Engine Temperature Sensor

[docc]

Altering the holder for the Engine Temperature Sensor has been the topic of much trial, error, and debate.

After almost eight years and over 55,000 miles, I decided to swap to the brass sensor holder (GU 30163301) and packed the housing with copper paste. Not that I needed to change it, but it was all the rage at the time and my Guzzichondria got the best me. I wasn't sure it ran any better, in fact I thought it got a bit worse, but went on to chase down a series of other, mostly electrical, issues.

I rode the Sport two years, around 14,000 miles, before switching back to the original plastic/composite/Bakelite holder with no packing paste. Right away, I felt she ran a bit better. Not a lot, but a bit.

So, taking the path less traveled, with the help of TFMKasRH (The Former Member Known as RatchetHack), I removed my thermister from its crimped housing using a Dremel with cut-off wheel (I figured I could JB Weld it back if needed since new sensors are 105 USD), fabricated a Delrin holder using a 22mm length of copper stud and 5mm air gap with stainless steel set screws.

I have only ridden it five or six times with a total of maybe 250 miles. But, here are my observations so far:

1) Low angle throttle transitions are much smoother. Rolling into a corner, I've become accustomed to some unsettling sputtering and burble which caused a little hesitation as the throttle was rolled on for the exit. This transition is notably smoother making for much smoother and reliable cornering transitions.

2) Idle and low throttle angles, especially at low (under 3000-3500) rpm are more stable. The motor has never liked to plod along under 3000 rpm (I know it's a bad idea for other reasons) always bucking and sputtering especially right at an indicated 2800 (my tach is 300 rpm high BTW). This sputter is entirely gone.

3) There is much improved high temperature sensitivity at traffic lights and parking lot maneuvers. It has been quite hot here and the first test ride was some spirited riding into town then sitting in traffic and navigating a parking lot at 96 degrees Fahrenheit (35˚C). There was nary a hiccup or sputter or stumble. With nearly 72,000 miles and ten years riding this bike, this is unprecedented.

4) My fuel economy is unchanged. TFMKasRH noted a 10% improvement which I thought would be great. Alas, so far, not so for me. Summertime I get 35.5 - 37.5 mpgUS (sure 40-41 at steady throttle with no cogging and hair shirt corner exits, but who wants to ride like that?); wintertime (under, say 60ºF) 32.5-34.5 mpgUS with the lower figures closer to my 40˚F/ 4,4˚C riding limit.



I am really enjoying the smoother, more stable character. On a zero to ten scale (zero= not running, ten= running phenomenally), my Sport is usually an 8-9; 10 on a cool crisp morning at 5000+ rpm, and 6-7 in hot traffic. (Well, there was that once or twice it was a zero!) Right now, she's giving me a solid 9-10 even with the heat.

With this much seat time for comparison, I'm looking forward to getting some time and miles for a more complete impression. Hey, maybe I'll end up with those other 3 or 4 mpg!

 

EDIT: August 21, 2014 - with almost 20,000 miles on this "modification", I can say it has performed flawlessly. I was concerned the thermistor would just burn up, but no. The wiring connector did come undone once, and the holder loosened one time. My fuel economy never improved, but the harsh running in hot conditions never returned.

[docc]

Woops, meant to add this:

[GuzziMoto]

Yeah, this one was beat to death at one point. His idea to lower the temp the sensor reports to the ecu to cause the ecu to add more fuel (or at least not take away fuel) was not a bad idea. It only got stupid when he kept insisting that his mod MUST be inproving the accuracy of the sensor because the bike ran better. If you have a problem with your Guzzi in hot conditions running lean then modifying the sensor to report a lower temp is not a bad idea. I personally would not do it but my bikes also have to run in cool conditions that may cause an issue with that. Plus it is a crude modification in that it adds more fuel everywhere and not just where you need it. But it is likely better then nothing.

I went with Power Commanders so I could adjust the fueling in some areas to add more fuel and to add less in other areas. But the ETS mod is much cheaper.

[Skeeve]

Yeah, this one was beat to death at one point. His idea to lower the temp the sensor reports to the ecu to cause the ecu to add more fuel (or at least not take away fuel) was not a bad idea. It only got stupid when he kept insisting that his mod MUST be inproving the accuracy of the sensor because the bike ran better. If you have a problem with your Guzzi in hot conditions running lean then modifying the sensor to report a lower temp is not a bad idea. I personally would not do it but my bikes also have to run in cool conditions that may cause an issue with that. Plus it is a crude modification in that it adds more fuel everywhere and not just where you need it. But it is likely better then nothing.

I went with Power Commanders so I could adjust the fueling in some areas to add more fuel and to add less in other areas. But the ETS mod is much cheaper.

