TPS Alignment, Idle Speed Adjustment: Beyond Zero=Zero

Brian, Thanks for the feedback. Robin also mentioned the sensitivity of the TPS when removed. I'm traveling but after I'm back will make some measurements and see if there is an easier method. Glad though that it has worked out. RB

I got out of work and did a run over the southern mountains yesterday. I believe I experienced surging on the 110Km/Hr Hiway run home from the Gold Coast.

I don't want to look into it further just yet, I now have a line on an R1150RT with under 50K Kms on it. That is a rare bike down here. There may be a deal to be done.

I got out of work and did a run over the southern mountains yesterday. I believe I experienced surging on the 110Km/Hr Hiway run home from the Gold Coast.

 

I don't want to look into it further just yet, I now have a line on an R1150RT with under 50K Kms on it. That is a rare bike down here. There may be a deal to be done.

Which 1150? what year?

2002 R1150RT silver, but it's 65K Kms. AU$7K.

I got out of work and did a run over the southern mountains yesterday. I believe I experienced surging on the 110Km/Hr Hiway run home from the Gold Coast.

Afternoon Robin

Surging is usually noticed at much lower speeds & very light throttle opening 3-4K RPMs (very light loading)

If you have surging at hiway speeds/hiway load then it is probably something else (like fuel starvation) or possibly a wind management issue.

Cheers DR, I did know that but when your test riding you get really picky

It's faultless round town and rides better than the R1200RT-P I test rode today.

Comment on Earlier Posts

In the earlier posts in this thread there are mistakes or omissions that I have found. The biggest error is the conventional understanding of what rotating the throttle does after a Motronic reset and recal. As I mentioned, this is a work-in-progress. I have a new recommendation for zero=250 later in the post.

New Measurements

I spent a few hours yesterday making measurement of the TPS and Motronic for small throttle angles and found quite a few interesting things. Among the things I learned (and although I can see vestiges of the factory paint), I see that someone performed zero=zero on my R1150 since there is no fully intact paint. The BBSs are about equal so I'm not too concerned but will go back and reset it at some point.

To make the measurements, I attached a good DVM between pins 1 and 4 or the TPS and connected the GS-911 to the diagnostic port. This way I could measure the voltage change versus throttle angle change as reported by the Motronic. I was careful to move the TPS very slowly (by rotating it with the throttle lock on), and find the trip point for each reading of the GS-911. It outputs angles with a resolution of 0.32 degrees. So the angles it registers are: 0, 0.32, 0.64, 0.96, 1.28, 1.92, etc. I also monitored the Idle Switch Off indicator reported by the GS-911. Lastly, I measured (and then calculated to confirm) how much physical rotation of the TPS per 10 mV (it's tiny) of voltage change on pin 1 (the high resolution wiper).

Here's what I've found:

1) TPS cal (rotating the throttle) does not have any effect on the voltage-to-throttle angle as measured by the Motronic. This was a surprise because we all "believed" that TPS cal (rotate the throttle twice) learns the idle and full scale positions. I set the TPS as low as 300 mV and as high as 375 mV, performed a full reset (10 minutes) and recal. In spite of that, the point at which the Motronic/GS-911 transitioned from reporting 0.32 degrees to 0.64, 0.96 and the point at which the Idle Switch reported OFF were the same whether I set the TPS to 300, 350 or 375 mV.

It now seems clear to me that the reason for setting the TPS is to position it to 0.48 degrees, which is always approximately 340 mV to 350 mV.

2) The A/D converter in the Motronic is 8 bits (expected) and the step size for each "bit" of the A/D is approximately 20 mV. The Motronic reported step sizes are 0.32 degrees, which equals 60 mV, which means each bit of the A/D is about 0.11 degrees (20 mV). That means one degree is about 200 mV (187.5 to be exact). So the angle to voltage conversion measured (and then calculated by me) below. Because the A/D has some tiny fluctuations (noise) these values can change by 10-20 mV.

0.00 degrees: 250 mV to 309 mV

0.32 degrees: 310 mV to 369 mV

0.64 degrees: 370 mV to 429 mV

0.96 degrees: 430 mV to 490 mV

Idle Switch Off: ~430 mV

3) Using a feeler gauge under the throttle stop, I measured that every 0.1 degree of TPS rotation is about 0.0025" of movement at the TPS locking screw. This means a 10 mV change in DVM reading is the result of about 1-2 thousandths of an inch of movement!

4) On my R1150 the A/D "noise" at small throttle angles seems very good, less than one bit. That means that sometimes the transition for, say, 0.96 degrees is at 410 mV and sometimes it is at 430 mV. 0.96 degrees is the point that the Motronic reports Idle Switch: OFF.

