engine oil temp sensor

[emry]

But missiles are much much much much KOOLER!!! Thats why we have EFI on a 1950's AL lump in the first place right?

 

Not that I want to cut this short of 40 pages, (sorry Greg, I didn't invest) how about a simple experiment Ratch. Measure the divided voltage of the sensor and the actual head temp at the same time. That way you can actually post some thermal inertia data that would make everyone happy. Then you could scientifically decide if you needed MORE mass or less MASS to the newly spec'ed sensor. (What if the head actually cools faster than the sensor!!!! OH DEFICATION!!!) Less mass and no plastic or brass holder.

 

Wait, mine is made out of copper. OH DEFICATION!!, what do i do? Elmer's Paste in the gap? I'm so confused but I refuse to re-read GIGABYTES of speculation. Sorry, I have to work.

 

Oh yeah, from the divided voltage you can calculate the other half of the resistor in the ECU. But only if you study Ohm's law really hard!!! For those who just want to know.

 

Ratch, I am glad you are attempting to at least spec out a sensor that would do the job the Italian's wanted it to do in the first place. I may rub you once and awhile but it is all in good fun.

 

I'll just thank Cliff for making a simple ECU that I can do what I want with. Like correct the poor engine mapping temp corrections.

 

 

I MADE 34!!! YES!!!! Had to edit that in....

[Guest ratchethack]

. . .how about a simple experiment Ratch. Measure the divided voltage of the sensor and the actual head temp at the same time. That way you can actually post some thermal inertia data that would make everyone happy. Then you could scientifically decide if you needed MORE mass or less MASS to the newly spec'ed sensor. (What if the head actually cools faster than the sensor!!!! OH DEFICATION!!!) Less mass and no plastic or brass holder.

Thanks for the input, Emry.

 

I beleive there are times when the head does cool faster than the OE sensor body, as noted in detail previously. It fits my understanding, and explains much of what I've observed.

 

The difficulty with what you've suggested above is that to be of any value, these measurements need to be taken SIMULTANEOUSLY OVER TIME and at all possible combinations of speed, load, and air flow (again, as mentioned previously). Otherwise you're looking at a "shapshot" of the point where highly dynamic readings happen to be at a point in time, long past the point where they have any relation with where they are in actual operation on the road.

 

I had much the same thoughts as yours before I fully considered one of the biggest of the many compromises we're strapped with here -- that being the heat flow problem and thermal inertia of the relatively high-mass OE sensor body. Those thoughts went away very quickly once I understood it.

 

The performance of the OE sensor is highly subject to wonky outputs when air flow is insufficient, and when the motor is fully warm and making off-on throttle transitions. I'm limited to testing only my Guzzi, no 2 being alike, but I believe the operating principles of the sensor/holder are the same for all, regardless of map and engine config.

 

Again, as posted numerous times, a more ideal way to take the proper measurements would be to data log all measurements on a timeline -- both CHT and sensor body temps along with sensor outputs in all possible transitions on the road being pretty ideal. Again, the only alternative to measuring all this on the road would be my oft-mentioned boatload of diagnostic gear connected by cable and wire to the Guzzi on a brake dyno in a wind tunnel.

 

This is just me, but I find either alternative above a just a mite prohibitive at the moment.

 

OR, alternatively,

 

Having identified the thermal inertia problem, road test a correctly spec'd sensor without the problem , and compare the symptoms on the road.

 

Will do, and will advise, soon as I get the oft elusive ROUND TUIT.

[Greg Field]

Excellent. One page closer to financial independence . . .

[dlaing]

The sensor I’ll be using is thermoplastic. The only metallic part is the thermistor itself. I presume you mean to ask about the holder? Haven't got that far yet, but no aluminum. Will fab this up, and likely use brass for the base part that screws into the head, and something relatively non-conductive for the rest of the holder. Copper would be preferred over either for the base:

 

Thermal conductivity

 

copper -- 231 BTU/(hr-ft-°F), or 10.15 watts/in-°C

brass -- 69 BTU/(hr-ft-°F), or 3.03 watts/in-°C

aluminum -- 36 BTU/(hr-ft-°F), or 1.58 watts/in-°C

 

SOURCE: http://www.engineersedge.com/properties_of_metals.htm

 

But then, this ain’t exactly rocket science, is it? There are many effective ways to deliver an enemy combatant to his 72 virgins that are more effective than a sling and stone, but no need to call for a laser-guided missile strike when a well-planned and carefully placed single head shot from a thousand yards will do quite nicely.

