My DIY EFI tweak

I wrote in a previous carb conversion post "My bike will stall if I ask it to pull from idle with no throttle. No throttle opening - no go. POP - stop. Turn the throttle a bit and it will lug down. Turn the idle screw in another half turn and it will mask this but it is there. Idle is idle and it does not make any power at the zero throttle point."

I believed this was deliberate leanout of the idle and off-idle fuel curve - probably to reduce fuel load to the catalytic element at idle. This probably is not true.

Yesterday I decided to misadjust my throttle position (using the hard stop) to replicate the flameout issue that is plaguing some of our members.

I first took a start point. With the engine cold and the battery held at 14.0Vdc (external power supply) my TPS wiper voltage was 995.1mVdc (black to orange/black on the throttle body electrical plug), my WFO (wide full open) voltage was 4.385Vdc. My reference voltage (orange) was 5.049Vdc

With 3 turns out on the idle screw my 09 TE510 would idle at 1850-1950. I feel this start point is a good one because I trust the Ibeat setup from the dealer, the bike idled at 1650 rpm with the 02 sensor in, and the bike acted very consistently. All my observations are with the full Power up Kit installed and no spark arrester in place.

Once the side cover was off the thottle body (the side with the cables and pulley) and I could manipulate the throttle shaft by hand a couple of things were obvious.

1) there is a flat spot right off idle. When slowly moving from idle to 3K rpm it is a blubbering point -when snapping the throttle open it is a bog-recovery spot if you start at that blubbering point. (much as if the accelerator pump shooters where too large on Holley automotive race carb)

2) by propping open the throttle idle point (I used a feeler gauges) this rich spot became wider and I could consistently replicate the flameout with the bike on the stand and in first gear. I minimized the test because I don't think my cam chain appreciated POP-bang-stop too much and it was obvious I was going the wrong way with the throttle plate. Idle at this point was about 2000 rpm with idle screw remaining at 3 turns out. Turning the screw in half a turn helped slightly and brought the rpm up to 2100 or so.

3) closing the throttle plate (by turning out the throttle plate hard stop) improved the throttle response. The action was noticeably smoother rpm increase in slow throttle movement, and quicker rpm increase with quick throttle movements. I backed out the screw until it idled at 1650 rpm with 3 turns out on the idle adjust screw. TPS wiper voltage was 949.0mVdc (hot engine) with battery 14.0Vdc.

I buttoned it up and went for a mixed ride. I deliberately fully closed/snapped open the throttle on slow uphill gnarly sections - lugged around (chain snatching the whole bit) some single track -maneuvered around at idle with no throttle (ashphalt).

The bike is snapper at low rpm and lugs about like 500cc is expected to. Backfiring on deceleration not increased. Gentle small throttle movements are more linear. Throttle action above 2500rpm is unchanged. I'd say my clutching in tight stuff is reduced 80%. Small movements (1/8 turn) of the idle speed screw makes a noticeable difference in crispness of throttle response.

One downside is the bike is more apt to bite in tight stuff - when you snap the throttle it will leap wherever it is pointing - and if there are some roots in the mix it will carry the wheel very easily. Bike reacts like it is lighter than it really is. (No joke)

=================

On the trail I backed the throttle body idle stop position screw out until it had no more tension - then in 1/4 turn. This insures the throttle plate is being held open at it's fully closed position.

The bike changed very little from before - Idle is 1650-1700rpm with the idle screw 2 7/8 turns out - starts in gear with a tap of the button - lugs around parking lot-style with no throttle - very linear slow movement throttle response, very snappy quick throttle response.

I will leave it like this until it is time for a Ibeat calibration - then I will increase the position to the original wiper voltage value before the test. I will also not use the 02 sensor at this new setting. I believe the computer takes an offset based on operating conditions with the 02 sensor in. I don't want that to happen with this out-of-specification TPS setting.

I didn't have a flameout problem - but my clutch hand is very happy with the results of this tweak. Bike is much more responsive and fun to ride.

Works for me!

