Injecting on open intake valve?

Tuning concepts, methods, tips etc.

Moderators: jsmcortina, muythaibxr

Injecting on open intake valve?

Postby TimoL » Sun Oct 29, 2017 5:07 am

Lately I've been experimenting on injection timing and so far the best results have been with the injection ending at 260 deg BTDC.
I got this result by looking at the AFR (when it's richest with the same PW) and throttle response.
The default value for timing was actually pretty close, 270.
To my understanding this when the intake valve should be open.

But this all goes against what I've been told. I've been doing some reading on the topic and generally best results are found when injecting against closed intake valve.

Here's a snapshot from the MAP logger with my observations:
MAP_vs_toothNo.png
MAP_vs_toothNo.png (111.81 KiB) Viewed 613 times


This all got me thinking if I should try bigger injector to hit the "injection window" better at higher loads.
Currently I'm using quite smallish one with the max recorded duty cycle of 82%.

The setup (same as as in the sig):
Microsquirt v3, FW version is Extra 3.4.2
670cc single cylinder 4-stroke
Flex fuel, mainly used with E85
Sequential fuel & spark, MAP phase detection
Honda NX650 Dominator - MS2 Extra, pure Alpha-N, seq fuel & spark, tuned for flexfuel
Subaru Leone 87 1.8T, MS2 Extra - project starting
TimoL
Helpful MS/Extra'er
 
Posts: 48
Joined: Tue Aug 02, 2016 9:16 am
Location: Finland

Re: Injecting on open intake valve?

Postby Dennis_Zx7r » Sun Oct 29, 2017 8:37 am

I also did some experimenting with this some weeks ago. My main thing to take away from this was that injection timing had no measurable effect whatsoever on my engine during idle that couldn't be attributed to different differential fuel pressures and varying deadtimes. I'm therefore of the opinion that adjusting injection timing for AFR while idling probably is of no real value on setups with ITBs or other intake systems where fuel pressure differential can't really be kept constant.
On the left you can see the AFR, corrected for EGOcorr. Injector timing on X-axis, with a Map(t) signal on the right as a reference (Ignore Z axis). Notice the correlation, absolute FP is fixed.

InjTiming_BenzDruckDiff.png
InjTiming_BenzDruckDiff.png (61.23 KiB) Viewed 592 times
My build (german, page 5+): Link
User avatar
Dennis_Zx7r
Experienced MS/Extra'er
 
Posts: 210
Joined: Thu May 26, 2016 1:25 pm
Location: Germany

Re: Injecting on open intake valve?

Postby TimoL » Sun Oct 29, 2017 12:11 pm

Thank you for sharing your results. I remember having the same kind of AFRs with different injection timings, maybe a little less variation. But the responsiveness and general "feel" of the engine was highly dependent on the injection timing.

When I was running untimed injection I used to play with the Accel Enrichment settings all time but it never felt quite right. After setting the inj angle to 260 I could disable the enrichment all together and the throttle response was way better than ever before.
Honda NX650 Dominator - MS2 Extra, pure Alpha-N, seq fuel & spark, tuned for flexfuel
Subaru Leone 87 1.8T, MS2 Extra - project starting
TimoL
Helpful MS/Extra'er
 
Posts: 48
Joined: Tue Aug 02, 2016 9:16 am
Location: Finland

Re: Injecting on open intake valve?

Postby PaulS » Sun Oct 29, 2017 1:46 pm

I've done a lot of work with injection timing on the siamese port A-Series engines. The only way to get fuel into the outer cylinders is through an open valve, otherwise it goes to the inner cylinder on the same port. Hence, I have a handle on whats needed in terms of injection timing.

I think that you have not taken into account the transit time in the port. Generally this is about 25-30 degrees of engine rotation for one inch distance from injector to valve. Could it be that you are injecting whilst the valve is open but it has closed by the time the fuel gets there?

I'm always intrigued by the term "best" to describe the results of any testing without clarification of what you are actually trying to achieve. In your case you are getting a richer AFR for the same pulse width. This means, assuming that the fuel amount remains the same, that you are getting less air into the cylinder. If you inject on a clsoed valve, the fuel has time to vapourise and this will reduce the air flow as the fuel expands. Best for emmissions and response maybe, but worse for power.
PaulS
Helpful MS/Extra'er
 
Posts: 81
Joined: Mon Dec 31, 2007 10:57 am
Location: Northants UK

Re: Injecting on open intake valve?

