Which turbo did you end putting in there?

Jeff's or the Greddy?

How did you end up liking the Bride Seats? And How tall are you btw, I need to get a new seat that will fit my tall self in the car with a helmet on...

Good to see you are getting the car running again.

 

I have Jeff's on order, but installed the upgrade. I will install Jeff's at a friends house as a car proj once I get it. Maybe this summer.

The seat is AWESOME!!!! More headroom than stock, no question. It is a little tight in the hind end, but that was expected. I will measure the seat-to-roof clearance once I remount the seat so that you can know how much space you have. I had to take it out while we installed my K40 and Blinder, and I am adjusting the mounting holes.

I will be ordering the pass side after I pay for paint.

 

Very Cool keep me posted, being 6'4" doesn't help having a helmet on top of that as well..lol

 

I installed my turbo with the manifold bolted on still. I only removed the pass. motor mount, dropped the engine as far as it could go, removed the studs from the manifold to add more space. It's still not easy, but I did it. That's all I got. GL.

 

Ah, I left the studs in. THree on top. Maybe that would have helped. I suppose I could have dropped the engine more. THat like would have helped. I am replacing the studs and bolts on the next go around this summer. Will try to remember this. Thx.

 

Well after much fiddling and on-again off-again work in the garage, I finally got the gumption to turn the key after nearly a year (well 10 mo.) on jack stands.

And what do you know, the car started right up without any hesitation.

So the new mods (after going under the knife) are a semi-return fuel pump upgrade, 780cc/min and 650cc/min injectors replacing the yellow bank, cold side upgrade as discussed, a bride gias driver seat (waiting for $ to by the pass side,) a K40 radar detector and a Blinder laser jammer. I think that is it.

I also pulled my int-x (if ya'll want it let me know) for the AP. (I also deleted my secondary air injector pump with thanks to Jeff for the blocking plate!! Note: that big honking resistor on the sec air pump mount is the fuel pump resistor. You need that.)

This raises some issues. I dont have a map from Jeff for the tune, I dont want to wait for one and I think I am smart enought to tune my car myself. This last part might prove my undoing, but I have started making my own *.ptm starter map this evening. I think I might be able to pull it off.

But for now, I just need to get the injectors scaled on a cribbed valet map so that I can drive the car to VEIP (emissions; I am 6 mo. late) and to the painter. After paint, I will worry about the tune.

 

Why don't you just download one of my FI calibrations?
It'll get you in the ballpark and you can at least drive the car safely.

 

Hey! Whats happining? I would have liked to say "hey" while you were in town, but deadlines are pending and the clock stops for no man. (that is fancy talk for "I had to work.")

Again, thanks for sending out that blocking plate early. You didn't have to do that and I appretiate it.

There are two reasons why I didn't want to use your services at the moment. First, I didnt think it was productive to clog up your email with an other request for a base FI calibration when I can do some work and reading at get to a lower quality but workable result. Second, if I install your maps, I wont be able to tinker with any maps that I am writing.

As for downloading availible maps, I will have to scale my injectors to the appropriate size. If the map is locked, I am stuck. (I also have no idea where I would download said map. I checked your site and this one with no luck.)

Now, maybe there is something that I am not understanding?

 

http://www.mazdamaniac.com/dl/apcal/

 

duuuudde..........

 

And, since you went to bigger injectors, the scaling would be off in a positive direction - maning too rich, not too lean.
So, it would be drivable and safe.

I'll be in Maryland again in the 3rd week of April (20th - 25th).

 

I want to post up two questions:

1) I forgot that I was not supposed to use synth for the first oil change and loaded up with Mobile 1 as usual. (I am not asking for comments on using synth or Mobile 1.) I forgot that there was a sticker on my upgrade suggesting that 3000miles should be synth free. I have NO intention to drain $30 of oil to put dino in there. In fact, since the dealer, there has never BEEN a dino oil in my car. The turbo might not last a year and I am pulling it when I get the 3071 anyway. However, I am curious WHY synth MIGHT be bad for the first oil change.

2) in AccessTUNER, am I correct to assume that bank 2 fuel injectors applies to P2 and bank 3 applies to secondaries?

 

I'll see you round then. I will bring beer.

I'll load that cali and see what I can see. I just need to get to about 3250 rpm to drive. Currently I start bogging around 2700 rpm with too much fuel. this is with cobb's base map (or their stage 1)

 

No.
Bank 1 is primaries, bank 2 is secondaries and bank 3 is P2.

 

FWIW that base map does what I need it to do. No cels, I can rev to about 3500rpm, and is easier to remove than expected. It is perfect to put in someone elses hands as they cant go nuts and hurt the car.

Now to figure out why there is an A table and a B table for the AF targets.... This is going to be fun.

 

I spent this weekend trying to figure out the AccessTuner Race (ATR) software. I think I made some good headway.

I first made sure that my MAF was scaled appropriately from about 4000rpm to maybe 5000rpm under MM's base calibration. I am not completely clear on this concept, but I am pretty sure that the idea is to look for differences in AFR in open loop from the AFR mapped. This air necessary to bring this difference back to zero is the additive inverse of the difference in air that the MAF is reporting from the true mass of air delivered.

Interestingly, even with a greddy MAF housing, my LTFTs have been pretty close to zero. I am not sure why this should be. I was under the impression that the greddy housing was smaller. Anyway, until I am forced to change, I will just run this one.

