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?

Now, there's a thought...

How about VSH (Variable Surface Heatsink)?

Now I am concerned that we will loose heatsink even before we reach page 40.... This will, of course, be discretely tucked in between the endless lines of nonsense produced by yours truly, so noone will ever notice. Go Ratchet, go!

Hmmmm, dunno... Ratchet has resorted to a nasty trick! Instead of adding new posts, he is changing existing ones. I guess we all need to chip in....

Actually, one would not find it anywhere, because basic principles of heat flow and thermodynamics speak quite the opposite. Nice try hiding behind pseudo scientific mumbo-jumbo, but there are a few that have, unlike yourself, actually read and comprehended these works and know when you are talking bullshit. Or, alternatively, prove that you are right! Why don't you run a calculation (if you can), that will explain the purpose of heatsink. and how did you select THAT PARTICULAR ONE. I hope there are other reasons than its $2 price tag. Remember how did you select the 500k resistor? Was it because it was cheap? Is that the "scientific" approach? Ho, ho, ho! It seems that the post #280 changes on daily basis! I bet a case of that the heatsink will soon disappear from it, just as air gap did (seemed to be a good idea back then), and thermo-goo found its way in (which was a no-no earlier), toghether with a 1k pot (is it really important that it is linear?) where previously 500k one was (blamed it on the multimeter, didn't you?). Pathetic! Is this what you call "science". It is just a proof that you have no cucking flue what you are doing. As far as heatsink goes, you have cylinder head with all its mass and surface (fins) attached to the other end of the sensor. If heat is in need of "escaping" that is the "open door". There is no need to provide an "emergency exit". Quite the contrary.

So, how does actually adding more MASS (heatsink) reduce THERMAL INERTIA and with it LAG TIME? I am still missing a dimensioning input for a heatsink. What does seat of your pants tell you? Small one? Big one? Or maybe n-one? ANALytical mind.... (you know)...

Puppetmaster indeed... On the other hand, this thread began to spill. We have considerable thermal inertia leakage with negligible lag time. Will someone mop it up? Or at least add a heatsink?

And if that's not enough, you can always splash a few buckets of thermal inertia on the bike. That'll fix it.

It is a NTC thermistor. We've been through that all over the last 30 pages. Well, how about an attempt (over 30+pages) to go for a consistent result, regardless of the bike, its configuration, conditions, time of the year and part of the world? I'd say it would be hard target to reach by linking everything to something as variable as weather. Why going that road? And I doubt very much that ratchet has in this particular case figured out what works on HIS bike. Or on any other bike for that matter.

A Carlsberg, Becks and Guiness sales rep walk into a pub, Carlsberg rep orders Carlsberg, Becks rep orders Becks, and the Guiness guy orders a Coke. They ask him: - How come you are ordering coke? - Well, if you are not ordering beer neither am I.

Well, look at this article: http://www.tetech.com/docs/thermoelectric_...ermocouples.pdf True, it is about thermocouples, not NTCs, but you'll get the gist... The following (important) statements are in there: I think especially this last statement provides answer to your question. Placement of the NTC inside the sensor apparatus is irrelevant, since you cannot influence it. That is why it is important to take care of #2 and #3.

"A" gets my vote! Reading a few posts back, I saw that we started with air gap, 500k pot, plastic washer and a heatsink, but without thermo-goo and considerations to change the map. Give it some 20 pages more and we will settle for no air gap, no pot, no washer, no heatsink, but thermo-goo + richer map.

WOW! Just how big of a culprit is it actually? How about a model, explanation, calculation, interpretation? What to hell, you can even throw a fistful of Philistines in the equation if necessary... Analytical mind....(well, you know)....

So you noticed that too. Funny how it seems the method to be acceptable now, when offered by "yours truly" and ridiculed when offered by someone else. It might, but it won't. I guess it all depends on the thermal inertia of the goo. That would be the same as adding an air gap (+ optional heatsink). Hard to get consistent results using that technique. Insulation MUST be applied on the outside of the adapter.

