Air Temperature Sensor - Improving Engine Performance?

[Highlander179]

Babo, will you be putting this to market?

LOL

[AndrewA]

Originally Posted by Highlander179

Babo, will you be putting this to market?

LOL

LOL is right.














I picked up a GS-911 today. A very cool tool indeed! It allows, among many other swell functions, a visual representation of exactly what the ecu sees as Intake temperature when more or less resistance is applied through the IAT circuit. The soft goal is to get the ecu to think that it's always -25C.





With 10 - 20K ohms of adjustability, I was able to spoof the ecu's Air Intake Temp as follows:

10k -3.0C
13k -5.3C
16k -6.8C
18k -9.8C
20k -10.5

Originally Posted by OwenM

The temp sensor on my bike reads about 7k ohms at 20 degrees C and 17k ohms at 0 degrees C. The thermistor is very non linear and I expect something like 40k ohms or more will be needed to simulate -25C.
I have just replaced my 10k pot with a 100k pot and I have marked temperature points for reference. The pot is mounted in the fairing for easy access while on the move and has a 6.8k fixed resistor in series so that cant set a temperature higher than 20C. This setup should give me enough adjustment to take the ECU as rich as it will go and with maximum advance just so I can see what it does.
The temp sensor is not in circuit at the moment, I’ll connect it when I work out what is needed.

Right. Good. My findings look similar to those. The difference between our two methods is that I'm testing with the bikes temp sensor in circuit, which I seem to remember reading provided around 10k ohms of resistance (at a specific ambient temperature).

Originally Posted by OwenM

It should be easy to see the responses time to changes in input with the GS-911 as it gives real time feedback from the ECU.

The response time varies greatly, and is non-linear. At temperatures that might be normal for riding, the GS-911 registers a change every second or so. As it starts to get much colder, the response time is much longer - perhaps every 10 seconds or so. Also, the ecu seems to stall when the spoofer knob is turned quickly - taking 30 seconds or so to register any changes. The ecu appears to be going through WTF is going on gyrations.


I added another 10k to the existing 30k and got the spoofed temp down to -18C. Tomorrow I ride!

I still need to put the GS-911 on a friends bike to get a baseline, but Im not sure what it will really mean, as his bike has a very free flowing exhaust (the assumption being that his device was "tuned" to accommodate his set-up).


Here is what the GS screen looks like in "Real Time Data" mode: (Is that battery voltage low?)

[AndrewA]

I got to take about an hour ride this morning. Below Ill compare my impressions of a stock R1200S (stock exhaust, filter...) and the same bike running with 20k ohms and 30k ohms of resistance applied to the IAT loop.



Firstly, I set the resistance to 20k ohms (-10.6C), and started the bike. The ambient temps are an unseasonable 69 degrees F.

Stock startup: The bike turns over a couple of revolutions, catches, and then fall into a rough, very lumpy idle. Throttle not responsive at all and would almost stall the bike if applied. After about 15 seconds of warm-up, it was fine.

I should note that I reset the idle stepper actuators with the GS-911 tool last night, so start up issues I may have had could have been due to mis-calibration.

@ 20k Ohms startup: Bike started immediately and fell into a nice, non-lumpy idle. Throttle was immediately effective. I could actually smell exhaust gasses, which I never could as stock. Something was happening!





Stock riding: Anything below 3k rpms, bike would take its time...at about 3.2k rpms, it would start to spool-up but just felt "flat". At 5.1k rpms, bike would start to rocket. No pops on deceleration from the exhaust.


@ 20k Ohms riding: Below 3k rpms, bike would pull strongly all of the way past 5.1 to redline. No discernible flat spots, nor kicks, anywhere in the rev range. Bike produced a very consistant popping sound on deceleration - not loud, but different.

It was a blast lugging the engine down below 3k rpms and whacking the throttle and then feeling the bike lurch forward. It reminded me very much of my R1200GS, and how fun it was to brake hard going into a corner and then simply opening the throttle to access the torque to get me through the turn. It was one of the features that makes the GS such a great corner carving machine - tractable power on demand...