Postby BigBlockMopar » Mon Oct 30, 2017 2:33 am

TimoL,
Are you also taking into account that an AFR reading is a 'late' result which just happens only after the exhaust valve has opened?
The time from Intake Opening AND injection to Exhaust Opening, and to the moment the O2-sensor is actually able to read the result should be considered.
Daily driver: 1973 Dodge Dart - 318ci engine / MS3x - ignition only. A518-OD transmission / 3.55 gears
-+-+-+-+-
Website: https://www.bigblockmopar.nl/
YouTube Channel: https://www.youtube.com/c/BigBlockMopar
User avatar
BigBlockMopar
Experienced MS/Extra'er
 
Posts: 181
Joined: Sat Dec 31, 2011 6:10 am
Location: Netherlands

Re: Injecting on open intake valve?

Postby TimoL » Mon Oct 30, 2017 9:02 am

PaulS wrote:I've done a lot of work with injection timing on the siamese port A-Series engines. The only way to get fuel into the outer cylinders is through an open valve, otherwise it goes to the inner cylinder on the same port. Hence, I have a handle on whats needed in terms of injection timing.

I think that you have not taken into account the transit time in the port. Generally this is about 25-30 degrees of engine rotation for one inch distance from injector to valve. Could it be that you are injecting whilst the valve is open but it has closed by the time the fuel gets there?

I'm always intrigued by the term "best" to describe the results of any testing without clarification of what you are actually trying to achieve. In your case you are getting a richer AFR for the same pulse width. This means, assuming that the fuel amount remains the same, that you are getting less air into the cylinder. If you inject on a clsoed valve, the fuel has time to vapourise and this will reduce the air flow as the fuel expands. Best for emmissions and response maybe, but worse for power.


Very good point on the travel time. I completely missed that. The injector is located approximately 20cm (~8Inch) from the intake valve. I guess the fuel would indeed hit the closed valve.

What lead me to believe that injection must be happening on the open valve is that even small change on the injection timing seemed to have big effect (on responsiveness). Therefore, the optimal injection window must be quite narrow, such as the intake being open.

Unfortunately I don't have any logs/statistical data to support this claim for "best" injection timing but let me assure you, the difference in throttle response is huge when changing the timing only by like 30 degrees. The AFR just happens to be the richest when the responsiveness is best. For my application the throttle response and low speed torque are number one priorities. Peak power is really of no concern.

I forgot to mention that the entire fuel spray is not directly hitting the intake valve (due to line of sight issues and limits on injector placement). Maybe 50 to 75% of it does. Approximately half of the remaining spray hits the inner curve of the port wall on a very low angle, thus bouncing and hitting the intake valve. At least that was the behavior when testing the injector placement with a spare cylinder head without any airflow in the port.
Honda NX650 Dominator - MS2 Extra, pure Alpha-N, seq fuel & spark, tuned for flexfuel
Subaru Leone 87 1.8T, MS2 Extra - project starting
TimoL
Helpful MS/Extra'er
 
Posts: 48
Joined: Tue Aug 02, 2016 9:16 am
Location: Finland

Re: Injecting on open intake valve?

Postby TimoL » Mon Oct 30, 2017 9:15 am

BigBlockMopar wrote:TimoL,
Are you also taking into account that an AFR reading is a 'late' result which just happens only after the exhaust valve has opened?
The time from Intake Opening AND injection to Exhaust Opening, and to the moment the O2-sensor is actually able to read the result should be considered.


Sorry, but I don't see how the "latency" of the AFR reading since I was adjusting injection timing and taking the avg AFR reading from about 1 min duration. Can you enlighten me on this?

But as on the AFR being the deciding factor on determining the optimal timing, that's not really the case here. Most likely I would have arrived to the same result just by comparing the throttle response between timings. The difference is quite noticeable.
Honda NX650 Dominator - MS2 Extra, pure Alpha-N, seq fuel & spark, tuned for flexfuel
Subaru Leone 87 1.8T, MS2 Extra - project starting
TimoL
Helpful MS/Extra'er
 
Posts: 48
Joined: Tue Aug 02, 2016 9:16 am
Location: Finland

Re: Injecting on open intake valve?

Postby PaulS » Mon Oct 30, 2017 10:11 am

With your injector placement at 8" from the valve you probably have a full cylinders volume of air in the port between the injector and the valve. Hence even if you wanted to, you cannot inject through the open valve.