5) If you want to readjust the TPS and throttle stops (because someone else changed them) zero=zero is the wrong procedure. It certainly leads to the throttle angle being opened too much (10 mV at zero degrees, then to 370 mV, which means the throttle is opened 1.92 degrees). Based on a 340 mV target for the TPS (confirmed with the GS-911 and equal to the mid-point of 0.32 degrees), the throttle is meant to be opened about 0.48 degrees and the rest of the air comes from the BBS. (I could be off a bit on the 1.92 degrees since there is no spec for angular rotation below 250 mV on the Bosch TPS.)

Looking at the TPS data sheet, I have an explanation for why the requirement to rotate the throttle twice after a Motronic reset on the R1150 (the R1100 may be learning it on-the-fly, or BMW may have fixed a problem by introducing the procedure on the R1150), since it doesn't seem to calibrate the end point. There are two potentiometers in the TPS. The fine-tps (0-23 degrees throttle) has it's upper and lower points specified. TPS-coarse has only it's upper point spec'ed. That means that rotating the throttle allows the Motronic to measure and learn the crossover point and the start of tps-coarse. It may be doing this by using the fine-tps to measure the starting point of the coarse-tps, thereby having a seamless crossover between the two.

Also, I now also can see why moving the TPS idle setting to 370-385 mV can affect engine operation, at least temporarily, especially on the no-cat configuration of the R1100. At that voltage, the Motronic is just starting to register 0.64 degrees, so the Motronic then goes up to the next 0.32 degrees more quickly. This would give a slight rich fueling bias for throttle angles below (I'm estimating) 5 degrees. I doubt that it is a permanent benefit but could be, very slightly around idle. (Going a little further with the reasoning, since zero=zero opens the throttle too much, it makes sense that a higher TPS voltage would get the TPS reading by the Motronic closer to the actual throttle-angle-plus-BBS air flow. On the Closed Loop Oilheads, Adaptation Values will fix this up, but as you know, adaptation is not perfect. Therefore, it is better to get it right from the start.)

Recommendation

The procedure I suggested earlier in the thread for restoring the throttle stops and TPS will work but now seems too complicated. I also realized you can set the TPS with voltage, something I didn't think was possible because of the conventional-wisdom of what rotating the throttle twice did after reset. I still believe that the GS-911 or LED (for R1100) is the preferred route for setting just the TPS because it is a Closed Loop process that includes the Motronic.

Although I don't advocate changing the throttle stops or TPS, for someone (like me?) who has had their TPS Lentini-ized by zero=zero and wants to get it set correctly, a zero=250 procedure can be used.

Begin by setting zero degrees throttle to 250 mV (instead of 10 mV recommended in zero=zero). Then the left throttle should be rotated to 340-350 mV, which will rotate the throttle open 0.48 degrees. Then just follow the rest of the Lentini procedure. That will undoubtedly require the BBSs to be opened more than two turns. I will experiment with this on my bike to see how many turns are required when I have time.

Next Tests

My next experiments will be to close the BBSs and see how much the throttle needs to be opened to idle at 1100 and at an AFR of 14.7 to 1. Then I will see how many turns the BBSs can be opened before the RPMs stop increasing.

Roger, I got my sight window and fresh oil installed and gave the R1100RT a bath.

 

The idle LED is ON only between .327mV and .386mV as measured with my Fluke meter on my TPS. Getting the LED to light by manually adjusting the wiper on the TPS is a real challenge but can be done. I burnt up a lot of time figuring out there is a range involved and without a meter on the TPS, landing the wiper within that range will be very difficult, increments are tiny.

 

I have the bike synced and idling at 1100 rpm but it's coughing randomly, not what I'd call a smooth idle. But it's almost beer O'Clock here and I have to go out tonight so it's all stop till tomorrow.

With the recent measurements in my post above, I've concluded that the MA 2.2 and MA 2.4 behave the same. Each has a nominal setting of 340-350 mV for center of the 0.32 degree range. Here are same measurements taken on Robin's R1100:

"The idle LED is ON only between .327mV and .386mV as measured with my Fluke meter on my TPS."

This is essentially the same range that I measured on an R1150 the other day.

That means that the LED confirms that you are within the proper range. Note that it is almost exactly 60 mV wide, the width of 0.32 degrees.

I can't think of a test to confirm that the function of TPS alignment on the R1150 is to align the two potentiometers inside the TPS module, but it is my best explanation, given that: a) TPS alignment doesn't affect the bottom of the range, and b) that the exact angle at WOT is not as critical.

The R1100 (without the procedure to twist the throttle twice) either a) has no such alignment or b) does it automatically as and when the throttle moves through the crossover point. I don't think we'll ever know.

