The RB26 drives its oil pump off a flat on the nose of the crankshaft, and on early cranks that flat is about 8mm wide. Nearly everything you have heard about RB26 bottom ends letting go at high rpm traces back to those two missing millimetres. Here is what actually fails, how to work out which crank is in your block without splitting the motor, and what the three real fixes cost here.
What Actually Lets Go Down There
The Collar Is Doing More Work Than You Think
The RB26 drives its oil pump straight off the nose of the crankshaft. No chain, no gear train, no intermediate shaft – the inner rotor sits on a flat collar machined into the crank snout, and that collar is the only thing turning the pump.
The problem is the width of that collar on early cranks. R32 GT-Rs and early R33s got what everyone calls the short nose crank, where the drive collar is roughly 2mm narrower than the one Nissan fitted from late 1995 onward.
Two millimetres does not sound like much, but spread across a face transmitting every torsional pulse from six cylinders into a hardened rotor, it is the difference between a part that lives and one that wears itself round.
Once the collar rounds off or the rotor cracks, drive stops. Not degrades – stops.
Harmonics, Not Just Revs
The lazy version of this story is that it breaks over 8000rpm. What kills the drive is torsional vibration – the crank does not spin at a constant speed, it winds and unwinds thousands of times a second as each cylinder fires, and the pump rotor sits on the very end of that whip.
Big cams, high boost, a light flywheel and a tired harmonic damper all push more of that energy into the nose. Which is why plenty of standard 400hp R32s do 200,000km on the original pump, and why a 600hp car with a 9000rpm limiter and a solid crank pulley can chew a drive in one season of club days.
Rev range is a proxy for the real variable, not the cause. If your build sheet has aggressive cams and a hardened damper, treat the collar as a wear item.
The Failure Nobody Gets a Warning For
Bearing failures usually give you something first – a knock on overrun, pressure that sags when hot, glitter in the filter. Pump drive failure gives you nothing. The rotor is fine at 7500rpm and in pieces at 7900, and the gauge goes to zero in roughly the same second the bearings do.
We have seen the aftermath twice at Ruapuna and it looked identical both times: full pressure on the out lap, a puff of smoke down the back straight, and a block that is scrap because a rod came through the side of it. The bill is not the pump. It is the block, the crank, and whatever the debris did to the turbos on the way past.
That asymmetry is the whole argument for doing this preventatively.
Working Out What You Have Without Pulling the Sump
Reading the Crank Snout From the Front
You do not need to split the motor to find out which crank you have. Pull the crank pulley and the front cover and the collar is right there.
The measurement that matters is the width of the drive flat on the snout, taken with verniers. The early short nose crank sits around 8mm across the collar; the later crank is closer to 10mm. If you would rather not pull the front cover yet, build date and chassis get you most of the way – every BNR32 and every R33 built before roughly the middle of 1995 is short nose, R34s are all long nose, and late R33s are the grey area where you have to look.
Do not take the seller’s word on an imported engine either. Plenty of the RB26s that land here have been apart at least once in Japan.
Telling an N1 Pump From a Standard One
The pump itself is easier to identify. The N1 unit runs a taller rotor set and a reworked inlet, and the give-away is the casting plus the part number stamped on the body. Sit one next to a standard pump and the difference in rotor height is obvious even if you have never handled one before.
Worth being blunt about a common misunderstanding. Fitting an N1 pump to a short nose crank does not fix the collar problem. The N1 moves more oil and its rotors are stronger, but the drive interface is still the same narrow collar, and the narrow collar is the bit that fails.
An N1 pump on a long nose crank is a genuinely good combination. On a short nose crank it is a better pump bolted to the same weak point.
What the Gauge Will and Will Not Tell You
Nothing on the dash predicts this failure, which is worth saying plainly because people fit oil pressure warning lights and assume they are covered. A pressure light tells you the pump has already stopped. By the time the buzzer sounds the bottom end has been dry for a revolution or two, which at 7000rpm is more than enough.
What a gauge is genuinely useful for is spotting the conditions that accelerate the wear. Pressure that spikes hard on a cold start, especially with heavy oil in a Christchurch winter, is loading the drive at exactly the moment rotor clearances are tightest. Keeping it under 3000rpm until oil temp is off the bottom stop costs nothing.
Cold start discipline will not save a marginal collar forever. It will buy you time to plan the fix rather than react to it.
