When you click on links to various merchants on this site and make a purchase, this can result in this site earning a commission. Affiliate programs and affiliations include, but are not limited to, the eBay Partner Network.
This winter has been kind of an odd one, car doesn't really need anything except tuning. no parts, no adjustments haven't really opened the hood much.
i had the map 84% tuned with no fuel pressure, so i figured re-tuning it with fuel pressure was going to be a big change, so ive been going after small chunks at a time.
i had it up to 91% and then found another part of the map and now its 58% again, lol
it did find a value it likes for idle and WOT though which is nice.
i got warned/told that i want peak injector duty around 85% max, to keep the injector driver cool/happy, which is sensible. the funny part is that the map doesn't look too much different way up top,
the FC Tweak has mapped where the car runs pretty well, duty cycle dropped from 1 million percent to 78%, and its still a bit rich of target (Target is 11.3, it hit 10.9),
but on the boost over that it has gone nuts and duty cycle goes super high. the thing is though, with my stock S4 turbo, i'm running as much boost as i can, it can't go higher, so it looks like duty is high, but it can't hit those cells
the other thing is that the map (in PIMS) has a 16000 row and an 18000 row, but my car runs 16500 to 17500, so i'm thinking of moving the 18000 row to 17000 it should give more resolution, in the spot it actually runs.
my power limit is basically the injector duty on the stock injectors, basically.
Part 2 is that i finally got off my butt and hooked up the Pre turbo backpressure gauge
my manifold is either just like the one ReTed has on site, or its actually possible that it IS that one
so it has a compression fitting for EGT, which made it really easy to install the brass tube to silicon hose to boost gauge (it didn't make it easy to figure out compression fittings though!)
pic of the gauge, if i had a longer hose it would have been nicer to have it closer to the tach, but i learned what i needed
mess under the dash is the Datalogit/wideband, there was some flap last year about the wideband signal being noisy, so i just put the WB on the DL,
the wires can't get any shorter, and the ground is shielded. we don't know, but the old setup ran to the engine bay and back, so it was really long, seems like that may have been the thing
also one of my favorite mods, the RHD glovebox in a LHD car, remember the FC interior is forced perspective to look square from the drivers seat, but that means that it isn't square!
anyways, the results. i'm not sure what i was expecting, but they still surprised me.
back pressure passes boost by 3000rpm (7psi boost), and then they go up together as boost goes up to about 13psi.
then after that boost falls off but back pressure just stays at around 15psi.
looks like this
boost (blue) peaks around 5200rpm at ~13psi, and then falls off Backpressure (green comes up and stays there)
i tried with and without the silencer in the exhaust, which is the biggest exhaust for the FC, and it made no difference, which was nice
a while ago i re-read the Hugh McInnes turbocharging book, and built a spreadsheet to kind of see where i am (it was the tip off that there was something wrong with the fuel system)
and so we can put my data on a compressor map. my car hits basically the choke flow of the turbo, which seems like its almost a vertical line, 13psi at ~5000rpm dropping to 7.75psi at 7000rpm, it looks like this
on the turbine side, i assume the Big Dumb Exhaust has about 2psi, it puts me here on a turbine map.
my calculation includes the fuel, so its all the stuff that goes through the compressor, adjusted for temp plus the fuel mass,
which is why it seems like there is more coming out than went in.
next step? well ordinarily i'd be shopping for the HKS T51 to go with my 94mm exhaust, but since i own a V Trim compressor in a set of stock housings, i think i will try that and see what happens
why do you believe back-pressure will stay the same with a boost increase?
i'm not sure what backpressure is going to do, i'd say its either going to go up or down, but its really easy to find out, so why not?
so last time i ran this compressor, boost was lower than the stock compressor, so i'm kind of expecting boost to drop, but we will see about that too
I was half joking about the turbo, but only half … you declined before is why I mentioned it, so no big deal.
.
true, i kind of decided that i felt like i gave (give?) a lot of advice about how to pick a turbo and stuff, and maybe i should just follow my own advice.
so i'm going through the process, and i don't think anyone is surprised the stock turbo is too small. (ive also never seen anyone actually measure it)
i'm sure your turbo is a good choice, but i don't know what i need yet.
it turns out when you suggested the turbo i needed an exhaust! 70mm downpipe into 2x50mm exhaust didn't flow... no turbo would have fixed that
why do you believe back-pressure will stay the same with a boost increase?
well, my thinking (certainly not based on anything empirical) was simply that he's changing to a bigger (possibly better/more efficient) compressor, but the hotside is still the same. downpipe and exhaust are also remaining the same. in writing this response, i suppose i just made an argument for increase back pressure, so is that what you're trying to illustrate?
just making sure I’m on the same page; it’s going to be the same turbine with a bigger compressor?