@ 30k ohms riding: Below 3k rpms, the bike seemed to lurch like at the lower setting, but didn't seem as "smooth". Power band from 3k rpms through 5.1k rpms was clearly flatter than at 20k ohms, but not anywhere near the stock flat spot. Still a huge improvement over stock.






Seat-of the pants tells me that the engine spooled up significantly quicker @ 30k than at 20K. Where at 20K the bike felt buttery smooth all of the way from low revs all of the way to redline, it simply did not seem as "fast". Smoother, yes, more fun at low revs, yes, but slower overall.

I suspect that something profound happened to the ignition timing between the 20k and 30k pot setting. Where at 20k, the added resistance clearly richens the mixture, it is not enough to call-up advanced ignition timing. At 30k, the ignition timing is called-up, hence the bike seems to rev much quicker.

I'll keep experimenting with the pot to see if I can find a balance between the smoothness I want and the quick spool-up.

[Guest24]

I absolutely love this Andrew. I'll wait until you pull the plugs stock and enriched spoof. Early returns are lookin' good partner.

[Phil Y]

Absolutely fascinating, Andrew. Good work! Keep it up!!

[AndrewA]

Today's notes:


21k ohms - Exceptionally smooth with no band anywhere in the rev range. Big, huge, torque down low. Feels like freight train. Seems to rev too slowly. pops once or twice on decel at 3-4k.

23k ohms - Very smooth - Spools faster still - no band at all - a joy at 4 - 5k - single "soft" pop at 4k

25k ohms - Revs even faster - very faint band - wicked fun on my favorite road - no popping.

27k ohms - Revs faster still. Slight band 3k - 5k. Still pulls from low revs with little drama. Feels like stock tune except flat spot minimalized, almost gone. Minimal exhaust noise on decel at 3k.

28.5 Spools very quickly, quicker than stock - thrusts from 3.5k rpm to redline- 4k rpm is the new 5.1k rpm! Not as smooth as at 27k. Has almost imperceptible power bands. Likes short shifting. Does not seem to pull as quickly from down low, seems to rev much quicker than stock. Faster than stock?

[Guest24]

Andrew, could you PM me the parts list and sources. I think I would like to build one a play with it in the spring on the HPS.

[JonyRR]

Once more, in the spirit of a long departed board member named shovelstroked:
I have read this thread with some amusement'.
Why?

there are bandaids and there are bandaids.

there are EPA and Euro regs, and very smart microprocessors with more money and engineering time backing them than any one of us will ever be able to bring to the table designed just to comform to those regs.

If you REALLY, really want to get everything out of your scoot you can get, at every throttle opening, RPM range and environmental condition, then you need a completely programmable CPU, preferably with a positive feedback loop (think 02 sensor). And of course, dyno and over-the-road testing.

barring that, I run a USB PCIII with an 02 gauge so I can see what's going on.
If I had to do it again, it'd be RBIII and dyno in a heartbeat.

You spend upwards of $15K for the scoot and farkles; you won't spend some bucks to dial it in perfectly.

In racing, that's called.........losing

Just my $.02...

[Guest24]

Jony, In my non-racing world I call it tinkering. There are very few methods for the backyard mechanic to play with the EPA directed fueling that starves an engine to death. This is very interesting to me the lover of analog ways to screw around with digital stuff.

[JonyRR]

Tinker away
It's the american way
Screwing with EFI closed-loop systems is the art of diminishing returns, though, without the tools to do the job..but tinker away, and you have my good vibes floatin' out to ya...

[jduke]

AndrewA, any updates?

[AndrewA]

Originally Posted by Nail24

Andrew, could you PM me the parts list and sources. I think I would like to build one a play with it in the spring on the HPS.

Better than that, I'll publicly post some detailed instructions with a recommended parts list shortly. Until then, the only items that are not available for most people locally are the Bosch connectors with pigtails.

You will need one of each.

Male Bosch Connector:
Home » Shop » Connectors / Harnesses » Bosch » Bosch Type EV1 Injector Connector MATE Connector Pigtail
Female Connector
Home » Shop » Connectors / Harnesses » Injector connectors » GM Delphi / Packard - Bosch EV1 Type Injector Connector Pigtail Push to seat

Total cost: $ 19.38 delivered via USPS.