I see that this thread made quite an advance recently (in terms of number of posts, advance in terms of general understanding what is going on is somewhat limited). Unfortunately, I am a bit pressed for time (traveling a lot lately) so I'll try to keep it short/quick. If you experience high fuel consumption on cold days, you MAY try to improve thermal contact between sensor and engine. If you do that, you MAY experience symptoms of running lean (like pinging). If you do experience these symptoms and have a possibility to alter map (myECU, PCIII), you SHOULD fix the map (real solution). If you don't have the possibility to fix the map, you MAY add a resistor in series to temp sensor. Values in the range of 100-500 ohm seem appropriate (band aid). If you add big enough heatsink to your engine temp sensor you will turn it into air temp sensor. You SHOULD NOT do that. If you replace plastic sensor adapter with a brass one it will have the same effect as adding heatsink to the sensor. You SHOULD insulate the brass adapter. If you increase air gap between engine and sensor, you will make it more sensitive to ambient temperature. In an extreme case, you will turn it into air temp sensor. You SHOULD NOT do that. And yes, for those not sincerely interested in neither Philistines nor Bohemians, thus not willing to read the exceedingly long rant about them - Philistines = Palestinians, Bohemians = Czech.

Who are these individuals and what about their integrity? Why trusting them, even if what they write is in good faith? Could we have a discussion about dogma? Isn't that what certain "individuals of integrity" would call "groupthink" followed by despiseful spat? I choose to trust depending of what I read not who it comes from, and if what is written makes sense I'll accept it regardless of who is it coming from. I guess this is where we differ. Please understand "my attempt to mislead" as a simple experiment. Now I know that there is at least one person reading these posts with his brain switched on. And which of my posts he will believe and which not he can best judge himself.

Well, EVERYONE is posting their information (I am deliberately not saying argumentation) as a fact. How do you know who to trust? Going by their name? Length of their posts? Some people will not trust anything, some will trust everything. And then, there are those who will ask that claims are backed with some arguments. How good these arguments are, and what can you do with them (in terms of trust) is really up to you. I dare to say here that I am amongst a very few that offered any kind of argumentation around their claims. How good it is, is really up for you to judge, but I am still offering you that choice unlike some others who tend to call one a "Philistine" (would that be an insult? ) for the sake of argumentation. However, if that is your preference, I understand. Coming back to the topic, I am not assuming anything about thermistor's range, and I can't see where did you read that. I KNOW that thermistor's characteristic is a logarithmic function and that it will eventually tail off so the increase of resistance will be negligible beyond certain point. I also KNOW that a semiconductor it is made off will decompose not far beyond 200°C, which limits their applicability to measuring fairly low temperatures (like engine temps). However, for the sake of this argument, let us keep it between 0-120°C. It is quite sufficient for what we have in front of ourselves.

Hmmm, I was sure I posted this already, but looks like the post is gone now. Therefore, once again. If it appears twice at the end, I'll try to remove the copy. The post is to show how resolution of the temp sensing setup varies depending on the preload resistor. Resistor values used in this example are 1.5 and 15kohm.

This data set overthrows two statements made in this thread: 1) NTC is most sensitive in 90-110°C range, and this is why it is selected for the application. I am still missing definition of "sensitive". Is it dR/dT? Is it dT/dR? Or something else? If it is dR/dT then the NTC is "loosing on sensitivity" with increase of the temperature. What is maybe not so obvious from this calculation is that the sensitivity of the setup (if that is dU/dT) is actually dependent on the value of the preload resistor. I will post a calculation illustrating that in following post. 2) NTC (or better temperature sensing setup including ECU) does not have sufficient resolution to distinguish temperature changes above 90°C. This is the one I made and is so obviously wrong that I am surprised that it took so long for someone to counter it. I kind of hoped Dan would easily cover it, but he chose not to for whatever reason. Pity, since that would put the discussion quickly back on track. You could have also covered it, but I did not hold my breath. In terms of fishing, call it a "bait". Glad that raz took it, as it shows how a sensible discussion could look like. One makes an unsubstantiated claim, and bang, there is a quality argumentation to either corroborate or overthrow if. Of course, there is an alternative involving Philistines, but I just fail to see how they fit in the context. What it tells me is that the temp sensing setup resolution at temperatures between 90-110°C is in average 1/36 or ~0.3°C. Sufficient for accurate measurement of temperature even in that area. Goes with countering the statement I made about inability of the setup to measure temperatures above 90°C. See #2 above. First of all, it is not assumed it is specified. This is what a design or a model starts with. Measurement or experiment is 2nd step in the process executed to confirm the validity of design/model. Which brings me to the question: "Where is YOUR model/design for air-gap or heatsink dimensioning which you are trying to prove by experimentation?" Second, what tells me what ECU does with that info is the OilT line. The RELEVANT line. Stock and MyECU lines have been posted here earlier. Stock OilT line tells me that WM adjusts mixture even in high temperature range, which I may choose to compensate for by adding a resistor (of certain value) in series to the sensor. MyECU OilT line tells me that myECU simply ignores what happens above 40°C ("choke open"), so no need to do anything there. Well, I did that. Twice. I gave you a calculation on resistor in series to sensor, and the one on influence of a heatsink on measuring temperature. Both may be wrong, but you will just not know until you have put the some numbers in YOUR calculator. Only then you can claim that there is something wrong with MY calculator.

