Some of you don’t quite understand the dynamics involved in dynoing a car. Let me state a few things that I have encountered with this car. Keep in mind that I have over 65 pulls in my car alone, and probably similar numbers for 3 other cars. We have tried several different methods to see what makes the most consistent results. (note: I said consistent, NOT best!)

3 pulls is never enough. Any baselining I have done requires 5 or 6 runs. Engine temp, oil temp, etc have to be at normal and optimal levels.

Everything is a factor. You can drive to the dyno, get strapped up, and take your runs. You can drive to the dyno, wait your turn, get strapped up, and take your runs. See the difference? 3 runs *might* be enough in the first situation if the strapping is quick. It definitely does not work well in the second situation.

To be consistent you must dyno this way every time, otherwise you run the risk of getting a low number for baseline (using 3 run method) and a possible high modded dyno later. This makes it look like you made more HP than you really did. For example if I take my 2nd and 3rd run from my baseline was 232 where my 6th and 7th run was 237-238. That’s a 5 hp increase with no modification. Then on runs 8 and 9 it starts going down again as the engine becomes heat soaked.

Next is the computer. This can be the most frustrating factor to deal with. We have seen mods show HP when swapped on the dyno (plenum, intakes, cats, exhausts, etc), but we have also seen where they did not do as we expected. Sometimes the computer seems to "like" the mods immediately and sometimes it may take 100 miles or so. I dont think anyone has figured this thing out yet. Usually when the results are not up to par, the A/F’s are off and the car needs to be run for an extended period of time before making the next set of dyno runs. Despite the popular consensus, resetting the ecu does not always help. Needless to say, in most our tests we prefer to run the car on the road for a bit between dynos. This also allows to make sure everything is ok with install. As always, allow SAE correction to handle the change in variables. It does a very good job!

The best thing to do is to talk to the dyno operator and see if he has a different option instead of the 3 run system. For $140 we usually can get two cars worth of dyno runs when we pay by time. That would be around 12-15 runs total. I know everyplace is different.

If anyone else wants to share their dyno trials and tribulations I would be interested in hearing it. Always looking for ways to get out off the dyno with a smaller bill!

neffster, good luck in the install and dyno runs!

 

Also, for those who have not installed a set of headers on this car, you may not have enough time in a day to do a pre dyno, install, then post dyno. A very experienced shop my be able to accomplish this if the dyno is onsite, but a shop that has not done many header installs on this car could take 6-8 hours for the install alone!! That does not give much driving time to and from a dyno shop. Some people are just not being realistic here. I also encountered the same thing Vandy speaks to when dyno'ing related to the car and oil being at the proper temps. I saw my best dyno run at 4-5 then it started coming down a bit.

 

If I can sit 10 hours in a treestand, someone can stay at a shop for 10 hours. Don't leave the dyno shop until the headers are installed and dyno'd, and then go get some miles on it (however many needed for the ECU to learn the mod) and then go back (whenever that may be, could be a day or two or three later) and get another dyno. Oh well though, the guy already gave up.

 

People have got to let go of the thought that the ECU will need time to learn a mod. This is just not true. The only thing that adjusts itself is short-term and long-term fuel trim and they work almost instantly. Long-term fuel trim will eventually settle to a certain value after some time ... but again, this has no major impact to open-loop operation when going WOT. If anything ... putting more milages on will likely hurts your output, as the ECU eventually keeps track of knocks and decides that the gas you run no longer deserves the aggressive timing curves. You are then down graded to the lower map.

Resetting ECU definitely has it's merrits. Again ... it is the only way to insure you are basing every single dyno session on the same ignition timing map. If you don't observe any gains from resetting ECU, then that simply implies you were not stuck on the lowest ignition map (a good thing). Simply perform the ECU reset ... let idle for a few minutes. If you must, drive lightly for a block and you're ready to dyno.