Some further observations:

Zero=zero opens the throttle vane 1.92 degrees when the lefthand stop screw is adjusted from 10 mV to 250 mV, however, the Motronic only reads 0.32 or 0.64 degrees. Then if you open the throttle another 1.28 degrees, the throttle vane is actually open 2.20 degrees but the Motronic reads 1.6 degrees. In other words your Motronic is always reading incorrectly lean.

Zero=250, as described above results in the actual throttle angle being the same as the measured angle as read by the Motronic, neither lean nor rich.

Ok, so I got a left TB from n5692g that matched my right TB except it was a dual cable setup. Swapping parts was pretty easy, just time consuming since I wanted to inspect things as I went along.

Once installed I went about setting things up.

1. Set zero to 250mv. My original settings definitely had the throttle way to open.

2. adjust the stop screw on the left TB until the LED just turned on. I went back and forth a couple of times and found there is quite a bit of hysteresis in the measurement compared to how far I could turn the stop screw before the LED would change.

3. Measured voltage with LED on and it was ~340mv.

4. Right TB I backed off the stop screw until it was off the throttle then just turned it in a little bit to just open it.

5. Started up the bike with the air intakes off so I could easily get to the stop screws.

The motor actually ran pretty smooth right off the bat and idle was about 900. No sputtering as I've had since I bought the bike.

After warming up I shut it down to attach my manometer. Sync was off by about 12" with the left in the lead so I backed off the stop screw on the right and got them pretty even. I fine tuned it with the BBS.

Next took the engine to 2500-3000 rpm and adjusted the right cable and got them within about an inch. See the attached video link. I'm letting the bike cool down so I can reassemble the tupperware then take her for a ride. Idle is around 1000-1050 now. MUCH better than before at 1200-1300.

As I started writing this I realized I didn't do one step which is remove the TPS, set it to the LED on, then with the bike running adjust the TB stops, but I don't think I need to do this now.

Evening Tom

Before buttoning up MAKE DARN SURE that BOTH side throttle cams lift off their idle stops at exactly the same time & both hit their wide open stops at the exact time (that is a good verification that throttle plates are tracking together)

So, that's what these bikes are supposed to ride like!!! Much smoother! Also, there was an on/off feeling to the throttle movement and that seems to be gone. Very smooth going from off throttle or light throttle to acceleration. Only complaint is it gets buzzy above 5000 rpm. Very happy right now though.

DR, good advice and I did check. As far as I can tell they are exact or damn close. When I was opening the throttle during the sync the vacuum between them tracked pretty well, with the right leading by maybe 3 or 4 inches and once at rpm it settled to within 1-2".

Tr,

Glad you got a good result and thanks for the feedback and detail. How many turns out were your BBSs? Another rider found that even with zero=250 his were only open about 2 turns.

In my revised procedure, I no longer recommend that the TPS be removed. Just close the throttle to zero degrees, set TPS to 250, mV, use left adjuster to 340-350 mV.

I worked on my TPS the other day and noticed the backlash too, about 15 mV. I found that I got a stable final result by adjusting the final TPS setting with the engine idling. I suppose it is due to shaft wear.

Roger,

Your information was extremely helpful in all of this. My final settings are ~2.5 turns left and about 3 right. I can stand to fine tune it and will probably do so after riding it awhile. When I checked the TPS voltage it was about 335-340mv with LED on. I did notice the LED was not on and would not come on when it was running. Maybe just a lousy connection with my setup, but it seemed odd.

I'll take the bike on a long ride to work tomorrow so we'll really see how it's running and any difference in seat vibration.

More good news. Rode to work this morning (>100miles btw) and the bike was actually a joy to ride. Very smooth with no more abrupt transitions off/on, on/off throttle. Deceleration is much easier to control now and I can enter a turn and open the throttle and have smooth power. Very happy now!

Happy to hear the TB worked out....and BTW... Thanks!

Another "thanks!" Roger -- I used your zero=250 approach with good results.

When I took my RT-P apart, I re-worked the throttle bodies and installed Dan Cata's replacement shaft bushings. For a variety of reasons (one of which being that the bike's original throttle bodies' stop screws had clearly been messed with by the prior owner), when I reassembled the bike I used another pair of throttle bodies just to get the bike on the road for its initial "shakedown" runs.

Today I re-installed the bike's original "Dan Cata'd" throttle bodies. Knowing I could not trust any of the settings, I backed off the left throttle body stop screw, set the TPS to 250mv, then advanced the stop screw until the TPS showed 370mv. On the right side, knowing the thread pitch of the stop screw and the target "starting" angle of .32 degrees, I calculated that the stop screw needed to be advanced from initial contact by 1/3 rotation (two facets of the hex). I also set the BBSs on both sides to 1 1/4 turns out.