MAT

Thanks for the info Mat. I will be doing some adjusting on my TPS on a couple of weeks and this info may come in handy. Keep us posted on what you find when you get hooked up to Ibeat.

Ken

Impressive work and good information. Thanks for sharing.

Thanks for the report


can you sum up your findings-
maybe I need to read this a few more times but it seems you basically changed the physical "closed throttle position" without changing or adjusting the TPS- I am unsure if you ended up opening it more than the starting point or less than your starting point- in the end it is 1/4 turn in from full closed.

Wouldn't this change you TPS reading now- and so it is out of callibration (whether improved or not) or maybe the TPS being calibrated is useless if the Physical closed is off?
So does your TPS voltage read different now?
Seams hard to replicate- but I am pretty new with EFI overall and don't understand much.

can you describe how you felt it was rich- you talk about a rich spot becoming wider- where does that come from- I am just not following where that info comes from.

Thanks!

I've played with this some, too, but not as "controlled" an experiment as you did. Most of mine was actually in the dark while whatching the header pipes change color.

I know that sounds a little crazy, but I could watch it come in and out of a lean spot pretty easy.

Here is a summary.

I played with the hard stop on the butterfly (throttle plate) of the TPS.
I took an electrical reading that represents the position I started with. I will put it back there before my next Ibeat session, or my next session with the 02 sensor in place.
I opened it further and it ran worst.
I closed it a bit and it ran better.
I closed it as much as the mechanism would allow and it ran best.
The TPS idle setpoint is presently out of specification.

These are my thoughts:

The full throttle position continues to be in specification, as is the span of values from that WFO position to the idle setpoint. (no new values have been accepted since calibration May 09, no new offset from the 02 sensor has been created since my last run with it in.)

Since all fueling and ignition maps have not changed - and the relative position of the throttle plate to that map has not changed (above the TPS idle setpoint) - no damage will occur.

Since my idle / operation continues to be above or at the stock 1650 rpm - no damage from lack of oil flow or coolant flow will occur.

No increase in engine temp or exhaust system temp is noticed - fuel economy is no worst. My need for clutching is down 80+%, and I find myself (generally) riding about 1 gear higher than before the adjust. No detonation or pre-ignition is noticed right down to the stall point.

My idle speed screw is best 3 turns out - 2.75/2.5 if offroad and humid and hot. The ideal postion for the conditions seems to correlate to starting in gear with no hesitation and a quick tap of the starter. If it fails to start immediately, then 1/4 turn in. If it starts but does not immediately jump to full idle, then 1/8 turn in. I only turn out the screw out to 3 turns to have the cold idle at 1650-1700 during warm up. There seems to be no downside to leaving it at 2.75 turns, but I like 1650 to warm it up, and it never builds enough heat on pavement to require it in further.

My conclusion of stock off-idle rich operation is a result of the combination of direct throttle plate control (hand on pully), exhaust tone (ear), hydrocarbon output (nose) and rpm uptake vs slow throttle plate movements (brain/ear/hand). It is a qualitative judgement only.


MAT

Thanks Matt

could the fact that you closed the throttle plate in combination with not readjusting the TPS result in your current state of tune- whereas if you adjust the TPS with Ibeat- that would then create a different state of tune. Right now wouldn't your TPS zero point be "TPS wiper voltage was 995.1mVdc" but now you closed it so the reading in mV would be less than before. If you recalibrate the TPS with Ibeat the new reading will be your zero point. Isn't it likely that your bike will run different than it does right now? better or worse. ? Maybe you should leave it as it is right now or record every thing and have it calibrated in its original settings to later be able to replicate what you now have?

this is a setting by the factory- maybe you did exactly as "they" should have done it- maybe we should all adjust this or check it? How come dealer techs don't check this when they set the TPS with the Ibeat- seams it should be part of the process?

I need to learn more about all this- sorry for the machine gun questions and less than rudimentary understanding on my part... I think you are on to something-

Document on video what adjusting you are doing...some (me) are more thick and need visual aids!?!?

My "1/4 turn open from fully closed" throttle plate position would have a wiper voltage less than the first mis-adjustmemt point. (0.9490 Vdc).

