Throttle Position Sensor

[dlaing]

Here is a table from Jeff Brannen

 

showing the dual linearity

His description of it is here

http://www.guzzitech.com/HD-TPS-Jeff_B.html

Jeff also made interesting posts here:

http://www.v11lemans.com/forums/index.php?...=2410&st=15

I agree with him on everything except for his not using air screws for balance.

[Ryland3210]

Here is a table from Jeff Brannen

 

showing the dual linearity

 

 

I agree with him on everything except for his not using air screws for balance.

 

Terrific!

 

That confirms two assumptions:

 

The TPS is non-linear, but within the 3.4-10.4 degree range I analyzed, it is linear.

The ECU does interpolate between the entries in the map.

 

I'm off on a long ride just now, and look forward to digesting the whole report later today.

 

Thanks for this info, dlaing.

[Skeeve]

Here is a table from Jeff Brannen

 

showing the dual linearity

His description of it is here

http://www.guzzitech.com/HD-TPS-Jeff_B.html

Jeff also made interesting posts here:

http://www.v11lemans.com/forums/index.php?...=2410&st=15

I agree with him on everything except for his not using air screws for balance.

 

All of which begs the question: is the 4k rpm flat spot caused by too much or too little fuel? If it's a lean stumble, then wouldn't it be interesting if the easiest fix is just to substitute the H-D tps in for the stock unit, rather than spending all the time & expense on a PCIII & mapping on an otherwise stock motor?

 

[Ryland3210]

[quote name='Skeeve' date='Sep 8 2007, 03:56 PM'

All of which begs the question: is the 4k rpm flat spot caused by too much or too little fuel? If it's a lean stumble, then wouldn't it be interesting if the easiest fix is just to substitute the H-D tps in for the stock unit, rather than spending all the time & expense on a PCIII & mapping on an otherwise stock motor?

 

 

Well, maybe, or maybe not. I've been involved in electronics, including specifying and purchasing many potentiometers over the years. Now, I'm choosing my words carefully: It would not surprise me to find that both of the potentiometers tested were made to the same specifications, and the difference in break point is simply typical production variation. Without access to the manufacturer's specifications, I remain unconvinced that the potentiometers are intentionally different. Even though they have different part numbers for the Harley and Moto Guzzi applications, the difference could relate something other than the transfer function. It would not be the first time I have seen the same part given different part numbers for different markets, or simply to set two different prices! I'm happy to be proven wrong about this, if the spec's are available.

 

In any event, this does not change the draft procedure. I have added a note at the end suggesting that the left throttle idle screw can be used for occasional minor adjustments of RPM without having to use the bypasses and monitor vacuum balance, once the whole procedure is done. I have tried this out earlier today at various idle RPM settings from 1000 to 1500 RPM without any problem.

 

So I think we are done with the procedure.

[docc]

I'm glad I asked . . .

 

I am getting some pinging at really hard throttle in the hottest ambient temperatures. It' good to know I've probably advanced my ignition timing with the higher TPS setting.

 

Still no account or confirmation for tachometer error . . .hmmmm . . .

[Ryland3210]

I'm glad I asked . . .

 

I am getting some pinging at really hard throttle in the hottest ambient temperatures. It' good to know I've probably advanced my ignition timing with the higher TPS setting.

 

Still no account or confirmation for tachometer error . . .hmmmm . . .

 

On my '04, I checked the tach against an extremely precise (relative to the tach) digital counter. It was spot on at 1100 and 1300 RPM.

 

What makes me wonder about your situation is that IF the tach is driven by a signal from the ECU (as it is on some automobiles I know of), which monitors RPM for its own purpose, is it possible your ECU thinks the engine is revving higher than actual?

 

If you look at the ignition chart, except for the anomaly at 3200 RPM (which I do not understand the reason for), the advance increases with RPM in the middle range, so your ECU would over-advance the timing if it thought the revs were higher than actual.

 

Another possibility is that the tach itself is calibrated incorrectly.

[Ryland3210]

The Procedure below for adjusting throttle balance, air bypasses, linkage connecting rod, TPS and RPM is the product of contributions of numerous members of the V11 LeMans members, for which I am grateful. An important goal of the procedure was to make it easy, logical, with minimum trial and error guesswork.

 

 

IDLE TPS/THROTTLE BALANCING TUNING September 9, 2007

 

First make sure the TPS is calibrated to 150 mv at fully closed as follows: disconnect the connecting rod, back off the right throttle idle screw and choke cam (make sure the choke cable permits full retraction of the cam (it didn't on my bike), then loosen the TPS clamp screws and rotate if needed. + - 5 mv can be obtained with a little effort.