.
yep. going from stock 44/63mm compressor to a 55/70mm V trim, aka 409179-0025. its going in a bored out stock compressor housing/backplate, and the three parts are literally the only thing that are getting changed.
last time i ran this, with the old exhaust, it did make more power up top, but it was also really laggy
unless the compressor efficiency is well improved my expectation would be for backpressure to stay the same or increase.
it could still make more power with the boost not dropping off. The same turbine wheel/housing can only swallow so much flow regardless. However, if it produces more power, then the extra exhaust mass has to be processed on the turbine side. The question is whether the turbine assembly was choking or whether the previous compressor was simply out of flow to keep up, as well as the efficiency improvement of the new compressor.
I don’t see emap decreasing without a substantial efficiency increase from the new compressor. The imap/emap delta relationship will likely be improved even if emap is the same or higher. So my bet is emap will be the same or higher, only supported by assumptions.
the turbo I was referencing is an EFR7163 fitted with a hybrid 1.25 A/R EWG housing; a hyped up 7670 in a B1 frame size. The turbine housing would probably be a B1.5 size though. Still a smaller package overall compared to the EFR7670.
.
i'm pondering about testing the ole K2RD turbo block-off plate.
back in the day the turbo on Pauls 20B Fc took a dive, so he made a plate to go in place of the CHRA, while it was being rebuilt
it still had the manifold turbine housing and DP/Exhaust, just no turbine wheel.
i don't have the plate, but i could build a CHRA with a bolt instead of the turbine wheel.
it would isolate the manifold/turbine housing from the wheel, as far as flow goes.
if i could get to all the bolts that hold the CRHA in i would do it for sure, but since i would need to take the turbine housing off i'm on the fence.
the S4 turbo is not a lot of fun to take off.
so i made 4 changes, and it pretty radically changed the AFR in vacuum and i'm not sure why.
1. i changed the Compressor wheel, housing and back plate
2. instead of a nut and bolt to hold the Turbo Inlet Duct to the bracket, i used a U nut
3. on the hose that vents the OMP nozzles, i added clamp
4. same hose i blew through it, and it had like a hairball in it, it moved air, but not as much a a 6mm hose should.
net result, stupid lean in vacuum, barely runs. i took an old map and just added like 5% fuel everywhere, and its good enough to start tuning again, but unexpected
clock was way off too, so i wonder if maybe the PFC forgot the map or something.
part 2 is when i took the belt off and casually tossed it to the side it bent the CAS, so it threw CAS codes yesterday, really fluke thing
oh so the 5th change i made was to remove one of the two brackets from my turbo heat shield, it was really hard to get too, and maybe it felt over constrained.
result was lots of noise, so much so that the knock sensor picked it up and the FC-Tweak threw a warning!
it was easy to put the second bracket back on, and slot the other one, to make it easier to put on
but if you tap it, it rings like a bell. the car has always had more knock activity than it should, and so i want to make the thing acoustically dead.
i feel like most people try to get these things to shield heat, but well here we are.
my friend had given me this stuff that is aluminum on each side with plastic in the middle, and it comes in sheets and can be bent, folded and cut really easily
so i bolted a cell phone size chunk to the shield and its acoustically dead, which is great.
but then it doesn't actually fit, cause the intake manifold is in the way
so i removed the black stuff, and instead used the piece of header wrap in the pic, which didn't fit either
after that i realized i had a spare turbo and manifold and why not use that as the mock up.
which let me bend the thing to be more closer to the turbo.
i ended up using a strip of header wrap on the inside, which is kind of an old school trick, although usually the old schoolers would drill a couple of small holes and use safety wire
i used gravity which, its probably going to be a problem later
its been tough to get the "new" turbo tuned in,
also with the CAS problems, it threw a CAS code, again.
last time, i just checked the clearance, and thought it was find, this time i pulled the whole thing apart and found this
when i threw the belt at the CAS plate, it bent in, and i guess the CAS sensor made that little burr.
so i got to spend some time filing it smooth, and then making really sure the plate was so flat that the straight edge looked bent.
