ADVrider

Go Back   ADVrider > Bikes > GSpot > GS Boxers
User Name
Password
Register Inmates Photos Site Rules Search Today's Posts Mark Forums Read

Reply
 
Thread Tools Search this Thread Display Modes
Old 09-28-2003, 05:18 PM   #1
Poolside OP
Syndicated
 
Poolside's Avatar
 
Joined: Apr 2003
Location: Long Beach, CA
Oddometer: 11,682
Bosch TPS Datasheet

Here you go JJ.

Bosch TPS Datasheet

- Jim
__________________

IICE Air Hotrod your GS  Fuel Injection  Tech Info  Buy  Order List  Installation

Poolside screwed with this post 05-09-2009 at 12:34 PM
Poolside is offline   Reply With Quote
Old 09-28-2003, 06:42 PM   #2
Frank Warner
Traveller
 
Frank Warner's Avatar
 
Joined: Mar 2002
Location: Sydney, Australia
Oddometer: 6,593
Thanks for that.

I've some data at the bottom of this page. it could help for those doing resistance readings.


Frank Warner
Green bottles found hanging on wall
Frank Warner is offline   Reply With Quote
Old 09-28-2003, 07:19 PM   #3
Poolside OP
Syndicated
 
Poolside's Avatar
 
Joined: Apr 2003
Location: Long Beach, CA
Oddometer: 11,682
Hi Frank. Lots of time collecting data, thanks for the effort and the link.

As far as measuring resistance; Do you mean to determine if their TPS is ok?

- Jim
__________________

IICE Air Hotrod your GS  Fuel Injection  Tech Info  Buy  Order List  Installation
Poolside is offline   Reply With Quote
Old 09-28-2003, 08:23 PM   #4
johnjen
Mod-U-Later
 
johnjen's Avatar
 
Joined: Nov 2001
Location: Seattle
Oddometer: 15,188
Yee Haaw

Way Kewl....

Where did you dig that up from???? And if that info is available what else might they have??? This is the type of documentation that is invaluable when trying to figure out just what the heck is going on our FI systems....

Thanks and can I have some more???

JJ
__________________
An illustrious quote from vermin:
"I lie to myself because I am the only one who continues to believe me."
johnjen is online now   Reply With Quote
Old 09-28-2003, 08:48 PM   #5
Ricardo Kuhn
a.k.a. Mr Rico Suave
 
Ricardo Kuhn's Avatar
 
Joined: Sep 2001
Location: Salt lake city, Utah
Oddometer: 12,201
Re: Bosch TPS Datasheet

Quote:
Said Poolside:
TPS Datasheet. Here you go JJ.

- Jim
great work mister holmes,,,,

I just hope I was able to understand any of the stuff they talk about....

I see in the future you and john J are going to become best friends
__________________
I'm looking for a cheap battery for my 11GS so I can get her running again..
Ricardo Kuhn is offline   Reply With Quote
Old 09-28-2003, 09:25 PM   #6
Frank Warner
Traveller
 
Frank Warner's Avatar
 
Joined: Mar 2002
Location: Sydney, Australia
Oddometer: 6,593
Re: Yee Haaw

Quote:
Said johnjen:
This is the type of documentation that is invaluable when trying to figure out just what the heck is going on our FI systems....

Thanks and can I have some more???
Umm

1)where are you starting from?
Have you read the stuf in the haynes manual for your bike... that should give a general how it works kinda thing... the bosh manuals are a little deeper but cost $$. And they are not model specific, so don't do a great job on the motorcycle side of things. But they do give insights on the operation of things.

AND

2) what do you want to know?

