Jump to content
IGNORED

GS intake tubes on RT-Spark Knocking at my door....


mrsoup

Recommended Posts

I have installed GS tubes on my RT. Like the extra grunt, dislike the spark knock. I did not change the CAT code, should I? What can I do to reduce the spark knock if I continue with the GS tubes? This bike has always had some knock in warm conditions, it seems to be significantly worse now. I sure don't want to ventilate the pistons!!

Link to comment

You haven't told us what bike you're talking about, but I had the same problem on my R1150RT and had to remove the GS tubes. The only thing I was able to do to get that same mid-range torque boost was to install dual-plugs.

 

Pat

Link to comment

The bike is a 2000 1100 RT, sorry I didn't include that information in the first post.

Is the motronic program really that much different for the RT than for the GS models? Is removing the GS tubes the only answer?

Link to comment
Joe Frickin' Friday

Got a '99RT. Put GS tubes on it in spring '02,at 40K miles. Knocked horribly at first. Tried a few engine cleaners - Techron, 44K, etc. - none of them worked well or lasted for long. Finally dismantled the engine and manually scraped all the deposits from the pistons and heads. No knock - at least for 10K miles. Then it started coming back. In spring '03, I took one head/piston apart and cleaned, but I also measured it: bore,piston, rings/gaps, etc. and found that no damage was being done. As it happens I got a new helmet that spring that was a bit noiser than the old one, which drove me to wear ear plugs all the time (instead of just on the highway). That does a lot for taking your mind off of it.

 

It only knocks a bit now. I run the highest octane I can find, and yes, I use the GS CCP. Engine is still trouble-free at 108K miles.

 

If the knock is really outrageous, try some engine cleaner (BG Products 44K worked when Techron wouldn't) to get it under control (i.e. only happening at high/max load, RPM >= 4K), and then whale on the engine regularly. Seriously, get in the twisties and run the engine hard; this will help remove a little more carbon and help keep it under control in the future.

 

And wear ear plugs.

Link to comment
Joe Frickin' Friday
The bike is a 2000 1100 RT, sorry I didn't include that information in the first post.

Is the motronic program really that much different for the RT than for the GS models? Is removing the GS tubes the only answer?

 

The GS uses different valve timing and of course has those GS intake tubes, so it breathes differently. It also has a lower compression ratio. AFAIK nobody's actually checked, but it stands to reason that the GS would utilize more retarded ignition timing (in addition to the lower compression ratio) to allow it to run safely on low octane gas - which is all that's available in some very remote areas and foreign countries.

Link to comment

GS tubes put you right on the edge of the detonation zone, so every little factor you can fudge in your favor matters.

 

The easiest to deal with is gas. Make sure you use the best gas you can get, for two reasons: reliable octane values, and engine cleanliness.

 

Chevron or Shell are my choices; they seem to have the best engine-cleaning additive package. Where you get your gas does make a difference. I've never had a tank of bad gas from a Chevron station. Arco, on the other hand, has scared the hell out of me a couple times with the engine noises it generated. Go to the high-volume stations to get the freshest, least-watered gas possible.

 

If push really comes to shove you can retard the timing a tad by adjusting the plate the hall sensors are on. I don't know, though, if retarded timing would rob you of the gains made by the tubes, and timing adjustments are not something that BMW intended.

 

Pilgrim

Link to comment

Not replying specifically to Pilgrim's post.

 

Why do the GS intake tubes make the RT engine more prone to detonation?

They don't make the mixture more lean...

 

Do they really make the cylinders to fill so much more efficiently (borderlining a supercharger effect) that the cylinder pressure just before ignition is high enough to detonate the mixture?

 

--

Mikko

Link to comment
Joe Frickin' Friday
Not replying specifically to Pilgrim's post.

 

Why do the GS intake tubes make the RT engine more prone to detonation?

They don't make the mixture more lean...

 

Do they really make the cylinders to fill so much more efficiently (borderlining a supercharger effect) that the cylinder pressure just before ignition is high enough to detonate the mixture?

 

That's pretty much it. On the real GS engine, the compression ratio is slightly lower to compensate for this, and as I speculated earlier, the GS CCP may also retard spark timing slightly.

 

FWIW, funny things may happen to the mixture with the addition of the GS tubes to an RT engine. The valve timing is not the same as the GS, so the breathing throughout the entire operating map is not going to be identical, either. If it's operating in open-loop mode at WOT, then it may in fact not be adding enough fuel at certain RPM's. But again, this is speculation; I don't know that anyone's ever measured this.

Link to comment
Not replying specifically to Pilgrim's post.

 

Why do the GS intake tubes make the RT engine more prone to detonation?

They don't make the mixture more lean...

