I thought there wasn't enough action on this section of the forum, so here we are again.

Things are definately much, much better. The car starts fine, and for the most part runs well in warm up mode. But last night when coming home from work, the car did a few jerky bucking misses when the ECT reached 144. The fuel ratio looked good, I didn't notice any goofy variations in the AF ratio. The timing looked good. Of course when this happens, it is too quick to see everything at once, and by the time you take it all in, the miss is no longer happening.

Here are the observations: 1. It was REALLY cold, (for us in the lower part of the continent). It was in the mid 20's. 2. At the time of the miss the intake port temp was reading 45 degrees while the ECT was reading 144, and the head had reached 180. That is a difference in intake port to head temp of 100 degrees or so. 3. The car was going up hill. Everytime this has hppened, it seems as if the car was either going up hill, or I was turning left. I don;t know if this is important, but I am begining to think maybe it is. I can't imagine what would be motion sensative that is also temp sensative. It might be just a coicidence. This is sounding like one of those old articles from Popular mechanics back in the '50's. Anybody remember that? Gus's Garage. Was that the name?

I am wondering if the filter on the surge tank has anything to do with this. Except I don't see even a flutter in the fuel pressure on the gauge. I was considering somehow routing hot air from the header to the intake until the ECT gets all the way to 180. And maybe hooking up the manifold and or TB heater would be a helpful thing. And, Since the stock plugs imporved things so much, I was thinking of now going to the ZFR5N plugs to see if that would eliminate this little gremlin completely. I could run them just for the winter. The car is working so well now, tht I could accept this, and live with it. But I don;t want to.

I just felt compelled to share this since the last post was so positive. I felt like I was living under false pretenses here. Hope you all had a nice holiday.
Sam

_________________

H'mmm.
Is there a datalogging function available?
Turning left I would disregard as a coincidence. Uphill, however, suggests something that is temperature and load sensitive. I wonder if maybe your next step might be to hook up the throttle body and manifold heaters and see if that'll reduce the temperature gradients across the system. Also...what thermostat temp are you running? My inclination would be to use a 195° or even a 205° unit in your setup.
Gus Wilson was the proprietor of the Model Garage. His eager-beaver young apprentice was Stan Hicks. The town skinflint who was always neglecting his cars and improperly repairing them, then demanding discounts after Gus & Stan fixed them correctly, was Silas Barnstable. I loved those stories! I'm glad you brought it up, for it made me do a Google search, and it seems we're not the only two fans. Many of the stories've been put up on the web! See here, here (non-obvious navigation — use the rectangular "The Stories" button up at the top), and here (navigation alphabetically by story title, not chronologically by year...odd!).
I have known factory port injection systems (such as Bosch KE-Jetronic at least as late as 1990) that had thermostatic air cleaner setups, which if not working would cause the car to exhibit driveability issues such as what you describe. But it seems to me probably easier to enable the TB and manifold heaters on an experimental basis, rather than starting from scratch to rig up a TAC setup. That said: What is the nature of these TB and manifold heaters? They are...coolant-heated? electric...?
I wonder if a "happy perfect" might be found on the spark plug front in NGK ZFR5FIX-11. This has longer electrodes than standard, not quite as long as ZFR5N, an iridium fine-wire centre electrode for extremely long durability and reduced arcover voltage requirements, and a corner-clipped ground electrode for spark unshrouding and preignition resistance. It's a relatively costly plug, at $6 or $7 apiece, but due to the very hard electrodes should last quite a bit longer than a ZFR5N. Factory gap is 0.044", which is just about perfect. I am currently running ZFR5FIX-11 in my truck at the moment. NGK stock number is 2477. Don't forget to delete the metal ring washers!

_________________
一期一会
Too many people who were born on third base actually believe they've hit a triple.

For the up hill / left turn to be significant, it is possible that if your surge tank isn't vertical or the pick-up is off center that this could cause the pick-up to suck air.

