GTSBoy

Don't try it. It will be shit. Better chance of damaging the engine than gaining any knowledge about the old one. Maybe, but unlikely, especially if the O2 sensor is working. You can use NissDataScan to look at the stock ECU. Cleaning the AFM is seldom necessary and seldom helps. The R35 coil & loom conversion should eliminate dud coils or plugs. But as you still have misses - look to the plug where the coil loom connects to the engine loom, and the CAS plug. Need quality connections there. Also look to injector plugs. And the AFM plug while you're at it. Also look to whether your injectors are clean and flowing evenly. Get them out and onto a testbench.

You'll fit right in. Acceptance is the first step.

R32 HICAS, so long as the CU is not itself faulty, is easy to keep working enough for the PS. Just unplug the smaller of the two loom plugs, on the CU, like I have already said about 4000 times on this forum. This works with all the stock hardware, or with a lock bar, or with a complete removal. I know, because I have done all of them. OK, actually, I never put a lock bar on because why would you put a lock bar on when you can get rid of the whole lot, including the stupid tie rod ends?

Oh, another question comes to mind though. What sort of shitty fuel are you feeding it. Mine, a turbo, has to be fed 98RON, so it doesn't ping itself to death. 98 in Oz is good, dense fuel with minimal oxygenated compounds. Meaning that the energy density is high, and if the O2 trim is working, you will get more MPG than with a lesser fuel. If you're using some sort of piss poor E10 blend with some other MBTE or other oxygenated additions, that could explain using more per mile.

Hence why I suggested some tests.

Yeah, nah. I had the actuator rod off it today. The arm will not move at all. Neither out, nor in. Yeah, you'd think so, but I've been thinking about that. Even when the actuator rod fell all the way off at the beginning of this saga, it would build more boost and faster in lower gears than it would in higher gears, and you'd think that that was the opposite of what should happen. But I strongly suspect that there is a thing with the gearing getting the revs to rise faster, that there must be some transient effect with the gas flow rate rising quickly, that you don't get with the more steady state case of the higher gears. Keep in mind - the gate is not shut in either of my weirdnesses. So things are not "normal". We normally think about a turbo spooling up (below the wastegate target) with the gate shut. I have all sorts of mental models running now where the gate is a little bit open, and having it stuck open allowing gas out while it should be going through the turbine has all sorts of weird effects (in these mental models). I'm thinking in the higher gears, the ex mani pressure builds to the point where enough gases spill out the wastegate to just prevent the pressure rising much more at all, or just creeping up, all the whole the revs are increasing and getting closer to the point where a gear shift becomes necessary.

And now the bloody flapper is jammed slightly open. Lazy as all f**k in 3rd-5th (very hard to convince it to make much boost before you run out of revs), but will make 12 psi in 1st, which shouldb't be possible with a 5 psi spring and controller set to <10 psi. Methinks I'm lucky it jammed where it did, rather than fully closed. It's actually a well weird situation. Careful examination of what's happening in the back of that housing when the gearbox comes out and all the other stuff (injectors, AFM retune) is supposed to be done, next week.

Keeping in mind that I am currently getting flat 10 L/100km on my RB25DET, which means >>500km/tank. I would test this car for vacuum leaks, check codes and AFM signal with NissDataScan as D says, check plugs, do a compression check while they are out.

Just weld it onto the hot pipe. If you want it right after the turbo, put it right after the turbo. Or rip the comp cover off and drill and tap a hole into it.

That's the spirit.

A HICAS fault really should not cause the engine to go into limp mode. These cars are so old that the systems barely talk to one another. The thump might be the lockoff solenoid valve for the HICAS (the safety valve) getting closed, to lock the rear wheels "straight". I say "straight", because when my HICAS used to shit itself before I took off and nuked it from orbit, it would lock up with the rear wheels wildy non-straight and you'd need 45° of steering wheel to drive straight. Take off and nuke it from orbit. First step is to pull the smaller of the two plugs out of the HICAS CU. See if it stops misbehaving. You can also (whilst leaving it plugged in) try to get the HICAS CU into diagnostic mode by doing the pedal dance, or by connecting a CONSULT capable diagnostic reader into the car and seeing what it has to say for itself. Can't do either of these with it unplugged though. 1st gen HICAS is a ballache. I would do a complete delete. I did do a complete delete.

Ah. I dunno why, but I skim read your earlier post and thought the manual switch was to force it on, not force it off. I'd be tempted to have both. So, your two in series, and another in parallel with those. Or, a master arm switch, followed by the manual force on and the thermo in parallel. 6 and 2 3s.

When sanding across a convex panel, you need to sand, as much as possible, at right angles to the curve. If the panel is compound convex curved (ie curved in both directions) then you are SOL and have to use "craftsmanship" to get it right.

So, 3 switches then?