Turned out these were quite good initial settings. After starting and holding the throttle open until I reached 5 bars (with a fan in front of the bike), when full warmed the bike would idle just below 1000 rpm, with my CarbTune showing near equal balance. I advanced the right stop screw just a hair to reach L/R balance, then brought both BBSs out another 1/2 turn to bring the idle up to 1050-1100 rpm.

Sync'd the throttle cables and set the throttle and fast idle cables' slack, and at least in the driveway she's running nicely. I hope to get out this afternoon to see how she does on the road.

Again, thanks -- without the zero=250 starting point, I would still be chasing my tail trying to figure out why zero=zero wasn't working.

Follow-up --

I took the bike out for a several mile ride, local and highway portions, as well as sitting at several lights idling.

Major difference! The frequent, irregular miss at idle with the other throttle bodies is gone. No hiccups anywhere from idle to 6K, accelerating and steady throttle.

I'm going to put the body panels back on and run it for a while. When I'm satisfied that she's in good "stock" running order, then I'll purchase an AF-XIED. I am confident enough that I have already moved the O2 sensor connector to a place I can reach it to install the AF-XIED without removing the tank again :

Here is a post from another forum, detailing a re-alignment of TPS and throttles on a 2002 R1150GS that is long but interesting.

Warning: this is a very long post!

 

Reader’s Digest version – my R1150GS throttle body idle position stop screws and TPS mounting screws had little or no blue anti-tamper paint on them when this bike was acquired. Several recommended methods for resetting these stop screws and TPS to their OEM positions were reviewed and many trial and error settings were tested. Best results of all were achieved using the Zero = 250mV method as described in an earlier post of this thread.

 

Beginning point: new (to me) 2002 R1150GS, 57K miles at acquisition time, all stock except no canister, with 4 prior owners. Acquired and rode it 3300mi to WA State Sep ’14 and made this 5-day trip without issues, avg mpg 38. General running condition: OK enough to make this trip, but not great… It felt like the engine needed a good tune as a minimum.

 

A winter project was undertaken to restore this bike to a proper and reliable running condition. The work that affected state of tune included valve clearance adj; rocker end play adj; replaced air filter, fuel filter, and all in-tank hoses and clamps; cleaned the fuel tank internals, air box, intake tubes, and engine intake ports interiors; fuel injectors sent off for professional cleaning and flow test; disconnected and lubed all electrical connections; and the engine compression, fuel system pressure and fuel flow were verified to be within spec.

 

During prior attempts to balance TB vacuum at idle using a TwinMax the BBS adj seemed vague and would not steady up at a balanced position. Also, because both TB shafts were rattling, the TBs were removed for thorough cleaning and to install Dan Cata TB rebuild kits. Found a lot of thick built up crud in the BBS ports that was likely part of vacuum balance problem. Original BBS setting position was almost closed. Also found and removed a small rock imbedded in the groove of the LH TB cable pulley that essentially shortened the LH throttle cable. No blue paint was evident on the LH TB pulley stop screw/lock nut and only a remnant on the RH TB stop screw and bracket. With no continuously intact OEM anti-tamper paint on the stop screws, the likelihood was that a previous owner or independent shop had adjusted them from OEM idle position. The BMW repair manual does not address adjustment of the throttle body idle stop screw setting and for the TPS setting refers to the use of the “BMW MoDiTeC” for setting after installation. The parts fiche for this bike shows the throttle body already populated with the butterfly, shaft, return spring, stop screw, etc as a single OEM part number and the nearest BMW shop advised that they replace TBs that have had the OEM stop screw positions altered before they’ll work on anything related to the TBs. Not wanting to go that route, forums were queried for information.

 

After much reading on this topic in the various BMW MC-related forums, several relevant threads, including this “beyond zero=zero” thread started by Roger were found. Having then read all the threads and articles I could find pertaining to setting the TPS and stop screws correctly for idle position, it was evident that virtually all recommended methods involve the following steps:

1)fully close LH TB butterfly verifying no contact at stop screw.

2)using a good digital volt meter (DVM) attach its positive (red) lead to a jumper wire connected to TPS electrical connector terminal 1 and attach its negative (black) lead to one of several varying recommendations for ground connection. Some say TPS terminal 3, some say TPS terminal 4, some say LH TB ground wire, some say engine ground, etc.

3)rotate TPS until an indicated TPS voltage of a particular value corresponding with the butterfly in the fully closed position (zero position) is read on the DVM and tighten the 2 TPS mounting screws. This is the second area of varying recommendations but the smallest recommended TPS voltage for fully closed butterfly position (zero position) is 10mV and the highest is 250mV.

