Greg Field Posted March 3, 2009 Posted March 3, 2009 At least the sensor would be in liquid as it was designed. Another important discovery! Well, I consider grease and assembly lube liquids. They work fine.
Greg Field Posted March 3, 2009 Posted March 3, 2009 30729331 is used both for air and oil temp sensors on 1100 Sport, I guess that is what G2G meant. Maybe Guzzi found out it was a bad idea, I don't know. No other model I know of used that for the oil. Also used on many of the California series.
Dan M Posted March 3, 2009 Posted March 3, 2009 30729331 is used both for air and oil temp sensors on 1100 Sport, I guess that is what G2G meant. Maybe Guzzi found out it was a bad idea, I don't know. No other model I know of used that for the oil. Wow, I've never seen that type of sensor in anything but an air box. Where do they mount it? Here's a 30729331
raz Posted March 3, 2009 Posted March 3, 2009 It's mounted through a little hole in one of the valve covers. I guess it gets an occasional splash of oil but when that doesn't happen, it has to settle for the temperature of the air in there. It also gets a considerable amount of cooling from outside but that effect may be more due to the whole valve cover being cooled than the sensor body - and maybe the oil does get a little cooler too. As long as I don't ride in rain or very cool weather, it pretty much agrees with my oil dipstick thermometer. Theoretically I think it's much worse than the V11 solution but I have no practical reason to alter it.
dlaing Posted March 3, 2009 Posted March 3, 2009 surprisingly rough curve! Accuracy of sensor is not shown, but any inaccuracy of sensor is more forgiving when the ECU interprets between 30 and 40 degrees at the 103 difference, right? °C ohms mV ADC Difference to next ten degrees -40 100950 4927 1009 X -30 53100 4863 996 13 -20 29120 4755 974 22 -10 16600 4586 939 35 0 9750 4333 887 52 10 5970 3996 818 69 20 3750 3571 731 87 25 3000 3333 683 48 30 2420 3087 632 51 40 1600 2581 529 103 50 1080 2093 429 100 60 750 1667 341 88 70 525 1296 265 76 80 380 1011 207 58 90 275 775 159 48 100 205 601 123 36 110 155 468 96 27 125 100 313 64 this row is 15° change so, (32*2/3=~21)
Dan M Posted March 3, 2009 Posted March 3, 2009 Theoretically I think it's much worse .... Well boys & girls we've been through 23 pages of theories, opinions, facts and fiction. We've studied charts & graphs. Mused to paste or not to paste. Brass or plastic. Add heat-sinks? Have an air gap? What does it do? Does it do anything? Can the computer read it? Make it adjustable? Somewhere in all of this, maybe in one of the several PMs from various people I've received through the course of this, it was even indicated that those sensors were just switches used to turn on a cooling fan!?!? Here's a summary as I see it. The sensor has a definite effect on fuel mixture. Moto Guzzi uses coolant temp sensors on air cooled engines and calls them oil temp sensors, and sometimes uses air temp sensors to test oil temperature. Fuel injection for an air cooled engine is an adaptation of systems designed for liquid cooled engines. The brass holder allows more heat to get to the sensor than the plastic and is less likely to break under a ham fist. Adding paste to the holder allows more heat to get to the sensor's probe. If the bike came with a plastic holder it was likely mapped with a plastic holder. Adding heat to the sensor will lean the mixture. A leaner mixture will improve gas mileage. Too lean a mixture can create some running problems. We still don't know if the plastic was used for a reason. Sensors have various ranges of accuracy that nobody can agree on. There. That should get at least somebody grumbling.
helicopterjim R.I.P. Posted March 3, 2009 Posted March 3, 2009 So if there is a flaw or fault with the oil temp sensor will it cause the ECU to go to limp home mode?
Guest ratchethack Posted March 3, 2009 Posted March 3, 2009 There. That should get at least somebody grumbling. Yep. That oughtta do it. (Doesn't seem to take much, particularly in Winter.) BUT WAIT! THERE'S MORE TO COME! Will advise, (Part X)
Dan M Posted March 3, 2009 Posted March 3, 2009 So if there is a flaw or fault with the oil temp sensor will it cause the ECU to go to limp home mode? I don't even know if the MM unit has a limp in mode.
gstallons Posted March 3, 2009 Posted March 3, 2009 There should be some kind of "limp-in" mode. The non O2 sensor engines use tps, ats,rpm,ckp and cts(aka oil temp) for inputs. For those of us that know open and closed-loop systems that is not many inputs. I don't know what this system does in the event of a sensor input problem?
Dan M Posted March 3, 2009 Posted March 3, 2009 There should be some kind of "limp-in" mode. The non O2 sensor engines use tps, ats,rpm,ckp and cts(aka oil temp) for inputs. For those of us that know open and closed-loop systems that is not many inputs. I don't know what this system does in the event of a sensor input problem? I'm thinking the only limping is what you'll be doing after pushing it. On a non closed loop system it probably depends on what you loose. CTS? It will go rich or lean if it is open or shorted. ATS? Not much. TPS? I think it will stall on throttle opening. CKP/RPM? No start. Start unplugging stuff and see
raz Posted March 3, 2009 Posted March 3, 2009 My impression is that the Weber-Marelli components are not bad at all (most if not all problems we have discussed are implementation flaws by Guzzi) and since the ECUs do know when a sensor is broken I would be very surprised if there isn't some error handling (they just need to assume a temperature, not much more to it). But it's just my guess. The only sensor you really can't do without is the phase/rpm sensor.
Dan M Posted March 3, 2009 Posted March 3, 2009 My impression is that the Weber-Marelli components are not bad at all (most if not all problems we have discussed are implementation flaws by Guzzi) and since the ECUs do know when a sensor is broken I would be very surprised if there isn't some error handling (they just need to assume a temperature, not much more to it). But it's just my guess. The only sensor you really can't do without is the phase/rpm sensor. I'd be surprised if it would stay running on throttle opening without a TPS signal.
raz Posted March 3, 2009 Posted March 3, 2009 I'd be surprised if it would stay running on throttle opening without a TPS signal. For a limp home mode it could assume a very low throttle and you'd have to try matching it... but they probably don't. The MyECU, if anyone is interested, can be put in sensorless (meaning temps and pressure) mode by flipping a dip swicth inside it. AFAIK there is no automatic cancelling of out-of-bounds inputs. A separate switch will disable the lambda sensor(s) so it just runs open loop. I tried to get Cliff to implement sanity checking for the lambda sensor input, so it would go open loop by itself if the signal goes insane, but I got no response.
dlaing Posted March 4, 2009 Posted March 4, 2009 So if there is a flaw or fault with the oil temp sensor will it cause the ECU to go to limp home mode? It would depend on the fault. If disconnected I think it would run lean. As an experiment it might be worth pulling the wires to see how it runs, but don't go burning valves! Dude! I thought you had died! Looking at your posting history, however, you have been back since November. A belated Welcome Back to You! Do you still have your Guzzis?
Recommended Posts