Originally Posted by jduke

AndrewA, any updates?

My two posts above represented two days (last weekend) and about 220 miles.

Nothing to update today...except some random thoughts....


After reading various "theories" of how BMW's closed loop / open loop operation are allegidly supposed to work, I was wholly expecting my homemade, bench designed "Spoofer" to be not much more than a placebo induced, temporary (meaning 1/2 mile or so) folly, and a waste of the $40.00 or so I have into the project.

Fact:: The Spoofer makes the bike run significantly differently, weather at steady throttle or at the pesky 3k - 5k rpm zone, or anywhere else in the rev range. The Spoofer is effective any time the throttle is moved. ANY TIME, every time. Does it work at "steady" throttle? I wouldn't know. Not really an issue.

Theories abound that the IAT loop is only used at start up, and once the engine is warm, other operations (closed loop/open loop) take over and ignore the IAT.

Fact: The Spoofer, being running through the IAT circuit, works always, and every time the throttle is moved. Period. Still.

There are claims that similar devices don't work on closed loop operation, or deceleration; how then would you account for the various "pops" on deceleration, at vastly different rpm, if the bike was only in closed loop? (by deceleration, I mean chopping the throttle closed at a high rpm (perhaps at redline or 8k), and then listening for tell-tail back-fire or pops as revs and mph decreased. The pops that I've experienced have always been well after, and at relatively low rpm. How could these pops have the left-over, unused fuel to develop at low rpm, if the bike was in closed loop operation? Where is the extra fuel coming from? Clearly by the time that deceleration brings the revs down to the 3k rpm - 4k rpm range, any excess fuel would have been exhausted. No? Doesn't really matter, does it?

Regardless, who really cares about deceleration? It is only when the throttle is moved or twisted that matters to most.

The "spoofer", if set within a very specific temperature range, is extremely adjustable - every 1k adjustment brings a very specific "feel" to the bike.

On day 1 of my testing, I ran the spoofer at its lowest (20K) and its highest settings (30k), and felt a profound difference from the two settings. This was the POF, or Proof of Concept run. On day 2 of my testing I was able to parse the initial range of my testing within the range of 20k to 30k of resistance, and each 2k felt like a different bike.

Devises similar to the the spoofer have been available though places like JC Whitney since the advent of fuel Injection. This is not something that has recently been discovered by Owen, Ben, nor even Dr. Robert Foster. There is no proprietary info here, just well known, 30 year old trickery.

Fortunately, it really works on the modern BMW!

Here's a basic web summary of the legacy IAT "technology". Please read the entire thing! Sound familiar? Does it?
IAT Resistor Mods vs Performance Modules vs Real Chips | Technical Domain


I believe that the difference between my device, and other devises, is the extremely low temperature that the IAT is spoofed to(-18kC @ 30l Ohms).

I'm guessing that an extremely cold spoofed temp tricks the BMW ECU into not going into closed loop operations, enriching the fuel/air ratio, and advancing timing.

:end random thoughts:

[OwenM]

I have an adjustment range of 6.8k to about 90K, which is a lot and I have now travelled over 1000 miles, a lot of that two up with luggage. My standard RT with Lennies sprockets does not seem to like too much resistance (super cold spoof) and seems happiest at about 20k (total including temp sensor) or about -5 degrees C while running in a true ambient of around 25C. Going to cold (high resistance) makes the engine run a little flat as one would expect.

The RT runs fine standard with no flat spots, the R12S with free flowing exhaust probably has more to gain from going colder, and from what Andrew has said it appears the R12S is much more sensitive to adjustment than my RT. Small changes on the RT produce no significant difference in performance or feel.

A 20k (Lin) pot in series with the temp sensor should provide enough adjustment for most bikes, if you want to go nuts a 50k Lin pot is the go and should be more than enough for any bike.

[boxercup]

Originally Posted by AndrewA

Better than that, I'll publicly post some detailed instructions with a recommended parts list shortly. Until then, the only items that are not available for most people locally are the Bosch connectors with pigtails.