Thanks for replying.

"The thermal inertia of a material is defined as a the square root of the product of the material's bulk thermal conductivity and volumetric heat capacity"

-- http://www.absoluteastronomy.com/topics/Vo...c_heat_capacity

 

Your Aluminum conductivity number is wrong. It should be 136

 

According to this site

http://www.ibrt.ch/electronics/cooling.html

copper has a conductivity that is 1.7 times that of aluminum

Copper has

a thermal conductivity of 400W / m*K

Aluminum has

a thermal conductivity of 235W / m*K

Which is a very different ratio than what you posted.

And furthermore,

copper has 1.4 times the Volumetric Heat Capacity of Aluminum.

copper has a heat capacity of 3.37 J / K*cm^3

aluminum has a heat capacity of 2.40 J / K*cm^3

 

So if I understand it correctly, we multiply and square to find thermal inertia we get:

Copper

(400 * 3.37)*(400 * 3.37)=1,817,104

Aluminum

(235 * 2.40)*(235 * 2.40)=318,096

 

So the Copper has far more thermal inertia,

But is low thermal inertia what we want in our sensor and adapter?

We want high thermal inertia from the engine (which we have) and we want high conductivity to the sensor tip and low VHC Volumetric Heat Capacity from the sensor and adapter so that it reacts quickly to the change in head temperature.

We want high thermal diffusivity!

http://en.wikipedia.org/wiki/Thermal_diffusivity

Aluminium 8.418 × 10-5

Copper 1.1234 × 10-4

Which makes Aluminum the more favorable for the material between head and sensor tip!!!

Aluminum will heat up faster and cool faster, closely following the thermal inertia of the cylinder head.

I don't know the Thermal Diffusivity for brass, but I'll bet it ain't what we want, excluding properties of weather resistance.

Aluminum may have some other advantages, like no dissimilar metal issues of differing heat contraction or oxidation.

Fascinating stuff!

But I still do have concerns about the sensor getting too hot

A longer probe and a ceramic or bakelite insulator would do the trick!

Or if your thermistor has a high enough temperature rating and the volts could be brought down so it never reads over 125C, you might be in business.

Would you care if are getting 200C readings that are likely off the ECU charts and out of the thermistor's recommended operating range. Might be a good reason to take voltage readings...

 

If your thermistor is metal, are you sure it is a thermistor and not a thermocouple?

[Guzzi2Go]

...SIMULTANEOUSLY OVER TIME...

 

Now, there's a thought...

[Greg Field]

Six more pages . . .

[GuzziMoto]

Six more pages . . .

 

Jesus guys, why are you so mean to poor Ratchet. He could be on to something. Clearly the CHT sensor is not accurate or the bike would run perfect. Why would you suggest that he needs to measure the reading it provides and compare that to what the actual cylinder head temp is when he already has proof that it is not reading correctly, the aforementioned running issues. Clearly part of the problem is it needs a heat sink, after all there are fins on the motor so there should be fins on the temp sensor or they'll never read the same. And clearly the little temp sensor Guzzi uses stock has way more thermal inertia then the motor does so you need to address that to. How is the stock temp sensor going to keep its temp the same as the motors? Obviously it won't. It's gonna heat up and get substantially hotter with out something to pull heat back out of the sensor into the motor, or is it keeping a fresh flow of heat going into the sensor so you don't end up with stale heat in the sensor(hence the need for the heat sink).

It's a wonder it even runs as delivered by Guzzi.

This from the same guy who said that if God wanted my Guzzi to have better brakes Luigi would have installed them at the factory.

[Guest ratchethack]

Your Aluminum conductivity number is wrong. It should be 136

Good catch, Dave. Thanks for actually paying attention. This evidently puts you in the company of a mighty small number indeed.

 

I stand corrected, and really oughtta be horsewhipped for a such gross error!