But, like Dan said, this does sound more likely to be a combination of high load and low temperature. I agree, try hooking up the throttle body heating lines and see what happens. You might be able to rig some way of bypassing them once the engine is fully hot, or just install a valve to turn it off for race use.

_________________
"Mad Scientist" Matt Cramer
'66 Dart - turbocharged 225
My blog - Mad Scientist Matt's Lair

I still say on the temp coolant sensor in wrong place, head is hot already. 34F difference is way off for computer to use. Missing is telling me the computer is shooting wrong mixture target at particular temps. The computer depends very heavily on the coolant temp sensor and MAP to set different mixture ratios then closes loop and monitors O2 sensor after set point is reached.

Cheers, Wizard

Wizard, I am not going to rule out what you are suggesting, and I certainly will plan on giving that a try if I can;t get it worked out any other way. But those maps and relationships you are talking about are maps that I must program. They are not preset at the factory. The problem is I don;t know how to program them to work correctly with any set up. The instructions tell me what the maps do in very technical terms, with no mention of how changing them affects performance, or drivability.

All the values in these maps are a numeral "one", which means no adjustment is being done based on differences in manifold temp and ect temp right now. That is the problem more than where the sensor is. . If I knew how to program it, I could set the curve to accomodate the placement of the sensor. I can set the curve you are discussing to be all out by the time the ECT is 100, or 166 or 205 or anything I want. And, I could make that decision based on the temp of the head, even though it was reading from the block.

The problem with putting it in the head, aside from the fact that I would have to start all over with the warm up maps is that the ECU would register a fully heated engine when it was not really warm all the way through yet. In my experience, the engine wants different things when the head is 180 and the block is 144 from what it wants when the head is 180 and the block is 166. Both of these situations happen before the thermostat opens. If I put the sensor in the head, the ECu would not recognize any difference between 180/144 and 180/166. The ECU would see this as the same thing, and I am fairly convinced the engine wants something different in those two conditions, if I can figure it out. I feel pretty certain that if I can learn this next step of programing, I can get this right. Tht being said, maybe you are right. I just don;t think it is that simple.

For the record, the miss usually happens at around 2000 RPM. I am going to look at both the VE map and the timing map, and see if there is a sudden, large shift in values there. If timing jumps, or VE figures jump there, then maybe that would explain a sudden miss.

I can data log, but it would be hard to do because so far it has been kind of random in its sppearance. I will still give it a try though. Unfortuantely on the manifold heater, I had to remove a core plug forom the bottom of it to get it to clear the Dutra duals. So, short of some Houdini tricks there, I am stuck with just the TB heater. I will do the tB heater. Fortunately, the weather warmed up today. We got your weather, Dan, for a day or two. You can have it back.
Sam

_________________

What bugs me about this idea is that the cylinder head temperature is overwhelmingly what matters as far as adjustment of fuel and spark for combustion is concerned. The block temperature isn't nearly so critical, because when combustion's happening, the fuel/air charge exists in a space consisting only of the piston top and...

...the cylinder head. I can't think of any factory EFI systems that use a coolant temp sensor located in the block; all the ones I know of are in the head, located near the thermostat in locations more or less analogous to the stock slant-6 location.

_________________
一期一会
Too many people who were born on third base actually believe they've hit a triple.

I hear you Dan. But the engine does seem to require something different as the block warms up. Maybe the differences are the result of the manifold/intake port temp that is the key, and not the block temp. In any event I still must figure out this phase of the programing.

Changing the sensor location would be an easy thing. I would have to drain the block, but that would not be that big a deal. The sensors are different. The ECU uses a two wire setup, and the gauge uses a one wire. This is so the ground can be all int he same place for the ECU sensors. I suppose I could rig a ground wire from the sensor to the ground lugs for the ECU. But I am not sure the two sensors are the same signal. I guess I can measure it huh?

Isn't this fun? There is always something fun to do.
Sam

_________________

I can't imagine it being a hard thing, but it strikes me you'll have to do a fairly large amount of reprogramming, else the computer will see the engine warming up a lot slower than it actually is, from the standpoint relevant to fuel and spark, which will probably do all the same bad things to driveability, economy, and engine life that a stuck-on choke does on a carbureted car.