it seems to have worked
the other problem i'm having with the new turbo is that, apparently finding a place where i can actually get it over 5,000rpm is tricky!
or maybe get it there enough for the auto tune to have an average AFR
up to ~5000rpm though, response changed, although the actual boost curve didn't, which is hard to explain. the stock turbo feels more like a turbo, and this V Trim feels more like an NA.
after 5000rpm boost goes through the roof, and it'll hold it to redline, which is potentially cool. its hitting fuel cut which also makes it weird to tune, i had it set to .95kg, but i moved it down to 0.9.
ideally i'd like to tune it before i just let it run 13psi. its funny that NOW i care about the afr, and not like 5 years ago when it was stupid lean...
i hooked up the gauge yesterday, and now i have some numbers, although what they mean, i'm not sure.
i really went down the turbo rabbit hole, and i felt like i knew what i was looking for, and well now having collected some data, i realize that there is more stuff i don't know
so here are the bullet points
Boost: from ~2000-5000rpm boost is the same as the stock turbo. auto-tune didn't do much, like 2% here and there. from 5000-7000rpm boost is up. the stock turbo would hit ~13psi at about 5500 or so, and then it was down at 7.75psi by 6800.
the V trim hits closer to 15psi, and it'll hold that to redline. i had fuel cut set at 0.95, and the stock turbo would occasionally just touch that. the V trim slams into fuel cut, and its kind of violent. so ive been lifting the accelerator to go to 7k
Pre Turbo backpressure: did not change. to recap, the stock turbo would cross over around 3k rpm, and BP would track with boost until the peak and then backpressure would just stay around 15psi up to redline. the V trim does exactly the same thing, however since manifold boost doesn't fall off the boost to backpressure ratio is way better. the stock turbo was 7.74psi of manifold pressure and ~15psi pre turbo at 7000rpm, the V trim is ~15psi preturbo and 13psi manifold.
Intake air temps: intake air temps dropped a little with the V trim, unexpected, but kind of nice. this time of year, 15c day, the stock turbo would be 38-40c, the V trim is 33-35c.
Fuel: i'm out of fuel again. over 6k, injector duty is in the 90's and it just goes super lean. on paper i've got enough injector, so maybe it wants more pump
Response: so the boost curve is the same, but the tip in response feels different. the V Trim feels more like its NA, its kind of natural and linear. the stock turbo feels more like a turbo, where there is a delay and then you get power.
acceleration up to like 5k is pretty similar, but after 5k the V trim is pretty serious. the stock turbo is much less demanding of the driver, potentially more fun, you can drive with more abandon. the V trim made the car faster.
Boost Control: with the V trim there is none, lol. the original setup of the car had an RE Amemiya ecu, and it was labeled 9/7k, and it would hit 9 and then the S4 turbo just falls on its face. but that kind of boost target seems like it would be pretty good. the S4 turbo can't do something that broad, and the V trim is just out of control, lol. also the V Trim wants to boost more than 9k. the PFC, has overboost fuel cut 0.25kg over the boost pressure, but the FD map sensor only reads to 1.1 (or something), so the most boost i can run and have fuel cut is 1.1 - 0.25 = .85
What's next? i'm not sure. the testing before this has really pointed me in a direction, and then this one kind of hasn't. well, except for the 86-91 Rx7 -> Fuel System -> Fuel Pump direction, lol
so in a complete change of pace, my brakes suck. the car stops fine, but the pedal felt like crap. they were kind of noisy too.
the pads that were in the car were a complete mystery. i did a lot of research because i wanted to be able to quantify something.
i ended up with these, and i'm not sure they are bedded in yet, but so far they are great.
super quiet, nice initial bite
i also adjusted the brake pedal, which was way off for some reason
i got out and did some tuning yesterday, and the brakes are great. in addition to being really quiet, good initial bite, they have more stopping power than the old pads, by a lot.
again i picked the endless pads because they had some stats and a Mu number, so if i didn't like something, i had some way to quantify it, and then try a different pad, but i'm really happy with these
two, i put the silencer back in the exhaust. i have it clamped on, because the stainless the muffler is made from is so hard i've spent a few hours drilling, and have a nice divot now.
if i get it clamped on ok, its been fine, but when i took it off to see what it did, i didn't get it back on right, and
so i bought another one, its just a cheap amazon thing.
at 30lbs a minute with the stock turbo the silencer didn't effect performance, which is pretty awesome.
