No-load engine testing

[WillGysi]

Hi everyone,

I'm a long-time reader of this forum (as it pops up in google whenever I search a question) but have one now that I haven't been able to directly find the answer I am looking for.

I am currently in a bit of an argument with a friend. We are building an engine with some design changes for our senior design project (mechanical engineering seniors) and will be testing our engine in comparison to the test results of a baseline engine. (ie: test engine, make changes, re-test).

I am trying to convince that we need to have a dyno of some sort to load the engine. He thinks that a no-load condition will be satisfactory.

We want to find our efficiency so determining HP,Torque, rpm, bsfc, bMEP, etc is ideal.

He wants to run the engine with no load and just see how long it takes to burn through a set amount of fuel.

What do you guys think here?

 

[SomptingGuy]

From a simulation point of view (possibly OT)...

Steady readings are fine for validating design requirements. Transient performance is much more interesting, especially with small TC engines.

- Steve

 

[140Airpower]

IMHO, Steady-state readings are the most reliable readings and are the readings that should be reported when characterizing an engine. Flash readings and inertial readings are by definition spurious and variable.
Dyno results depend on a lot of engine variables that can change as readings are being taken if the engine is not brought to equilibrium. Likewise, environmental variables contribute to dyno results and have to be accounted for. Ambient temperature, pressure and coolant temperature can be vital and can affect results in unpredictable ways if an engine is not at equilibrium and running on the edge of detonation. Merely applying compensation formulas may not give correct results.
It is generally the case that flash power readings are significantly higher than steady-state. A justification for relying on flash readings is that they may more closely model what happens in a drag race or at stop lights.

 

[WillGysi]

When speaking of steady state testing, are you guys referring to no-load or are you suggesting using something such as a variable load dyno (water brake, etc)?

 

[140Airpower]

I am referring to a loaded condition.

 

[patprimmer]

Generally steady state is the load is varied until the engine is held at steady rpm at WOT, although it might be tested with load varied to give steady rpm at steady manifold vacuum or steady air flow or even steady fuel flow to help maximise fuel efficiency at cruise.

Regards
Pat
See FAQ731-376 for tips on use of eng-tips by professional engineers &

http://eng-tips.com/market.cfm

for site rules

 

[BigClive]

I have seen a lot of cam testing done at no load (or even in "motored" engines) - but otherwise the only useful testing at no load is idle fuel consumption etc.

 

[mattsooty]

Testing at 'no load' conditions makes absolutely no sense to me.
Even when doing work at idle PAS/AC/Generator loading rigs are used, to give representative conditions ie zero clutch torque with significant accessory loads.

If you decide to continue you really must ensure that spark/air/fuelling are consistent across tests, do not leave it to the idle speed controller!

The best solution I could suggest is to get the biggest controllable loading rigs (as said above) and use these to impart a drag torque on the engine.

Ms

 

[SomptingGuy]

The only no load testing I've done that made any sense was NVH work - secondary forces from an I4 gasoline at high speeds. Scary work to be honest. Should really have work a helmet.

- Steve

 

[mattsooty]

Interesting point Steve, I wouldn't have thought of NVH working their black art in this sort of condition. I guess a motored dyno would be better?
Even still, a helmet is only going to save so much......

Another thing for me is that, bar a few exceptions, gasoline engines only have a low idle speed controller, with a torque reducing Nmax limit. Whereas a diesel has a low & high governor, a small but important difference?

MS