_________________
一期一会
Too many people who were born on third base actually believe they've hit a triple.

I still have to find how to program the other warm up schedules. Here are the maps I haven't figured out yet. 1. VE correction coefficient vs estimated intake port temperature; 2. ECT- VE correction VS map/RPM; 3. ECT- IAT difference; 4. MST-IAT difference; 5. port air ignition compensation; 6. ECT Compensation power modifier: These are all maps that can be programed. And I dont; know how changes in these affects the behavior of the engine. I am pretty sure one or two of them come into play here. Any clues would be helpful, but the text really does not tell you much. The descriptions are kind of like this: Blue can be modified to be more blue or less blue. It is a color that looks like blue. The more blue you add, the bluer it gets. Values can range from a blue value of 2 to a blue value of 6.

No kidding. That is what you feel like when reading the manual. If the tech guys ever come back from wherever they have been, I can ask them about these screens one at a time,but I certainly cannot ask them to give me a complete education in programing this over the phone or via e-mail. Unfortuantely, no one has written a book about Gen VII.


Sam

_________________

Does that make you feel blue????

_________________
Ed
64 Valiant 225 / 904 / 42:1 manual steering / 9" drum brakes

It makes your face feel kind of red, and then your ears turn blue as the steam starts to come out. There are a bunch of forumlas in there that I need to actually plug some values into in order to see what effect changing one value has on everything else. I really should go read the MS Mega Manual again. My guess is there is information there that might add some insight.

It is not that there are no values in these pages. It is just that they are default values for a small block chevy. Things are so different on a V style engine. In some cases the values are all "ones" which adds or subtracts nothing.

Sam

_________________

I think you are right about the sensor location. You guys who say change the location of the temp sensor are onto something. I am really scratching my head here, trying to read, read, ponder, read and reread the manual and figure this out. Here is the deal as I can figure it. This is not for the faint of heart.

There is an Estimated Intake Port Temp (IPT) coefficient that is multiplied times the VE table, which changes the fueling, and I can set this at different values for different ECT. The really key word here is ESTIMATED. Re-read that last sentence. There is currently a nice, pretty curve plotted in my configuration file which is likely for a small block Chevy. So the curve is likely wrong for a slant no matter what.

The complexity of this entire deal comes in because the intake port temp(IPT) must first be calculated by a formula within the ECU. This value is first estimated with in the ECU by comparing the IAT (intake air temp) with both the ECT temp and the Intake Manifold Surface temp (MST) at various engine operating conditions. . The estimated value is influenced by values that the user records into their respective screens for both the ECT and the MST. You can vary the values with the varying estimated air flow through the engine.

The purpose of the two tables is to give greater weight to one of the sensors depending upon the amount of air flowing through the engine at the time. This assumes that as the air vlocity increases, the temp of the air begins to dominate in both cases. In both the table for the ECTvs IAT and MSTvsIAT, I can tailor the value from zero to one for each amount of airflow to determine which sensor gets more weight in the final calculation of IPT (intake port temperature.

The question is which sensor is effecting the actual intake port temp the most? And, once this is looked up and calculated, this value is then is applied to the table that determines how this IPT is going to affect the VE table based on ECT.

So ECT comes into play twice here. Once in determining which sensor gets the higher weight in the formula, and the ohter in determining a coefficient that will get multiplied times the base VE table.

One of the problems here is that these tables all have default values that may or may not be correct, but are designed to simply "work". Like the limp home mode in an OEM ECU. And I have tried to work out warm up tables in myapplication that are being influenced by these default values which also get multiplied into the final fueling strategy. So other things are having an impact that I may not even know about.

Now, back to the point: since both of these formulas mentioned here assume the coolant temp is going to be the same as the head temperature, I suppose you guys who are urging switching of the ECT temp sensor might be right. And, further more, If I am going to re-plot the warm up curves, it seems as if maybe I should zero out everything that would effect the base VE table and then go back square one once again by starting with the base VE tables and figure all else from there once that is pretty well established.

