alexj

Turbotapin don't let them rebuild the diff, they don't know what they are doing and will get it wrong. Pick up the pieces and find someone else or worst case do it yourself.

you could look them up on the wiring diagram but its easier to just unplug the blue relays until the circuit you are interested loses power.

There is a connector to unplug the ignition switch from the wiring harness about 4 inches away. Unplug it and see if you still have acc powered on, if you do its the relay thats broken. Otherwise its the switch. On the bottom of the ignition barrel is a tiny screw, undo that and the switch with the wires attached comes apart from the barrel. When you have it out of the car you test the electrical switch itself.

Also the haltech and CAN analyser need the same byte order aka bigendian/littlendian setting. Confusingly this setting can sometimes also be labelled motorola/intel.

You may be mixing up the serial uart baud rate with the CAN bus speed. The haltech and CAN analyser need the have the same CAN bus speed (1mbs) and the CAN analyser and the Dash need the same serial baud rate.

I have tested the front torque gauge on the bench, the input from the ATTESA ecu is 0-5v. There are four terminals. As suggested above there is a fixed 12v across one pair of opposite screw terminals and the "sensor" 0-5v and sensor ground on the other two terminals. The copper track on the plastic "circuitboard" leading to the sensor 0-5v terminal is labelled "ETS" from memory (on the R32). The calibration units for the display on the ic7 dont matter really but you could use 0-50% or 0-288psi on a r32. If its a r33 then it would be 10-50% or 25-288psi because of the minimum pressure they run (that i guess isnt shown on the gauge). Based on my tests the front torque gauge meet doesnt totally accurately reflect the front wheel torque or centre clutch pressure in some conditions like the "fast start" behaviour when stationary, when the guage doesnt show that the centre clutch has been activated. But its close enough unless you are data logging for the track in which case drill/trap a pressure sensor on the back of the transfer case.

I tested the rollover valve in mine, and maybe unsurprisingly after 30 years it wasnt functional so i put in a motorsport one. Until then, for as long as i had my car, it had the canister removed and a bolt blocking the pipe that went to the canister in the engine bay. I hadn't checked and it didnt cause any noticable problems. It was pressuring the tank though.

I have old school-ish AP Racing CP5555 on mine, looks a bit period and they do the job on track, but not a great road car option imo. On the early 35 brakes are 380mm which is a better option, still tight on 18", needs te37 or similar with lots of clearance. So those would the ones to go for i reckon. I bet there is some kind of man maths i can come up with that the r35 would be better for the road that the Alcons; service parts, brake pad choices, road style dust boots etc. But mainly, massive gold brakes ftw.

Just to add to confusion, I think ATI had their part numbers on their website back to front for the R33 vs R32. Saw this info on a Facebook group iirc, a US drag racer noticed the error when he was figuring why he kept throwing PAS belts. ATI confirm the error so they may have fixed it by now. But there still may be duff info out there based on their old catalog. You can use the info from Ross that BK shared to check it before fitting. I have an ATI r32 pulley on my 32 but I use a 33 pas pump, so i should have fitted the ATI r33 pulley, and they aren't aligned but i have almost stock rev limiter so i dont seem to throw belts...

Reassuring to hear you rate the hioctane sump Duncan - no one in the uk has heard of them as far as i can tell. The RIPS sump is shallow and safely tucked and it does seem to fix the straightline oil surge you get with a stock sump but I was still getting oil pressure drops in long, high g, high rev left handers. My ecu has a very conservative oil pressure cut set up so I've not had a failure with the RIPS sump yet, but I does hit the ecu cut occasionally once the oil's hot and the boost is turned up. The Trust sump actually seemed to work better in that respect but you could see it was just a matter or time until you run a sausage curb at the wrong angle and rip it off. Hopefully the hioctane cast sump is a good mix of the two; decent capacity, a bit deep, with flappy gates.

Since its being discussed, here's a few pics of the hioctane sump extension going on my car this week to replace the RIPS sump it had. Hoping the swing gates and bit of extra depth will control the oil surge a bit better. To trust sump was worked perfectly by too little ground clearance for me in the long run.

I had one r32 abs pump leak. Vacuum bleeding seemed the trigger for that but the seals get cooked from the heat and age so it probably in the post for everyone. There should be a decent supply of not so cooked abs pumps though if you want to replace like for like, just get one from a z32 and deal a little some wiring. The z32 has the pump in the boot so doesn't get cooked.

I have a gtst which came with abs and that started to give an inconsistent brake feel so i took it out. And switched to the smaller non abs master cylinder to reduce the chances of snatching the brakes and locking up. I haven't driven it hard enough to really notice the difference other than that pedal is the same every time now. My 32 gtr factory abs worked well on track until it didn't. Failed in the wet on track and put the car in the tyre wall. I took the abs out and couldn't build any confidence in the brakes unless it was totally dry. So i put the bosch m4 system in and its immense, unlike any road abs i've tried. The bosch m5 is now way too expensive. But there is the option of retrofitting the bmw mk60 system which is available in three and four channel versions and the right model can be reflashed to the vehicle parameters of your choice.

