Transient boost pressure (turbocharged gasoline) 1

[SomptingGuy]

If I subject a gasoline engine to a step change in throttle input (almost closed to wide open) at constant engine speed, what would be the expected shape of the boost pressure vs time curve (assuming no change in wastegate position and stoichiometric operation throughout)?

My thinking is that it should be a step followed by an s-shaped curve. The step is caused instantly by the throttle openning. The gradient should then increase with time as the operating points of the compressor and turbine gradually move from the low to high efficiency parts of their maps. And then the gradient should decrease as the turbo speed settles. This gives a nice smooth s-shaped curve.

Does anyone have any experience of these sorts of transient measurements? Or any useful theories about what should happen and why?

 

[GregLocock]

Why not model it in Lotus' engine simulator.

I'd expect a massive bog if you maintained stoich throughout, in reality.

Cheers

Greg Locock

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips.

 

[SomptingGuy]

My WAVE model shows a "bog" - i.e. as S-shaped boost pressure & torque curve vs time. I was wondering if people had measurements to verify or whether it was purely a function of the quality of the turbine & compressor maps used.

 

[GregLocock]

You'd get a bog on a NA EFI car at stoke, no need to worry about turbos.

Cheers

Greg Locock

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips.

 

[Warpspeed]

If the throttle is almost closed, that presumably means no boost to start off from. Instantaneous opening suggests it will occur much faster than any appreciable engine or turbo Rpm change.

What you are left with is an almost stagnant airstream at around atmospheric pressure up stream of the throttle, and an almost stagnant downstream flow at some reasonably low depression.

Snap the throttle open, and what happens ? A sudden spike in flow and a rapid increase in induction pressure, but it will probably depend mostly on the total volume between throttle and inlet valves, and peak flow capability through that throttle.

A really good reason to run individual throttle bodies if possible.

 

[Rob45]

Don't forget that the inertia of the turbo prevents it from spooling up just as quick as one might like.

 

[NickE]

here is actual test data from my car.

2.0L 8:1 compression TD04-G (IIRC). 4th gear pull.

RPM   Th(%)  MRP(psi)  Time(BTDC)   AFR
2812    0      -11.8      11.5      23.63
2858   100       6.4      21.0      11.82
2897   100       8.1      19.0      11.82
2925   100       9.1      17.5      11.82
2955   100      10.7      16.5      11.36
2971   100      12.2      15.5      11.24
3016   100      12.9      15.0      11.24
3064   100      13.6      15.0      11.24
3082   100      13.9      15.0      11.24

Notes: Gear ratio in 4th is .973:1, Final drive 3.9:1. Average delta time between data points is 165msec. AFR reported using a narrow band (OEM) O2 sensor, IE not really accurate away from stoich.

Sorry I dont have the time to pull up data going from some throttle position that gives stoich AFR at cruise and then the changeover to WOT (wide open throttle).

I have some difficulty with your constraints:

1) constant RPM is only going to happen with a dyno that can increase load as output power/torque is increased. I dont really see that condition as having much value in the real world. And I dont know how you would test it either.

2)The AFR must go rich upon boost/WOT or else the motor will ping itself to death (at least that's what I hear).

 

[SomptingGuy]

many thanks for sharing your data. To answer your question:

1) constant RPM is only going to happen with a dyno that can increase load as output power/torque is increased. I dont really see that condition as having much value in the real world. And I dont know how you would test it either.

By constant RPM, I meant reasonably constant RPM, like you'd get in a high gear tip-in. Your's looks fairly constant - do you have an elapsed time column to go with it?

Again, many thanks.

 

[NickE]

I'll Log one on my way home from work tonite, I may be able to post from home, not sure.

You're looking for 5th gear, around 2800-3000 rpm? I'll also normalize the time collumn for ya.

 

[NickE]

SomptingGuy- Sorry that project got put on hold for a bit, I should have another log for you by next week.

 

[NickE]

2.0L 8:1 compression TD04-G (IIRC). 5th gear pull.

(msec)	RPM	Th(%)	MRP(psi)AFR	Timing
   0	2834	15.7	-1.2	14.68	33
 200	2842	98.0	 4.2	15.83	25.5
 421	2881	98.4	 6.8	11.93	20
 591	2867	98.4	 7.8	11.93	17
 761	2906	98.8 	9.3	11.59	16
 961	2904	98.8	10.4	11.47	16.5
1172	2934	99.2	11.6	11.24	15.5
1342	2931	99.2	12.6	11.24	14.5
1532	2981	99.2	13.5	11.24	14
1692	2993	99.2	13.9	11.24	13.5
1893	2992	99.2	14.1	11.24	13.5
2073	3025	99.2	13.9	11.24	14
2233	3062	99.2	13.8	11.24	14.5
2414	3064	99.6	13.5	11.24	14.5
2614	3072	99.6	13.5	11.24	14.5
2824	3079	99.2	13.5	11.24	14.5
3004	3112	99.2	13.5	11.24	14.5
3165	3133	99.6	13.6	11.24	14.5
3335	3150	99.6	13.6	11.24	14.5
3535	3185	99.6	13.8	11.24	14.5
3735	3183	67.5	13.6	11.24	14.5
3906	3190	   0	 6.2	11.24	20
4076	3182	   0	-1.7	11.47	31
4276	3183	   0	-5.7	25.35	23.5
4486	3183	   0	-7.3	23.75	13
4677	3159	   0	-8	  23.75	11.5

Notes: Gear ratio in 5th is .738:1, Final drive 3.9:1. Average delta time between data points is 187msec. AFR reported using a narrow band (OEM) O2 sensor, IE: not really accurate away from stoich.


Hope thats what you were looking for.



Nick
I love materials science!

 

[SomptingGuy]

Hope thats what you were looking for.

Absolutely! Couldn't have asked for more. Thanks again.