With respect to dyno .... find yourself a load-based DynaPack and be done w. it. The repeatability is extremely high with these machines any day any time. No need to strap down or worry about slipping, tire size, wheel mass, etc. The absolute number will not be comparable to DynoJets, but it's the delta you are interested. SAE correction is a must. Also use an OBD-II reader to monitor coolant temp and make the 1st run at 190deg, then 2nd at 195-200 then the 3rd at 205-210 to preserve maximum consistancy. The ECU will start backing off timing w. increased coolant temp. Likewise ... it will also take out timing before it reaches operating coolant temp (above 160deg).

It's not as black magic as one would think. If the delta shows greater than +/- 3% (precision of a DynaPack) of baseline then you're on to something.

 

Respectfully I disagree. I know from my experience that most of what you say just does not hold true.

 

Without revealing too much about the ECU and getting in trouble for it. I can assure you this is in fact how the ECU behaves. This is in fact how most modern ECU behaves as well.

The fact of the matter is ... if you're unable to reproduce dyno numbers within +/- 3% any day any time ... then you're not doing something right.

 

Furthermore ... please reference This Thread

Your response is certainly expected and no offense taken or given here. Really. I'm simply here to inform. Feel free to ask questions regarding ECU ... if you respect that. I will divulge as much information as I believe will not compromise both myself and my source.

Sincerely,

 

All I know is what I have experienced over 150 dyno runs on various Z/G cars and from what Technosquare tells us. If your source is Nissan, I would like to talk with you more offline, as Nissan is historically very tight lipped about ecu specifics. For this reason we have to rely on third party discoveries.

The ecu is not total voodoo, but I know there is something more than just simple mapping going on. I have seen some layers of the hex code, and there is way more info there than the old school ecu mappings.

Thanks!

 

Uh guys... I'm not doing the dyno...

 

Respectfully ... if it's dyno experience you're looking for, then I certainly don't lack any it; most of which have been with top tier racing developments (i.e. Champ Car, ALMS, Formula SAE etc.) Although not exactly for VQs, I have also observed, participated as well as organized a fair share for the G/Z as a hobby (natually, as I also own a VQ). The key is a standarized scientific approach to testing (something I'm well trained at). Miniminzing x-factors.

I won't elaborate much more in regards to my source(s), as I'm already trading on thin ice (so I was told). I've already semi-slipped once. I'd gladly discuss in private.

Again, there is NO algorithm for defeating mods or to learn a mod for that matter. The only learning exists is for the sake of stabilizing emission and corrective trims, so they don't go skewing off to instability. This is also standard engineering practice to compensate for the lack of possible sensor recalibration evey time power is cut to the ECU. It also addresses the minute (yet sometime significant) manufactureing tolerances from one sensor to the next. There are more than 5-degrees (or types) of sensory drifts ... most of which are not addressed and designed into the system as slack.

The key to engine output is ignition timing ... assuming A/F is rougly in the desired ballpark (need not be EXACT). Eveything about today's ECU is about taking away timing every chance it has. You want reliability ... this is what it takes. There are more maps in the ecu devoted to this than others. One for coolant temp, one for intake temp, etc. etc.

So to quickly sum it up ... you must reset the ECU so they are defaulted to the same inital value and base it off that. Often, that is actually the most aggressive settings you'll ever see. Dyno pulls must commence at the same predeteremined coolant temp as much as possible. This is not advocating you will make the most power doing it this way (although that is usually the case 9.9 out of 10 times). Instead, you are simply normalizing the many x-factors that exists in the ECU. Comparing from the same page(s) ... sort to speak.

It makes little sesne for consecutively dyno measurements to get high and higher (as often reported). Those numbers are not bogus, they are simply invalid data until coolant temp reachs optimal point.

The whole DynoJet vs. DynaPack is a whole another debate. I'll admitt I'm very much pro-DynaPack for very good reasons. But, I will say it is definitely possible to gain good consistant results from DynoJets for (WOT measuremenst only). But, it require much more attentions to detail, which unfortunately most DynoJet shops do not practice. Life is simpler working with a DynaPack.

 

No problem Neff ... one of these days!

 

Interesting info on our ECU's. Any good news on the horizon about tuning for the 2005 ECU's?

 

Moved.