 

Do they really make the cylinders to fill so much more efficiently (borderlining a supercharger effect) that the cylinder pressure just before ignition is high enough to detonate the mixture?

 

--

Mikko

 

Mitch didn't state this directly, so I will.

 

Yes, they do lean out the mixture. The engine does not have an airmass sensor; it depends on a map of various sensor inputs (throttle position, ambient temp, etc) to calculate how much air it's getting, then shoots fuel based on that calculation.

 

As you noted, there is a mild supercharging effect brought about by the increased intake velocity through the narrower, longer tubes. That extra air is unrecognized by the Motronic, so it injects the same old amount of fuel, just like it had good sense. Thus, you wind up with a slightly leaner mixture. Since the thing is tuned lean for emissions anyway, that can push it into the deto zone.

 

Pilgrim

Link to comment

But that assumes the FI operates in open loop mode.

Isn't the mixture kept at very close to stoichiometric based on feedback from the Lambda sensor? confused.gif

In what conditions (Mitch mentioned WOT) is the FI open loop?

 

Mikko

Link to comment

I did the same mod to my 97RT, but changed the CAT plug. I never had the spark knock until I bumped my engine timing up three degrees. At the same time I also increased the intake valve clearance. This gave me a great boost to the mid range power, but killed total top end. To make a long story short, back your engine timing off a couple of degrees and or run premium fuel.

I also have a Techillusion fuel enhancer. I enriched the fuel curve a little which helped.

Link to comment
Joe Frickin' Friday
But that assumes the FI operates in open loop mode.

Isn't the mixture kept at very close to stoichiometric based on feedback from the Lambda sensor? confused.gif

In what conditions (Mitch mentioned WOT) is the FI open loop?

 

WOT is one of them; the engine is usually made to run slightly rich at WOT, which you can't easily do with closed-loop operation on a switcher-type O2 sensor. Ed (Mr. Shovelstroke) has suggested that in fact the engine runs open-loop most of the time, with closed-loop being the exception rather than the rule.

 

To resolve my own curiosity, I've been meaning to tap into the O2 sensor harness with a volt meter and tape the meter on top of the fuel tank while I ride. During closed-loop operation, the mixture is adjusted rich/lean (and back) a few times per second, with sensor voltage flipflopping between about 0.2 and 0.8 volts (with 0.8 volts corresponding to a rich mixture). This should be readily apparent, even on a digital voltmeter (with the bar readout at the bottom of the display). I'll post the results here if I ever get around to doing this. crazy.gif

Link to comment
ShovelStrokeEd

AFIK, the engine only operates closed loop between roughly 2 and 13% throttle and at RPM's above about 2K or so. Beyond that it just goes with the map with the map ending at around 60% throttle. It doesn't actually end but the pulse widths can get no larger as, with increasing RPM's the frequency of pulses is such that it over rides the pulse width. Strange but it works.

 

I also think there is a need for some terminology correction here. Most of what you describe as spark knock is really just pinging and, as Mitch already mentioned, it will do no harm to the motor. True knock, or detonation, which is pingings evil brother, will do significant damage to the motor but it is extremly rare to encounter this on a properly functioning, stock engine. One of the first signs would show as aluminum flakes on the interior insulator nose of the spark plug. Usually from the piston tops or combustion chamber walls. This is serious stuff but I very much doubt that is what is going on. Detonation will almost always cause a severe loss of power as well and that has not been described.

 

I have run motors, on race gas, with compression ratios as high as 14.5:1 and static spark lead of 40 degrees and basically either at idle or WFO and have only encountered detonation once. Grenaded the motor in short order. I wouldn't worry over much about an 11:1 street motor. Buy some good ear plugs and ride on.

Link to comment

Ed, thanks for making the distinction between pinging and detonation. For a while I was bit puzzled how the GS tubes could ever make the motor to detonate. They don't.

 

So how about the pinging then. What is pinging?

Noise/shock pulses from uneven and at times violent flame burning through a uneven mixture of fuel/air? But the flame is still ignited by the spark?

 

And this situation is made worse by higher cylinder pressure because of better intake efficiency.

 

--

Mikko

Link to comment
ShovelStrokeEd

Close.

 

Pre-ignition is the actual word for pinging. It can be caused by glowing carbon deposits in the combustion chamber or on piston tops, an overheated exhaust valve, spark plug glowing due to wrong heat range and phase of the moon.

 

The additional flow won't have any dramatic effect on cylinder pressure but rather mixture which can exacerbate most of the above.

Link to comment

As Mitch stated, the knock is only under 'load', meaning whack the trottle open to pass in the higher gears. It does not always do it, but happens a lot. I have tried to modify my riding style to minimize this condition with some success, down shifting helps.

