916 Fitted With SPS Cams
Something we hadn’t done before was fit 996SPS cams in a 916 engine. This job was done with the engine out of the bike (actually an engine swap) so we pulled the heads and checked out the pistons, barrels, big end plug and gears we could see. This particular 916 engine had around 45,000km on it and is an original ’94 model P8 ECU twin pick up engine (the chassis we fitted it to was a P8 equipped chassis).
I was suspicious of the big end plug, but it was fine so we didn’t do any further bottom end disassembly. We refitted the cylinders without base gaskets and ran the old style thick head gaskets (just like my 888) which gave us 0.95 1.00mm squish and 11.8:1 compression. The SPS cams were set to 104/107 centrelines, which is inlets advanced 15.5 degrees from spec and exhausts on spec. I wasn’t sure if 104 was going to be too far advanced given the 916 is somewhat valve restricted, but the aim of this was to make max power under 10,000 RPM so I gave it a whirl. This gives timing as below:
This bike was fitted with a 50mm Termi full system and std 50mm 916 throttle bodies. We also had the option of 54mm twin injector throttle bodies but I really wasn’t too interested in the agro that goes along with the required low throttle remapping when bigger throttles are fitted. Otherwise it’s a std ’94 engine, heavy flywheel, clutch and all. The bike it’s in has enlarged air inlets in the front fairing without screens, but std filter tubes and filters.
We took the bike to the dyno and did some fuel runs then some spark runs to see what it liked. To map it properly it then went to Broadford a few days later where we remapped it on the track. There wasn’t much remapping done under 4,000 RPM, but we did lots of work above that, most of which involved leaning out the original 916 P8 eprom. It was that rich in places that it was actually bogging down, which really surprised me.
The graph below compares this bike to two other 916. One running std 45mm exhaust with Termi mufflers, std cams set to 112/108 and tuned as required. The other is an as delivered 916SP with the Termi mufflers fitted. The SP has much longer duration cams (G inlet, A exhaust), 1mm bigger valves and 45/50mm half system. It’s another Excel generated graph as that’s the easiest way to show it I generally don’t bother with a final ‘best power everywhere’ run these days, I just find out what I need to and head back to work. Red is SPS cams, blue is std cams, green is SP. This graph goes to 10,000 RPM, whereas the SP has another power peak at 11,000 RPM of around 113 Hp.
I expect the dip around 5,000 RPM for the SPS cammed bike is due to the 50mm exhaust, as most dips in that RPM range are usually exhaust related in some way. The increased compression due to not fitting base gaskets, going up to 11.8:1 from 11.3:1 with the std 0.3mm base gaskets, will affect the power output a little. On my 888 this made between 1 and 3 hp difference from 6,500 RPM upward. More between 7,000 and 8,200 and then again from 9,500 to 10,500 RPM than at the actual 9,000 RPM power peak. So given the 8 hp peak increase I’d be guessing 6 or so of that is due to the SPS cams.
The 104 inlet setting seems about right too the power peak is at 9,500 RPM, a touch higher than the std cam bike. I could see from the Dynojet air/fuel trace that the bike was going lean just before the rev limiter, and the mapping on the track confirmed this. So I had another excuse to lower the rev limit, settling on just over 10,000 RPM.
The owner was quite impressed (and surprised) with the way it went though. And especially impressed with the results of the few hours I spent remapping it. It would wheelie coming out of corners still leant over, which I think he particularly enjoyed.
The remapping was done using one of Duane’s (Ultimap) Corsetec twin lambda sensor controller box that I’d borrowed for the S2R track day and one of the data loggers Duane sells. The data log is then read and processed using some other hardware and/or software of Duane’s, creating a correction file that (using more of Duane’s software) is dumped directly into the eprom file, which is used to burn the next eprom. You can have the bike in, data log read and a new eprom fitted in less than 10 minutes, making the process quite quick if track access allows. Apart from a couple of delays unfortunately due to waiting for the ambulance to collect a fallen rider from the track it all went very smoothly.
Both the main fuel map and offset fuel map are modified at the same time when you have the two lambda probes fitted, speeding up the process and giving a real time feel for the improvement. Using the single probe Motec box I have means you have to tune the main map then move to the offset. This often means the bike will get a little worse before it gets better (if the required offsets are large), stressing out the owner who thinks it’s all going backwards.