Buddy just got a Dyno that measures BSFC!

[theunchosen]

I kinda figured he was into racing of some sort otherwise its kind of backwards to own an engine dyno.

[tasdrouille]

Originally Posted by dremd

[...]but unfortunately it is an engine Dyno[...]

Oh well, I can't read. Where's the slap on the forehead smiley when you need it...

[cfg83]

TestDrive -

Originally Posted by TestDrive

...
It's not labeled, but I believe the line of heavy dots across the top is the BSFC curve at WOT.

Yes, this was the consensus in the other thread :

The BSFC chart thread (post 'em if you got 'em)

Originally Posted by Duffman

Yes, min BSFC should be at the same RPM as peak torque as is in this case.
I agree.

CarloSW2

[cfg83]

dremd -

Originally Posted by dremd

...

As a bonus here is the blue Supra backing up to the Dyno at Hennesey last month. That's a friends Road Race Z06 on the dyno in the background, heads, cam 56x rwhp; smooth as can be.




I'll snap some pics of the engine Dyno next time I'm over.

This is *the* car my wife wants. What are the specs on the Supra?

CarloSW2

[Deezler]

Originally Posted by jimepting

A single full power pull traces a pattern across a portion of the BSFC contour.

Correct, kind of. This contour that you speak of is just a single line across the top of the operating range on the chart, as others have noted.

Originally Posted by jimepting

...BSFC data is for full throttle and that is virtually never the situation when one is attempting to find the best fuel economy.

This is wrong. BSFC = brake specific fuel consumption. In layman's terms this means fuel consumed per unit of power produced. You still have a value for this when the engine is at part throttle. The contour plots that are often posted here are in fact values of BSFC across the entire operating range of the engine, zero to full throttle from idle to redline. Obviously this could encompass an enormous amount of data points. But generally speaking you only need to map the engine at a handful of load points (okay maybe a dozen), every 100 rpm or so. Then toss the data in excel or matlab and crank out a contour plot. If you are looking for max efficiency of your engine this is not a hard thing to do for the quality of data that you'll get.

Test drive, whether the engine in the BSFC chart you posted was at WOT is not a function of engine rpm. If you reference the top line of the data range, that is exactly what those two circle data points are, at WOT. The chart contains all data from zero load to full load across the entire rpm range of the engine. The vertical axis is engine load (torque). We need a more solid BSFC explanation thread that we can link to. Maybe I'll start drafting one.

Any type of sensor that generates a voltage or current signal can be used to measure a given parameter of engine operation... as long as you know how to interpret the data. If you have equipment to measure cylinder pressure, its an extremely valuable tuning tool, though not easy to acquire.
Obviously its a little tricky to pull an engine from a vehicle just to throw it on a dyno stand. But if you're building up a project car, repairing a bad wreck, etc, why not set it up on the dyno to shake it down? Much easier than finding problems later in the vehicle.

Congrats on the dyno acquisition, hope to see some data posted here some day.

[TestDrive]

Originally Posted by Deezler

Test drive, whether the engine in the BSFC chart you posted was at WOT is not a function of engine rpm. If you reference the top line of the data range, that is exactly what those two circle data points are, at WOT. The chart contains all data from zero load to full load across the entire rpm range of the engine. The vertical axis is engine load (torque). We need a more solid BSFC explanation thread that we can link to. Maybe I'll start drafting one.

Your explanation matches my understanding.

I'd love to have a link to cite as an authoritative source explaining exactly how a BSFC map (not curve at WOT) is produced.

And ideally several real world examples of interpreting BSFC maps to achieving maximum FE. Something like -

1. Here is the BSFC map for V engine.
2. V engine is in W car:
   • curb weight
   • Cd
   • Cda
   • Crr
   • X speed manual trans with gears y1, y2, ... yX
   • Z differential/final drive gear ratio
3. Therefore, on level ground accelerate (thusly ...) from 1st to 2nd.
4. Accelerate (thusly ...) from 2nd to 3rd.
5. ...

[Deezler]

I did calculations something like this for my engines BSFC map here:
http://ecomodder.com/forum/showthrea...t-1466-13.html

Also a post at the bottom of page 12.