This sounds cool. But, as per usual, anything really complicated like that, that I take on, constitutes a steep learning curve that I never actually "learn". I just get it working, install it and start using it. Then, when it breaks a few months later, I've got no memory of what I did, how I did it, or even the bloody IP address I gave it. I'm having more or less exactly that problem right now. My Proxmox machine was discovered to be non-responsive, and the VMs on it were ditto. Power cycle it to a stream of loud Dell beep codes - which was unpopular at midnight, I can tell you. Move it to the lounge room so I can HDMI it to the TV for a screen, see the boot messages complaining about the /PVE/data having all sorts of LVM problems, saying it needs to be fixed manually. And I'm like....great. How am I going to work out what I need to do? At least it wasn't the stupid Silicon Power SSD shat itself. Mental note for all readers - DO NOT BUY SP SSDS! They are shit. I have had one take out an entire VM setup already, and this one could have done the same. Maybe it actually has. Cheap Chinesium shitbaskets. I won't cheap out again, even on play projects with no budget. Because my time budget having to fix shit is worth far more than the cost of proper parts.

I don't like "actual computers" for in car use. They take time to boot up, have OS annoyances, and so on. Arduinos etc are ready to go a few seconds after power on, don't mind being agressively powere cycled, because everything is non-volatile, don't mind being shaken and stirred.

Being a race car, and being in the era of the Arduino, one would think it would take little effort to build a controller to do the spraying based on a real physical measurment. Waaaaay back in the dim dark AS days, JE "designed" (as in, he had help) a microcontroller based intercooler spray system. It watched the difference between a temp sensor stuck on the core and one in the free air in front of the cooler, and if the temperature difference exceeded a (settable) threshold, it would activate the sprays. Thus, it only ran water when there was an actual need for water. If you stop to think about the actual physical things that are going on in that stack of coolers, there's probably at least a couple of triggering conditions one could come up with, and one could probably even run one pump with more than one solenoid valve, to allow water to be placed where it is needed, or at all points at once (if it is needed at all points). We're in the age of science baby. But.... I suspect that intercooler water sprays are on the forbidden list in most circuit classes, no? So only good for Targa type stuff?

That's going to be nice to look at.

IMO, flush with the front (and by front, I mean the face that is closest to your face, when you're looking at the flywheel).

That's not completely truthful now is it? Any flex fuelled factory cars also had the option to run on steam.

So.....wire it up appropriately. You can't use the resister pack with those injectors anyway.

My first car was a HG. I'm very familiar with them. A mild cam upgrade is a good idea. The 186 is a very flexible engine - meaning it has good torque from down low. You can give up a little torque down low for quite a lot more excitement in the mid range, and a bit more up top - but they are not exactly a rev monster. You need to upgrade valve springs at the minimum. For a bigger cam, you'd want to make sure it wasn't still running the original fibre cam gear. That would be unlikely, given that most of them shat themselves in the 70s and 80s, but still within the realms of possibility. Metal cam gear required. Carbies are a huge issue. The classic upgrade was always a Holley 350, which works, but is usually pretty bad for fuel consumption. The 186S had a 2 barrel Stromberg on it that was very similar to the one on the 253, and is a reasonable thing if you can find one, and find someone to help you get it set up (which is the same issue with setting up a 350 to work nice). The more classic upgrade was twin sidedraught CD type carbs, or triples of same, or triple Webers. The XU-1 triple Webers being the best example. You can still buy all this stuff new, I think, but it's a lot of coin to drop. And then the people able to set them up are getting fewer and further in between. There's still some, but it used to be everyone's** dad and uncle could do it. **Not everyone's! But a lot. All in all, I wouldn't get too carried away with the engine. Anything you do to it without a full rebuild for power and revs will only make it slightly faster. I am all in favour of a complete teardown rebuild, with nice rods and pistons, 10 or 10.5:1 compression, and a clean port job with at least a big enough cam to run 98 with that compression, if not bigger. And if I did that to a dirty old red motor, I'd want to inject it too, which I'd struggle to fight against the devil on my shoulder that would argue for ITBs and trumpets. But the bills would start to mount up, and it will still never make stupid power. OK, a few people still know how to build absolutely mental red motors, courtesy of the work that went into HQ racing and modern knowledge being applied. But even a 300HP red motor is no match for an RB20 with a TD06. So you have to decide what it's worth to you. I'd just put a set of 6>2>1 extractors, a 2.5" exhaust and an electronic ignition conversion/dizzy on it and just run the old girl like the fairly slow old girl that she really is.

Stock RB fuel pressure is near enough 43.5 psi, so the latency in that table at 31.6 will be close. You can see that 7 or 8 psi equates to about 0.4µs extra latency. So if you wanted to interpolate between the 31.6 and 39.9 psi values you could say you're going up about 2 psi out of those 8, so add about 0.1µs, which is barely worth talking about and is quite possibly wrong because ideally you would fix the latency while running at the appropriate conditions on the dyno, with a wideband sniffing its butt.

Also true. But imagine not wanting new injectors? Imagine wanting to use 30 year old injectors?

Just get rid of the orignal low impedance injectors, replace with modern high impedance injectors, and you won't need the resistors. Wire direct, as per any other engine.