AF-XIED Beta for Husqvarna TR650

[big-t]

I have just finished an AF-XIED install and also hooked up an innovate LM-2 to my stock Terra.I'm impressed with the simplicity of this unit and how well it works.I have posted up a few charts over on ADV and hope to run a few more once it has some more time to adapt.

Post # 72....and yea I know Its spelled wrong....... http://advrider.com/forums/showthread.php?t=1002255&page=5

 

[drzcharlie]

It will be interesting to see how my charts compare to yours I have an LC2 on the way. My TR has the POD mod, cannisterectomy and catectomy done to it. I currently run the AF-XIED but will attempt to tune with the LC2 alone.

 

[nwrider1]

Last time I posted I had recently installed the AF-XIED on my Terra set on #5. After running it for awhile on that setting, I was experiencing stalling off of the line and when starting from cold, it would not continue running on the first try. After moving to setting #6 the failure of the motor to continue to run on the first attempt at a cold start has been cured and I've not had further stalling issues. There are no apparent flat spots except when rolling off of the throttle slightly it feels like the throttle is being closed more than it is. The motor is a bit more vibey than with the Wiseco Fuel Controller. Even though I think the Wuka King on my Strada is gradually being modified by the ECU, it's still better that what I've tried so far with the Terra.

 

[drzcharlie]

My LC2 "should" be delivered today. I have a whole day of VA appointments starting at noon so I won't be able to do the install today. I am looking forward to finally be able to see what the MM is doing and target my AFR's to smooth out the lean spots. My AF-XIED worked brilliantly, it really did what I was told it would do. I still have lean stumble on cold start though. I have found that if you let the engine warm for about 3 minutes the stumble is nullified. Other than that it works great. My bike liked the F7 setting and my fuel economy was decent. On a recent trip I got between 55 & 57 mpg depending on my wrist position. The bike has no decell pop, it has a very linear throttle, pulls like a tractor on steroids and the engine is smooth as silk.

But being the tinkerer I am I am going to send back the beta AF-XIED and install the LC2 to experiment with it.

 

[mag00]

My LC2 "should" be delivered today. I have a whole day of VA appointments starting at noon so I won't be able to do the install today. I am looking forward to finally be able to see what the MM is doing and target my AFR's to smooth out the lean spots. My AF-XIED worked brilliantly, it really did what I was told it would do. I still have lean stumble on cold start though. I have found that if you let the engine warm for about 3 minutes the stumble is nullified. Other than that it works great. My bike liked the F7 setting and my fuel economy was decent. On a recent trip I got between 55 & 57 mpg depending on my wrist position. The bike has no decell pop, it has a very linear throttle, pulls like a tractor on steroids and the engine is smooth as silk.

But being the tinkerer I am I am going to send back the beta AF-XIED and install the LC2 to experiment with it.

Surprised to hear you are not keeping the AFxied.

Have you looked into the SLC PP2: http://www.14point7.com/products/sigma-lambda-controller-pure-plus-2

Below is from their website.

Features:​

​

SLC PP2 utilizes the Bosch LSU 4.2 Wideband O2 Sensor - a cost efficient Wideband O2 sensor providing 0.01 Lambda accuracy, fast response, and rated life of 100,000 km

​

Compatible with all fuel types

Ultra wide AFR/Lambda range, 0.68[Lambda] to 100[Lambda], 10[AFR] to 100[AFR] equivalent for gasoline

​

0.01 Lambda accuracy without the need for any calibration

Although not required (or recommended), Free Air Calibration is supported

​

O2 Sensor Temperature Monitor

Did you know that the most common cause of inaccuracies and premature O2 sensor failure is attributed to the O2 sensor temperature being improperly controlled? Only 14Point7 branded Wideband O2 Controllers are able to monitor O2 sensor temperature

​

Programmable Linear Output allows easy interfacing with gauges, fuel management systems, and data-loggers

​

Our innovative Output Sequencer feature means our Wideband O2 Controllers to have the most accurate Linear Output on the market. You can be sure that any device connected to the Linear Output will receive the most accurate AFR/Lambda signal possible

