low throttle opening glitch

[luhbo]

Here is a picture from the worn track:

 

 

If you follow the leads you can also recognise at wich point the slope changes. And you can see the "510" of course.

 

Hubert

[Guest Jeff in Ohio]

The .pdf I have is areas copied from a blueprint for the TPS. It says nothing of a third party part number. The reason the curve is shaped like that is quite simple. The ECU is an 8 bit system. It coverts the analog signal from the TPS into a digital signal. Since it is an 8 bit system, there are 256 EVENLY spaced steps that can be used for data. Since the fuel injection sensors work on 5 volts, that means each step is about .0195 volts. That is, the ECU can not resolve a change in the TPS output if that change is less than about 19 millivolts. That being known and that the steps are equally spaced, the 1st part of the TPS output has a higher slope so that for physical rotation of the TPS, a greater change in signal is sent. What this means is that at low throttle openings, more values can be mapped in the ECU. Small changes in the throttle are thus seen by the ECU in the critical part throttle and low throttle areas where small changes in throttle opening need to be resolved. In the upper region, the fine steps are not required, because the operator is pretty much putting twisitng it close to wide open to really accelerate at that point.

 

The reason for starting at 150mv is simple also. Looking at the graph, Marelli starts it's output spec at 3% of input. Well, 3% of 5 volts is..... 150mv. Below 3%, Marelli does not place a hard and true specified output for the TPS. Below 3%, the TPS could be linear, non-linear, or nothing. According to the graph, Marelli doesn't care what it is. This is good practice because you really do not want to use a sensor at the very beginning or very end of it's range. The 3% allows the sensor to get 'off the bottom' and up into a useable specified range. Most sensors tend to do funky things, or get non-linear, or loose resolution at their extremes.

 

As an aside, I sent this chart to a known maker of Guzzi chips a long time back and his reply was "it should be linear". Well, by looking at the graph, it obviously is not. Doesn't that make you feel good? A chip based on an assumption that the TPS provides a linear signal to the ECU? Before I received the drawing, I used the technoresearch VDST and a multimeter to plot degrees of throttle opening versus voltage and I generated very much the same graph, so it is accurate......

 

here is the PF4C the Harley uses....

[luhbo]

Jeff, thank you for this very clearly and understandable written statement. Right on the spot.

 

I think starting at 3% is important to keep the output above zero. The PF4C graph shows the tolerances. 1% is already lost for manufacturing tolerances, then there are of course tolerances applied to the input voltage, tolerances in the total assembly TPS to shaft and so on. I'd say 3% is already right on the edge.

These potentiometric sensors normaly dont have probems with linearity because of being used at the end or beginning of their range. The resistance should be a function of homogeneity, thickness and width of the printed resistive paste.

 

What you have said about the coherence between the steep output curve and the narrow TP steps in the map at low opening angles makes absolutely sense.

I'd take this as a forward pass and herefrom deduce an urgent need to adjust the TPS exactly to the values the book tells us. Otherwise there is a very good chance that the ECU might not fall back in idle mode for instance or do other unexpected things.

 

The TP rows are staggered like this: 2° - 3.67° - 4.18° - 5.18° - 6.77° - 8.16° ...

 

I think this could lead back to where the topic started: a slight glitch while adjusting the TPS could quite easily lead to glitches while slowly opening the throttle.

 

Hubert

[Cliff]

I'm impressed at how well the TPS wears. I spend too much time look at what value it returns and even under the smoothest road conditions there is quite a lot of movement being transferred into the TPS.

 

I try and reduce the wear by moving my hand outwards so I have part of my palm stabilised against the bar ends. I find this is gives a stable output - necessary whe trying to tune on the road and dyno.

[Guest ratchethack]

A few years ago I posted my surprise at finding an amazing quantity of accumulated corrosion between the TPS and the TB. The cavity that's formed between them is open to every bit of rain, wash water, and road debris that can fall into the opening on the outboard side. It holds water that will pool up in there like a small cup. Amazingly enough, the bottom of the TPS is also wide open to this cavity. I figured this amounted to a ready-made sauna due to trapped water, cleaners, and what-have-you in there cooking off with engine heat, creating an occasional steam bath for the wiper and contacts. This can't be a healthy thing for its long-term condition. I cleaned all the white flaky corrosion out, including the heavily corroded butterfly shaft, oiled it, carefully gave a shot of contact cleaner through the entirely open bottom of the TPS, and re-installed it with a fillet of silicone sealant over the opening. The TPS can still be adjusted by flexing the silicone slightly. I haven't had a single problem with it in 30K miles, and as always, the Guzzi runs like a Champ.