This is what I mean Dan - http://www.dpguzzi.com/efiman.pdf Look at the page 12 of the PDF. True, different part types, but identical NTCs.

HA!!! WELL DONE RAZ, WELL DONE!!! Finally someone puts a few numbers "on paper" and comes up with a plausible explanation. I am fishing for this kind of argumentation for quite a while now, but it seems that our "scientists" are more inclined to writing essays on philistines, sincerity and seat-o-pants experimenting methods then to pulling out their r(d)usty calculators and doing things that an 8-year old could do. Once again, well done!!! Everyone, how about some rough models/calculations on influence of air gap, heatsink, resistor, whatnot? The ones that include some numbers rather then descriptions of your seats of pants?

We have a situation where same thermistor type is used for measuring both air and motor temperature. Obviously, this will be in a different temp/resistance hence accuracy ranges. If accuracy was of concern for both applications, wouldn't different sensor types be specified? I think that on a liquid cooled engine, the task of air temp sensor would be to fine tune EFI (I think it matters if it is -10°C or +40°C outside), whereas the task of a cooling liquid temp sensor on a warmed up engine is to switch fans on or off. Hence the same sensor can be used for both applications, since the requirements on the accuracy are quite different. Earlier I wrote that emission tests are done at engine temperature between 60-80°C. I interpret this as freedom to "freeze the map" at that point, rather then continue to lean out. So I don't see the point of fine tuning past 80°C point as the engine (at least ours) would benefit from running richer than it is running. That also explains why the same sensor can be used for oil temp measurement. It simply does not matter. The map should be constant at high temperatures. Sure an attempt to "optimize" was made in the early map, but this can be attributed to an error "on the specer's part".

Upright sitting position? No?

I didn't say you do. I wrote the sentence as a precaution, since some people seem to take asking questions as some sort of a personal attack. This is clearly not a case here. Happy to see we can have a sensible discussion without worrying of hurting each others feelings. You are right about that. I am looking at the entire range. I've seen that the datasheet mentions 5% accuracy but says nothing about its whereabouts, which led me to believe that entire range is meant. Is there a convention about accuracy/certain range which make it redundant to be explicitly mentioned? True. I believe you are aware of the fact that the same sensor type is used for air temperature measurements? Clearly this would be in a different range, 10°-40°C, rather then 90°-110°C. Any thoughts on that? I think I understand what you are saying, and I could agree with it if we were talking of analogue electronics where stable behavior of the system would rely on careful composition of components with adequate characteristics. It is quite different with digital electronics where elegance and cleverness of a good design gave way to brute force of crunching numbers. And this is where I loose you. Looking at the capabilities of an average microcontroller equipped with a 10-bit A/D converter, I am inclined to think that in the "most interesting" temp range (>80°C), ECU would be unable to read out sensor's value accurately, no matter how sensor itself is accurate. As I wrote earlier, a change of resistance of 100 ohm would result in a change of a single bit in ADC. 100ohm at 90°C covers a span of >10°C, so ECU can only guesstimate the temperature in steps of 10°C. This may be corrected using clever circuitry, but I don't see how would that work over extended range. Look at this: A simple voltage divider + low pass filter. Would you expect to see something different here?