AND

3) what do you want to do with the information? Building another EFI system or even modifing the one you have is not a backyard job unless you have the skills and knowledge for it. Roadside get you to the pub (or home) is another matter.
__________________
Regards Frank Warner
motorcycles BMW R80 G/S 1981, BMW K11LT 1993, BMW K75 G/S
http://www.geocities.com/fwarner_au/
Frank Warner is offline   Reply With Quote
Old 09-28-2003, 10:27 PM   #7
johnjen
Mod-U-Later
 
johnjen's Avatar
 
Joined: Nov 2001
Location: Seattle
Oddometer: 15,188
Re: Re: Yee Haaw

Quote:
Said Frank Warner:
Umm

1)where are you starting from?
Have you read the stuf in the haynes manual for your bike... that should give a general how it works kinda thing... the bosh manuals are a little deeper but cost $$. And they are not model specific, so don't do a great job on the motorcycle side of things. But they do give insights on the operation of things.

AND

2) what do you want to know?

AND

3) what do you want to do with the information? Building another EFI system or even modifing the one you have is not a backyard job unless you have the skills and knowledge for it. Roadside get you to the pub (or home) is another matter.
Mostly the details as they apply to our version of the moronic brain.. Stuff like input signal parameters and their interactive effects on operation, acceptable voltage and current limits for sensors, decision maps, error fault codes and how to read them, consequences of sensor faults with respect to operational modes, standard setup parameters for sensors to name just a few. You know, the easy stuff that BMW had to determine in order to get the system to function at all and for when things go sideways...

I've read the Bosch FI book and it is woefully short on details. Every Haynes manual I've ever read leaves me wondering if I can really trust what it's telling me.

Every FI system has certain design parameters that once understood allows for optimizing the entire system. These days the range of optimization is much narrower due to the epa etc but it exists non the less...


JJ
__________________
An illustrious quote from vermin:
"I lie to myself because I am the only one who continues to believe me."
johnjen is online now   Reply With Quote
Old 09-29-2003, 05:49 AM   #8
Poolside OP
Syndicated
 
Poolside's Avatar
 
Joined: Apr 2003
Location: Long Beach, CA
Oddometer: 11,682
Quote:
JJ said:
I think I follow what your doing. The minimum output voltage from the TPS as used in our GS's is not zero but [some amount of] volts (as per the spec sheet).
Let's clarify what we are talking about, to be on the same page, you know? There are THREE 'zeros'. See if these descriptions work.

1. One is the mechanical zero of the throttle butterfly. With the BSS (butterfly set screw) loosened enough to let the throttle return spring 'seat' the butterfy in the throttle bore. This data point is an assumption. We do not know if BMW uses this 'seating of the butterfly' as their zero reference for their 5° setting. But we will discover that as we go.

2. Another 'zero' is the electro-mechanical zero of the TPS. This is not a mechanical stop.

3. And the third is the TPS 'zero degree output/input voltage ratio'. The voltage the TPS puts out, as a ratio of its inout voltage, when it is resting at item 2 above.



Quote:
JJ said:Also realize that as soon as you break the paint seal on the BSS any BMW dealer has no choice but to replace the TB. BMW does not provide any other option.
Yea, I hear you. Maybe this is for the broken 'telltale' crew, who want to get the TB back to square 1?

And as an objective, we are trying to allow a TB that has been misadjusted, or maybe a used one, to be set as near as practical to factory specs. Sound right?

These ok with you JJ?

- Jim
__________________

IICE Air Hotrod your GS  Fuel Injection  Tech Info  Buy  Order List  Installation

Poolside screwed with this post 09-29-2003 at 06:10 AM
Poolside is offline   Reply With Quote
Old 09-29-2003, 11:15 AM   #9
johnjen
Mod-U-Later
 
johnjen's Avatar
 
Joined: Nov 2001
Location: Seattle
Oddometer: 15,188
Quote:
Said Poolside:
Let's clarify what we are talking about, to be on the same page, you know? There are THREE 'zeros'. See if these descriptions work.

1. One is the mechanical zero of the throttle butterfly. With the BSS (butterfly set screw) loosened enough to let the throttle return spring 'seat' the butterfy in the throttle bore. This data point is an assumption. We do not know if BMW uses this 'seating of the butterfly' as their zero reference for their 5° setting. But we will discover that as we go.
It seems as though that IS what BMW is refering to in the specs. They reference the stock setting of the butterfly to 5 degrees as the 'rest position'.