[guzzista]

So, my question for Jeff is: If the TPS is indeed set at 150MV (key on engine off and right throttle plate shut) , in your experience what is the average reading at idle for the v11's?.The reason that I am asking is that I have noticed ( a bit of ) improvement when the TPS is set at 500 mv instead of the "got to be max at 475" advice from my dealer....this is after 2 TPS units, both the original ( condemned by the dealer) and the warranty replacement which I received later...

[dlaing]

That's great, indeed! I'm afraid I'll never get behind the secrets of this PC stuff.

 

If one has already double cheated the ecu by dynoing the PC on the basis of a non-spec TPS setting why should he then cheat the whole construct again by shifting the basis the third time?

 

I always thought the PC allows it to enrichen or lean out the map exactly as needed?

 

Hubert

I would not mess with the TPS unless he wants to go back to the tuning link....although 25mv probably won't upset the PCIII map.

When most PCIII maps are set by Tuning Link the indicated air to fuel ratio is optimized only at fuel cells where the bike is under significant load, ie. cruising to WOT.

The partial throttle popping experienced by Guzzista does not take place in the map cells modified by Tuning Link.

Experienced Guzzi PCIII mappers like Todd Eagan modify the map appropriately to provide a better fueling at low throttle openings.

To fix the problem, simply add a little fuel to the map where the problem is suspected....but only add just enough to get rid of the PROBLEM.

Getting a CO reading at idle and adjusting to non-US spec would also be a good idea.

[luhbo]

... Amazingly enough, the bottom of the TPS is also wide open to this cavity. I figured this amounted to a ready-made sauna due to trapped water, cleaners, and what-have-you in there cooking off with engine heat, creating an occasional steam bath for the wiper and contacts....

...

 

Only on first sight. The TPS is of course sealed! What you saw is just the opening for the throttle shaft.

 

To others I'd strongly recommend not to 'seal' the mentioned gap. Water, cleaning fluids, gasolin e.a. will find their way inside, they have time for this, but they will never again find a way out!

 

Hubert

[luhbo]

I'm impressed at how well the TPS wears. I spend too much time look at what value it returns and even under the smoothest road conditions there is quite a lot of movement being transferred into the TPS.

 

I try and reduce the wear by moving my hand outwards so I have part of my palm stabilised against the bar ends. I find this is gives a stable output - necessary when trying to tune on the road and dyno.

 

The most challenging thing for a potentiometric TPS is the so called Ditter-Test. Very small movements for a long time. This kills most products actually available.

 

I never had a problem while autotuning the map. The software is so smooth that it always works, regardless whether I make minor "mini-twists" or not. Trying to keep the throttle rock steady for some time can lead to a damned dumb feeling in the right hand. And it prevents me from making suitable outlines for my cornering.

 

Hubert

[luhbo]

So, my question for Jeff is: If the TPS is indeed set at 150MV (key on engine off and right throttle plate shut) , in your experience what is the average reading at idle for the v11's?.The reason that I am asking is that I have noticed ( a bit of ) improvement when the TPS is set at 500 mv instead of the "got to be max at 475" advice from my dealer....this is after 2 TPS units, both the original ( condemned by the dealer) and the warranty replacement which I received later...

 

When I remember right then the proper idle setting is 550, at least for the Titanium ecu. At 475 you are running it too lean then.

 

How many miles do you have on your Coppa Italia?

 

The TPS might be cheap, but its an industrialy manufactured product, manufactured and sold by well known and reputated companies in large quantities. I think it's quite unlikely that you can get 2 of them in such a short period and both are already corrupt when new!

 

I suppose it's more a problem on your dealer's side.

 

You may have noticed, I have 60000 km plus on the clock now, that's around 40000 miles. I'm not amused about the TPS failure, but still it's quite a distance for a leisure product. A motorcycle nowadays is technically spoken just a leisure product. Built for quick fun, colorfull and cheap. I feel sorry about that, but it seems to be a fact.

 

There is a cartoon from MotoMania: One fellow kneeing besides his opened Laverda engine, the other one standing besides him. "I have never had my knee down to the street. And you?" "Me? Always!"

 

Hubert

[Guest Jeff in Ohio]

So, my question for Jeff is: If the TPS is indeed set at 150MV (key on engine off and right throttle plate shut) , in your experience what is the average reading at idle for the v11's?.The reason that I am asking is that I have noticed ( a bit of ) improvement when the TPS is set at 500 mv instead of the "got to be max at 475" advice from my dealer....this is after 2 TPS units, both the original ( condemned by the dealer) and the warranty replacement which I received later...

