Alright, what the hell - im going to throw this out there.
First off I'm new to the site. From the Honda and LSx camp trying to learn to help a ecotec friend out.
I've been reading about ways around the Alpha-N fueling issues in a boosted car. I tuned my Corvette on HPtuners using the Speed Density upgrade. Is was easy. But then realized - oh, they don't offer that for the ecotec ecu.
So then there is the 2 bar, 2.5 bar whatever fake. Read about it... seems like it does work but still has issues. Resolution, etc.
Then I'm thinking ok, let the stock ecu run the stock injectors along with HPtuners for tweaks and something like eManage or AEM FIC with additioinal injectors for boost situations. Seems like over complicating things... then I start thinking, bear with me here I will try to explain; Ok, your car is at; 4000 rpm, 10 psi, 100% tps. The car needs X amount of fuel (lets call this value 100). The fuel requirements are being met by; factory ECU seeing 100% tps and 4000 rpm and delivering 60% of the X value. AEM sees 10 psi and 4000 rpm and delivers 40% of the X value. Things are good. But what if your loading the car higher say your in a higher gear on the freeway going up hill. You reach 4000 rpm and 10psi but are only at 60% tps. Now the AEM still sees 10 psi and 4000 rpm delivering 40% of X, but the ecu sees 60% tps and 4000 rpm so it delivers say 40% of X. If that makes sense.... I'm need to look at how the PE tables work again, maybe this isnt as big of an issue as it seems?
Seems like we just need to go standalone and be done with it.
But then I was like hmmm.... what the hell?!? Why not just run a damn map signal to the TPS pin on the ECU and force the damn thing to run as speed density?!?!? Both sensors output 0-5 volts, the ECU wouldnt know as long as you remade the TPS vs RPM fuel table from scratch but who cares? Has this been tried? On low boost cars, or maybe even with a 2 bar fake type setup this could be interesting?
Disclaimer, I'm a newb with Ecotec stuff - go easy on me lol.
Discuss...
94 Civic - 10.82@134
99 Corvette - 11.71@125
First of all. Speed density uses the MAP sensor to determine fuel, not the TPS. So you're forcing it to run in Alpha-N mode. Which might be why you're having trouble with fueling. Another problem is that the MAP sensor reduces voltage as vacuum decreases (when you open the throttle). The TPS sensor increases the more you open the throttle. So doing what you're doing means that the computer sees you holding it wide open when you're idling, and seeing you idle with the throttle open. I'd sure hope that neither your Civic or Vette are wired anywhere close to this because it will cause a very short engine life.
Don't worry about the E-manage, it's junk.
Don't worry about spraying extra fuel in the intake manifold as it's not made to have liquids in it, could be prone to puddling.
I think you're overthinking things. What exactly are you trying to do to this Ecotec car, and what's been done to it? Just start from there.
i find it amusing that SHOoff has nothing better to do but follow me around & be an unhelpful dick in even cross-forum. - Jon Mick
Quote:
But then I was like hmmm.... what the hell?!? Why not just run a damn map signal to the TPS pin on the ECU and force the damn thing to run as speed density?!?!? Both sensors output 0-5 volts, the ECU wouldnt know as long as you remade the TPS vs RPM fuel table from scratch but who cares? Has this been tried? On low boost cars, or maybe even with a 2 bar fake type setup this could be interesting?
Heres the thing the ECU can not be made to read boost, not like that. Both the TPS and MAP sensor's are 5 volt sensor's, so doing what you are saying would not change anything. The ECU is not going to know it is the MAP sensor giving it that voltage. It will still think it is the TPS sensor.
What is the goal of this motor on boost? Is your friend putting another stock motor in?
You can tune for boost on TPS. Many have done it and made good power and drivablity.
FU Tuning
John Higgins wrote:
You can tune for boost on TPS. Many have done it and made good power and drivablity.
While I would agree that you can "attempt" to tune for boost solely via TPS, you are never going to get the drivability perfect and make the most of the system without a boost refernced fuel input, even if its an FMU.
To the OP - in order to do what you are saying, you would need to break down the VE tables into raw voltages and then correlate those voltages with the signal output from your new MAP sensor. For example, if 0V = 0% TPS, it now needs to equal 1 bar vacuum. That is assuming you can get the hardware to function, and that it doesn't screw up the other various tables/multipliers in HPT that we do not have access to.
Overall I think its a neat idea, but without FULL access to the ECU via HPT, its going to cause more headaches than alpha-n + FMU
Quote:
While I would agree that you can "attempt" to tune for boost solely via TPS, you are never going to get the drivability perfect and make the most of the system without a boost refernced fuel input, even if its an FMU.
I'm not saying drivability will be prefect, it wont. It would be good.
I have tried tune a boosted car that was a speed density 97 ECU with a FMU and HPT. Big waste of time. I was told I would waste hours fighting the FMU. Ryan you were right! When we finally disconnected the FMU tune went so much better. I will say it was a cartech FMU that has a couple adjustments on it and not a normal FMU. Maybe with a regualr FMU it would have went better, but not sure.
FU Tuning
Quote:
First of all. Speed density uses the MAP sensor to determine fuel, not the TPS.
I know this, that was my point.
Quote:
Another problem is that the MAP sensor reduces voltage as vacuum decreases (when you open the throttle).
That is backwards - MAP voltage increases when you open the throttle.
Quote:
and seeing you idle with the throttle open. I'd sure hope that neither your Civic or Vette are wired anywhere close to this because it will cause a very short engine life.
LOL, yeah - I'm trapping 134+ in the quarter and my ecu thinks im at idle... come'on lol.
Proof it not wired crazy; datalog from a run.
Quote:
I think you're overthinking things.