The Three Fixes That Actually Work
Long Nose Crank and an N1 Pump
The cheapest genuine fix, assuming the motor is already apart, is to put the later crank in it. The long nose crank drops into an R32 or R33 block without drama – journals and stroke are the same, so shells, rods and the rest of the rotating assembly carry straight over. Pair it with the matching pump and you have the arrangement Nissan settled on once they worked out the early one was marginal.
Here, a good used long nose crank out of an R34 or a late R33 tends to land between NZD 800 and NZD 1,500 through the JDM importers. Budget for it to be checked, polished and rebalanced, because a used crank of unknown history is not something to install on faith.
If you are importing one yourself, used engine components are a biosecurity item – MPI’s rules on imported vehicle parts mean anything carrying soil or oily residue can be held, cleaned at your cost, or turned around at the border.
Collaring the Crank You Already Have
If your crank is otherwise excellent, and plenty of them are, the second option is to keep it and fix the nose. The collar modification means machining the snout and fitting a hardened steel collar over it, giving the pump rotor a wider and harder drive face than the factory one ever had. Nitto in Australia popularised the approach, but the machining is the part that matters more than the brand on the box.
This is not a job for a general engineering shop. The collar has to run dead true to the main journals, the interference fit has to be right, and the crank has to be rebalanced afterwards.
Expect NZD 600 to NZD 1,000 for the collar work, plus balancing, and add a couple of weeks if the crank has to travel between islands.
Billet Pumps and Where the Money Stops Making Sense
The third route is a billet aftermarket pump – Nitto, Tomei and a handful of others make them – combining a stronger housing, better rotors and a wider drive on a collared crank. On a 700hp-plus build that sees the limiter in the nines, this is the sensible answer.
On a 450hp street car it is money that would do more good elsewhere. A long nose crank and an N1 pump, assembled properly, will comfortably outlive the rest of the engine at that output. The billet pump is not buying reliability you are missing, it is buying headroom you are not using.
Rough NZ pricing at the time of writing:
– N1 pump, new: NZD 1,200 to 1,600
– Good used long nose crank: NZD 800 to 1,500
– Collar modification plus balancing: NZD 800 to 1,300
– Billet aftermarket pump: NZD 2,000 to 2,800
None of those include the labour of having the motor apart, which is why doing this during a planned rebuild costs a fraction of doing it on its own.
Getting It Right on Assembly
Clearances, Pickup and Priming
The pump is only half the job. Check the rotor-to-housing clearances against the manual with feeler gauges before it goes on, even on a brand new pump – they do occasionally come out of the box at the wrong end of tolerance.
The pickup gauze and the sump baffling matter more on an RB-series engine than on most, because the standard sump is shallow and the pickup is easy to uncover under sustained cornering load. If the car sees track work, a baffled sump belongs on the same invoice as the pump.
The pump also has to be primed with assembly lube before the front cover goes back on. A dry rotor set will cavitate on first crank and can take a long time to pick up prime.
What to Check Before It Sees Load
Before the engine fires, crank it with the coils and injectors disabled until the oil pressure gauge lifts. If it has not lifted after twenty seconds, stop and find out why rather than lighting it up and hoping.
Then run it in like the crank collar is the expensive part, because it is. Keep it off boost and under 4000rpm until it has been through a few heat cycles, change the oil early, and cut the first filter open to look at the media. Metal in there is telling you something while it is still cheap to hear.
If your car carries an LVV plate for the engine or the induction, an internal rebuild on its own does not normally disturb the certification – but if the work changes anything the plate describes, talk to your certifier before the next WOF is due.
None of this is exotic engineering. It is a known weak point with three known fixes, and the only genuinely bad decision is finding out which crank you have after the fact. If the motor is coming apart for anything else, the collar is the cheapest line on the invoice – and if it is not coming apart, at least go and measure the snout so you know what you are running.
3 Comments
Measured mine last weekend after reading this – 8mm on the dot, and it’s a 93 R32 so no real surprise. Been taking it to 8000 for three years thinking the N1 pump had me covered. Crank is coming out over winter.
Fair call on billet being overkill at 450hp, but I’d still rather pay for it once than pull the motor twice.
The cold start bit deserves more attention than it usually gets. We had a 33 lose a pump drive at a Manfeild club day and the owner swore black and blue he never revved it cold, but he was blipping it on the trailer at about 4 degrees every single time. Any thoughts on whether dropping to a lighter oil over winter actually helps the drive, or just shifts the problem onto the bearings?