning purposes.

This is no different than Randy programming the Unichip. He has programed the chip based on a few cars with certain mods. When people have put it on their car, they have received different increases. Some as different as 5hp. This is to be expected. The chip was only tuned on a set of cars with a set of conditions. People have taken these chips and gotten them further tuned to their cars and have thus seen increases comparable to others.

I think looking at the dyno and making it as an absolute measurement is pushing the tool beyond it's intended uses.

 

It's a fairly simple matter to monitor and report intake air temp and cooolant temp before and after each run.



As I recall, Helix doesn't start a run unless intake temp is 135 degrees or lower; not sure what they do about coolant temp

 

I'd like to hear someone's response to KatanaPilot's question about break-in. Perhaps this has a lot to do with it? Has anyone seen a correlation between cars which have been broken in a certain way and the later performance of said cars?

-mike

 

The adaptive functions of the ECU really only affect part throttle cruising conditions and idle, once you have your foot into it you are relying on open-loop operation and fuel maps because the narrow-band O2 sensor is not able to accurately determine the fuel ratio at the optimum richer full throttle operation. Full throttle optimum is often around 11-13:1 where the typical O2 sensor has a range of 13-15:1 at best. Things will change once cars start using the newer, wide band O2 sensors and the ECUs are able to track the optimum fuel mixture throught the powerband.

Most inertial dynos have the disadvantage of being affected by rotational mass as well as too short of a test run. Unlike a brake-type dyno which actually provides a constant resistance to the engine output at various RPMs, the inertial dynos measure the acceleration of the large mass flywheel (roller) in relatively short time. Take a stock MINI with runflats and S-Lites and make a couple runs for an average, then replace them with Centerlines and non-runflats. Run the car again and you will see a horsepower increase. You've done nothing to the engine yet here is an increase due to the lowered rotational mass of the wheels (you've essentially made the roller in the dyno 'virtually' lighter). Say for sake of argument that you have netted a 5hp 'increase' by swapping the wheels. The real-world performance of the car would not increase as much as if you added 5hp at the wheels, but you can most likely notice faster transient acceleration than before.

Cars like the MINI with airlow-dependent intercoolers are subject to a wide variety of testing variation among different dyno setups when it comes to providing airflow through the intercooler. Some people may only use a simple box or stand fan in front of the car where others might use a more directed, separate flow into the intercooler. This makes a big difference. It works best to benchmark your car prior to doing mods, mod the car, then test your car on the same dyno, with conditions a close as they were before. Testing your car on one dyno then on another after mods is full of a lot more potential error.

 

--->KatanaPilot and mlebeau: You both bring up an excellent and previously undiscussed idea; varying break-in methodology may or does result in varying engine performance. Traditionally, "proper" break-in is important in obtaining maximum performance and/or longetivity of the engine. I wouldn't count out non-optimal break-in habits accounting for some of the variability.

--->jlm: you are totally spot on that it is critical for each dyno run to have the exact same IAT and ECT (among every other variable, duh) to assist in assuring relevant comparisons between runs and cars.

--->dotBob; sorry but your dyno plots aren't really helpful, as we don't have all the pertinant information regarding each run's parameters (atmospheric (because the Dynojet uses a blanket SAE correction) and vehicular). Without that information, just like I said to jlm, comparisons cannot be accurately made, even on the "same dyno, same day"

 

Ryephile, alternatively, there has been some evidence (valid or not) that 'proper' break-in actually results in less overall power. There are some who firmly believe that breaking an engine in hard is the best way to get the most out of the engine in the long run. This is another possibility.

-mike

 

I never said which way was "proper"

 

Maybe next time someone hosts a dyno party, we should come up with a questionnaire..... Ask about things such as how did you drive your car in the past x days (to account for adaptation), how did you break-in your engine (follow the book/like I stole it), take note of coolant temp, boost level, outside air temp, etc etc...... Ryan and everyone else, maybe we can start some ideas about what questions we could ask?

There are definitely stronger cars and weaker cars. If we can find some correlation between break-in procedure and the dynoed numbers, (or any other factor with the dynoed numbers) then I think this exercise would be worth it.....

 

The fundamentals are important: proper SAE correction, and noting the gearing, IAT and ECT. MAP would be a nice bonus, as would driving style, break-in routine, etc.

 

Short and Long Term Fuel Trim values are available via an OBD-II scanner. I don't know what other adaptation variables there may be, but I think these get reset when going through the odometer button reset routine....

 

Here is a couple of links showing some issues with dyno testing. Also has some interesting info on the weakness of SAE corrections.

http://www.depac.com/applications.htm
http://www.depac.com/applications.htm#CF


Also the SAE correction factors used on dynos is the SAE J1349. This correction factor is not suitable for forced induction engines. Many dynos do not have proper correction factors for forced induction engines.

 

...so what is your proposed solution?

 

Ryan,
Just a little more enlightenment about variables and chassis dynos from someone you might believe:

http://www.caranddriver.com/article.asp?article_id=8020

Bill

 

from this article, sounds like a low tech approach with some corrective
factors would be closer to the real world truth on hp available when
actually driving the car on the road....use a g-tech pro or
similar (I have a tazzo vpc myself). and correct for car weight,
tire/wheel weight, temp (ambient, intake, coolant) and wind if
any....aerodynamic losses should be about the same for MC
and MCS's, and can be estimated, and driveline losses
(comparing same transmission models) should be similar except for
change due to weight factors.

 

--->grnmene and dgszweda1: I think the tangential topic of dyno accuracy has been beaten to death. While it is a useful and relevant topic, it is not the main point here. Do either of you have a practical (read: no cherry picking), more accurate way to measure engine output? Otherwise, you've been preaching to the choir. I understand both your concerns that accurate measurements equal better information to base a deduction for the thread thesis, but for now, humor me and postulate the potential reasons for differences in actual output among MCS's.

