Quote:
Originally Posted by AeroMcAeroFace
Julian isn't here to defend this, throttle stop testing is like coast down testing, susceptible to environmental effects. Doing multiple consecutive A-B runs a few minutes apart will mean that the wind is likely to be the same for all tests or at least averaged.
So from your data, we can say that multiple consecutive A-B runs on a warmed up car has no issues other than pumping losses.
Pumping losses depend on the engine, a six cylinder has virtually no pumping losses due to constant crankcase volume. But even if the pumping loss increase is there, it is a very small, virtually insignificant amount increase of the total drag and even then you need to have a reduction in aero/rolling drag to get the engine speed to increase. So it may not be accurate to 1% but I think it is more reliable than coast down tests.
|
1) I've identified eleven ( 11 ) unknowns.
2) Pre-conditioning the test vehicle to achieve thermal equilibrium at any given ambient temperature will certainly whittle away at this list.
------------------------------------------------------------------------------------
Beyond the warmup, the following issues remain:
3) Asynchronous A/F ratio optimization signals are rendered impossible. The BSFC can drift.
4) Asynchrounous ignition timing optimization ( best torque spark advance ) signals are rendered impossible. BSFC can drift.
5) Asynchronous EFI frequency and duration optimization signals are rendered impossible, due to drifting Oxygen sensor voltage signals being over-ridden by a non-participating TPS prompt. The ECU may default to 'Open-Loop', power-enrichment without veto from the TPS.
6) Any presumption as to 'constant torque' is a presumption.
7) Powertrain mechanical efficiency is predicated upon ' transferred power' and is not a constant.
8) No provision for knowing vehicle performance at velocity-2, is available for comparison the the baseline velocity-1. They're all unknown quantities.
* Rolling resistance road-load power
* Engine accessory loads which vary arithmetically with velocity
* Pumping-losses are non-linear, having to do with hydrodynamic, tribological losses, which vary with lubrication viscosity, as the square of engine rpm, oil pump, and with water pump hydrodynamic drag, which also varies geometrically. If say, the Insight develops it's maximum Bhp @ 6,000 rpm, and we choose 2400-rpm as our test velocity, even a 50-rpm increase at velocity-2 will create a 4.2% rise in pumping losses alone.
* BSFC
------------------------------------------------------------------------------------
It can be a setup for false-positives, or false-negatives.
Any 'true' effect from an aerodynamic modification could be lost in the 'noise' of the unknowns.
And we know that, typically, a smaller load-load fraction leads to a higher BSFC, even when all engine management components are 'communicating.'
-------------------------------------------------------------------------------------
Hucho tried to inform us in his 2nd-Edition, and it is he would cited Sovran's SAE Paper 830304 for those interested enough to check it out.
Why the 'select panel of experts' went into the ditch is beyond me.