Quote:
2. Another 'zero' is the electro-mechanical zero of the TPS. This is not a mechanical stop.

3. And the third is the TPS 'zero degree output/input voltage ratio'. The voltage the TPS puts out, as a ratio of its inout voltage, when it is resting at item 2 above.
The Bosch TPS spec sheet calls for a .05 voltage ratio at the minimum for both of the 2 voltage outputs. This results in .0055 volts as the minimum output as measured when the TPS is adjusted to it's 'zero' position.

Quote:
Yea, I hear you. Maybe this is for the broken 'telltale' crew, who want to get the TB back to square 1?

And as an objective, we are trying to allow a TB that has been misadjusted, or maybe a used one, to be set as near as practical to factory specs. Sound right?

These ok with you JJ?

- Jim
Getting the TB/TPS back to square 1 is only one of the objectives. As the engines age, changing the ratio of TPS voltage to butterfly angle becomes necessary in order to compensate for wear and tear of the engine. What we are in fact doing is changing the idle richness. As long as we keep the TPS voltage in the 'acceptable' range while at closed throttle and vary the amount of air we can change the overall mixture at idle. R-Dubb has let me know that this adjustment is very important in order to get the bikes to idle correctly especially as the engines pile on the miles.

Also remember that what we are fussing with here really only affects idle conditions and very small throttle opening running... And as soon as the engine is up to operating temp and the throttle is opened, the moronic brain shifts into closed loop mode and all this gets overridden as the O2 sensor kicks in.

What would be a very helpful number to know is what the factory setting of 5 degrees yeilds with respect to the 2 output voltages when the butterfly is fully closed... This is the factory bias setting... I have run this setting from 0.01 all the way up to 0.9 volts and now have a range of 'useable' bias voltages. Having the factory number would help put these into perspective.

JJ
__________________
An illustrious quote from vermin:
"I lie to myself because I am the only one who continues to believe me."
johnjen is online now   Reply With Quote
Old 09-29-2003, 06:24 PM   #10
Poolside OP
Syndicated
 
Poolside's Avatar
 
Joined: Apr 2003
Location: Long Beach, CA
Oddometer: 11,682
Great, we can align on a couple of things -

---------------------

1 - We are clear about BMW's butterfly zero reference point. And we both make the same 'prediction' regarding what we think BMW is doing. Maybe that is two things we align on.

Jim - This data point is an assumption . . .
JJ - It seems as though that IS what BMW is refering to . . .

And it's a good thing too. It is good not to 'know' before we know. The data will tell us what is what.

---------------------

2 - We have the same objective as well. Smashing.

JJ - Getting the TB/TPS back to square 1 is only one of the objectives.

Jim - And as an objective, we are trying to allow a TB that has been misadjusted, or maybe a used one, to be set as near as practical to factory specs.

I see that you and your fellow authors have a lot of ground to cover. Our 'first steps' of setting the throttle butterfly to 5°, or any other degree, by using the TPS voltage will be some of the 'procedures' you will use to reach your Grand Objective. My destination is Oz. The land of driveability. And I will surely be following your lead through many places on the way.

R-dubb may find some use for repeatably setting the throttle butterflies to 0°, 10° and any point inbetween.

---------------------

3 - And we want to know a factory setting.

JJ - What would be a very helpful number to know is what the factory setting of 5 degrees yeilds with respect to the 2 output voltages when the butterfly is fully closed... This is the factory bias setting... I have run this setting from 0.01 all the way up to 0.9 volts and now have a range of 'useable' bias voltages. Having the factory number would help put these into perspective.[/i]

That would be good to know JJ. We could test a lot of new (as in recently delivered) bikes, but it may be of no use. Many dealers (all the ones around here) set the TPS to a higher-than-factory output voltage. So if those new bikes were checked, that is what would be seen.

---------------------


Other useful info -

The the ECU is running in closed loop at idle. I have confirmed this from instrument readings. AND you can too. Confirm it yourself with your own tests. You can use either a DVM or a scope.