4)while counting screw turns, rotate the LH TB cable pulley stop screw until a certain TPS indicated voltage for the butterfly in the idle position is read on the DVM and then tighten the locknut. This is another area of varying recommendations but the smallest recommended open position is 100mV higher than the zero position voltage and the highest recommended open position is 400mV higher than the zero position voltage.

5)Using the same number of screw turns as in step 4, set the RH TB pulley stop screw the same number of turns.

6)Reassemble everything, verify all cables have slack and cable sleeve tips are in their adjuster sockets.

7)Start engine, warm to 4 bars and complete vacuum balance at idle and at 2500 rpm.

 

When disassembling the throttle bodies I began with the RH TB and found a remnant of blue paint on the stop screw, lock nut and bracket that suggested it may in fact be where the screw had originally been positioned by OEM. The paint was cracked and broken in places but, again, it looked like it aligned so this position seemed like a good baseline for where the stop screws should be set. By counting the 2+5/6 ccw rotations until no contact with the stop and by measuring the clearance from butterfly to bore at top of TB bore (0.003-in), there were 2 repeatable references for restoring these settings. Further, the amount of butterfly movement from fully closed position to this 2+5/6 turns of the stop screw idle position was measured at the top center of the butterfly by dial vernier. From the top center of the intake end of the throttle body to the top center of the butterfly, from fully closed to idle position the butterfly movement measured .041-in. Using an online angle calculator the open angle at idle position was ~2.65 deg from zero position (fully closed). Interestingly, when I later disassembled the LH TB, which had no blue paint evident anywhere, the stop screw turns, and butterfly clearance to bore were the same as the RH TB so this was more evidence that this was where the stop screws were supposed to be for idle position.

 

After thorough cleaning of the throttle bodies Dan Cata TB rebuild kits were installed without difficulty. Very complete kits by the way. For those considering rebuilding their bike’s TBs, these kits are recommend. Because it wasn’t clear whether or not the idle stop screws and TPS had or had not been tampered with, after the TBs were fully assembled, experimentation began with different butterfly idle positions using the LH TB, beginning with the stop screws and the TPS in the same setting position as when disassembled, with the TPS connected to the bike’s connector. Specifically, butterfly movement compared to TPS voltage output within the above mentioned range of recommended idle positions beyond the fully closed position was looked at using the various recommended TPS voltage settings for the fully closed position. It was found that with the TPS mounted where it had originally been positioned, with the butterfly fully closed (stop screw backed out and not in contact with the pulley stop) the TPS voltage was 6mV (as low a reading as it would go without forcing it). After adjusting the butterfly to the original idle position with the stop screw, it was evident that the butterflies on this bike had been far more open at idle position than even the 400mV highest in the range of recommendations! Now what?

 

Knowing that this bike ran well enough to make short work of a 3300 mile trip across country without issues, I was convinced the correct setting of the stop screw for idle position had to be further open than any of the recommended setting methods indicated. So, I decided to start where the TPS and butterfly idle positions had been when first disassembled, and where the BBS had been set, which was less than 1/2 turn. After assembly it started and an idle vacuum balance was attempted, noting that the BBS adjustment now had some affect but seemed to have little effect in terms of magnitude of change. It would take a good half-turn to show TwinMax needle deflections. Also, the engine was idling at much higher rpm than before (about 1600rpm) and the BBS had little effect because of the amount of air coming around the butterflies. To reduce the idle speed I knew the butterflies would have to be closed more at idle position but didn’t know by how much.

 

It was fairly clear by then that the OEM TB adjustments had been altered, possibly to compensate for poor injector performance and restricted air intake. The pre-cleaning flow test and spray patterns on the injectors were indicative of clogging in the injector upstream filters and the injector flow rates and spray patterns were outside of 15% of equal for the two. The post-cleaning results were well within spec and within .05% of equal. Also the installed air filter when acquired was a K&N that had been too heavily oiled. The combination may have led a prior owner or independent shop to attempt adjusting the TBs to overcome the operating conditions caused by these component problems. As mentioned earlier, part of the winter project work included professional injector cleaning and replacement of air filter with a Mann dry paper element filter and now the idle rpm was ~50% too high at the old settings.

 

Further trial adjustments were made with gradually increasing TPS voltage at fully closed butterfly and gradually reduced butterfly idle angle using the stop screw but whenever the butterfly idle position was higher than 400mV above zero position, idle was still too high and BBS had little effect on lowering idle rpm until they were nearly closed. More trial adjustments were made varying the TPS voltage output with the butterfly in gradually less open positions from closed position.Eventually, with the idle stop screws turned 1-1/2 turns, the TPS voltage was adjusted to an indicated 350mV and the engine started and appeared to be running much better and the idle speed could be controlled by BBS adjustment. I thought I had it at least close to right. I borrowed a GS-911 and ran a cold start real time data log and sent it to Roger for his review and comment, fully expecting he’d agree that I had it right… Wrong!