 

If it is 'pinging' that I am hearing I will not be too concerned. I have seen holes in piston from serious detonation (spark knock) and that is obviously something to avoid.

 

I will install the GS CCP to see if that helps though. I always run 92 or 93 octane fuel, the best that I can find.

 

Thanks for all the feedback and technical analysis.

Link to comment

To resolve my own curiosity, I've been meaning to tap into the O2 sensor harness with a volt meter and tape the meter on top of the fuel tank while I ride. During closed-loop operation, the mixture is adjusted rich/lean (and back) a few times per second, with sensor voltage flipflopping between about 0.2 and 0.8 volts (with 0.8 volts corresponding to a rich mixture). This should be readily apparent, even on a digital voltmeter (with the bar readout at the bottom of the display). I'll post the results here if I ever get around to doing this.

 

 

 

 

 

Mitch,since the o2 signal orginates at the sensor and works all time(as long as it is at operating temp)there would be no way to tell whether the ECM was in closed or open loop just by monitering the signal.

 

Howeveeer,if you were to hook up your voltmeter to the o2,connect a small o-scope to the injectors(possibly another voltmeter ,if its sensitive enough to read PW at speed)Tape in a small propane bottle with a lead to each vaccum nipple on the TBs.Crank it up and run it to the speed/throttle opening you want to check and add a LITTLE propane to enrichen the mixture and watch the meters.O2 should respond first with a rich mixture,then the PW should decrease if you are in closed loop.If nothing happens,you would be in open loop.Simple Huh grin.gif

 

 

Oh,make sure your insurance is paid up. tongue.gif

 

 

 

Ed (Mr. Shovelstroke) has suggested that in fact the engine runs open-loop most of the time, with closed-loop being the exception rather than the rule.

 

 

I'm not buying into this just yet, for a number of reasons,the first one being that its contrary to all the current info that I have read over the years and also the primary reason for computerized engine control is Emissions control.Granted the resulting performance is a big plus,but without the emissions laws,we probably would not be anywhere near what we have today.I just can't see the mfgs running open loop most of the time given the emissions requirements.

 

 

Ed,I'm not saying you are wrong,but would just like to see some documentation in terms us non-engineers can read. grin.gif I try to learn something new every day......I have to,because I keep forgetting something everyday confused.gif

 

 

Where is Knappy when you need him ??? bncry.gif

Link to comment
Joe Frickin' Friday
Mitch,since the o2 signal orginates at the sensor and works all time(as long as it is at operating temp)there would be no way to tell whether the ECM was in closed or open loop just by monitering the signal.

 

You're right that the sensor is always working, but the nature of the output will vary depending on what the Motronic is doing to the mixture composition.

 

With a "switcher" O2 sensor, the transition between high/low output voltage happens over an extremely narrow range of mixture composition, basically right at stoichiometric; it's pretty much impossible to produce a mixture that will cause the sensor to hold some intermediate voltage. In CL mode, the mixture is deliberately dithered back and forth across this transition point; you'll see the sensor output voltage flip-flop rhythmically from 0.2 to 0.8 volts, kind of a square-wave output. In OL mode, the Motronic makes its best guess as to the correct fuel quantity - and has been discussed, may even bias it slightly rich, and keep it there. When that happens, the sensor output voltage would be expected to park at 0.8 volts and stay there.

 

So that's the difference I'm expecting to see:

 

CL: flip-flop-flip-flop

OL: flatline

 

Ed,I'm not saying you are wrong, but would just like to see some documentation in terms us non-engineers can read.

 

That's what I'll be shooting for with the DVM test, if/when it happens. crazy.gif

Link to comment

CL: flip-flop-flip-flop

OL: flatline

 

 

OK,I see where you are going with this. thumbsup.gif It should work in theory,but in reality I don't think will give a cut and dried vision of whether its in open or closed loop.Based on what I have seen,the only time the signal parks,is when it cools at an idle.Usually it just gets erratic as the computer guesses at the mixture.Thats based on automobile sensors I have watched,the bike could be different.

 

I know that if anyone can get it worked out,you will be the one. clap.gif

Link to comment
ShovelStrokeEd

Big,

I got that info from a guy over on the Pelican board who worked for Bosch when they were developing the last couple of generations of the Motronic. You are partially right in that the O2 sensor and closed loop operation is primarily for emissions control but wrong in thinking that it works all the time or the bike spends very much of its operating range in closed loop mode.