The Output Sequencer provides a dual level precision voltage signal to the Linear Output. This allows you to easily check that the AFR/Lambda seen by any device connected to the Linear Output is accurate, and if not, to generate a 2 point calibration to achieve perfect accuracy

​

Programmable Simulated Narrowband Output allows you to install a Wideband O2 Sensor in place of your Narrowband O2 sensor

​

Built in Multi Parameter Display of AFR/Lambda, Boost, Exhaust Gas Temperature,Intake Air Temperature, Fluid Temperature, Fuel Pressure, and Oil Pressure​

​

32 mb of onboard memory for storing upto 3 hours of datalogging

​

Native USB connectivity means no more fooling around with expensive and unreliable USB-to-Serial converters

​

Built in RPM Pickup - 0[RPM] to 20000[RPM]

​

Exhaust Gas Temperature probe input, 0[F] to 2500[F], 0[C] to 1400[C]

​

Boost/Vacuum sensor input, 0[PSI] to 44[PSI], -30[Hg] to 88[Hg]

​

Intake Air Temperature sensor input, 0[F] to 300[F], 0[C] to 150[C]

​

Fuel Pressure sensor input, 0[PSI] to 150[PSI], 0[Hg] to 250[Hg]

​

Oil Pressure sensor input, 0[PSI] to 150[PSI], 0[Hg] to 250[Hg]

​

Fluid Temperature sensor input, 0[F] to 300[F], 0[C] to 150[C]

​

Dual 0-5 Volt inputs

​

A Two-Year Warranty that is twice as long as any competing product

​

60 day money back guarantee

I heard the price is reasonable $225

 

[drzcharlie]

I may well end up back at the AF-XIED when all is said and done. I started installation of the LC2 this afternoon. Our TR's do NOT have a lot of room for extras anywhere. It took me about an hour and half just to come up with a viable place to stash it. Tomorrow I need to solder on a plug, build a resistor bank and finish the install.

 

[mag00]

Charlie, how is this LC 2 supposed to work? What will it do for performance? Am I missing something or is it just a data logger, and not anything to do with enhancing the performance?

Did you get a chance to look at the SLC PP2? I think I may try one of these. I have not been able to get any questions answered as of yet from this company. I Emailed and phoned this morning, no reply.

 

[drzcharlie]

Magoo,

I will be able to target AFR's using the serial input/output from the unit. The unit is connected via serial port to a laptop and then ridden to gather AFR information and then you can use "Logworks" software to target trouble spots. What I am really going for here is to see what the MM ECU is doing during normal operation of the motorcycle. Once that log is finished I will know more about where to target my changes. If we can make sense of this information it may help all of us . Firstly we all know about the famous lean stumble but there are lean spots in the upper portion of the power curve as well. My first challenge here is to get it installed and I working towards that now. After coffee it's off to gather bits for the installation.

 

[mag00]

I got a reply back from 14point7.

Here is an excerpt:

​

My question is, will your product allow me to eliminate the narrow band sensor, and have adjustability of the o2 output to the ecu? ​

Yes, the narrowband output should work without problems.

That was the extent of the reply from Alan. There was more to the email. I will fire off another email since I cannot get them on the phone, I have many questions.​

​

What I'm thinking is to run the dobeck style fuel controller with one of these PP2's. Since the o2 doesn't really work for the first minute or so, the dobec will add the fuel in the beginning and the PP2 should feed the ecu the modified values to keep the A/F proper at the other rpms once warm. In short, let the dobeck get adapted out. Where it matters, cracking the throttle, should not get adapted, but the rest should adapt to the stock MM map but spoofed by the PP2 values.​

​

​

​

Effectively creating an accelerator pump using the dobec, but adjusting the lean using the o2 spoof of the SLC PP2. This solution starts to get up in price. About $475 total. It should work better than the pc5 or other, because it keeps the stock map in play and the adaptions for altitude, fuel quality, and other inputs. ​