 

There were several line items of data in the cold start log that I didn’t at the time either fully understand or appreciate the significance of that indicated problems. After Roger’s gently stated, “not even close” reply I considered options and decided to use his recommended butterfly idle position of 100mV open beyond 250mV set at fully closed butterfly, which was both the maximum recommended value of TPS voltage setting for fully closed position and the minimum recommended value of TPS voltage change from fully closed to the idle position. I dreaded the thought that a lot of time would likely have to be spent to zero in on the correct values with several iterations of changing these settings to somewhere between where Roger’s method would have these settings and I where I had thought it was close to right. Wrong!

 

The TPS and stop screws were set as Roger had recommended. His method has you connect DVM red lead to TPS terminal 1 and black lead to TPS terminal 4 using jumper wires. Turn on ignition and ensure the LH TB butterfly is fully closed before rotating the TPS to an indicated 250mV reading, and then tighten the TPS mounting screws to lock in place. Then, while counting turns, adj the stop screw to an indicated 350mV and tighten its locknut. Then turn RH TB stop screw the same number of turns and lock in place, followed by setting both BBS to 2 turns open before starting engine.

 

When resetting the LH TB, the stop screw “turn count” to attain a TPS voltage of 100mV beyond the fully closed position voltage of 250mV was just over 1/2 turn (~45deg past 1/2 turn to be exact or 225 deg cw rotation from first contact with the stop). Wanting another way to confirm that the RH TB was set equally, a dial vernier was used to measure the distance the butterfly moved from fully closed to idle position, as done during initial disassembly to determine where to put them back to the old settings. Total movement now was only 0.0060-inch, which equates to an open butterfly angle at idle of only 0.388deg or ~15% of the old angle at idle! With a bright light shining in one end of the TB just a very thin ring of light could be seen all around the butterfly. The RH TB was then set to the same small “thread count” and the same 0.0060-inch difference was measured between fully closed and idle positions. At the time I thought this barely open throttle position at idle was not going to be enough to even allow the engine to start. Wrong!

 

After reassembly all cable slack was verified (this is very important to do), the start button was pressed and it started! It was idling at only 800-850 rpm with no throttle assist but very smoothly. The former fairly significant engine vibration at idle was completely gone. My fast idle lever had taken a dump somewhere during the prior runs and I didn’t want to take the time at that point to look into it. An aftermarket hand throttle stop “cruise control” was therefore engaged to hold rpm at ~1500 for warm up, adjusting the throttle downward once or twice as the idle speed increased during warm up to 4 bars. The GS-911 data logging failed to record on this first run with new TPS/stop screw set points but the idle vacuum balance with BBS (which were now quite sensitive to adjustment), and for higher rpm with RH TB throttle cable adj (which also was now sensitive to adjustment) was done using TwinMax. Then, some balancing of vacuum during throttle transitions was done using both BBS and RH TB cable adj, followed by both idle and 2500 rpm balance being tweaked back to center with minor adj. During this vacuum tuning process the GS-911 real time data was running and I noticed the indicated throttle angle was flipping from .32 to .64 degrees. TPS mounting screws were loosened a bit and light taps on the TPS mounting flange nudged the voltage reading from 350mV to 340mV at idle. The GS-911-indicated throttle angle was then holding steady at .32 degrees. The idle rpm was still a bit low after warm up so both BBS were turned small and equal amounts open to increase idle until 1100 +/- 50 rpm was steady with vacuum balanced. The engine was idling very smoothly and there were no exhaust burbles or popping during throttle down transitions. Right on!

 

After a cool down period assisted by fans the engine was started again and a cold start data log was taken, the data file was sent to Roger and I went for a ride. The before and after change in engine feel and performance was pronounced. Throttle response was now very “lively” on both acceleration and deceleration. The “seat of pants” feel was more power available and more quickly in a broader range of rpm. Specifically noteworthy was the pull that was now available at lower rpm in the higher gears. On hard acceleration the front suspension now extended near fully and getting up to speed quickly on freeway onramps was no longer a challenge. Also noted was dramatically reduced vibration throughout the rpm range. Right on!

 

The only “negative” was the hunting or surging felt in 3rd gear doing ~35mph at ~3000rpm. This surging had been there before but now it was more pronounced. I’ve since acquired an AF XiED to hopefully diminish this surging and will install it sometime soon.