 

Here is the rub. The real purpose of the whole O2 sensor, which is a flip/flop btw, and closed loop operation of the Motronic, is to keep the catalytic converter lit up. It is the cat converter that does the emissions removal, not the fuel maps. The unfortunate thing is that cats are strange beasts (not refering to felines here). They only operate properly over a very small range of A/F ratio. Something on the order of 14.7:1 up to 15:1 or stochimetry or a fuzz leaner. Leaner than that and they stop working, richer and they overheat.

 

Closed loop is only used during part throttle cruise and pretty much quits in any other mode, (passing, WFO, idle). IOTW, any mode that requires a richer fuel mixture to produce power. To really test exactly when the thing switches from closed to open loop you would need a couple of channels of data logging and a wide band O2 sensor. Monitor throttle position voltage, O2 signal and engine RPM and you should quickly see the point at which the mixture is allowed to go richer than stochiometry and stay there. I suppose looking at the flipper type and seeing when it rails would work but it would be much less precise.

Link to comment
Joe Frickin' Friday
I suppose looking at the flipper type and seeing when it rails would work but it would be much less precise.

 

confused.gif Why would it be any less precise then the other method you describe?

Link to comment

 

 

Closed loop is only used during part throttle cruise and pretty much quits in any other mode, (passing, WFO, idle). IOTW, any mode that requires a richer fuel mixture to produce power.

 

So,if I jump out on the interstate,get to speed,set the cruise,and ride to the other coast,I'll be in closed loop all the way,except when I accellerate hard enough to drop into open loop.So that will leave me in closed loop 95 % of the time as opposed to the 15% you are refering to???? grin.gif

 

 

It just occurred to me that we might be on opposite ends of the same horse grin.gif

 

 

 

 

 

Once closed loop is obtained,the only reasons for it to drop back to open loop would be:

 

Exceeding the WOT specs.(measured by TPS voltage.not sure where that point is.Maybe that 60% of the throttle opening.What does that relate to in speed,probably way over the limit on high gear.) tongue.gif

 

Hard accelleration.(causes a momentary lean mixture condition at the o2 sensor when the throttle is snapped open.gradually goes back to closed loop when the o2 starts to toggle)

 

 

At idle when the exhaust cools,o2 will get erratic and will drop to open loop.

 

 

So where am I going ??If I'm chasing David across the Dragon,grin.gif see-sawing on the throttle and whacking it wide open as I come off the corner(not smooth like him bncry.gif).Yea,I'm in agreement with being in closed loop only 15% of the time. smirk.gif

 

 

However.....

 

 

The rest of the time I'm applying the throttle smoothly,probably not exceding the WOT requirements for the most part and probably staying in closed loop 90% of the time.

 

 

 

 

Depending on how the bike is ridden,it could either be in closed loop 15 or 90 %of the time,or actually anything in between.

 

 

 

Whadda think ? grin.gif

Link to comment
Joe Frickin' Friday

From Ed's description, it sounds like he's saying that only about 15% of the engine's operating range (1050-7500 RPM, 0-100% throttle) is covered by CL operation; it just happens to be the 15% of the range where the typical streetbike engine spends 90% of its time.

Link to comment
ShovelStrokeEd

Close enough.

 

The throttle position for closed loop is from about 2% open to 13% open. 60% is about where the map stops changing pulse width. A completly different proposition.

 

Yes, for the most part, you will not exceed 15% throttle opening when steady state cruising at even 80 mph and thus remain in closed loop mode for considerable periods of time. Chasing anybody around on twisties will result in open loop most of the time as throttle openings tend to be larger and the fuel cutoff will interfere on trailing throttle.

Link to comment
  • 11 months later...
I have installed GS tubes on my RT. Like the extra grunt, dislike the spark knock. I did not change the CAT code, should I? What can I do to reduce the spark knock if I continue with the GS tubes? This bike has always had some knock in warm conditions, it seems to be significantly worse now. I sure don't want to ventilate the pistons!!

 

I would recheck the BASIC TIMING. It may be a little advanced and the problem did not show up until you changed the induction scheme.

Chris smile.gif

Link to comment

I am very happy to report that I have virtually no ping except on the hardest pulls.

I use the cheap gas too, 87 octane.

 

I'm no angel on the throttle either but I do try to keep the revs up.

2000 1100rt with stock green CCP (not a GS one)

 

Having done the swap in the cooler CA "winter" I'll need to wait and see what hot weather yields.

Link to comment

I'm not nearly as smart as these guys, but I can tell you that the GS tubes made my '97 RT ping like crazy, no matter what else I tried. I ended up taking them off. In fact, my '02 GS also pinged like crazy so I put in RT intake tubes, which cured it.

 

Several people on the Advrider board have claimed that replacing the left side cam chain tensioner with the updated version caused their pinging to disappear.

Link to comment

Archived

This topic is now archived and is closed to further replies.

×
×
  • Create New...