What's most likly happening is that you are injecting into moving air that then stops moving as the inlet valve closes, the fuel sits in the port and vapourises before moving into the cylinder on the next inlet valve event.
PaulS
Helpful MS/Extra'er
 
Posts: 81
Joined: Mon Dec 31, 2007 10:57 am
Location: Northants UK

Re: Injecting on open intake valve?

Postby TimoL » Mon Oct 30, 2017 11:26 am

PaulS wrote:With your injector placement at 8" from the valve you probably have a full cylinders volume of air in the port between the injector and the valve. Hence even if you wanted to, you cannot inject through the open valve.

What's most likly happening is that you are injecting into moving air that then stops moving as the inlet valve closes, the fuel sits in the port and vapourises before moving into the cylinder on the next inlet valve event.


I'm still puzzled by that narrow injection window required for good responsiveness. I guess the moving air & low pressure could play big part in this. To rephrase the original question: maybe the key is to inject during the intake valve event, not necessarily injecting through the open valve. This again makes me wonder if I should experiment with bigger injectors? Price for one injector is not that high anyways.
Honda NX650 Dominator - MS2 Extra, pure Alpha-N, seq fuel & spark, tuned for flexfuel
Subaru Leone 87 1.8T, MS2 Extra - project starting
TimoL
Helpful MS/Extra'er
 
Posts: 48
Joined: Tue Aug 02, 2016 9:16 am
Location: Finland

Re: Injecting on open intake valve?

Postby Dennis_Zx7r » Mon Oct 30, 2017 11:50 am

TimoL wrote:Unfortunately I don't have any logs/statistical data to support this claim for "best" injection timing but let me assure you, the difference in throttle response is huge when changing the timing only by like 30 degrees. The AFR just happens to be the richest when the responsiveness is best.

I think this may be more due to the fact that a richer mixture is less sensitive to throttle blips. Is this still the case after adjusting the tables?
My build (german, page 5+): Link
User avatar
Dennis_Zx7r
Experienced MS/Extra'er
 
Posts: 210
Joined: Thu May 26, 2016 1:25 pm
Location: Germany

Re: Injecting on open intake valve?

Postby mrx79 » Mon Oct 30, 2017 12:40 pm

Dennis_Zx7r wrote:I think this may be more due to the fact that a richer mixture is less sensitive to throttle blips. Is this still the case after adjusting the tables?

Interesting idea, especialy for dynamic... it would be interesting to know if the responsiveness is the same with different SOI's but constant AFRs.

I would also throw the time of the MAP sampling for the PW calculation also into the equation.
The later your SOI is, the higher the chance to get a later/higher MAP (only applies to dynamic situations and also only if you are running speed-density and not 100% alpha-n).
Honda CRX B16A1 Turbo | MS3 in construction | pre1.5.1 beta7 Firmware | 24/1 Dual Wheel | COP ignition | 725cc ID injectors
mrx79
Experienced MS/Extra'er
 
Posts: 273
Joined: Wed Oct 29, 2008 8:26 am
Location: Frankfurt / Germany

Re: Injecting on open intake valve?

Postby TimoL » Mon Oct 30, 2017 12:44 pm

Dennis_Zx7r wrote:
TimoL wrote:Unfortunately I don't have any logs/statistical data to support this claim for "best" injection timing but let me assure you, the difference in throttle response is huge when changing the timing only by like 30 degrees. The AFR just happens to be the richest when the responsiveness is best.

I think this may be more due to the fact that a richer mixture is less sensitive to throttle blips. Is this still the case after adjusting the tables?


I retuned the tables after changing the timing so result AFR is same as before. Besides, I tried richer idle mixtures and various accel enrichment settings long before changing to sequential. Nothing I ever came up with was even remotely close to as it is now with seq.
Honda NX650 Dominator - MS2 Extra, pure Alpha-N, seq fuel & spark, tuned for flexfuel
Subaru Leone 87 1.8T, MS2 Extra - project starting
TimoL
Helpful MS/Extra'er
 
Posts: 48
Joined: Tue Aug 02, 2016 9:16 am
Location: Finland

Re: Injecting on open intake valve?

Postby TimoL » Mon Oct 30, 2017 12:51 pm

And as stated in the sig, I'm running pure Alpha-N. Using MAP signal as a fueling factor (whether it was SD, ITB or AN with multiply MAP) totally killed the throttle response. I spent about 3 months with each tune and finally settled with pure AN.
Honda NX650 Dominator - MS2 Extra, pure Alpha-N, seq fuel & spark, tuned for flexfuel
Subaru Leone 87 1.8T, MS2 Extra - project starting
TimoL
Helpful MS/Extra'er
 
Posts: 48
Joined: Tue Aug 02, 2016 9:16 am
Location: Finland

Re: Injecting on open intake valve?