I haven't measured the voltage because it's exact position (electrically speaking) is not critical to the EFI - it is below the Ibeat idle setpoint of 0.9951 Vdc and is read as 0% throttle.

Now that the idle speed screw has more influence on the low speed operation, it is also apparent the EFI system is making sizable compensations for engine temperature (or overtemperature). It is not a "set it and forget it" adjustment. Maybe the timing is being pulled back as well? Measuring that will not be quite as easy as TPS voltage.

When the EFI controls both fueling and spark from multiple sensor inputs - it's difficult to get inside the programmer's head to tweak what the electronics are looking for to operate the way I am looking for.

Bike runs great though - different shades of great depending riding and enviromental influences - but great nevertheless.

MAT

Thanks Mat
That helps- if less than zero point=zero point as far as the system is concerned....
I went through the manual last night- there's no mention of the throttle valve closed adjustment. Maybe its in the 50 page Ibeat Manual- but I don't have one.

OK- so next questions Mat...
what year model cc is your bike?
what Exhaust and what insert (none, us sparky, EuroDbkiller 8" long, US db surpessor ~3" long)
gearing-
fuel- race or premium
airfilter

it seems if someones is way off this is something to check and I think your method of turning the throttle adjust out till there's no pressure and then turning in 1/4 turn is a good standard setting.
In this (unrelated to Mikuni efi) efi manual it almost describes exactly the same procedure.

I think if someone is at 4 turns out or 1 turn in on their Idle Bypass Screw they can be assured that this may be the real adjustment that is off. Less damage/out of spec to the "start/warmup(choke) knob" and the TPS being off.

I don't understand how to check the TPS voltages (what wires exactly) need to look at more... But I think I am going to see how many turns from 0 or no pressure on the throttle adjust screw mine is at- out of curiosity-

I think you found- a contributiong factor to the Weak bottom/ response/ flameout issue with some peoples bikes. I don't think I have a real problem- but want as you describe:

Thanks again

Measuring the TPS voltage is not difficult - but you do need a very accurate electrical meter (likely digital).

Get two very small pins (I raided my wife's sewing stash), and carefully slide them under the rubber seals in the base of the TPS connector until they stop. Of the 4 wires on the connector - do this to the black one, and to the "orange with black stripe".

Turn the ignition on - do not start bike - and measure the voltage across the two pins. It is slightly more accurate if you wait a couple of minutes since the battery voltage will settle.

With no throttle applied = this is your idle set point voltage.
With full throttle applied = this is your full throttle set point voltage.

Now you can try whatever you want on the idle hard stop - and you can easily go back to the "before" position. My bike REALLY didn't like having the idle stop set "more open" than the IBeat calibration. I don't know if a mechanical reference point (ie 4 1/8 turns in from touching) is accurate enough to put you back EXACTLY where you started.

For myself - "exactly" isn't important because I run with the plate as closed as possible WITH NO O2 SENSOR IN PLACE. I would not attempt any tweaks on the TPS with the o2 sensor in place - I believe the EFI computer will attempt to compensate for this "error" and I'm not equipped to deal with that fun stuff.

My bike - 09 TE510 with full power up kit installed and no spark arrester installed. Geared up with backpack/tools I weigh 270 (or so) pounds. I had 13/50 gearing.

This last 2 days I had 14/50 gearing and spark arrestor installed. Rode about 10 hours of mixed. Spark arrestor made very little difference to the performance on the tight single track - I'm probably going to leave it in. The taller gearing didn’t cause any issues, and I will save the 13 tooth front for when I am breaking new trails. The crisper throttle and better lugging improved my endurance, and generally used the clutch to remove engine braking or save a shift.

I hard-wired a manual override to the electric fan circuit. I manually engaged the fan once we hit the tight first gear stuff, and this eliminated the need to turn in the idle speed screw to keep the throttle response crisp and maintain lugging ability. This makes me think the fan can maintain an operating temperature that is low enough that the EFI is not changing the fueling to compensate for the heat - or - whatever component in the system (and they are all located right over engine) that gets pissed off when it's hot isn't getting too hot anymore. The fan uses less than 2 amps of current while operating, and I leave it running if I stop for less than a couple of minutes.(engine off)

I'm glad I have the switch and I think it is a good tool to minimize hot-engine flameout.