After I didnt adjust my MAF calibration, I went ahead and opened the "Stock Style Mode MTv100" map and started poking around. It became obvious pretty quickly that the first thing to do was scale injectors, then RESCALE the fuel tables. I rescaled the injectors, but noticed no change in behaviour of the car. The fuel tables were much more difficult to understand.

The biggest issue blocking my understanding of how to tune, was understanding "Load." I am still not completely clear on calculated load, but my assumption is that load is some how linked to VE. While I dont have any way to crib responses from the car while boosted under control of some other piggyback, nor un-boosted, i did follow "load" while driving out of boost. It looks to me that the AP reported "load" is a percent of output at a theoretical 100%VE. The "Calculated Load" in AccessTuner Race, looks to be a fraction of the same number. If this were the case, then I would be boosting into ranges of a map that would need to be scaled much higher than 1. In fact, my very very rough guess would be that boosting to 14lbs would require a map scaling to at least 2 (this rough guess is not only based on a known error (that doubling boost doubles mass flux) but seems to be incorrect altogether as I achieved loads of 170 at only 8 lbs or so of boost.) I then scaled my fueling maps to 2.

Scaling to 2 has some advantages as the evenly spaced Calc. Load steps are simple tenths. I kept the tenth map columns and transfered them to my new map. I filled in everything with 0.73 that I thought MIGHT be in boost. At this point i had a very, very poor understanding of load instead of just a very poor understanding. I hadn't, at the time, equated the load on reported by the AP to the calc load on the maps. I then took the car out and repeatedly tried to get the AFRs to drop just prior to boost. This never happened so I pushed the low lambdas deeper into the lower load and tried again, but with no success.

At this point, I figured the relationship between load and calc load, and started understanding where I would want to enrich the map. I figured that anything under 3500 rpm would not be able to get into high load, i.e., the bottom left and upper right part of the maps are somewhat unused. I also noted on the AP the basic cruise load, and load just prior to boost, in a very rough way. Using this info, I could put my low lambda points.

I took the car out and expected to get good AFRs. Unfortunately, I noted that just prior to boost I was getting AFRs around 19!!! This was nearly instantanous and at low RPM and low boost (ebc set at 3.9lbs). I remembered right then that there were limiters on load in the ATR. I went back in and set the load limiters up to near or at 2 and headed back out.

BINGO! Car MOVES! AFRs under boost are nice and flat and close to what I set them (I set to 10.7; I am getting about 11.1.) Boost holds to decent revs. (Figure 1 is Load vs AFR.)

On the downside, my transitions into boost are crap and I get a consistent bog.

I will post up my map for ya'll to look at.

There are major issues that I am not sure about. For instance, why do I get uncontroled flow at high load (Figure 2: Note the "Bulge" in airflow at high load. The graph is filtered for rpm >4k, TPS 30-100%.) Also, I have got to figure how to transition from closed to open loop.

I will say at this point that I have NO idea what I am doing, so dont go use it or this method on your car. Your car will certanly explode, or at a minimum, be permanantly damaged.

 

do you have access to a logger to plot your rpm vs load vs AFR ?
You will need this - I have found a chart like shown below invaluable for tuning




once you get this pretty right you will find you can pretty much use the same lambda values in all cells at each rpm row from 100% up

 

You mean like this? Sure. I figured using plots that eliminate time give you more data from multiple runs and make it a little neater. It takes more time for me to set up these multiple plots as well.




BTW, I am not sure what is going on in the afr chart. I just randomly pulled a run. Not sure what was going on.


So just set the target boosted AFR to every cell above 100%? What about in between?

Thanks

 

looking pretty good - the spike from Aux port needs to be knocked back a little - hit every point on the 6500rpm row and above 100% load and drop the lambda value by 3-4 % for that row only. You can lean it out 5-6k range also but take it a few % at a time

You can leave everything below 100% unless you notice it running rich somewhere

 

Thanks man! Will try it out and post back. Might take a week or so, tho.

 

here is what they mean by load, according to SAE standards:



oh and for airflow:

http://www.translatorscafe.com/cafe/...ond-%5Bg/s%5D/

most compressor maps are displayed in lb/minute, so you can trace where you are on the map if you convert the units.

 

That is a good read. I am thinking that I need to scale my map out to 2.2-2.3 just for safety because I would like to reach about 330g/sec air flow.

Of course, I have several other issues to sort. I would like to get the MAF cali'ed over a larger flow profile (for stable and accurate AFR) and the injectors compensated (for accurate AFR), the AFR zeroed onto about 11.5 in boost (maybe 11.9? final), the closed-open transition evened out and the AFRs up to about 15.0 in cruise.

 

330 g/s should be well within the range you already have ?

 

Yeah, but there is something(s) funky going on that I dont understand.

Without any error or some other behavour that I dont understand. I should be at around 150% load at my current max flow, which I am, but it looks like I have excursions to 300 g/sec even as flow drops (see load vs flow). If you look at my time trace, you can see load dropping while flow increases. IATs as far as I can remember were near ambient, but I will log them next time. Is this a MAF scaling issue? A separate behavior? If I dont understand what the hell is going on, does it makes sense, to be safe at 330 g/sec and 10 psi (just on the edge of the chart, IIRC), to buffer of 20-40%? These would be calc. loads near 2 on average.

 

have a look at the chart I posted - load drops away as rpms increase .