Now,
Back on topic; Variability of the MCS engine output:
1) Variability in sensor manufacture (IAT, MAP, ECT sensors)
1a) ECU code reaction to the tolerances and variability of above sensors
2) The style in which the engine was broken-in
3) [to a lesser extent?] Manufacturing tolerances regarding cylinder head casting and machining, as well as supercharger manufacture and related hardware.

OK, so far based on what we've deduced, the only way to really measure a difference would be to data log the IAT, ECT, and MAP on every MCS (in the same road, driving style, and atmospheric, lol) and then directly compare. I suppose one possibility would be; a crappy blower would have less boost and/or a higher IAT?

alright I'm burnt out now

 

well it's not just on the Dyno...it's lining up and racing too...you can tell bad driver or not...intake, pulley, and exaust should pull on a stock car.....

part of me still belives in break-in procedure.... last night while waiting for the garage door to open I reved a little..and noticed that although stock...my car was back firing..now I know the MINI runs rich...but I thought this was a bit much for a stock car...maybe this is why maybe I have a richer running motor than those I have raced

ofcourse I was making 160 whp...and ran into a guy yesterday on the highway with a intake, pulley, borla and pulled on him...heh heh...that's fun but 20 to 30 hp over other cars is alot I too am curious cause I want to get another mini but I would be annoyed if the one I got was 30 hp less than the one I have...

 

What was your break-in procedure?

 

Ryan,
I am really not trying to nit-pick (or cherry pick). But the reason this thread got off on dynos is that there is contention concerning your original hypothesis: There is a major difference in HP between similar MCS's, whether stock or modified.

I've only been trying to point out that before you can make that claim, you have to show that your measurements are, indeed, accurate. Without that as a basis for your hypothesis, it is a waste of time trying to solve a problem that may, or may not, exist. Furthermore, without accurate measurements, how do you know when, or to what extent, you've solved the problem?

Be that as it may be, there is no easy answer to this whole topic. I do have a suggestion, which may, or may not, be pratical, but short of putting all these engines on an engine dyno, it may be the easiest.

For the moment, lets assume that you can get repeatable, if not totally accurate, measurements. Using the two cars that have exhibited the largest variation, start swapping parts (one at a time). Easy ones first: sensors, ECU's, etc., and see if the variation follows the parts. Of course, if the variation is due to several things, you may or may not notice any difference with any one of them. And once you get to the head and blower, things get a bit more complicated, but it all depends on how far you want to take this.
Bill

 

well after taking it easy on the drive home...ten miles or so...then I drove normal except I didn't go over 4500 rpms...except around 250 miles on the highway I was crusing and down shifted an went thoguht three gears and reved it out to 7000 every gear...then at about 110 mph or so I just let off and let the car slow it's self down...I would do this a few times while in this range.....then I took it easy again for the most part...till about 500 miles...Just gave her a taste of some real driving every now and then...after a 800 miles I stoped worrying about the break in...I still wasn't driving that hard but I was watching it as much....

OH also...I would vary the rpms when on the highway...for example....
if you are in 6th...I would accelerate to 75-80 or so....not fast just giving enough gas to do it....then I would let off and let the engine slow the car back down to 65 or whatever the speed limit...65 here in AZ...this is vauge I know...but this is a general Idea...Also, I would drive in different gears constanlty...and if I was in sixth and it was 3000 rpms the nI would down shift and now it would be 4000 I would switch back and forth between gears making sure I gor 2000, 3000 , 4000 all given time running in the rings...It's a game to me....always watching listening ...not giving anything too much or too little.....I did the same thing with my ducati and again had a 20 hp increase over what was considered normal...some of my associates say I am just lucky...I got good motors...I say it's an engine it needs certain things...

If you beat it too much you will have trouble down the road...if you baby it too much you will have trouble down the road....find the balance....I feel I have...well at least it works for me...heh heh

 

I've read this entire thread and have to agree with the folks that claim:

Now, given that, I'd like to take a slightly different spin on this topic... I'd like to throw out that I'm sure there is some difference in HP between similar MCS's, however I'm not convinced it's "major", and in fact I'd consider some variation to be expected.

If one were to contend there should be no difference in HP between similar cars, are we saying each and every car that rolls off the line would have identical HP & torque numbers? I know in theory you'd think this is possible, however in reality I'm sure everyone can agree there are many factors that come into play that would say every car produced off the line has some HP & torque variations from day 1. I would expect there is a "tolerance" level and I would expect variation within the tolerance to be normal. Whether that means all MCS's that are produced are within 5% of average, 10%, ?? I don't know. I'm just saying I would be extremely impressed if any car maker could produce cars that consistently produce the same exact HP & torque numbers across the board.

Another point of view I'd ask you to consider is, would you expect a brand new MCS and a 2 year old MCS to produce the same HP numbers, assuming both originally produced the same numbers when they came off the line? I'd think a variation of time, mileage, abuse, etc would mean 2 cars such as these would produce different numbers.

What about type of fuel used: octane, ethynol content, additives, cleaners, etc. Wouldn't 2 cars that use slightly different fuels possibly produce different numbers?

What about type of oil, viscosity, weight?

All in all, I guess I'm just saying I'd be much more surprised if everyone's Mini's dyno'd the exact same numbers. I'd be surprised if everyone's Mini's dyno'd within 5% of each other! I'm less surprised that the dyno numbers are seemingly coming out with 10-20% variations. Given my theory that there is some variation right off the line, plus the other factors I mentioned above, and then adding a little more for variation between dyno machines, people running them (i.e. human error), climate/temp/humidity variations, etc. I'd say 10-20% is just about right...