Improve drivability in 5 mins

[antonch]

Since I switched over to new much bigger injectors (req_fuel=5.1ms), my car had problems idling well. It would idle but was very unresponsive at 600rpm. Then I switched over to 2 squirts per cycle. Now the injectors are firing once every revolution. This is done by combining injectors in Injector Output screen and set the divider to two.

Immediately, the car started up smoother. Idle was much better and the car accelerated smoother from idle. To my surprise the lambdas didn't change at all. This tells me that the injector model is working well (TS diode+0 ms opening time). As you are squirting twice as often, the error should have doubled.

From my understanding this improved the car's response because without cam sync, depending on how the car started you were either injecting fuel when the intake valve was open (poor vaporization and mixing) or when the intake valve was closed (good). This works as a compromise. I plan on configuring cam sync to work and switch to full sequential. But for now batch works much better than semi-sequential.

[Sprocket]

Switched over to two squirts per cycle from??..

The way I see this is, the required fuel value calculated is the fuel required by one cylinder for the given size of injector and does not take into consideration the type of injector sequence being used except for sequential injection.

Leaving the size of injector alone for now and using a calculated required fuel value, if you run batch fire on a four cylinder engine with a divider of 1, the required fuel value as calculated is 4 times as big as it should be, or, your injectors are 4 times as big as they need to be, or, the VE values are 4 times less than they need to be. You are only using a quarter of the injectors capebility and idle could suffer. You could possibly sumise that if the injector was already big for the engines requirements, the error has now multiplied by 4?

With the same required fuel value, setting the divider to 2, the calculated required fuel value is twice as big as it needs to be, or, your injectors are twice as big as they need to be, or, the VE values are twice as less than they need to be. You are now using half of the injectors capebility and idle will have improved over the above.

Same Required fuel value and sequential ,divider set to 1, the calculated required fuel value is the required fuel, the injector size is as originaly sized and the VE values are representative of the injector. you are now using the full capability of the injector.

Going back to the injector size. It is important to select the correct size injector for the expected full load demand of the engine, while taking into consideration which 'type' of injection you wish to use, sequential, semi sequential, or batch. Most 'calculators' I have come across will give you the size of the injectors to use on your engine, its basicaly telling you the required fuel in cc per cylinder. If running batch with a divider of 1, the injector size could be divided by 4. If the divider is set to 2 on batch, the injector size could be divided by 2, and if running sequential, with the divider set to 1 and simply use the calculated injector size. In doing this, the injectors are matched to the engines expected full load demand.

One other thing on injector sizing, most people say to use a max injector duty cycle of 80%. Fair enough, but does this leave enough room for EGO correction or Warm Up and Acceleraion enrichments for it to be safe? I know that Rover rated their injectors at roughly 60% duty. It may be possible to have a 10% EGOc with 5% warm up enrichment and 5% acceleration enrichment at the same time?? theres the 80%. If the original max duty was 80%, you can see that you might already have maxed out the injector.


Its all retrospective, as you can usualy tune the system by changing many settings, the problems arise when you use injectors sized way too big for the engine in the first place. Really big injectors dont like really small pulse widths.

I read that the idle PW needs to be in the region of 2ms for a decent idle quality, fair enough, and probibly a good rule of thumb. I know you can get the engine to idle reasonably well with PW of 1.2ms, but there must come a point where the injector just doesnt like anything less (ramp up time?), resulting in poor idle and 1.2ms must be getting close.

LOL, waffled on for too long and probibly got it all wrong, bit of a brain dump im afraid

Please discuss, and sorry to have hijacked your topic

[antonch]

Switched over to two squirts per cycle from??..

From the standard semi-sequential 1 squirt per cycle (once every 720*).

The way I see this is, the required fuel value calculated is the fuel required by one cylinder for the given size of injector and does not take into consideration the type of injector sequence being used except for sequential injection.