The pcv and auto tune would be a bit fussier, because it takes the advanced Mag Marelli out of the equation to a great extent. ​

​

The Dobec style controllers do not adapt on the fly, you have to manually change a setting​

The AFxied, only fattens the mix in varied amounts, across the whole rpm range, but lets the computer work for adaption.​

​

​

Booster, wuka, eruption, spoof ait, and some of that is learned out, probably not all, but some, which would bring the a/f at highway back down, unlike the afxied.​

​

​

Running the Eruption with the PP2, might be an option, but I am not a fan of the Temp reading wrong. ​

A new stripped down version of the PP2 can be bought for $140. Might be a nice option for the tech savvy. That puts a good fix at $160 (plus o2 sensor) if it works as stated.​

​

​

​

I am curious on what your LC-2 project will reveal.​

​

​

Think I'll have some coffee also, as I head to the bearing supply store and my machinist. ​

Dave​

 

[engineerk9]

Could you expand on this? I'm going to be doing the same mods soon....

I found that with the Pod-mod the AIT no longer sees the correct air temperature
The AIT needs to re-located at the air inlet, front of the airbox

 

[engineerk9]

I got a reply back from 14point7.

Here is an excerpt:

My question is, will your product allow me to eliminate the narrow band sensor, and have adjustability of the o2 output to the ecu? ​

Yes, the narrowband output should work without problems.



That was the extent of the reply from Alan. There was more to the email. I will fire off another email since I cannot get them on the phone, I have many questions.

What I'm thinking is to run the dobeck style fuel controller with one of these PP2's. Since the o2 doesn't really work for the first minute or so, the dobec will add the fuel in the beginning and the PP2 should feed the ecu the modified values to keep the A/F proper at the other rpms once warm. In short, let the dobeck get adapted out. Where it matters, cracking the throttle, should not get adapted, but the rest should adapt to the stock MM map but spoofed by the PP2 values.



Effectively creating an accelerator pump using the dobec, but adjusting the lean using the o2 spoof of the SLC PP2. This solution starts to get up in price. About $475 total. It should work better than the pc5 or other, because it keeps the stock map in play and the adaptions for altitude, fuel quality, and other inputs.
The pcv and auto tune would be a bit fussier, because it takes the advanced Mag Marelli out of the equation to a great extent.

The Dobec style controllers do not adapt on the fly, you have to manually change a setting

The AFxied, only fattens the mix in varied amounts, across the whole rpm range, but lets the computer work for adaption.



Booster, wuka, eruption, spoof ait, and some of that is learned out, probably not all, but some, which would bring the a/f at highway back down, unlike the afxied.


Running the Eruption with the PP2, might be an option, but I am not a fan of the Temp reading wrong.
A new stripped down version of the PP2 can be bought for $140. Might be a nice option for the tech savvy. That puts a good fix at $160 (plus o2 sensor) if it works as stated.



I am curious on what your LC-2 project will reveal.


Think I'll have some coffee also, as I head to the bearing supply store and my machinist.
Dave

Are you saying that the ecu ultimately adapts the AF-XIED out ?
Quite complex this ecu challenge
I think we all agree:-
That it's basically the O2 sensor setting at 14.7 which is our major challenge
With the Pod-mod & more open exhaust, the gas flows are significantly different
At higher speeds there is less pressure in the airbox, (unless we seal it) which is possibly why the ecu was set so rich at higher rpm to in some way compensate

Magoo You're doing a great job

My 11 ohm resistor in series with the re-positioned AIT, certainly improved things, however the ecu will ultimately adjust it out, the O2 sensor rules

Be great to be able to reprogram the ecu

I wonder if there is anyone who can do that, without using "add-ons"

Even on an exchange basis

Please give me more info regarding your unit ?

 

[Roger 04 RT]

Are you saying that the ecu ultimately adapts the AF-XIED out ?
Quite complex this ecu challenge
...

Please give me more info regarding your unit?