Congratulations to anyone who got through this post at all, and more Kudos to anyone who understands it. I know the theory, but still don;t know how varying the values in these tables actually affects the way it will drive. But I haven't given up, nor do I intend to. I do think I will change the sensor location. Do you think I can just change the wiring and leave the sensors in place even though one is a single wire, and the other is a dual wire? One is an Autometer, and the other is a GM.

Sam

_________________

Here is a question you don;t have to be an EFI guy to answer. Does the engine want more fuel or less as the Intake Port Temp goes up? And, shouldn't the intake manifold temp be included in the formula to determinethe final fuel amount? The manifold has a temp sensor, and it is feeding values to the ECU.

As it stands now, the tables are configured to pretty much eliminate the influence of the manifold temp. I assume this is because on an SBC the manifold and coolant temp are pretty much the same. In the IPT (Intake Port Temp) tables the engine coolant is given a value of "0ne" for all engine airflow conditions, and the manifold is currently given a value of "zero". This means that as it is currently configured, the intake port temp is calculated as a function of only intake air temp and coolant temp . Manifold temp is disregarded entirely. It seems as if the coolness of the manifold on a slant should be a part of the formula. I can make these values anything I want from "zero" to "one". In this way also, I can make the manifold temp have an impact upon the VE as the temp rises over a long period of time, independent of coolant temp in an attempt to keep the mixture from going too lean after a long drive as it currently does.

Another curious thing, the intake port temp is calculated to be somewhere between 50 and 120, depending on the temp of everything else in the system. I have run the calculation by hand with various hypothetical values in to see how it functions. Those temps seem awfully cool to me. But maybe the air rushing through it keeps this temp down. So again, does the engine want more fuel, or less as the aintake port temp goes up?

I appreciate anyone who reads these posts, and encourage you to comment even if you don't completely understand this. My explanation is, of course, more for my benefit, as writing it down helps clarify my understanding, which is stil pretty shakey. But I am hoping that one of the resident Slant geniuses will see something that they think would be helpful to add; such as "Maybe you should move the temp sensor up to the head".

Sam

_________________

cooler air is denser air, you want more fuel
with cooler air, fuel is less likely to atomize, so more fuel to compensate

hot air is rarified air, you want less fuel
with hot air, fuel is more likely to atomize, so less fuel to compensate


that's what I think at the moment.....

_________________
Ed
64 Valiant 225 / 904 / 42:1 manual steering / 9" drum brakes

Warm up miss at specific temp (head) can be timing, or fuel ratio need a little twreak in that area for that particular temp if your computer supports this flexiblity. Mine had a short warm-up power loss take off that confused mechanical automatic (3 speed and it shifted rapidly back and forth when engine fell flat in middle of shift (thud thud!!), let up gas a bit stops this. This only 1 or 2 light stops after first 1 mile warm up. I had to advance ignition timing a bit, 10 deg, wrong computer is the reason I suspect, must had been specified for 12degrees ignition and P.O. had computer placed in my caravan that called for base 6 degree ignition. Carb 2.2 BTW, all OK now till I get weber installed instead of the gas-sucking 5220 holley POS.

Oh, KEEP IN MIND: the Iron conducts heat less than aluminum so count on having much HOTTER combustion chambers while rest of head and coolant isn't warm enough to equalize the temps quickly enough. Aluminum is excellent thermal conductor and transfer heat to coolant much easier and tracks coolant temps more closer and in the process chambers are cooler (greater detonation resistance and permits compression ratio advantage over the iron heads usually.

The BEST heat conductor: silver and copper. Too heavy and too soft, stress cracks easily, expensive, copper hard to cast in detail and stringy when attempting to machine.

I see this issue all the time with steel and aluminum heatsinks in electronics. Some cheap-ass electronics used steel sheet stamped U shaped or plain plate heatsinks.

Cheers, Wizard