I've got a new sequential gearbox on the way. it has the holinger/gm26 spline style input shaft. I currently have a osgiken twin plate with a sprung centre which works nicely. Osg used to offer this clutch as a special order with a centre for the holinger spline. But for some reason osg aren't responding at the moment. Which in japan usually means "no". osg make holinger spline unsprung centre for their less friendly triple plate clutches. But i wanted to stay with a sprung centre. Does anyone else make a spring clutch available with a holinger centre? (This is for a 400kw r32 track car)

It sounds like 4wd "wind up" you get when front and rear diff ratios are different. If the ratios are the same, then do the wheels have the same rolling radius front/rear? If they were a long way out you might get something like this. Also, would be crazy, but... that front diff looks pretty dry, you are putting oil in there aren't you?

Googling it, it looks like penrite do both a GL5 and a GL4 version. Apparently the GL4 is the one you want.

Wow that makes a huge difference. The wire labelling is so much clearer, makes everything easier to follow. I'll be printing these out to replace my dogeared prints of the old pdf/scan. Thanks for sharing.

Better quality scans of the wiring pages would be brilliant if you ever had some time on your hands! I have sent myself cross-eyed a few times trying to decipher the "normal" scan.

Thanks BK. That millers stuff was what my workshop chose. I should have researched it myself. Thinking about it I had similar issue ages ago when I put Mobile 1 SHC in my old RB20 box - that's GL4/5 oil too. So how much work is it to strip this suspect box the the point where I could see if the synchros where actually OK? Does it make any difference that its a series 3 spec box?

Small update. Had a track day on Monday after swapping back to the old box running heavy shockproof. Good as gold. So it's not the clutch mechanism or hydraulics at least. I can't seem to see any difference in the gear change speed going back in the logs but i need to look at it a bit more methodically.

I may be wrong but i did some maths recently; at high rpm an upgraded oil pump is moving around a quart a second. So you can size the accusump for the length of protection you need. A long corner on track might make for a few seconds of oil surge to make up for.

Thanks Duncan, that's some good experience-based advice, I really appreciate it. Not many genuine specialists left in the UK sadly. Driving style is in the back of my mind as a cause. I generally shift fairly slow and I have a fair amount of mechanical sympathy. I don't religiously heeltoe though, and might have started to get a bit quicker shifting recently, with a bigger turbo now, trying to keep it on boost. I have logs going back a few years I could see if I've got a bit keener over time. What would you look for in the data logs? I dont log the clutch switch but maybe could look at the time it takes for the revs to drop for the next gear? I actually have a sequential on order, long lead time though, so the 5 speed will need to see me through at least the next few months.

I'd like to hear opinions on this (cross post from the end of a different thread) I ran a standard r32 gtr 1991 5 speed on track for years. it had heavy shockproof in it that whole period. No problems at all. I bought a brand new series 3 box from nissan via rhdjapan and my garage put "normal" Millers CRX 75w90 NT oil in it. It ran fine for a few track days and changed the oil. now after around 5-10 days on track (so 10-20hrs of hard use) it crunches on all upshifts over 4500rpm or so. Only double de-clutching stops the crunch on high rev changes. Downshifts are ok. When it started crunching I got the garage to inspect the clutch (os twin plate) and bleed the master and slave cylinders. They couldn't see anything wrong. They feel if the clutch wasn't fully disengaging (which was my first diagnosis) it would have problems at low speed, which it doesn't. They feel the synchros are toast and it needs a rebuild. What killed the new box so quick? Was it the wrong oil? The first fill on the new box and the oil change were both Millers CRX 75w90 NT which are described as for "Manual transmissions, synchromesh ... which require API GL4 or GL5 performance"... I'm going to throw the old high mile box back in, with heavyshockproof as before. But I'd rather not ruin the synchros my old box too, so what would you suggest as diagnostic steps to make sure it isnt an issue with the clutch or something else that is the root cause?

So here's an odd one that I'd like to hear opinions on. I ran a standard r32 gtr 1991 5 speed on track for years. it had heavy shockproof in it that whole period. No problems at all. I bought a brand new series 3 box from rhd and my garage put "normal" oil in it. It ran fine for a few track days and changed the oil. now after around 5-10 days on track it crunches on all upshifts over 4500rpm or so. Only double de-clutching stops the crunch on high rev changes. When it started crunching I got the garage to inspect the clutch (os twin plate) and bleed the master and slave cylinders. They couldn't see anything wrong. What killed the new box, wrong oil? The first fill on the new box and the oil change were both Millers CRX 75w90 NT which are described as for "Manual transmissions, synchromesh ... which require API GL4 or GL5 performance"

My gtst has nismo bushes in the standard fucas. This is just an attempt to fit something that could have longer life than the standard bushes which wear out fairly fast. There isn't any adjustment for camber with just the nismo bushes but the drop in ride height from the teins adds a useful little bit of negative camber, which was a good enough for my needs. The whiteline tension rod bushes have an eccentric adjuster to add a little bit of caster. Nismo also do a kit for the gtr which moves the holes in the inner mount bracket of the fucas. Which helps a bit with the bad geometry. That bracket would work on a gtst. I can't remember if the rest of the kit is gtr specific. I think it is as it also has a longer lower arm iirc. Sounds like gtsboy had a particularly bad time with poly bushes, I didn't, but that was with very different conditions and usage, and your situation will be more like his than mine. I'm not sure you need adjustable fucas for you application anyway tbh.