Measure or watch the voltage from the O2 sensor at idle. It climbs and falls in the characteristic closed loop waveform. The low voltage is in the area of 100mV, and the high side is in the area of 800mV.

To see where and when closed loop begins, the exact high and low voltage is unimportant. The voltage change cycle is the indication of closed loop ECU operation. That is what to look for.

Also, the ECU begins running in closed loop as soon as the O2 heater warms the sensor to operating temperature. And that is not dependant on engine temperature. Do the test yourself starting with a cold engine. The output does not change for a brief time. Then the sensor output begins the lean to rich cycling shortly after startup.

The ECU's lean to rich cycle time is approx 20% slower at idle. If you hold the engine at a higher idle, 3K say, the cycle time speeds up. Whenever you see this O2 voltage cycling, the ECU is running in closed loop.


---------------------

And . . .

If you will allow me, I will try to explain the datasheet a little in the next post.



- Jim
__________________

IICE Air Hotrod your GS  Fuel Injection  Tech Info  Buy  Order List  Installation

Poolside screwed with this post 09-29-2003 at 06:46 PM
Poolside is offline   Reply With Quote
Old 09-29-2003, 06:38 PM   #11
Frank Warner
Traveller
 
Frank Warner's Avatar
 
Joined: Mar 2002
Location: Sydney, Australia
Oddometer: 6,593
I think ... (so all this is just metal rambling ... no justification :): )

The brain uses the TPS and engine rpm to do a rough map of fuel to inject (along with temperature, air pressure etc)
The TPS and engine rpm give an indication of air volume (or mass) going into the engine ... as these bikes have no air flow sensor they simply calculate it from the TPS and engine speed...
The CO sensor is used to correct the rough map (thus closing the loop and giving (almost) perfect air/fuel ratio.

The rough map is only important when things (like the CO sensor) are cold .. and when that occurs you’re usually also affected by the engine temperature too ... so things are usually rich.

In short I don't know why the TPS has such a reported effect on the motors running when you have a CO sensor ... possibly the control loops PID constants are wrong ... and control things are not my cup of tea
.

Humm reading through that maybe the brains calculation of air volume (or mass) into the engine is very sensitive to the TPS being correctly set... hummm yep that would make sense ... still leave the surging down to the PID being wrong but if the TPS is more correct then the correction by the CO sensor is less so any error in that correction is also less.. if you follow.

Frank Warner
Backpacker selects Bob Marley at pub jukebox

Frank Warner screwed with this post 09-29-2003 at 06:43 PM
Frank Warner is offline   Reply With Quote
Old 09-29-2003, 09:09 PM   #12
johnjen
Mod-U-Later
 
johnjen's Avatar
 
Joined: Nov 2001
Location: Seattle
Oddometer: 15,188
Quote:
Said Poolside:
Great, we can align on a couple of things -

SNIP

3 - And we want to know a factory setting.

JJ - What would be a very helpful number to know is what the factory setting of 5 degrees yeilds with respect to the 2 output voltages when the butterfly is fully closed... This is the factory bias setting... I have run this setting from 0.01 all the way up to 0.9 volts and now have a range of 'useable' bias voltages. Having the factory number would help put these into perspective.[/i]

That would be good to know JJ. We could test a lot of new (as in recently delivered) bikes, but it may be of no use. Many dealers (all the ones around here) set the TPS to a higher-than-factory output voltage. So if those new bikes were checked, that is what would be seen.


JJ response in BOLD

The range of acceptable TPS bias voltages for the ECU at idle runs from .300 (+/-) to just under .400 volts. Anything outside this range will be considered a fault by the ECU. I am not sure what the ECU does about this fault because it can't ignore the TPS voltage. My bike stock was set to .315 volts and I had them set it nearly as high as is possible and the voltage was .390. As long as the ECU is happy with the TPS voltage at idle, the butterfly opening can be 'adjusted' to vary the idle mixture. This, in a back handed way, (via the 'correct' bias voltage) is one way of determining what the butterfly angle is set to, in terms of rideability (not measureable degrees of butterfly angle).