I was analyzing the effect of Cd reductions and a gearing swap on expected fuel economy during 5th gear cruising. (my commute is all highway).

But your query regarding where to shift gears based upon integrated fuel consumption is very interesting. It would be much more difficult to analyze as you'd need the hard data from the BSFC map. In my case I simply recreated the plotted lines and made assumptions in between....

Maybe theres a matlab script that can take scanned data and convert to real data for a contour plot.

[MechEngVT]

TestDrive (and others),

The dark black dots and dark line across the top of the chart you posted is the WOT torque curve for the engine. The vertical axis is torque (i.e. engine load) and the horizontal axis is RPM. That's your typical full-load power pull from a dyno.

The rest of the chart is a 2D projection of a 3D map where BSFC in g/kW-hr is the "z" axis normal to the x-y of RPM-torque. In full 3D (which Excel or even better Matlab could do, but it would be hard to read) this would look something like a funnel cloud tornado with high values along the 0-load RPM range and a deep valley centered at 2500 RPM/ 125 Nm. The circles drawn on the chart are incremental elevations like a topographical map of this 3D chart and the lines would then be isoconsumption (?) lines.

To create such a chart you would have to perform multiple pulls at various loads. You would have to measure steady-state fuel flow at fixed speed/torque points. Every 500-1000 RPM would be good enough to start and it would take nearly forever. It looks like the chart posted has values every 250 rpm and 3 Nm or so.

Using the same data you could also turn this into an isoefficiency plot by calculating the indicated horsepower (IHP) from the heating values of the fuel used and comparing that with the measured power (calculated from torque and RPM) to determine the percentage efficiency. It would look similar to the BSFC islands.

[Deezler]

Originally Posted by MechEngVT

the lines would then be isoconsumption (?) lines.

To create such a chart you would have to perform multiple pulls at various loads. You would have to measure steady-state fuel flow at fixed speed/torque points. Every 500-1000 RPM would be good enough to start and it would take nearly forever. It looks like the chart posted has values every 250 rpm and 3 Nm or so.

I suppose it should be iso-specific-consumption lines as they are all referenced to unit power production..?

On a dyno it's much easier to hold rpm constant and do sweeps through the load range vs trying to hold a load and sweeping through rpm. Data points every 500 rpm or closer would really be needed for accuracy.
If you figure on 500 rpm and 12 load points per speed, with engine stabilization for about 5 minutes per point, you're talking about 11 speeds * 12 points * 5 minutes = 11 hours of data acquisition after the engine is fully warmed up! yikes. You could probably get away with engine stabilization for only 2 or 3 minutes per point though.

More sophisticated dyno software can automatically map engine performance with a set it and forget approach. Let the engine run on its own, provided it doesn't hit any safety limits....

[jime57]

Originally Posted by TestDrive

Is Quick BSFC question(geared towards folks with ATs) - Post#19 the post you're referring too?

I've seen the numerous instances of the assertion that "BSFC maps are based on WOT", but I've yet to see an authoritative citation to backup it up. I'm no expert, but I find it difficult (if not impossible) to believe that data in the 500 rpm -1000 rpm/15.6 Nm - 31.2 Nm range of this NOT ATYPICAL BSFC map/chart was generated at WOT!

It's not labeled, but I believe the line of heavy dots across the top is the BSFC curve at WOT.

I've been off the air for a while, so pardon the delay. Yes, it is post #19 that I was referencing. Your arguement is a convincing one and on additional study, I think you are correct. Sorry to add the confusion

Most of my dyno work was on an older Go Power engine dyno while doing development work on BMW 2L engines. Since the engines were for SCCA sprint racing, I really didn't have much interest in the full BSFC contour and never really tried to develop one. I sometimes found the single valued BSFC data that came out on the printouts helpful for tuning, but in sprint racing the engineer is mostly interest in the maximum power and the width of the power band.

It will be interesting to see some contour data come off the dyno.

BTW, not sure how my link ref got screwed up, but this is the correct link I meant to provide:

Browser Warning