 

After Roger’s positive review of the cold start data log, success was declared, the tank was topped off and I took a long ride, some county roads, some changing elevations, some freeway, etc. The Motronic had been reset and I was being attentive to getting as much steady throttle time in varying rpm ranges and varying gears to support the adaptive learning process of this ECU. The delayed drop in fuel level bars suggested increased mileage and the low fuel warning light didn’t come on until 235mi on the odometer. Previously, the warning light was coming on no later than 190mi. The calculated mileage on this first full tank was 8mpg better than prior average. It will be interesting to see if this holds up. The only place imaginable the excess fuel could have been going prior to this readjustment of the idle position and TPS is out the pipe. The spark plugs hadn’t shown fat running indication but I had noticed a fairly dense black soot coating at the muffler’s exit port before I cleaned it.

 

So, my conclusion after this trial and error process of setting the throttle stop screws and the TPS position is that Roger’s zero=250mV method works and works well. Had I ignored that this bike ran well enough to cross the country without problems and ignored its TPS and stop screw settings as they were when it did that, this would be a much shorter post! Had I followed this zero=250mV method for the first trial set up, that would have taken maybe an hour or so and the tuning process after initial setup would have taken maybe 30 minutes. I follow Roger’s deduction that zero=250mV of TPS rotation with butterfly fully closed (zero position), and agree. I also follow his calculation process that led him to recommend a stop screw setting for butterfly position at idle of 100mV of TPS indication beyond the 250mV setting at zero position (TPS voltage at idle should be 350mV), and agree. However in this case, as mentioned above, 90mV over zero position (TPS voltage at idle was changed to read 340mV by slightly rotating TPS and tightening mounting screws) kept the GS-911-indicated throttle angle steady on 0.32deg. Note that resetting the TB stop screws to a TPS indicated 340mV at idle position is likely the more correct way to move off the flip point between .32 and .64deg. I just took a “short cut” and moved the TPS a tiny bit to read 340mV. Regardless this probable transgression (sorry Roger), my ’02 R1150GS engine is running smoother and performing better than it may every have been, and that’s enough validation of this zero=250 method for me.

 

I’ll add here that my nearly 40 yrs of experience, mostly successful, in tuning multi cylinder, multi carburetor motorcycle and auto engines was of little use toward trying to figure out how to tune an engine where you have no adjustable control of fuel flow and no tech manual prescribing the settings for the TPS and the throttle stop screws. Setting the TPS to the correct voltage output to correspond with the correctly set mechanical butterfly angles at both fully closed and idle positions is critical for this engines ECU to determine the timing and duration of the appropriate fuel delivery signal to the injectors. There’s little doubt that the TB stop screws and TPS positions are factory set by Bing (OEM) to a precision degree by bench test methods, likely using calibrated air flow measuring devices and other precision instruments. It is also likely that trying to replicate this level of precision by users in the field will not be as accurate. However, the Motronic ECU appears to be capable of dealing with such inaccuracies, as long as they are within a tolerable error band. After all, this bike ran adequately across country at cruising speeds of up to 85mph with TB adjustments way out of whack and with partially fouled injectors and air filter. Truly marvelous engine to have done that!

 

...

Although most everyone in this forum has a good procedure for synchronizing their R1150's throttle bodies, I recently used the full BMW Repair Manual procedure and found it excellent for achieving good balance and proper cable tension.

The main problem with the BMW procedure is that it’s written for someone with a BMW test station. Below is the procedure, directly from the manual but with edits for clarity and to replace BMW specific and outdated terminology. There are also some Ed. Notes that I’ve added with practical clarifications.

BMW R1150RT Repair Manual

Procedure 13 60 110

Checking throttle-cable play, checking and adjusting idle speed and throttle-valve synchronization

- Notes and Attention (in red) from BMW Repair Manual

- Ed. Notes (in blue) and other edits by me for clarity--e.g. Choke changed to Fast Idle Lever, Carburetor changed to Throttle Body

- Headings (e.g. STEP 1, …) added by me for clarity

STEP 1: Preparation

- Test-drive the motorcycle until the engine is warm.

- Remove the left side panel.

- Remove the right side panel.

(Ed. Note: BMW Manual indicates remove only the left side panel and then small access panel in on the right side.)

Oil temperature: ..................at least 90 C (194 F)

Ed. Note: Adjuster is at the Handlebars

- Push back rubber caps on throttle and fast-idle cables.

- Use the adjusting screws to increase the play of throttle and fast-idle cables.

(Ed. Note: Two turns will create about 1 mm.)

Fast Idle cable free travel: ........................................approx. 1 mm (0.0393 in)

Throttle cable free travel: ........................................approx. 1 mm (0.0393 in)

Ed. Note: Adjuster are at the Throttle Bodies

- Turn the adjusting screws to increase the play of the cables for the left and right throttle valves.

Throttle valve cable free travel: ....................................... approx. 2 mm (0.0787 in)

(Ed. Note: Two turns of the TB adjuster is about 2 mm.)