Other useful info -

The ECU is running in closed loop at idle. I have confirmed this from instrument readings. AND you can too. Confirm it yourself with your own tests. You can use either a DVM or a scope.

Measure or watch the voltage from the O2 sensor at idle. It climbs and falls in the characteristic closed loop waveform. The low voltage is in the area of 100mV, and the high side is in the area of 800mV.

JJ response in BOLD

How do you 'know' this is closed loop mode by just observing the O2 sensor? My understanding of this sensor is it will generate voltage in a linear relationship to the amount of the 'target' gas present at the sensors tip. This voltage is generated regardless of whether the ECU is paying any attention to the voltage or not. The response time of the generated voltage is measured in ms. (Can you get the spec sheet on this O2 sensor?)
Once the O2 sensor is warmed up the voltage generated will follow the gas present at the tip in a cyclic manner, since the gasses pass by the tip in just such a manner (and which speed up as the rpm's rise). In short just because the O2 sensors output is readable does that necessarily mean that the ECU HAS to pay attention to it?



To see where and when closed loop begins, the exact high and low voltage is unimportant. The voltage change cycle is the indication of closed loop ECU operation. That is what to look for.

Also, the ECU begins running in closed loop as soon as the O2 heater warms the sensor to operating temperature. And that is not dependant on engine temperature. Do the test yourself starting with a cold engine. The output does not change for a brief time. Then the sensor output begins the lean to rich cycling shortly after startup.

The ECU's lean to rich cycle time is approx 20% slower at idle. If you hold the engine at a higher idle, 3K say, the cycle time speeds up. Whenever you see this O2 voltage cycling, the ECU is running in closed loop.


---------------------

And . . .

If you will allow me, I will try to explain the datasheet a little in the next post.

JJ response in BOLD

Yes indeed, lets see if we both 'see' the same thing.



- Jim
__________________
An illustrious quote from vermin:
"I lie to myself because I am the only one who continues to believe me."
johnjen is online now   Reply With Quote
Old 09-29-2003, 09:56 PM   #13
johnjen
Mod-U-Later
 
johnjen's Avatar
 
Joined: Nov 2001
Location: Seattle
Oddometer: 15,188
Quote:
Said Frank Warner:
I think ... (so all this is just metal rambling ... no justification :): )

The brain uses the TPS and engine rpm to do a rough map of fuel to inject (along with temperature, air pressure etc)
The TPS and engine rpm give an indication of air volume (or mass) going into the engine ... as these bikes have no air flow sensor they simply calculate it from the TPS and engine speed...

JJ sez
This could also explain why they used the dual output TPS. It's output has much better resolution from 0 to 23 degrees of rotation, right where you need it the most.


The CO sensor is used to correct the rough map (thus closing the loop and giving (almost) perfect air/fuel ratio.

The rough map is only important when things (like the CO sensor) are cold .. and when that occurs you?re usually also affected by the engine temperature too ... so things are usually rich.

In short I don't know why the TPS has such a reported effect on the motors running when you have a CO sensor ... possibly the control loops PID constants are wrong ... and control things are not my cup of tea
.

JJ sez
Open loop mode (when the engine is cold or a sensor has faulted etc) MUST use the TPS voltage, even if it's out of range, since it's the ONLY input relating to desired power control. As such it just might have an underlying 'master control' function that might even influence the pulse duration of the injectors when in closed loop mode. My hunch is that it acts as an overall fuel richness bias, which is further dialed in by the O2 sensor (or not depending on which mode the engine is running in, open or closed). This (at least to me) explains why we do observe different running conditions depending on where the TPS is set at idle. Of course this is all just speculation on my part and it's also why I'd really like to get a hold of all those operational parameters I listed previously...

Ahh, what is PID? Is it the open loop operational data map?


Humm reading through that maybe the brains calculation of air volume (or mass) into the engine is very sensitive to the TPS being correctly set... hummm yep that would make sense ... still leave the surging down to the PID being wrong but if the TPS is more correct then the correction by the CO sensor is less so any error in that correction is also less.. if you follow.

