joshuaho96

Yeah, but in the US if you do this it's almost guaranteed you get the infamous black goo buildup in intake ports and E85 fuel stations while relatively common are nowhere near as readily available as E10 premium unleaded. E85 is also not regulated that well out here. Lots of places "E85" is actually as low as E50 in the winter to reduce cold start unburned fuel emissions. E85 especially doesn't do well with short trips, Mike Kojima at MotoIQ documented how his daughter's EJ25 running E85 experienced stuck rings from being driven short distances and the crankcase was filled with a thin layer of condensation + oil emulsion as well despite very aggressive oil change intervals.

I would never run E85 without an ethanol content sensor. Just not worth the headache. Having to reflash the tune and only refill the tank when it's very nearly empty is super annoying.

High ethanol concentrations all the time is bad for the fuel system. Not because ethanol has some inherent issue assuming you've done your fuel system setup correctly, but because PIB isn't soluble in ethanol and it contaminates the entire supply of fuel in practice. It is soluble in gasoline so you want to run a tank of normal pump gas every so often to flush all of that out before it causes an injector or something to clog.

I really find it hard to believe that boost control demands enough compute that it needs its own module. Especially if the only real logic is closed loop boost control + adjustable boost target based on a dial or ethanol content.

Isn't the whole point of this integrated circuit thing to put a bunch of hardware all in one little package? Integration also brings its own benefits, at least in theory. If you were to put an electronic wastegate or do an N54-style vacuum wastegate that is normally fully open but closes under vacuum integrating everything allows you to do tricks like full wastegate bypass at cold start to warm the cats or in eco-mode under a certain torque request to reduce exhaust pumping resistance. Or full bypass as a limp mode if a certain amount of knock is detected. Or managing the compressor so it never reaches the surge point. None of this is strictly necessary per se but standalones generally have good enough boost control. PID control isn't exactly some crazy difficult problem to solve.

10 years is really pushing it. At 7-8 years you need to inspect frequently for signs of dry rot.

Yeah, it's more a personal pet peeve. I would rather go standalone purely to avoid putting in a Profec or whatever other standalone EBC.

I would not replace the gauges. A well setup tune should never require any of that. I'm of the opinion that if you can't safely hand over a car to your parents to drive it's not fit for purpose and more engineering work should be done. Honestly I'm not a fan of external boost controller boxes either but if you have to you can shove the controller into the glovebox or something once it's set up. Tee it off the OEM MAP sensor. Also you probably want to put a two pin plug on the new solenoid, bolting it to the OEM location works. Just separate the wastegate line and compressor outlet so it will work on a 3 port solenoid, kevboost was walked through how to do this on an RB26, same idea just different locations on an RB25. You may have to remove the restrictor in the OEM hoses if they're still in there.

This piece is obscenely discontinued. I honestly hate the frameless window design because the rubber gets stressed and the guides as well every time you open and close the door.

This would be a new pump with new gears. I'm just unclear on whether it's a good idea to run more oil pump flow if you don't actually need said flow. Oil level is set a minute or so after shutting off a warm engine so wouldn't the high RPM oil level in the sump end up lower all things equal? Plan is OEM clearances, main concern in my mind is whether the OEM pump can keep up with the flow requirements of any additional oil coolers.

Right, but if you replace the pump gears + put a spline or sine drive gear on the crank on a Nismo/OEM/N1/etc pump at that point do you really still want more flow/oil pressure? Let's say this is a the aforementioned "keep it simple" build, no more than ~400 kW at the crank.

I really don't understand the point of aftermarket oil pumps if your particular engine doesn't need more oil pressure. As far as I can make sense of it the problem seems to be cavitation from sucking air, maybe the pump gear design with how it interfaces with the OEM crank, and maybe the backing plate screws wanting to loosen themselves. How does flowing more oil fix these issues?

Save yourself the headache of an alarm. Immobilizer on the signal side of the main EFI relays so it won't start is good enough. DIY killswitch is the same thing except instead of a nice passive system you need to remember to switch it on and off.

Yeah if the goal is to drive something then modifying a car is a bad idea.

Is it just me or are those injector ms values insane? I swear from staring at Consult logs it's nowhere near that on factory 440cc much less 980cc.

For a daily just run Akebono ceramic pads. That's what they're optimized for. Obviously don't try to run it on track or anything like that.

OBD1 is a catch-all term for pre-OBD2 which could have basically any connector design and protocol. I really don't understand this attitude against using data logging for issues that can be multi-causal and subtle. A no-start situation or something like that sure the scan tool does nothing but if it runs like shit it can often be a sensor issue. When a friend of mine reliably had issues with his N54 not able to stay running eventually I narrowed down the issue to a faulty fuel pressure sensor. Only way I could do that was a scan tool to look at the fuel rail pressure and see shadow codes for fuel pressure. Then check against a pressure gauge. Is reflexively firing up the data logger probably a bad habit? I guess, it can encourage people to not listen/observe but on these old OBD1 cars just taking a log really doesn't do anything by itself. You still have to synthesize everything you know and think about what it all means.

Breather for it. If it's anything like the GTRs it should run up to the top of a bellhousing bolt.

What basics? With these OBD1 ECUs it's not like the ECU will give you any real clues, it's not really doing any sanity checking of sensors/sensor data. My experience is actually a lot of sensors tend to be the source of the issue first and foremost, and checking everything gives you a direction to investigate. MAF/O2/STFT/LTFT should all kind of make sense. You can try little tricks of the trade like tapping the MAFs to see if it triggers a stumble, things like that. But once you've done a compression test knowing whether you have an air leak or fuel delivery problem is kind of hard to narrow down without sensor data. And you don't really know if all your sensors are making sense unless you can interrogate the ECU directly about what it's seeing for sensor data.

I'm aware, but unless you're actually seeing the voltage the ECU is seeing and you're able to verify the sensors are actually working I find it hard to just trust STFT/LTFT. I will say, logging the ECU comes naturally to me because it's one of the lowest effort methods of diagnosis and I do similar things in my day job all the time. Staring at 20+ charts looking for something that isn't quite right isn't for everyone. NDS1 allows you to log almost everything so that's normally what I do and then sort out the data later.

I agree once misfires happen it's basically worthless, but the hope is you can see what's happening in the run up to that.

If it's reading full rich prior to a misfire that gives one directional hint, if it's already reading lean, etc. If it's reading pretty cleanly stoichiometric then suddenly drops out from a misfire that suggests it's not air mass estimation that's the problem. Could be ignition, could be something more subtle. Could be the CAS has decided to start dropping out at random or the drive pin is worn leading to excessive lash and trigger errors. LTFT can tell you the same but it's slower to react and if this is a recent issue it might not have stabilized. STFT stuck in one direction vs fluctuating back and forth can be used instead but I like to read O2 voltages anyways and interpret directly. If the O2 voltages make no sense in general or are super slow to react it could also be a failing O2 sensor. There's no real error correction for failing O2 sensors in these cars.

What do LTFTs look like? What do MAF voltages look like? O2 sensor voltages/AFRs? Knowing O2 voltages especially when it sputters would help a lot.

TPS sensor needs to be adjusted such that idle voltage is 0.45V, then relearn the TPS if applicable. From what I recall in the R32 GTST and similar ECUs they care far, far more about the actual TPS closed switch but later models seem to only keep track of voltage.

Oh yeah, always always always turn over an engine by hand first.