 

Next, close the bypasses, keep the right throttle idle screw backed off to put the connecting rod in tension, removing any backlash, and balance the throttles at idle using the connecting rod adjustment. Screw in the left throttle idle screw if the idle is too low to maintain. Do not use the choke for this purpose, because that would put the connecting rod in compression, introducing backlash.

 

Now adjust the left idle screw for a TPS reading of .518 volts. + - .005 (corresponding to 3.4 degrees physical opening) can be obtained with a little effort. (Some riders have been known to also subsequently physically readjust the TPS (not the idle screw) to lean or richen the entire throttle range. However, loosening its screws and offsetting the TPS to a higher voltage, e.g. .539, will fool the ECU into adding more fuel, but it will also fool the ignition timing table.)

 

Next open the bypasses to obtain the idle RPM at 1100 to 1200 while maintaining balance. Bypasses should be open 1/2 turn or more. If not, back off the idle screw to reduce the TPS in steps of 15 mv and open the bypasses to compensate until they are opened 1/2 turn or more. Check balance at midrange RPM as follows:

 

A When checking balance at cruise RPM, make any fine correction needed using the connecting rod adjustment, then:

B. Check balance at idle RPM. If OK, done, if not, rebalance at idle using the air bypass screws, and go back to step A.

 

Option: Adjust the idle mixture trim potentiometer under the label of the computer for best idle quality, or use a gas analyzer if available. However, according to Guzzijack, "not an option on the V11Sport as it uses the 15M ECU - manual adjustment of the idle mixture potentiometer is only applicable to bikes with the P7/P8 or 16M ECUs - 15M idle mixture adjust is only possible via factory or aftermarket software."

 

Once this procedure is completely successfully, in the future, minor changes in idle speed can be made simply by adjusting the left throttle idle screw. Since the throttle plates have been balanced, backlash between them has been eliminated, and bypasses have been properly adjusted to maintain balance at idle, these should be stable for many miles.

[John A]

My TPS started acting up after 90K miles so tonight I put in a new Harley TPS. I set TPS at wide open throttle and dont much care what it is at idle.I could only get 4.77 volts at WOT. For this particular bike it likes 4.84 at WOT. If I left the screws out of the Harley TPS it would go up to 4.93 but the slots wouldn't line up with the screw holes. So I put in my spare Guzzi TPS and set it to 4.84 at WOT. No troubles works great again,Just thought I'd let you all know.

[dlaing]

All of which begs the question: is the 4k rpm flat spot caused by too much or too little fuel? If it's a lean stumble, then wouldn't it be interesting if the easiest fix is just to substitute the H-D tps in for the stock unit, rather than spending all the time & expense on a PCIII & mapping on an otherwise stock motor?

 

Interesting idea, but the TPS model won't help, at least not at WOT, which is where it shows up in the dyno chart.

But mid throttle is a possibility, although unlikely.

[Ryland3210]

[quote name='Skeeve' date='Sep 8 2007, 03:56 PM'

All of which begs the question: is the 4k rpm flat spot caused by too much or too little fuel? If it's a lean stumble, then wouldn't it be interesting if the easiest fix is just to substitute the H-D tps in for the stock unit, rather than spending all the time & expense on a PCIII & mapping on an otherwise stock motor?

 

 

 

It's simpler than that. Look at the fuel delivery chart at 4000 RPM, large throttle openings. It leans out compared to delivery at the higher and lower adjacent RPMs

 

There are places in the charts like this, where fuel and timing entries make no logical sense to me. For example, at 4,000 RPM, if I had a Tune Boy, the first thing I would do is set the fuel delivery at the midpoint between the adjacent columns. My predictions is that your flat spot goes away.

 

Here's another: between 5.18 and 21.46, at 3200 RPM on the ignition chart, timing is retarded compared to higher and lower RPM's. Why? If there isn't any preignition at the lower RPM, why retard the advance at 3200?

[motoguzznix]

All of which begs the question: is the 4k rpm flat spot caused by too much or too little fuel? If it's a lean stumble, then wouldn't it be interesting if the easiest fix is just to substitute the H-D tps in for the stock unit, rather than spending all the time & expense on a PCIII & mapping on an otherwise stock motor?

 

The 4k flat spot is not influenced by the richness/weakness of the mixture.

I rectified the mixture and got better torque everywhere but the flat spot basically remained.

Remapping or a PC is the only way to adress mixture changes. The TPS is a measuring device, not for mixture tuning. Changing the calibration of the TPS changes the adresses in the map over a large area. This can cause a lot of unwanted side effects.

[Ryland3210]

My TPS started acting up after 90K miles so tonight I put in a new Harley TPS. I set TPS at wide open throttle and dont much care what it is at idle.I could only get 4.77 volts at WOT. For this particular bike it likes 4.84 at WOT. If I left the screws out of the Harley TPS it would go up to 4.93 but the slots wouldn't line up with the screw holes. So I put in my spare Guzzi TPS and set it to 4.84 at WOT. No troubles works great again,Just thought I'd let you all know.