JJ sez
Could you state that in a different way? I'd just like to make sure I understand it fully...

Also, for both Frank and Jim,
Have you noticed that as the engine decelerates it seems to jump from one 'map' to another? When I've noticed it the engine seems to slightly change the amount of engine braking and it occurs at relatively low rpms (say 1.5 to 2.5K) It doesn't manifest all the time but it does seem like the ECU is shifting from one 'map' (closed loop) to open loop as the rpm's approach idle. Have either of you noticed this?

Also, after doing the work on valve settings, TB synching and additional tweaking, I think that surging is an interactive combination of multiple factors that can be reduced to insignifigance, by careful adjustment of the parameters.


Frank Warner
Backpacker selects Bob Marley at pub jukebox
__________________
An illustrious quote from vermin:
"I lie to myself because I am the only one who continues to believe me."

johnjen screwed with this post 09-29-2003 at 10:02 PM
johnjen is online now   Reply With Quote
Old 09-29-2003, 11:47 PM   #14
fritz
Adventurer
 
Joined: Sep 2001
Location: Karlsruhe, Germany
Oddometer: 39
Quote:
Said johnjen:
How do you 'know' this is closed loop mode by just observing the O2 sensor? My understanding of this sensor is it will generate voltage in a linear relationship to the amount of the 'target' gas present at the sensors tip. This voltage is generated regardless of whether the ECU is paying any attention to the voltage or not. The response time of the generated voltage is measured in ms. (Can you get the spec sheet on this O2 sensor?)
Once the O2 sensor is warmed up the voltage generated will follow the gas present at the tip in a cyclic manner, since the gasses pass by the tip in just such a manner (and which speed up as the rpm's rise). In short just because the O2 sensors output is readable does that necessarily mean that the ECU HAS to pay attention to it?
Doesn't the fact that the sensor output is cycling prove that the ECU is reacting to it? If it were not, the output should remain constant.
__________________
Michael Neudatschin
'97 R11GS @75k km
'89 XT600 @85k km
'82 CB900 @150k km
Karlsruhe, Germany
fritz is offline   Reply With Quote
Old 09-30-2003, 12:31 AM   #15
johnjen
Mod-U-Later
 
johnjen's Avatar
 
Joined: Nov 2001
Location: Seattle
Oddometer: 15,188
Quote:
Said fritz:
Doesn't the fact that the sensor output is cycling prove that the ECU is reacting to it? If it were not, the output should remain constant.
No.

The sensor voltage is cycling because the exhaust gasses that trigger the sensor are only present (in quantity) during the exhaust phase of the engines rotation. That's why the sound of the exhaust goes bup, bup bup...

And even IF the gasses were constantly passing by the sensor the ECU couldn't and shouldn't affect the sensor output. If it did the sensors voltage wouldn't be useful because the precise relationship between the amount of the trigger gasses and the output voltage would no longer exist. That is the LAST thing the ECU would need, inaccurate input from it's sensors.

AND the sensor is feeding a data INPUT to the ECU NOT an output from the ECU.

In other words, an ECU has to assume that the data being received via an input port is accurate and reliable. The programming can setup out of range parameters that will tell the programming that the sensor is 'bad' and it will either ignore the sensors input or in other ways not fully trust the sensor. This is in part why the ECU has an acceptable range of input voltages for the TPS at idle, if the TPS is outside the 'good' range it 'knows' there is a problem and acts accordingly.

JJ
__________________
An illustrious quote from vermin:
"I lie to myself because I am the only one who continues to believe me."
johnjen is online now   Reply With Quote
Reply

Share

Thread Tools Search this Thread
Search this Thread:

.
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is On

Forum Jump


Times are GMT -7.   It's 01:29 PM.


Powered by vBulletin® Version 3.8.5
Copyright ©2000 - 2014, Jelsoft Enterprises Ltd.
Copyright ADVrider 2011-2014