 

This is one of the reasons I post this stuff here -- as a check. I b'lieve this is the first error of its kind that I've made in this thread, despite the continuing cavalcade of wildly ignorant and false accusations with nothing credible to back them up, which will no doubt be continuing on as before. I've corrected my post. My most humble apologies to the entire Tech Forum!

Which makes Aluminum the more favorable for the material between head and sensor tip!!!

Aluminum will heat up faster and cool faster, closely following the thermal inertia of the cylinder head.

I don't know the Thermal Diffusivity for brass, but I'll bet it ain't what we want, excluding properties of weather resistance.

Aluminum may have some other advantages, like no dissimilar metal issues of differing heat contraction or oxidation.

Fascinating stuff!

You've raised what appears to be a valid point, Dave. Now this'll take a bit more noodling than anticipated before proceeding here (but not much). I note that the OE plastic holder uses a copper base between the head and sensor tip, so both it and the brass holder were evidently spec'd for more desirable properties other than thermal diffusivity. No, I (still) don't accept the previously oft-presented idea that the OE plastic holder was spec'd primarily on the basis of low cost. IMHO, considerations of dissimilar metal galvanic (electrolytic) response, expansion/contraction, and oxidation are insignificant here, and have no bearing on any aspect of the function or performance of the holder. Let's not clutter up the valid considerations here with irrelevant and insignificant information, as if we haven't had enough of that already.

But I still do have concerns about the sensor getting too hot

A longer probe and a ceramic or bakelite insulator would do the trick!

Or if your thermistor has a high enough temperature rating and the volts could be brought down so it never reads over 125C, you might be in business.

Would you care if are getting 200C readings that are likely off the ECU charts and out of the thermistor's recommended operating range. Might be a good reason to take voltage readings...

WHOOOOOOOOAH there, Dave! Now the old speculation train has jumped well free of the track (again), and I'm afraid it's headed down yet another dirt road. The thermistor I'll be using has an output chart that follows the OE thermistor within <10% error, as noted previously. I'll post this later, so a direct comparison can be made at 10ºC intervals.

 

Any cylinder head going over 200ºC (393ºF) has far more serious problems than anything a head temp sensor has anything to do with. Being engulfed in a gasoline fire, or parked too close to a smelting furnace might be among them. . .

If your thermistor is metal, are you sure it is a thermistor and not a thermocouple?

A thermistor IS NOT a thermocouple, and n'er the twain shall ever meet. You can look that up in your favorite source, Wikipedia, where I'll venture that even the anonymous open sources there might be able to draw a proper distinction. But by all means, do use what you find there as the basis for looking up a valid distinction by taking the next step, and check the original sources (and others). The Web is full of 'em (but you've got to look).

[Greg Field]

Any of you ever run coon hounds?

 

There is no more glorious sound on earth than a brace of them on the scent and then at the base of the tree when they've cornered one.

 

Being imperfect creatures, dogs sometimes blow it and sit there barking at an empty tree. I believe they sense they've blown it, so the tone of their bark changes, and it becomes a quite unpleasant racket. I sense that tone here.

 

Who knows? It's been nearly 30 years since I chased hounds through the pitch-black woods of Wisconsin. Perhaps my ears are off?

[dlaing]

A thermistor IS NOT a thermocouple, and n'er the twain shall ever meet. You can look that up in your favorite source, Wikipedia, where I'll venture that even the anonymous open sources there might be able to draw a proper distinction. But by all means, do use what you find there as the basis for looking up a valid distinction from a reputable source. The Web is full of 'em (but you've got to look).

from wikipedia

"Thermistors differ from resistance temperature detectors (RTD) in that the material used in a thermistor is generally a ceramic or polymer, while RTDs use pure metals."

Since yours is metal, maybe it is an RTD????

But I think both the thermistors and RTDs are covered in a heat resistant plastic or a ceramic.

Thermocouples are also made of metal!!!

PS, no need to horsewhip you over a typo or any other mistake you willing to fess up to.

Maybe you mean the sensor tip is metal, and you can't actually see the thermistor, like with our OEM sensor????

How do you know it is metal?

