Quick Fuel Injector swap and idle speed adjustment

[QuantiQualification]

Please subscribe @ https://www.youtube.com/@QuantiQualification?sub_confirmation=1

 

[m in sc]

.. actually, after you get it to fire, you can let it run a bit and it will start adjusting down the injector pulses on its own. the ecm still thinks you have 'x' injector in there and needs to correct for 'y' injector, you need about 5-10 mins of idling for it to settle down, and restart it a few times. then ride and again, any change will take 40-50 miles to basically be totally settled in w a stock ecm and a efie . after a bit, you may have to adjust the throttle stop back down again since you adjusted it.

 

[QuantiQualification]

@m in sc The 10 hole is a beast of an injector. I have around 50 miles on it so far. It definitely has enough fuel to protect the motor with the stock ecm + EFIE. I am learning some really interesting things about the ECM now that I have an operating AFR gauge running. I think 80% to Wide Open Throttle is pure open loop and also open loop below around 30% throttle with no adjustment from factory defaults regardless of what Closed loop has learned. With the Takegawa G1 at low throttle, it was running in the 17 to 1 area and right at 13.0 - 13.2 AFR upto around 80%. Above 80% closing on 55 mph, the engine started to crap out. I was thinking this was because the fuel injector couldn't put out enough fuel but it could be the ecm backs off on the fuel call as it switches to open loop. Enter the 10 hole Keihin. Now, it runs 12.5 afr at low throttle, 13.0-13.2 mid throttle and 12.5 at WOT. What gives? Would love your thoughts m in sc. Could open loop really be pure unadulterated factory with no interpolation from the closed loop mapping? Just to test the theory, I did disconnect the EFIE and tried again with around a 15 mile run. Now (after several miles of learning) mid-throttle ran in the late 13's to mid 14's afr with both top and bottom unmoving at 12.5.

P.S. I do have a PCX150 on order so will try that next.

 

[dmonkey]

Do you know the cc/min ratings on the Takegawa G1 and the Keihin 10 hole, or how did you land on trying the Keihin 10 hole for your BBK build?

 

[QuantiQualification]

Takegawa does not publish the CC rating on the G1, unfortunately. I think the 10 Hole was made for a CB1100. Not sure the cc rate on that either. I got that injector from @m in sc who graciously gave it to me. I am thinking I need an injector in between the two, which I am hoping is what the PCX150 injector is. I would love to have the inector shoot between 13.0 and 13.5 at top speed WOT. The hp takes quite a hit at the 12.5:1 it currently is running.

 

[QuantiQualification]

Do you know the cc/min ratings on the Takegawa G1 and the Keihin 10 hole, or how did you land on trying the Keihin 10 hole for your BBK build?

Figured the CC ratings out.
The Keihin is a NNJ 10-Holes / 235CC/min Flow Rate
The Takegawa is a 5HT 6-Holes / 115CC/min Flow Rate (Crap, this is a very low flow rate)
I will be getting a PCX150 with a 140CC/min Flow Rate
And . . . I will order a 8-Holes 170CC/min Flow rate as well

 

[m in sc]

12.5-12.8 is actually ideal at wot typically. put more miles on it, it will adjust. you need to be IN that rpm and throttle position to learn.

heres a screenshot of my power-commander 6 map on my cb1100, just fyi. (ignore 0% throttle i was working out some firmware issues w dynojet @ the time)

 

[QuantiQualification]

@m in sc I am becoming less and less convinced that the stock ECM actually "learns" anything. Now, having watched the behavior of the 10 hole 235cc/min injector for well over 120 miles and a 140cc/min PCX150 injector for around 100 miles. This is what the two injectors appear to do according to my narrowband (inaccurate) AFR. This behavior did not change from the beginning to the end of my testing. Below a certain speed and throttle setting the AFR would just read RICH with the 235cc/min injector which makes sense since the thing is way over need at those speeds. Beyond a certain speed and below around 80% throttle (closed loop), the 235cc/min will go to 12.5:1 anytime the throttle is shifted. It will then, relatively quickly, around 3 seconds climb up to 13.0:1. It will do this every time. At full throttle (open loop), the AFR will peg at 12.5:1 and pretty much stay there rock solid. The problem with this is that the engine feels like it is being a bit drowned and 80% throttle at 13.0:1 will actually give a higher top speed and more power feeling than wide open throttle at 12.5:1. It literally feels like you hit a wall when going to open loop.
Similarly, and yet differently, the 140cc/min injector has similar behavior but wildly different numbers. Again, this is at the beginning and through to the end of testing. Below around 80% throttle (closed loop), the 140cc/min will go to 17.0:1 anytime the throttle is shifted. It will then, relatively quickly, around 3 seconds drop to 13.0:1. It will do this every time. At full throttle (open loop), the AFR will vary between 12.8:1 at 25mph (in 4th gear) 13.3:1 at 35mph, 14.4:1 at 45mph, 13.9:1 at 50mph and 13.6:1 at 55mph. Again, it will do this EVERY time. No problem or wall feeling at wide open throttle but running on the ragged edge of too lean for my setup with heat no doubt building. Would prefer for the max AFR to hit around 13.5:1.
My theory now is that the base fuel map is FIXED for both open and closed loop (I know it sounds crazy). During closed loop, the ECM adjusted to the targeted AFR fairly quickly but it keeps relying on the base fuel map every time the throttle position and speed are shifted and then settles after a few seconds. Open loop has much more generous amounts of fuel to protect the engine, but again, does not change based on "experience".
I am probably wrong about all of this but this sucker behaves like it is not learning.
I will be throwing in a 170cc/min injector next and will report back.