You will need one of each.

Male Bosch Connector:
Home » Shop » Connectors / Harnesses » Bosch » Bosch Type EV1 Injector Connector MATE Connector PAigtail
Female Connector
Home » Shop » Connectors / Harnesses » Injector connectors » GM Delphi / Packard - Bosch EV1 Type Injector Connector Pigtail Push to seat

Total cost: $ 19.38 delivered via USPS.
My two posts above represented two days (last weekend) and about 220 miles.

Nothing to update today...except some random thoughts....


After reading various "theories" of how BMW's closed loop / open loop operation are allegidly supposed to work, I was wholly expecting my homemade, bench designed "Spoofer" to be not much more than a placebo induced, temporary (meaning 1/2 mile or so) folly, and a waste of the $40.00 or so I have into the project.

Fact:: The Spoofer makes the bike run significantly differently, weather at steady throttle or at the pesky 3k - 5k rpm zone, or anywhere else in the rev range. The Spoofer is effective any time the throttle is moved. ANY TIME, every time. Does it work at "steady" throttle? I wouldn't know. Not really an issue.

Theories abound that the IAT loop is only used at start up, and once the engine is warm, other operations (closed loop/open loop) take over and ignore the IAT.

Fact: The Spoofer, being running through the IAT circuit, works always, and every time the throttle is moved. Period. Still.

It is running with a static slow to react value only! It is not dynamic! Trade offs will abound! Each Setting will have a sweet spot!

There are claims that similar devices don't work on closed loop operation, or deceleration; how then would you account for the various "pops" on deceleration, at vastly different rpm, if the bike was only in closed loop? (by deceleration, I mean chopping the throttle closed at a high rpm (perhaps at redline or 8k), and then listening for tell-tail back-fire or pops as revs and mph decreased. The pops that I've experienced have always been well after, and at relatively low rpm. How could these pops have the left-over, unused fuel to develop at low rpm, if the bike was in closed loop operation? Where is the extra fuel coming from? Clearly by the time that deceleration brings the revs down to the 3k rpm - 4k rpm range, any excess fuel would have been exhausted. No? Doesn't really matter, does it?

YOU ARE EXPERIENCING A LEAN BURN SYNDROME
The unburned fuel is popping in the exhaust header due to incomplete combustion.

Regardless, who really cares about deceleration? It is only when the throttle is moved or twisted that matters to most.

The "spoofer", if set within a very specific temperature range, is extremely adjustable - every 1k adjustment brings a very specific "feel" to the bike.

On day 1 of my testing, I ran the spoofer at its lowest (20K) and its highest settings (30k), and felt a profound difference from the two settings. This was the POF, or Proof of Concept run. On day 2 of my testing I was able to parse the initial range of my testing within the range of 20k to 30k of resistance, and each 2k felt like a different bike.

Devises similar to the the spoofer have been available though places like JC Whitney since the advent of fuel Injection. This is not something that has recently been discovered by Owen, Ben, nor even Dr. Robert Foster.

Yes, I knew about the resistor trick in the eighties! Nothing new!

There is no proprietary info here, just well known, 30 year old trickery.

Fortunately, it really works on the modern BMW!

Here's a basic web summary of the legacy IAT "technology". Please read the entire thing! Sound familiar? Does it?
IAT Resistor Mods vs Performance Modules vs Real Chips | Technical Domain

Straight From eBay!


I believe that the difference between my device, and other devises, is the extremely low temperature that the IAT is spoofed to(-18kC @ 30l Ohms).

Do Not Include

I'm guessing that an extremely cold spoofed temp tricks the BMW ECU into not going into closed loop operations, enriching the fuel/air ratio, and advancing timing.

"Steady State" riding will ALWAYS!! put the Motronic 2.2, 2.4 and the BMS-K into "Closed Loop' operation - regardless of what resistance you are using.

:end random thoughts:

End Empirical Thoughts!

Greetings Andrew,

Nice work!

You promised tuning 2.1! A Power FRK Module Clone<>You provided J.C. Whitney!

A Power FRK Module clone? You are not even close!