 

Interesting experience.

If the TPS is calibrated at WOT, and left to its own devices everywhere else, then logic dictates it will not be optimum at low to medium throttle openings.

It suggests that the engine is fairly forgiving of deviations from optimum, which is not surprising, based on the charts. However, while WOT performance is probably the same, less than ideal economy and part throttle performance would be expected. It is difficult to tell how performance is affected at part throttle. A variation of only a few degrees at, let's say, 10%-20% open can make a significant difference without being detected or measured when riding. Only a dyno test would make this easy to compare.

 

I'll stick to calibrating at the idle, and just make sure the TPS reaches 4.87 at WOT. Why 4.87? Because the chart from TuneBoy shows the highest entry of 84.56 degrees, which corresponds to 4.87 on the MPH chart.

 

 

 

The 4k flat spot is not influenced by the richness/weakness of the mixture.

I rectified the mixture and got better torque everywhere but the flat spot basically remained.

Remapping or a PC is the only way to adress mixture changes. The TPS is a measuring device, not for mixture tuning. Changing the calibration of the TPS changes the adresses in the map over a large area. This can cause a lot of unwanted side effects.

 

Please be specific about your first statement. Obviously mixture does affect power output, and the relatively lean mixture at 4000 RPM with standard tuning can explain a reduction in power output.

 

What do you mean by "rectified the mixture"?

[Guest ratchethack]

Obviously mixture does affect power output, and the relatively lean mixture at 4000 RPM with standard tuning can explain a reduction in power output.

Not very much, John -- and only within a relatively tiny range, assuming A/F is off to start with at a given RPM. If it were possible to "map out" every flat spot or dip on torque and power curves, this would certainly be convenient, but this is nowhere near the case. Many many other factors contribute to the shape of torque and power curves that are not related to A/F in the slightest. As Ernst has mentioned, the infamous 4K RPM "flat spot" is NOT a function of A/F.

 

As an example of this, (proven in the case of my own bike) replacement of the stock crossover with a Stucchi crossover nicely "fills in" the all-stock "dip" on the torque and power curves with NO change in A/F wotsoever.

 

Check Doug Lofgren's charts and comments here:

 

NOTE: This isn't the series of charts I was looking for (Doug's stuff is suddenly not that easy to find) but they illustrate the principles involved.

 

http://www.visi.com/~moperfserv/mgv11ex.htm

[Ryland3210]

Not very much, John -- and only within a relatively tiny range, assuming A/F is off to start with at a given RPM. If it were possible to "map out" every flat spot or dip on torque and power curves, this would certainly be convenient, but this is nowhere near the case. Many many other factors contribute to the shape of torque and power curves that are not related to A/F in the slightest. As Ernst has mentioned, the infamous 4K RPM "flat spot" is NOT a function of A/F.

 

As an example of this, (proven in the case of my own bike) replacement of the stock crossover with a Stucchi crossover nicely "fills in" the all-stock "dip" on the torque and power curves with NO change in A/F wotsoever.

 

Check Doug Lofgren's charts and comments here:

 

NOTE: This isn't the series of charts I was looking for (Doug's stuff is suddenly not that easy to find) but they illustrate the principles involved.

 

http://www.visi.com/~moperfserv/mgv11ex.htm

 

Without quantifying things, it's all subjective, and we could carry on that kind of discussion all day without advancing the cause of objective reasoning. "very much", "relatively tiny range", "rectified", "basically", and similar subjective terms are not enlightening. I look forward to motoguzznix answer.

 

F/A mixture affects output. That's clear. Obviously other factors do too. No one denies that. It's obvious. I want to know what motoguzznix experience is, and which other factor(s) he believes caused the flat spot. Furthermore, his message is not clear on how he "rectified" the mixture and to what extent. We may learn something from that.

 

In any event, I think this discussion should move to another thread on the subject of flat spots.

[luhbo]

.... Obviously mixture does affect power output, and the relatively lean mixture at 4000 RPM with standard tuning can explain a reduction in power output.

.....

 

Think of what you've written here. In the map you can only see the amount of injected fuel, or, even more correct, the basis for the pulsewidth calculation at this map point.

 

You cannot see any sign of how much air is delivered to the engine at this point (depending on balancing, exhaust, airbox size/shape and many more). Even a wideband O2 sensor can't give a 100% correct answer to the question of how much fuel gets correctly burned and how much of it gets lost.

 

And before we forget mentioning it: of course the PF3 and the PF4 are different sensors. They may be cheap (in production), but such a difference is never caused just by tolerances.

 

Hubert