 

'Moto, you're combining assertions that weren't related, & misquoting as well. His mod was specifically targeting the problems that many forum members have reported w/ excessive fuel consumption in cooler weather, while simultaneously running too lean in hot weather. His proposed mod to wire in a pot so that he could dial in his own fueling corrections was an initial foray into resolving this issue, a "brute force attack" if you will. The later replacement of the hot water sensor used by the factory in a "less-than-ideal-despite-official-factory-part bodge" by a GM air-temp sensor placed closer to the head so that it can react more quickly to cylinder head temp changes was entirely inspired and seemed to correct the empirically derived faults with the factory system.

 

The fact that he's still getting slagged for his vociferous support of his mod that certain individuals w/ personal axes to grind tried to belittle says more about our community than it does about the validity or efficacy of the mod. Thanks to Docc, we now have another data point on the "the factory setup is a craptastic bodge & the v11lm.com community can come up w/ a better solution" graph.

 

Ride on!

[GuzziMoto]

'Moto, you're combining assertions that weren't related, & misquoting as well. His mod was specifically targeting the problems that many forum members have reported w/ excessive fuel consumption in cooler weather, while simultaneously running too lean in hot weather. His proposed mod to wire in a pot so that he could dial in his own fueling corrections was an initial foray into resolving this issue, a "brute force attack" if you will. The later replacement of the hot water sensor used by the factory in a "less-than-ideal-despite-official-factory-part bodge" by a GM air-temp sensor placed closer to the head so that it can react more quickly to cylinder head temp changes was entirely inspired and seemed to correct the empirically derived faults with the factory system.

 

The fact that he's still getting slagged for his vociferous support of his mod that certain individuals w/ personal axes to grind tried to belittle says more about our community than it does about the validity or efficacy of the mod. Thanks to Docc, we now have another data point on the "the factory setup is a craptastic bodge & the v11lm.com community can come up w/ a better solution" graph.

 

Ride on!

I'm sorry, I didn't realize that saying "His idea to lower the temp the sensor reports to the ecu to cause the ecu to add more fuel (or at least not take away fuel) was not a bad idea." and pointing out where the debate got stupid was slagging him. I thought it was a positive comment. I apologize.

[docc]

"Where are we going and what are we doing in this handbasket?"

[Kiwi_Roy]

I think reading Ratchets many postings was one of the reasons I decided on MyECU so I could freely make adjustments to the effect the different sensors have, it's still a black box, but at least it has a window.

At one stage I tried changing the TPS voltage on the fly to see how it effected performance, now I have access to map values I can see I probably didn't change the voltage enough to have a noticeable effect.

Roy

[Dan M]

Woops, meant to add this:

 

Thanks for posting your results Docc.

I did quite a bit of work with RH picking & modifying GM sensors that ranged the same as the MG sensor as well as side by side testing.

My bike is also much smoother in the former trouble range (3000-3500RPM small throttle opening with a hot motor) regardless of ambient temp. Tank range has increased as well.

It is a shame people insisted on arguing about it at the time without trying the mod or testing anything themselves.

[luhbo]

....

It is a shame people insisted on arguing about it at the time without trying the mod or testing anything themselves.

 

It's not a shame, it's a must. Because 98% of this entire thread were nonsense.

 

As GuzziMoto said: it might be ok to cheat the ecu by altering one or more sensor signals, simulating a cheapo PC this way. Some resistors and a switch will do for that. But everything else in this context, in these postings, was as I've already said just plain nonsense.

 

Hubert

[Skeeve]

I'm sorry, I didn't realize that saying "His idea to lower the temp the sensor reports to the ecu to cause the ecu to add more fuel (or at least not take away fuel) was not a bad idea." and pointing out where the debate got stupid was slagging him. I thought it was a positive comment. I apologize.

 

My turn to apologize; I focused on your next line where you said, "Where it got stupid..." and went on to combine assertions from different experiments. I wasn't really accusing you of slagging him, just that he got slagged in general at the time for his sharing the results of his experiments.

 

All water under the bridge now. When I have time I'll likely follow in Ratch's footsteps and swap out the sensor for something w/ less time-delay/greater responsiveness. Other, more pressing projects intervene however, so it'll have to wait.

[Dan M]

I've already said just plain nonsense.

 

Hubert

 

Very true.

 

The argument from the beginning is no resistance values have been changed. Sure, if you added resistance or installed a sensor with greater resistance values you would be enrichening the mix or cheating as you say. That has not been done here so how is it cheating the ECU? Docc is using the factory sensor with the brass removed for faster response and the bike runs better. This is cheating? This is nonsense?

[GuzziMoto]

Very true.

 

The argument from the beginning is no resistance values have been changed. Sure, if you added resistance or installed a sensor with greater resistance values you would be enrichening the mix or cheating as you say. That has not been done here so how is it cheating the ECU? Docc is using the factory sensor with the brass removed for faster response and the bike runs better. This is cheating? This is nonsense?

