Part 4 - D-Day….Dyno Time
We’re now the Wednesday before Christmas. Amanda and Troy at Northampton Motorsport had very kindly juggled things to get the car on the dyno for an afternoon slot before the festive break. An early morning departure from home to get up there in good time to again remove the rear bumper and prep the car for strapping down.
The clients that had booked the dyno for the morning actually overran a little (they actually had a full day booking but *usually* only needed a few hours to systems check each new build car going on there). We didn’t get the car strapped down until about 15.30, so to say I was getting a little twitchy was an understatement. However, Troy had said that he was fully prepared to work late into the evening (now that his facility is ultra quiet and not upsetting neighbours). Fingers were cable-tied crossed that we’d have a straightforward session.
The first thing I noticed when looking over Troys shoulder, was just how sensitive the new BMW eThrottle pedal is. Just resting your foot lightly on it, you can see a voltage change in SCal (Syvecs' mapping s/w). In retrospect it’s little wonder the old mechanical setup was feeling like it was disconnected; that same level of foot pressure wouldn’t have amounted to even taking up the cable slack, nevermind registering a reading in the calibration.
The injectors dialled in very easily. In fact, it was all pretty uneventful. Drawing comparisons to the previous Injector Dynamics 1050’s, what we found previously was that the duty cycle on such a large injector was sub-optimal in the opinion of both Fearnsport and Northampton Motorsport. If memory serves it was about the 4-5% mark at the lowest end of their operating window, the suspicion being that very light throttle openings were being compromised with too much fuel delivery, even at this low duty cycle. But, what I would say in the balance of fairness is that we were also battling the throttle pedal. As with any well conducted scientific test, only change one thing at a time. In the real world, with limited time and (frankly) budget, we had to change both the injectors and pedal, see how it all performed and set it up as well as possible.
Something that was manifesting itself was a driveability area at the 1700-2200 rpm range, with quite a significant stuttering when going from steady state and then applying light throttle, e.g., the amount of throttle you'd use in heavy traffic to keep pace, or when “politely” overtaking something. It’s actually something that bothered Troy more than it did me, I was more concerned with the ham-fisted and clumsy pulling away from a standstill. What Troy was finding was that he could make changes to cater for the light throttle issue but that would compromise heavier throttle loads, and vice-versa. I’m probably doing a very poor job of conveying what Troy and Alex were struggling with, I’m no calibration expert so please don’t go critiquing what I’m writing here, it’s merely a recollection rather than verbatim testimony. Would we have solved the issue just with a change of the throttle? Maybe, but doubtful. Collective thinking was that the midrange stutter wasn’t pedal related.
Some years ago it was met with a healthy dose of scepticism that I had chosen a 1050cc injector. Not doubting that there are people like Scott Winders getting the results they're after, and I'm certainly not going to enter into a "debate" over it. There's an infinite amount of variables with these projects, different combinations of parts that WILL change how engine behaves, we all know this. These guys between them look after some immensely impressive and highly tuned toys for some very demanding customers…think 918's (not a typo), CGT's, F40 Evo's, all flavours of very pedigreed RSR's, ex Le Mans 993 GT Evos, 'Ring record holding 911's, etc... so I'm not about to second guess their experience and knowledge. They're the ones in the driving seat and calibrating this thing; if they felt that swapping the injectors would yield better results then I’m going along with it.
What I will also say is that we're now at the top-end of the operating spectrum of the new Bosch injector, and on WOT we're probably beyond the comfortable duty cycle. You only find this by spending time (and money) on the dyno. Not a problem for us with a road car. Any more than a few seconds of using WOT and it gets silly on the UK roads. Very seriously, jail sentences are a serious consideration if you want to use more than a few seconds of WOT. If this was a track car, or pulling a 30 seconds at time of WOT down the Mulsanne, then we'd need to go for a half-way house option and reduce that duty cycle. But the Syvecs is far more tolerant (better injector drivers) than a lot of ECU's (e.g my previous MBE system which would've popped the drivers very easily). Finding out how these current injectors performed was all done 3 days before Xmas; literally no more time to start experimenting with injectors at this point, that's for another day. My personal thoughts are that a 750cc injector would be the Goldilocks option. Maybe with a future exhaust system change I’ll add a set of slightly larger injectors and see what happens.
So…..did we actually eliminate that light throttle driveability stutter? I’d say it’s 90% gone. Is it something that bothers me? No. Lucy and I have now been out for a couple of long drives, mixed driving down some motorway, fast and (very) slow country lanes. It’s there still.....but if I didn’t point it out, you’d likely never notice.
Once Troy knew what to look for after using the car as his daily, he spent a lot of dyno time refining the calibration. It did immediately feel better but not entirely gone. I briefly mentioned Alex above; I’d not met Alex before on my previous dyno sessions…no idea how, but there you go. Great guy, who Troy speaks extremely highly of. Young and sodding clever when it comes to calibration and creative ideas. Together they decided to focus (or refine?) PID techniques (Proportional Integral Derivative) to control that particular rev range tighter. Upping the Lambda sampling rate to sharpen up the fuelling. Again, not being a calibration guy, this was all starting to get way over my head. I understood the principles that Alex was telling me, but his propeller was clearly spinning hard, so I let them crack on and see what they could do.
Results seemed to come thick and fast at this point, Troy and Alex looking far happier with things as the evening progressed. Next job was to do some full power runs. The Syvecs does have a very sensitive approach to knock detection, but with some fine tuning (teaching) and pushing the limits a little more than we did 2yrs (although still very safe) and adding a little more ignition timing, we netted a nice gain in both hp and torque, right across the rev range.
Couple of videos to break the text up...you've seen similar before....note the smart new tiled floor, done as part of their refurb work...means we no longer need to use heatproof matting to protect the old painted floor from my screamer pipes!
Apologies for the audio attenuation...was just quickly shot on the iPhone. Gives you an idea anyway.
As you can see below, the red line is from the last session back in 2019, the black was this session. Headline peak numbers are +12bhp and +18lb/ft, but better still is the midrange increases of about +26lb/ft at 3,000-4,000. Peak hp is now 538.1, and peak torque 516.4lb/ft. That should be "adequate"
At this point it’s getting late and time to call it a day. Troy and Alex are happy with things, notwithstanding a real world test drive being satisfactory. A long drive home for me beckoned (in the M2C, not in the Porsche) with a plan to come back up the next day to do that test drive and all being well, to drive it home!