MAT

can you post some pics of the wiring for the fan override wouldnt mind doing it to mine so i to can turn it off and on in singles.when its turned off does it revert back to being part of the efi circuit

dave

Has anyone seen documentation on the factory recommended procedure for adjusting the throttle hard stop screw? It isn't in the service manual or iBeat manual.

My guess at the procedure was to...

Loosen lock nut and back the stop screw out.
Ensure no tension was on the throttle cable.
Screw in the stop screw until contact with throttle cam.
Screw in the stop screw additional 1/4 turn.
Hold stop screw while tightening lock nut.
Calibrate TPS zero point with iBeat.

I figured that this would keep the butterfly plate from contacting the throttle body bore. I haven't taken the throttle body apart to see if damage or wear would occur if the stop screw was out too far.

I'll look into seeing. if my local service manager can track down the factory procedure. It may take awhile to get a response from Husqvarna.

The hard wire override to the fan is easy. I tapped in a switch from the power feed to the fan relay across to the fan feed - essentially bypassing the relay without changing the relay function. If the Computer turns on the fan with the manual switch off - the fan will start to operate. If the manual switch is on and the computer turns on the fan - no change (fan is already running). If the switch is on and the motor is off, the fan continues to run. The fan draws less than 2 amp of current.

The relay can be either of the two relays on the left side of the frame backbone (tank off with you standing at rear fender facing front fender). The power feed is Yellow/Black (12V fused and unswitched) and the fan feed is White / Black. If you use a pin to remove the spade connectors from the factory relay plug, you can solder directly onto the crimp section of the spade and still reassemble everything. I used the smallest automotive flexible plastic conduit I had to run the wires up past the steering head. I used the conduit because I didn't research what else was on the circuit and shorting that 12V feed to ground and blowing the fuse is probably a bad thing when way back in the bush.

MAT

I have not observed the Ibeat cal process, but from what I have read, the idle stop is physically set to allow for a 100.2% Ibeat wide open throttle reading - final idle rpm is set with the idle air bypass thumbscrew.

Someone who's actually done it is the one to get info from.

MAT

I checked my tps volts- I am at .998 volts closed and 4.63 volts open.
My bike was calibrated on the I/beat in Mid May this year-
My bike is an 09TE450-
My Idle Bypass screw is set to 4 1/4 turns out.
My bike ides at 1850.

So if you turn the idle bypass in idle goes up- right?
So this makes me believe that my throttle valve is open too far...

I will try and see how many turns it takes to have no pressure on the screw- ideally then turn it back in 1/4 turn.

Thanks guys- I am doing some riding this weekend- I don't know if I want to touch it yet and screw things up- but I'll see if I have time to do it then test it out.

Ok- I adjusted the Throttle stop screw- it takes an 8mm open ended wrench and I think the bolt is a 2.5mm Allen wrench (bottom of bolt)

My throttle stop bolt was about 3 turns from not stopping the throttle- I kept turning until it no longer turnd the throttle wheel - I went back and forth a few times to find that zero point. I then turn the throttle stop bolt 1/4 in.

I started the bike- it idles once warm at 1350rpms- with the current 4 turns out on the bypass screw. I adjusted it to 2.5 turns and checked the idle.....1650rpms.... sounds about right huh...

I then ensured the bike was warmed up and rode it around on the pavement at off idle twists- it performed much better off idle- noticeably

Thanks Mat, I will certainly have a better idea after this weekend but I think I am going to have an even better time. I encourage everyone to check this adjustment- I brought this up in a couple circles and everyone kept saying it shouldn't matter because the bypass makes up for it- well- its not meant to make up for being WAY OFF... I think 3 turns is way off- and if anyone elses bike idles at 1850 at 4 turns you should check this out- not all will be off but I was surprised this is not part of normal I/beat calibrations- but once its set it should be good.