Postby whittlebeast » Sat Nov 04, 2017 3:02 am

I did a test on this and posted the method and results on my Youtube channel. It got about 700 views in the first 12 hours.

https://www.youtube.com/watch?v=DLcsaLWKZMk

Andy
User avatar
whittlebeast
Super MS/Extra'er
 
Posts: 2151
Joined: Tue May 04, 2004 8:20 pm
Location: St Louis

Re: Injecting on open intake valve?

Postby TimoL » Sat Nov 04, 2017 8:39 am

whittlebeast wrote:I did a test on this and posted the method and results on my Youtube channel. It got about 700 views in the first 12 hours.

https://www.youtube.com/watch?v=DLcsaLWKZMk

Andy


Thanks for the link, the video was very informative. I've also checked your video on accel enrichment.
I would really like to see some analysis on injection timing vs transient behavior. At the moment I'm unable to do any testing my self as the engine rebuild is not finished.
Honda NX650 Dominator - MS2 Extra, pure Alpha-N, seq fuel & spark, tuned for flexfuel
Subaru Leone 87 1.8T, MS2 Extra - project starting
TimoL
Helpful MS/Extra'er
 
Posts: 48
Joined: Tue Aug 02, 2016 9:16 am
Location: Finland

Re: Injecting on open intake valve?

Postby Perkele » Fri Nov 17, 2017 2:31 am

whittlebeast wrote:I did a test on this and posted the method and results on my Youtube channel. It got about 700 views in the first 12 hours.
https://www.youtube.com/watch?v=DLcsaLWKZMk
Andy

Great and informative video, but doesn't MS use injection timing number as BTDC so bigger numbers makes injection earlier oppose to the image in your video?
Or have I understood something wrong?

And I take it that you used start of injection on you testings?
Perkele
Helpful MS/Extra'er
 
Posts: 131
Joined: Tue Feb 17, 2009 11:15 am
Location: Finland

Re: Injecting on open intake valve?

Postby TimoL » Fri Nov 17, 2017 4:59 am

Perkele wrote:
whittlebeast wrote:I did a test on this and posted the method and results on my Youtube channel. It got about 700 views in the first 12 hours.
https://www.youtube.com/watch?v=DLcsaLWKZMk
Andy

Great and informative video, but doesn't MS use injection timing number as BTDC so bigger numbers makes injection earlier oppose to the image in your video?
Or have I understood something wrong?

And I take it that you used start of injection on you testings?


I've also assumed that the injection timing is given as degrees BTDC, just like the missing tooth location. I guess I will have to verify this with an o-scope when I've finished the rebuild.
Honda NX650 Dominator - MS2 Extra, pure Alpha-N, seq fuel & spark, tuned for flexfuel
Subaru Leone 87 1.8T, MS2 Extra - project starting
TimoL
Helpful MS/Extra'er
 
Posts: 48
Joined: Tue Aug 02, 2016 9:16 am
Location: Finland

Re: Injecting on open intake valve?

Postby racingmini_mtl » Fri Nov 17, 2017 8:10 am

Injection timing numbers are definitely BTDC. That's the same as ignition timing. So higher numbers mean earlier injection.

Jean
jbperf.com Main site . . . . . . . . . . . . . . . . . . . . . . jbperf.com Forum
Image
racingmini_mtl
Super MS/Extra'er
 
Posts: 8012
Joined: Sun May 02, 2004 6:51 am
Location: Quebec, Canada

Re: Injecting on open intake valve?

Postby TimoL » Fri Nov 17, 2017 8:30 am

racingmini_mtl wrote:Injection timing numbers are definitely BTDC. That's the same as ignition timing. So higher numbers mean earlier injection.

Jean


Ok, thank you for clearing this up.

-T
Honda NX650 Dominator - MS2 Extra, pure Alpha-N, seq fuel & spark, tuned for flexfuel
Subaru Leone 87 1.8T, MS2 Extra - project starting
TimoL
Helpful MS/Extra'er
 
Posts: 48
Joined: Tue Aug 02, 2016 9:16 am
Location: Finland


Return to Tuning

Who is online

Users browsing this forum: Bing [Bot], LAV1000 and 3 guests