Leaving the size of injector alone for now and using a calculated required fuel value, if you run batch fire on a four cylinder engine with a divider of 1, the required fuel value as calculated is 4 times as big as it should be, or, your injectors are 4 times as big as they need to be, or, the VE values are 4 times less than they need to be. You are only using a quarter of the injectors capebility and idle could suffer. You could possibly sumise that if the injector was already big for the engines requirements, the error has now multiplied by 4?

With the same required fuel value, setting the divider to 2, the calculated required fuel value is twice as big as it needs to be, or, your injectors are twice as big as they need to be, or, the VE values are twice as less than they need to be. You are now using half of the injectors capebility and idle will have improved over the above.

Req_fuel assumes full sequential operation. If you run batch you would inject twice as much fuel. A very nice touch from VEMS is divider feature. Change it from 1 to 2 and it effectively changes your req_fuel.

Same Required fuel value and sequential ,divider set to 1, the calculated required fuel value is the required fuel, the injector size is as originaly sized and the VE values are representative of the injector. you are now using the full capability of the injector.

After about 50% duty cycle it really doesn't matter if you are running batch or sequential.

Going back to the injector size. It is important to select the correct size injector for the expected full load demand of the engine, while taking into consideration which 'type' of injection you wish to use, sequential, semi sequential, or batch. Most 'calculators' I have come across will give you the size of the injectors to use on your engine, its basicaly telling you the required fuel in cc per cylinder. If running batch with a divider of 1, the injector size could be divided by 4. If the divider is set to 2 on batch, the injector size could be divided by 2, and if running sequential, with the divider set to 1 and simply use the calculated injector size. In doing this, the injectors are matched to the engines expected full load demand.

You are right most calculators are very crude. They don't take injection sequence nor opening time into account. But again most calculators are rated for HP instead of fuel flow so you are guessing the BSFC number anyway. I've seen one very nice calculator that takes injector opening time into account. 1ms at 7000rpm can be a significant part of duty cycle.

One other thing on injector sizing, most people say to use a max injector duty cycle of 80%. Fair enough, but does this leave enough room for EGO correction or Warm Up and Acceleraion enrichments for it to be safe? I know that Rover rated their injectors at roughly 60% duty. It may be possible to have a 10% EGOc with 5% warm up enrichment and 5% acceleration enrichment at the same time?? theres the 80%. If the original max duty was 80%, you can see that you might already have maxed out the injector.

Do you push your car when its cold and Warm Up enrichments are enabled? 

Its all retrospective, as you can usualy tune the system by changing many settings, the problems arise when you use injectors sized way too big for the engine in the first place. Really big injectors dont like really small pulse widths.

I read that the idle PW needs to be in the region of 2ms for a decent idle quality, fair enough, and probibly a good rule of thumb. I know you can get the engine to idle reasonably well with PW of 1.2ms, but there must come a point where the injector just doesnt like anything less (ramp up time?), resulting in poor idle and 1.2ms must be getting close.

I heard that 2ms figure somewhere else too. But I think the newer injectors are much better and more responsive than older kinds. My high-z injectors idled at ~2ms. After I switched to batch its about ~1ms. My AFRS haven't changed which tells me that injector model is working well. By the way, don't forget that there is injector closing time which counter-effects open time. So open time in the model is really open time - close time.

LOL, waffled on for too long and probibly got it all wrong, bit of a brain dump im afraid

Please discuss, and sorry to have hijacked your topic

Not offended at all, I like discussing technical stuff:)

[rob@vems.co.uk]

The idle PW is dependent on the size of the injectors the engine and its volumetric efficiency at idle.

[Jorgen]

Hi,

It is interesting that you see a benefit of running the injectors in batch fire mode. I have yet to see this.

There is however a small region where you don't want to inject at low loads, actually two spots a few degrees apart. The engine will be very rough when idling with that injector timing, adding a second squirt 360deg off will greatly improve idling if the injector timing is set up like this.

Maybe we could see something like this if we have problems with fuel rail pulses as well.

Jörgen