it actually sounds better with the silencer in it, which is also pretty good.
however at ~40lbs a minute with the V Trim turbo it is effecting things, which is actually sort of good.
the old ECU i had for the car was setup to run 0.9 and then 0.7, which seems pretty good to me. the stock turbo was basically 0.9/0.5
and the V trim was boosting higher, like 0.95+, and the silencer lowered this a bit, instead of being wildly out of control, it just boost creeps now
in the long term a bigger wastegate is in the cards.
and three, i had set the suspension up so the rear was higher than the front, and before Sevenstock i lowered it a turn. it looked better, but it had so much toe in in the rear if you were on the gas it wouldn't turn!
when i was putting the brake pads in, i just decided that the floor wasn't very flat, and so i just set it so that all four corners are 7 turns up from the bottom. it wasn't a big change, but it rides way better for some reason.
while i was doing that i looked at the toe adjusters and they were just in the middle. ive had this car so long i'm kind of forgetting what i've done to it, i guess.
when i finally do make it to the alignment rack, its got corner weight scales too, so it will be pretty neat.
but until then i just moved the rear toe a notch on each side. its much better.
it seems crazy to do that, but like i didn't know where it was before.... i might try one more notch
the car has come a long way, it actually is really quiet and smooth. handling is closer to what i want, but its not quite there yet.
power i think is really close too.
the only two things, really are the noisy diff, and the clutch which just sucks. its an ACT Extreme w/ Pizza Shooters and its both high effort and it wants more travel than the stock system will give, so it just sucks all the way around.
two, i put the silencer back in the exhaust. i have it clamped on, because the stainless the muffler is made from is so hard i've spent a few hours drilling, and have a nice divot now.
if i get it clamped on ok, its been fine, but when i took it off to see what it did, i didn't get it back on right, and
so i bought another one, its just a cheap amazon thing.
I actually tried one of those silencers before I did my recent exhaust work. It did help make things a bit quieter, but all of the "high heat" paint on mine turned to ash and then the entire thing warped so bad inside the exhaust tip that I destroyed it on removal. It was worth a try though. Like you said, just a cheap Amazon part, but it was surprisingly effective and I didn't feel any issues with flow..
broad strokes (too long didn't write?), this replaces the old NZ made Datalogit.
the Power FC has an output that would/should have gone to an Apexi cable, and then you could use Apexi's tuning software to tune the thing.
as is typical, they have an English version of the software, but it was only given to Power Excel shops, of which there were 3 or 4 (the local one didn't even open it!)
so to fill the void the datalogit was reverse engineered to take the PFC's data stream and decode it and then be able to make changes and stuff.
it does work, but its slow and clunky and old and unsupported and it just makes tuning the thing really goofy.
the PFC is willing and able, but the datalogit needs a serial cable, or adaptor and then it needs to interface with windows, and the whole thing is like getting a haircut by a one armed midget balancing on a stool.
the DL-340XB fixes a lot of this, and then does more neat stuff on top of that. firstly the Datlogit logs things that were unknown when it was made, but are now known. however since its old crap, we just have to know that ?? is BAC valve duty. which shows how unsupported it really was. 2, the DL340 has a USB out, so there is no serial to USB converting, windows seems to not be very good at this, so this is big
three, it logs twice as fast, so instead of getting ~15-20 datas per second, its going to be like 40. just to bring this one home, the 20 year old ECU will spit out ~40 datas per second, but we were only getting 20 because the rest of the chain sucked
fourth there is some kind of timer in there, so when windows hiccups the software can make the data show up where its supposed too,
5 and this is actually a big one, it has shielded cables for the O2 sensor. any noise on the o2 sensor signal reduces the resolution. the DL340 comes with shielded cables
also one of the things that gets streamlined with this is that instead of hooking the wideband to the Datalogit and trying to get two streams of data to match up, this new box uses the PFC's o2 sensor input, which simplifies it to one stream.
there is more too, there is a playback of the log, etc etc
i've heard things about the Apexi software, and this DL340 box has all of those features, which is pretty cool
update #2 is monumentally stupid, but the parts aren't here yet.
its related to my high knock numbers, i'll post pics, but i was watching a playback of one of my logging runs, and the high knock happens at like 3000rpm cruise
the other neat thing about the playback is that you can see the lag between throttle input, and afr
02-03-24, 10:29 AM
you declined before is why I mentioned it, so no big deal.