Think of your ECU as a control system. The primary inputs are RPM and Throttle Position. After that, the ECU monitors Air Temperature, Barometric Pressure, Battery Voltage and Long Term Fuel Trim so that it can get fueling approximately right when the throttle is moved rapidly or at wide throttle angles. (Oil temperature is monitored and at low temperatures enrichment is provided, only while the ECU is not-in-Closed-Loop during warm up.)

Then think of the O2 sensor as a reference standard at lambda=1 (which is 14.7:1 if you burn pure gas and, interestingly is 14.1:1 if you're burning E10 fuel). It is used to see if the result of the fueling calculation has produced an exhaust oxygen content equal to lambda=1 (or another lambda value if you use an LC-2 or AF-XIED or something else like at 14point7). If the fueling result isn't the lambda value specfied by the O2 sensor, a Short Term Trim is created. If the Short Term Trim is large enough or persists long enough, then a Long Term Trim is created.

So the bottom line is that Closed Loop Adaptive Fuel Control is complex, but since the O2 sensor is used for control, it does not get "adapted out".

 

[engineerk9]

Think of your ECU as a control system. The primary inputs are RPM and Throttle Position. After that, the ECU monitors Air Temperature, Barometric Pressure, Battery Voltage and Long Term Fuel Trim so that it can get fueling approximately right when the throttle is moved rapidly or at wide throttle angles. (Oil temperature is monitored and at low temperatures enrichment is provided, only while the ECU is not-in-Closed-Loop during warm up.)

Then think of the O2 sensor as a reference standard at lambda=1 (which is 14.7:1 if you burn pure gas and, interestingly is 14.1:1 if you're burning E10 fuel). It is used to see if the result of the fueling calculation has produced an exhaust oxygen content equal to lambda=1 (or another lambda value if you use an LC-2 or AF-XIED or something else like at 14point7). If the fueling result isn't the lambda value specfied by the O2 sensor, a Short Term Trim is created. If the Short Term Trim is large enough or persists long enough, then a Long Term Trim is created.

So the bottom line is that Closed Loop Adaptive Fuel Control is complex, but since the O2 sensor is used for control, it does not get "adapted out".

Hi Roger my AF-XIED arrived this afternoon

A couple of questions if I may?

Dyno tests this morning indicated that after nearly 1,000klm my 11 ohm resistor in series with the AIT had adapted out as expected & predicted
Now tomorrow when I install the AF-XIED would you suggest leaving the resistor in or remove it ?

Secondly, batteries - any comments on Lithium batteries for the Terra ?

Thirdly, how is your unit progressing? Will it be a plug'n'play?

Many thanks for your input into this "mob", I don't think that there are very many Husky's here in Australia

 

[Roger 04 RT]

Hi Roger my AF-XIED arrived this afternoon

A couple of questions if I may?

Dyno tests this morning indicated that after nearly 1,000klm my 11 ohm resistor in series with the AIT had adapted out as expected & predicted
Now tomorrow when I install the AF-XIED would you suggest leaving the resistor in or remove it ?

Secondly, batteries - any comments on Lithium batteries for the Terra ?

Thirdly, how is your unit progressing? Will it be a plug'n'play?

Many thanks for your input into this "mob", I don't think that there are very many Husky's here in Australia

Hi EK9,
If it were me I'd remove it since the amount temp offset produced by a fixed resistor varies with the absolute temperature. However, when you remove it, there will be a leanness for a while. That leanness will be corrected after a while.

I don't have experience with Litium batteries but the forum posts that I've read suggest that can work but often aren't a good match in motorcycle applications.

I'm not building a lambda-shifting device although I have experimented with many: LC-1, LC-2, 14point7, Wideband O2, and AF-xied. Since the only unit that A manufacturer was willing to make plug 'n play was the AF-xied, I helped the Nightrider develop the software and hardware that would allow it to run on non-Harley Davidson motorcycles by providing specifications and then testing. I have also tested and data logged Boosterplug, Fuel pressure changes, Power Commander and EJKs to record what they actually do and don't do.