[dlaing]

Any of you ever run coon hounds?

 

There is no more glorious sound on earth than a brace of them on the scent and then at the base of the tree when they've cornered one.

 

Being imperfect creatures, dogs sometimes blow it and sit there barking at an empty tree. I believe they sense they've blown it, so the tone of their bark changes, and it becomes a quite unpleasant racket. I sense that tone here.

 

Who knows? It's been nearly 30 years since I chased hounds through the pitch-black woods of Wisconsin. Perhaps my ears are off?

So what have you been chasing?

You lost the scent of the coon as soon as you wrapped the crap in duct tape.

Which is better, your method or Guzziology's?

[Greg Field]

Dave R. has never owned a V11. I ride one nearly every day. I'll let you infer your own answer from that.

 

If I were to rig up a variable input to fool the ECU it would be a variable voltage input that bypasses the sensor entirely. All the sensor does is supply a voltage to tell how hot the engine is. The engine is able to run on a leaner mixture far before that silly sensor tells the ECU to lean out the mixture. Set it to the "I'm cold" voltage for easy starting and then switch it to the "I'm hot" voltage a few minutes later. That would net you the best milage. This is the only approach to what you guys are trying to accomplish that nets any sensible amount of advantage over the sensor holder (either one) with either sold tip contact or good heat-transfer goo. This would be a "round tires" approach. Can you dig it? Or must we sit through 40 more pages of bloviation?

[Guzzi2Go]

42 pages.

 

- What is the answer to the Ultimate Question of Life, the Universe, and Everything?

- 42

...

- What was the question?

- What do you get if you multiply six by nine?

[dlaing]

Dave R. has never owned a V11. I ride one nearly every day. I'll let you infer your own answer from that.

 

If I were to rig up a variable input to fool the ECU it would be a variable voltage input that bypasses the sensor entirely. All the sensor does is supply a voltage to tell how hot the engine is. The engine is able to run on a leaner mixture far before that silly sensor tells the ECU to lean out the mixture. Set it to the "I'm cold" voltage for easy starting and then switch it to the "I'm hot" voltage a few minutes later. That would net you the best milage. This is the only approach to what you guys are trying to accomplish that nets any sensible amount of advantage over the sensor holder (either one) with either sold tip contact or good heat-transfer goo. This would be a "round tires" approach. Can you dig it? Or must we sit through 40 more pages of bloviation?

I am half way to digging it!

Everything so far seems to be a half-assed solution.

Cooling fins get too cool at high speeds or wet weather.

Adding goo can make it too hot, when hot.

Not adding goo can make it to cold, when cold.

Brass adapter makes it too hot when hot, but improves it in some conditions.

Duct tape is major bodgy and could make it too hot in some conditions, but blocking the wind will make it run more consistently at varying speeds and in varying weather.

An adjustable variable voltage input is a great idea, but if you forget to use it properly, it is a lousy solution....

 

Increased Thermal DIFFUSIVITY will only make a slight difference, but combined with conductive goo and insulation it could add up to a fairly substantial improvement.

I still suspect the sensor is getting too hot. I know for sure it is too cool much of the time.

I can change the map, but that does not help others, and there are still consistency issues.

A volt meter or AxeOne could provide a better clue.

I'd love to see how hot Pete Roper could get one and show the temp with Axe One.

I guess I could try a meat thermometer...

[mike wilson]

Again, as posted numerous times, a more ideal way to take the proper measurements would be to data log all measurements on a timeline -- both CHT and sensor body temps along with sensor outputs in all possible transitions on the road being pretty ideal. Again, the only alternative to measuring all this on the road would be my oft-mentioned boatload of diagnostic gear connected by cable and wire to the Guzzi on a brake dyno in a wind tunnel.

 

Your local college/high school/university may have just the equipment you are looking for _plus_ an excitable, nubile student wishing to undertake something almost exactly like this for either a project or other academic effort. We use dataloggers that can log four channels, will fit in a jacket pocket at a push and can log temperature, volts and current and, with a bit of manipulation, could be made to monitor things like road speed.

 

http://www.sciencescope.co.uk/logbook_gl.htm

 

Just a thought........

 

Keep up the good work.