Ed. Note: STEP 1.5: Zero=250 mV Procedure Placeholder.

-Resetting the throttle stop screws is specifically cautioned against by the BMW procedure.

-Procedure might be useful if someone has already tampered with TB stop-screw adjustment.

-Zero=250 mV procedure to be added later.

STEP 2: Adjust Idle Speed and Left/Right Idle Syncronization

- Connect hoses of your throttle-body vacuum-synchronization tool (e.g. Twin Max, Harmonizer) to the vacuum ports on the left and right throttle bodies.

- Start the engine.

! Attention: Do not allow the engine to idle for longer than is necessary, because the trim panels near the exhaust could be damaged.

(Ed. Note: Use a fan to move air past cylinder heads and oil cooler.)

- Adjust idle speed by turning the air bypass screws while checking that the throttle bodies are balanced.

Idle speed: ........................................ 1,100 ±50 rpm

(Ed. Note: A hot idle speed above 1100 rpm, results in leaner AFR readings during cold-starting, this based on data from GS-911 logs from several R1150R/GS/RTs.)

Note: Make sure that both throttle valves are closed.

(Ed. Note: firmly resting on the throttle stop screws)

! Attention: Do not tamper with the sealed stop screws of the throttle valves, as otherwise the idle-speed volume flow will have to be reset by the manufacturer.

STEP 3: Remove Free Play from Throttle Body Cables

- Carefully turn the adjusting screw of the left throttle valve and reduce play until the reading of the vacuum synchronization tool changes.

- Turn the adjusting screw very slightly in the opposite direction until the reading returns to its original value.

- Tighten the locknut to secure.

Note: Make sure that the reading does not change when you tighten the locknut. Adjust throttle cable play at throttle valve so that no play is perceptible, but the throttle valve reliably contacts the stop screw (no strain on cable).

- Adjust the right throttle valve in the same way.

Note: Zero play can cause the throttle valves to rattle.

(Ed. Notes:

-One quarter to one half turn of the TB adjuster results in 0.25 to 0.5 mm slack.

-Considering this warning, and the instructions above to "adjust so that no play is perceptible" if the throttle cable adjusters on the throttle bodies require adjustment later in Step 5, either the left or right (as appropriate) should be turned clockwise, so as to not remove the "imperceptible play”.

-The BMW procedure says to use engine rpm or engine vacuum-balance to indicate when the throttle cables are fully tensioned and have started to move. A much more sensitive indication can be had by monitoring the TPS voltage between pins 1 & 4. When the voltage starts to increase, the throttle is moving. This will work for the left TB, Fast Idle Lever and Throttle Twistgrip. Because the adjuster for the right TB doesn't move the TPS, it won't work for that adjustment.)

STEP 4: Adjust Throttle and Fast Idle Lever Free Play

- At the right handlebar, use the adjusting screw to adjust play of the throttle cable.

Play of throttle cable: .....................................approx. 0.5 mm (0.0197 in)

Simplified Zero=250 mV Procedure

(N.B. This procedure should be considered a last resort to solve problems not resolved by the TB synchronization above.)

--Loosen all throttle and fast idle cables

--TBs and BBSs must be clean

--Set TB BBSs to 2.5 turns (just a starting point) and make them equal.

--Center and fully close both TB butterfly valves

--Set TPS to 250 mV and lock TPS screws (no further adjustment of TPS required)

--Open left TB and count stop screw turns after contact until TPS reads 340 mV. Lock left stop screw (no further adjustment).

--Set right TB stop screw open same number turns as left.

--Start engine, warm up to at least 3 bars.

--Balance at idle using only right TB stop screw, do not touch BBS. RPM does not matter

--Lock right TB stop screw.

--Follow TB alignment procedure above

Just an endorsement to roger 04 rt's antidote to the zero=zero method where I messed with the throttle body stop screws (because I thought I was doing the bike a favor), on an 1150 rather than an 1100 doing a tuneup. 

Roger you saved my weekend!  The bike now runs as well as I could expect for 75,000 miles with little maintenance before it came to me.

No surging, no snatchy off idle throttle response, idle is nice and relatively stable (might slowly waver by half a tachometer needle width, if that makes sense) - all of which were the opposite only a day ago and it sounded like a tin can full of old nuts and bolts in a paint shaker machine.

Things changed were...

cleaned throttle bodies and BBS ports / needles (seriously full of junk in there)

Autolite 3923 copper plugs 

valve clearances were set ( waaaaay out of specification - could have driven a truck through the gaps)

careful setting of the TPS using a regular DMM and throttle bodies as per this thread - i used an electronic vacuum balancer that reads in a resolution of 2 mm hg.

I actually had fun during the test ride it was that much of an improvement!