I've enjoyed this board because there are so many riders willing to experiment.
RB

 

[engineerk9]

Hi EK9,
If it were me I'd remove it since the amount temp offset produced by a fixed resistor varies with the absolute temperature. However, when you remove it, there will be a leanness for a while. That leanness will be corrected after a while.

I don't have experience with Litium batteries but the forum posts that I've read suggest that can work but often aren't a good match in motorcycle applications.

I'm not building a lambda-shifting device although I have experimented with many: LC-1, LC-2, 14point7, Wideband O2, and AF-xied. Since the only unit that A manufacturer was willing to make plug 'n play was the AF-xied, I helped the Nightrider develop the software and hardware that would allow it to run on non-Harley Davidson motorcycles by providing specifications and then testing. I have also tested and data logged Boosterplug, Fuel pressure changes, Power Commander and EJKs to record what they actually do and don't do.

I've enjoyed this board because there are so many riders willing to experiment.
RB

Hi Roger, & I'm one of them, my passion
I've been testing the AF-Xied this morning with the 11 ohm resistor in series with the AIT, I've now removed it, whilst waiting to reset the ecu, I took time-out for a coffee & chat
I ran tests on the AIT, between 0degC (melting ice) & 100degC (boiling water) - it's not a linear change which surprised me
Between you & Nightride you appear to have done a great job
I'll put a few 1,000 klms on it & report back

The Battery Guy sells a lot to Dirt Bike riders who don't expect a great life, just appreciate the weight
Apparently at less than 9.0volts kills them, they'd possibly work if there was a voltage cut-out at 10-11volts, preventing further drain

Cheers
Keith

 

[engineerk9]

Hi EK9,
If it were me I'd remove it since the amount temp offset produced by a fixed resistor varies with the absolute temperature. However, when you remove it, there will be a leanness for a while. That leanness will be corrected after a while.

I don't have experience with Litium batteries but the forum posts that I've read suggest that can work but often aren't a good match in motorcycle applications.

I'm not building a lambda-shifting device although I have experimented with many: LC-1, LC-2, 14point7, Wideband O2, and AF-xied. Since the only unit that A manufacturer was willing to make plug 'n play was the AF-xied, I helped the Nightrider develop the software and hardware that would allow it to run on non-Harley Davidson motorcycles by providing specifications and then testing. I have also tested and data logged Boosterplug, Fuel pressure changes, Power Commander and EJKs to record what they actually do and don't do.

I've enjoyed this board because there are so many riders willing to experiment.
RB

Hi Roger thank you for your knowledge
Since fitting the AF-XIED, left on 7, I've been doing a number of 80-100klm runs, then an hours cooling down (coffee break) then another 80-100klms
As the ecu "unlearns" the spoofers & "learns" the new parameters - as you predicted
Certainly, the motor appears to be getting more responsive & less vibration each time I start - I don't know how long I have to run this process
There still seems to be some minor detonation at approx 4,500rpm during full throttle acceleration in top gear, when pulling hard (it's easy to ride around, more a point to note), this could be leanness, which as you suggest will be corrected out
Now that I'm more than happy with the performance, as soon as the new airbox arrives, l'll tidy everything up & enjoy the ride

 

[Roger 04 RT]

Good to hear you're progressing. I think you should try setting 8 sometime soon. It won't reverse the adaptation that's already occurred, it will just add to it.

 

[engineerk9]

Good to hear you're progressing. I think you should try setting 8 sometime soon. It won't reverse the adaptation that's already occurred, it will just add to it.

Okay, cool bananas Wednesday

 

[engineerk9]

Good to hear you're progressing. I think you should try setting 8 sometime soon. It won't reverse the adaptation that's already occurred, it will just add to it.

I have another Offshore Contract, be back in Sydney in a month, will be riding then
Meanwhile someone, somewhere might come up with the perfect solution
How does it get any better than this ?

 

[engineerk9]

Good to hear you're progressing. I think you should try setting 8 sometime soon. It won't reverse the adaptation that's already occurred, it will just add to it.

Hi Roger l see in an Australian mag adverts for the EJK Units, for the BMW, have you any comments ?