 

Acgtually, your dealer should be telling you to set it at 3.6 degrees using VDST software. If a dealer is setting things based on millivolts, I would ask him the question "where is your factory authorized software?"

 

By the way, 3.6 degrees on the VDST software IS 525 mv. 475 is too low. At that setting, the ECU could be almost 3 steps behind where it actually needs to be on the map (see above explaination where 1 step is 19 millivolts).

 

if you have not read this yet, then please do so. I have used my procedure on may bikes including V11's with good results. I have also gotten a lot of positive feedback from others like "wow, so that's how it is supposed to run.."

 

http://www.guzzitech.com/EVTuneup-Jeff_B.html

 

 

The procedure was developed by piecing together various tidbits from Guzzi manuals (which always leave at least one step out or accidently reverse something in the translation from Italian to English) and good engineering knowledge.

[docc]

 

 

if you have not read this yet, then please do so. I have used my procedure on may bikes including V11's with good results. I have also gotten a lot of positive feedback from others like "wow, so that's how it is supposed to run.."

 

 

 

With 44,000 miles on the Sport I've tuned it about ten times. Maybe more including retuning it after the dealer services that had less than ideal results on the running.

 

Having used the factory manual, Guzziology, internet forums, heresy, and voodoo I can say that the Jeff in Ohio Method yields the very best and most consistent result.

 

Thanks to Jeff! And thanks for the Degree opening-Millivolt conversion chart as well!

[Guest ratchethack]

Only on first sight. The TPS is of course sealed! What you saw is just the opening for the throttle shaft.

 

To others I'd strongly recommend not to 'seal' the mentioned gap. Water, cleaning fluids, gasolin e.a. will find their way inside, they have time for this, but they will never again find a way out!

 

Hubert

Hubert, we discovered previously that although we have the same model Guzzi, you got different fork springs than I got from Mandello. I suspect this may not be the only difference between US-issue and Euro-issue Guzzis. I don't know what your TPS looks like, but if yours is "sealed", mine is most definitely not the same.

 

The bottom of my TPS is OPEN -- that is, when held upside down, you can look into the open "drum" area of the TPS through two openings in the bottom of the plastic housing and see the contacts on the flex strip.

 

For ref., on top of the TPS appears: Magneti Marelli PF3C/00 3239BE

 

The way I have sealed off the opening between TPS and TB, the only POSSIBLE way that water, gasoline, cleaners, etc. can now find their way into the cup-shaped cavity below the TPS is if the entire bike were submerged in liquid -- long enough for air bubbles to work their way out first. The only other avenue for moisture entry is past the seal of the rubber boot over the 3-pin connector. Even direct spray on the TPS will not allow moisture entry now, and there is NO WAY any kind of road debris, dust, etc. can now get inside that space.

[luhbo]

Hubert, we discovered previously that although we have the same model Guzzi, you got different fork springs than I got from Mandello. I suspect this may not be the only difference between US-issue and Euro-issue Guzzis. I don't know what your TPS looks like, but if yours is "sealed", mine is most definitely not the same.

 

The bottom of my TPS is OPEN -- that is, when held upside down, you can look into the open "drum" area of the TPS through two openings in the bottom of the plastic housing and see the contacts on the flex strip.

 

For ref., on top of the TPS appears: Magneti Marelli PF3C/00 3239BE

...

 

Ratchet,

 

what shall I say? I think I'll just beleave it this way. Funny enough, indeed. Probably the -/00 stands for the two US-wholes.

 

Hubert

[Guest ratchethack]

One thing I figured out is that because the Guzzi motor takes such a LONG time to break in, the TPS needs re-setting several times during break-in. The engine got considerably looser and smoother even after 10K miles, right up until about 20-25K, when it seemed to stabilize. During this time, the idle RPM would gradually creep up, and I had to adjust it back down to 1100-1200 RPM. Lowering the idle RPM threw the TPS setting out of the low end of the 500-525 mV adjustment range. A few times when I did this it was well below 500 mV after re-setting the idle, around 480 mV, as I recall. I re-set the TPS each time I dropped the idle down. Throttle response on the low end got a little better every time I did this. I reasoned that as the engine continued to loosen up, the original low angle throttle openings were resulting in higher and higher RPMs than when the PC III map was zeroed out and installed. I don't know how many times I did this, but I think it was 3 or 4 times. To my way of thinking, it makes more sense than the technique of "pegging" the TPS to 150 mV with throttle plate closed, because it's relating TPS setting (and therefore fuel delivery) to RPM, not degree of throttle plate opening, which is not directly fixed to RPM until the end of the 25K mile break-in. TPS readings tend to vary a bit with temperature (ambient as well as engine), but it's been pretty stable since about 25K miles.