 

[m in sc]

its closed loop, this has been confirmed by dhm and mmntbhx. I would take into what your afr gauge is reading with a huge grain of salt as its picking up voltage from the signal where? at the o2 wire? from the ecm? remember, the o2 sensor with the efie is lying to the ecm. the -only- way to know what its truly reading is off of an independent wideband, that's just a reality, ...buuuut maybe its the injector. even the power commanders do this. But, i do agree, that 10 hole i sent may not be able to be trimmed down enough to get you where you need, so you may need to go to an 8-hole injector, which is where i landed on the grom w the 30mm tb, 164, cam, and tyga exhaust & ported big valve (2 valve not 4). .02

 

[QuantiQualification]

Agree that it is behaving in closed loop below 80% throttle or thereabouts. I am just thinking that it looks at throttle position & rpm, looks up a fixed table, gives fueling then reads the O2 sensor and adjusts. Rinse and repeat every time the tp and rpm changes . . . and it never updates the fixed table. Same goes for open loop above 80% throttle except there is no reading the O2 sensor or adjustment from a fixed table. The EFIE device is amazing, by the way. Uses a transformer to change the voltage and it only affects voltage on the green wire period. It does not leak or suck voltage on the white wire. I confirmed this in a number of ways before connecting the narrowband AFR gauge to the output of the O2 sensor in series before the signal gets to the EFIE. I will post a video with diagram etc. on the AFR connect once I settle on a correct fuel injector. The 170cc/min is an 8 hole so might be good for my setup. I can also see why the 140cc/min is working so well for the 143cc BBK crowd. It probably is a great match. I also can now see why an actual autotune makes sense 1) More accurate AFR values, 2) Real adjustment of the table and real-time adjustments of the AFR in closed loop, and 3) ability to remain in closed loop at all throttle positions. The current setup lacks 1) very accurate AFR values, 2) only real-time adjustments of the AFR in closed loop but with a potentially static table, and 3) no O2 sensor feedback so no AFR control at high throttle positions beyond fixed table.
Strangely, after all is said and done, EFIE + Fuel Injector tuning is very reminiscent of carburetor tuning except jets instead of fuel injectors on the stock ECM.

 

[m in sc]

it adjusts up top, im 100% sure of it. if not, i would have blown mine up a looooong time ago.

I dont trust your afr gauge 1 bit. Not being disparaging, just saying.

 

[dmonkey]

My theory now is that the base fuel map is FIXED

This is technically correct. The base fuel map does not change without a reflash. The "learned" changes are stored elsewhere in non-volatile memory in a range that is usually referred to as "adaptation memory". There are limits to how much the ECU will "learn", the larger injector may have put it over a threshold of what deviation it will store from the fuel map.

 

[Shoot870p]

You guys are so far above my understanding on this! But I’m really enjoying the discussion. I may just pick up a nugget to keep in mind.
thanks and please continue on.

 

[m in sc]

dmonkey has it right. it feeds off of what the lower areas read, but still is considered closed loop since it is, in all reality, dynamic, based on input from the 02 sensor then interpreted and adjusted. if it wasn't it would have pretty much immediately completely fouled the plug at wot with that big injector. remember, i did ride my trail to work and back with the one i sent you, mines a 143 so it was too much for that, (figured it might be 'ok' w a 164, hence why i sent it) but it never fouled it. when i put the pcx150 back in, it def ran lean up top at WOT at first due to the adjustments it made from the 10 hole injector being in, but then adjusted back down in about 20 miles or so.

 

[dmonkey]

Haven't done this on the CT125 myself, but the service manual has instructions for doing a throttle position sensor reset. Might be worth doing that when you replace the injector again. The SCS connector mentioned is just a loop of wire to connect two of the plug terminals, should be green and blue if it's the same as the other miniMotos. You could DIY with wire or a paperclip of the appropriate gauge same as shorting the engine oil temperature sensor terminals.

Trail-125A-Service-Manual.pdf

View and download Trail-125A-Service-Manual.pdf on DocDroid

www.docdroid.net

Here's a video showing it, but in Japanese and they struggle with it a bit and don't do a great job of showing the check engine light flashes. Still it's something to illustrate the process.

 

[m in sc]

that wont solve the issue. the injector is too big for his setup. but its good info nonetheless.

 

[QuantiQualification]

This is great info guys. Will definitely contemplate doing a throttle position sensor update since I have a new throttle body on it versus what came stock. Thanks for the suggestion @dmonkey . I have decided to make a video that I will share here, once I get all the injectors in, showing the AFR gauge's response to different Fuel Injectors. I will start with the big guy from @m in sc and then work my way down until it is optimal. I will have a 235cc/min, a 190, a 170, a 140, a 115, and the stock 90 on hand! In the mean time, here is how I attached the AFR gauge.

Please subscribe @ https://www.youtube.com/@QuantiQualification?sub_confirmation=1

 

[JPMcGraw]

Im diggin the socks and sandals shtick and the effort you put in. Good work.

 

[QuantiQualification]

Ha ha, thanks! Got to be comfortable when working on the beast!