The Power FRK Module uses no resistors!

We use 3 types of Diodes, Capacitors, Transistors and _____________!

The Power FRK Module is dynamic! Resistance tuning is static and outdated!

Thus:

Power FRK Module - Tuning 2.0 - Dynamic - No Dyno Necessary! FROM A SMOOTH STRONG IDLE TO REDLINE - Smooth and Powerful!

Don't include the Power FRK Module in your suppositions. It does not relate to your device at all!

Again nice work!

[boxercup]

Power FRK Module Partial:

[AndrewA]

Here’s the parts list in total. Other than the Bosch connectors, one could find everything they need at their local Radio Shack store.

For the R1200S, I believe that the sweet spot remains between 22k ohms and 29k Ohms of resistance. I will be redoing my box to include the 10k adjustable pot and one of the 22k ohm fixed resistors. I like the 10k pot because it is easy to dial it in. I tried to hit individual resistance numbers with a 100k pot and had extreme difficulty - move the knob just a hair, and the ohms jump way to much - the 10k pot will allow extreme granularity.

As Owen mentioned above, different bikes appear to like different resistance levels. I would imagine his RT, a GS, or R would need minimal resistance. A R1100S maybe a little more, and the R1200S, more still. A HP2??? What mods are installed are also key. A good starting point might be the 20k pot that Owen recommended, or a 10k pot and a 10k resistor. Is your bike fully modded? Then maybe the 10k pot and a 22k....




Ballenger Motorsports
Bosch Type EV1 Injector Connector MATE Connector Pigtail
Home » Shop » Connectors / Harnesses » Bosch » Bosch Type EV1 Injector Connector MATE Connector Pigtail
Bmotorsports Part # CONN-85571
$ 5.99 + freight

Bosch EV1 Type Injector Connector Pigtail Push to seat
Home » Shop » Connectors / Harnesses » Injector connectors » GM Delphi / Packard - Bosch EV1 Type Injector Connector Pigtail Push to seat
Bmotorsports Part # CONN-85660
$ 5.99 + freight



Radio Shack
10K-Ohm Linear-Taper Potentiometer
10K-Ohm Linear-Taper Potentiometer - RadioShack.com
Radio Shack PN 271-1715
$2.99


10 K Ohm Resistors
10K ohm 1/2W 5% Carbon Film Resistor pk/5 - RadioShack.com
Radio Shack PN 271-1126
$ .99 /five pack.

22k K Ohm Resistors
22K ohm 1/2W 5% Carbon Film Resistor pk/5 - RadioShack.com
Radio Shack PN 271-1128
$ .99 /five pack.


Project Box – 3x2x1 (You could also use an Altoids mint tin, or an "Ice Breakers" mint box)
Project Enclosure (3x2x1") - RadioShack.com
Radio Shack PN 270-1801
$ 2.29

Other than some wire, a knob for the pot (which you can also get at Radio Shack - make sure it is the kind that screws tight so it wont move) and something to protect the cables (I used heat shrink tubes), nothing else is needed.
As far as tools; a soldering iron, wire snips, and a multi-meter (so you can test and read the resistance of your creation, which you then mark on paper that’s taped to the project box as reference points).

I’ll post-up a wiring diagram shortly.

[throttlemeister]

Originally Posted by boxercup

Power FRK Module Partial:

Pretty smart solution to the power problem.

[roger albert]

Nice!

[jduke]

I do appreciate all the work AndrewA has put into this, and I know I sound like I'm beating a dead horse, but I sure do wish AndrewA would take Robert up on his return policy and try his device and then the FRK and come back to us with some kind of feedback on the plus's and minus's of one versus the other.

[Highlander179]

Originally Posted by jduke

I do appreciate all the work AndrewA has put into this, and I know I sound like I'm beating a dead horse, but I sure do wish AndrewA would take Robert up on his return policy and try his device and then the FRK and come back to us with some kind of feedback on the plus's and minus's of one versus the other.

or...

You could always buy the FRK and ship it to AndrewA, pay him to do an analysis, then have him return it...

or since his work is spelled out here... you could do it yourself.