Docc has modified his temp sensor so that it gives a different value then it used to, apparently lower as a lower temp reading would cause the ecu to provide a richer mixture and that is what it sounds like is happening. Good for him. He realized he had an issue with the way his Guzzi ran (though at least part of the issue may have been self inflicted by improving the heat transfer to the temp sensor when it seems like it was working well to begin with) and he took steps to correct it. The hows and whys of this have been beat to death and I doubt it is worth continuing the debate.

Good for you Docc. Glad to hear your bike runs well. My wifes V11 runs well too (with the stock ETS). My Griso has an issue with getting too hot and I have considered a similar mod to slow the heat transfer to the ETS so that it gives the ecu a lower temp reading. I don't think the ETS is the root of the issue and I realize that modding the ETS would be treating the symptoms and not the cause, but the bike has been gone over by a few mechanics and no one has yet figured out why it detonates when hot. It is not the only Griso to have this issue, but most do not. So clearly it is not the design of the temp sensor causing the issue. But modding it may be a band aid fix to help the bike run better.

[luhbo]

Very true.

....

 

Hi Dan, acting the smart ass again or what? Please keep things fair and quote other members correctly. Thank you.

 

Anyway, here's another one out of my "My favorite quotes" collection:

 

OK, Hubert. Confession time. You're right. Of all the photo's I've taken of the underside of V11 TPS's, turns out none are open. All are sealed, exactly as you insist. I made up the bullshit (your term) about cleaning the winding and wiper contacts through the open bottom out o' thin air, just to see if anyone was paying attention 3 years ago. I got Docc to go along with the Grand Charade way back then (and ever since), 'cause he's (nearly) as loopy and demented as I am, and well. . . this is the kind of thing loopy, demented people do for amusement.

 

Sure had you goin' for quite awhile, though, before you finally called us out, didn't we?

 

Hubert

[Dan M]

Docc has modified his temp sensor so that it gives a different value then it used to, apparently lower as a lower temp reading would cause the ecu to provide a richer mixture and that is what it sounds like is happening. Good for him. He realized he had an issue with the way his Guzzi ran (though at least part of the issue may have been self inflicted by improving the heat transfer to the temp sensor when it seems like it was working well to begin with) and he took steps to correct it. The hows and whys of this have been beat to death and I doubt it is worth continuing the debate.

Good for you Docc. Glad to hear your bike runs well. My wifes V11 runs well too (with the stock ETS). My Griso has an issue with getting too hot and I have considered a similar mod to slow the heat transfer to the ETS so that it gives the ecu a lower temp reading. I don't think the ETS is the root of the issue and I realize that modding the ETS would be treating the symptoms and not the cause, but the bike has been gone over by a few mechanics and no one has yet figured out why it detonates when hot. It is not the only Griso to have this issue, but most do not. So clearly it is not the design of the temp sensor causing the issue. But modding it may be a band aid fix to help the bike run better.

 

What Docc did was cut the brass away that was insulating the thermistor. That did not change the value, it just removed the lag time to read the actual temp. A lag time that was as long as minutes.

 

How many people have used PCs, Tune boys or other means to correct fueling and while at it were compensating for inaccuracies of a very, very slow sensor?

[GuzziMoto]

What Docc did was cut the brass away that was insulating the thermistor. That did not change the value, it just removed the lag time to read the actual temp. A lag time that was as long as minutes.

 

How many people have used PCs, Tune boys or other means to correct fueling and while at it were compensating for inaccuracies of a very, very slow sensor?

 

I understand what Docc did. The debate is about what effect that had on the system.

See, that is where the debate gets fuzzy. You say that the brass is insulating the actual sensor. In my experience, brass is a horrible insulator for heat. It actually conducts heat better then air, which is what you are "replacing" the brass with. I have yet to see anyone post any actual measured data that supports the theory that removing the brass and adding air increases the response time or the accuracy of the sensor. Or any evidence that the stock set up is "a very, very slow sensor". So then it comes down to opinion and belief, and that does not make for something worth debating.

Docc (and Hatchet Wacker before him) has made a change to his ETS that alters the value the sensor is sending to the ECU and that is changing the way his Guzzi runs. I don't have an issue with that. But there is no evidence that he has "improved" the accuracy or response time of the sensor. The only evidence is that his bike now runs better, and by the description of the change in running it sounds to me like it is running richer. Based on what I know of the Guzzi ECU that means the moddified ETS is telling the ECU that the motor is cooler. This makes sense to me as what was done to the ETS (remove the brass and replace it with air) I would expect to lower the temp reading of the sensor. If you choose to draw a different conclusion from that action/result, fine. But unless you have some evidence to support your claim......