I probably am, I agree. I'm not sure what his goals are just yet. But either way I just thought this would be something interesting to consider?
Quote:
While I would agree that you can "attempt" to tune for boost solely via TPS, you are never going to get the drivability perfect and make the most of the system without a boost refernced fuel input, even if its an FMU.
I'm no expert in the ecotec world - but I agree with you 100%
Quote:
For example, if 0V = 0% TPS, it now needs to equal 1 bar vacuum.
Yeah, thats it exactly - I still think its an interesting idea. Your right about how this could affect other tables of coarse. I just thought it would be a cool idea to bounce of you guys.
94 Civic - 10.82@134
99 Corvette - 11.71@125
Again my thinking is basically to have a VE table that fuels the car based on MAP not TPS. Yes, you can make a car run boosted using TPS but as other said - its not perfect.
Consider this; your cruising at 3800 rpm and floor the car. In TPS based fueling - You shoot right to 100% TPS and simply ride that row out as TPS stays constant and RPM increases. There is no accounting for boost (at least I dont see it? maybe im wrong). Your fueling is that same at 1psi as it is at 10psi. Correct?
Then ponder this TPS MAP hack whatever we want to call it.; You floor it and shoot down to a zero psi row (the old 100%TPS row) Yet since we are fueling based off a MAP sensor - the voltage will continue to rise with manifold pressure. SEEING boost and moving down into rows that can fuel for it.
Timing would work like the fake 2 bar...
And as a side note, most Honda factory MAP sensors are like a 1.7 bar ish (they max out at 10.5 psi). Would offer better resolution than a 2 bar or 2.5bar (since we have limited rows in the VE table).
Looks good on paper, but yes could have plenty of issues in actuality.
94 Civic - 10.82@134
99 Corvette - 11.71@125
The main problem with your idea is this. The changwe that needs to be made is not what sensor is connected to what wire at the ECU, but what the ECU uses to determine fueling. This means changing the way the ECU is programed from the factory, whcih means having access we do not have, and will never have. You can connect the map to the TPS wires at the ECU, but the ECU is still going to think it is a TPS.
Same as what you did with your Corvette. You changed the OS file so it could read boost.Samething would need to happen on this, but opnly option for that is GM reflashed which is locked and can not be tuned.
FU Tuning
Quote:
The main problem with your idea is this... what the ECU uses to determine fueling.
Tell me then, if the ECU doesnt primarily use the VE table to determine fueling, what does it use?
Quote:
This means changing the way the ECU is programed from the factory, whcih means having access we do not have, and will never have.
Yes that would be the correct way to do things. We all agree on that. But yes, its probably not going to happen.
Quote:
You can connect the map to the TPS wires at the ECU, but the ECU is still going to think it is a TPS.
Who cares what the ecu thinks it is? I'm just saying - think outside the box. All that matters is that we would now have a 2 dimensional primary fuel table that refrences RPM vs MAP. Increasing pressure in the manifold will move the target cell down on the table. What happens when you open the throttle on a stock car? TPS voltage goes up, target cell moves down on the VE table, and manifold pressure increases. On n/a aswell, its all related. All the values on the table would need to be changed of coarse...
Quote:
Samething would need to happen on this, but opnly option for that is GM reflashed which is locked and can not be tuned.
Thats the 2.5 bar reflash? Think I was reading about that. That really stinks that it is locked, seems like that os could've made things for everyone alot easier.
94 Civic - 10.82@134
99 Corvette - 11.71@125
i'm just curious how the ECM would react with the open loop/ closed loop switch over. It uses TPS as the primary threshold to engage open loop.
you could force open loop all the time, and just run off the VE map, but you also run into issues with the multiple VE maps..
there's 3 the car uses primarily to run while you're accelerating. idle and decel are two different stories, since they actually run on the MAP sensor... basically anytime there's zero throttle input, the car reverts to speed density.. but don't get any ideas, these maps end at 40kpa.
but again, i'm not sure how they'll react without seeing ~0 volts on the TPS sensor to engage them.
its a good idea, just not sure how the other tables will like it... and you do lose an awful lot of resolution since the TPS scale isn't linear, its exponential.. so in the higher voltage ranges you'll have larger and larger gaps.
for instance, the high rpm VE map goes from 50% throttle directly to 100% in the next row.
There are too many modifiers/tables that the sensors correlate with to even think of doing something like what you are thinking. Bottom line is, its not going to work...
P&P Tuning
420.5whp / 359.8wtq
And the biggest problem may be...one cannot know just what will be affected, nor how to combat it. The software just doesn't go deep enough.
Bill Hahn Jr.
Hahn RaceCraft
World's Quickest and Fastest Street J-Bodies
Turbocharging GM FWD's since 1988
www.turbosystem.com
DaFlyinSkwir(LS61) /PJ/ OEM+ wrote:
its a good idea, just not sure how the other tables will like it... and you do lose an awful lot of resolution since the TPS scale isn't linear, its exponential.. so in the higher voltage ranges you'll have larger and larger gaps.
for instance, the high rpm VE map goes from 50% throttle directly to 100% in the next row.
This would be a problem. To clarify, the TPS voltage vs. output is linear, but we get an exponentially scaled VE table axis.
Still, I would be curious to see how it would react with a 2 bar MAP wired up, just for shts and giggles you know? I think having two rows for fueling in boost is better than none...
I guess tomorrw we are gonna try this on my car, so you will know lol. Im not boosted but at least if it works and we can get the car to run decent and whatnot we will know if it works.
^^ good luck. I think it'll cause more problems than it solves, but I'd be happy to be proven wrong in this instance.
John Higgins wrote:
I feel your pain.... Can't wait to see the "My car won't run,whats wrong"/"My sh!t blew up" thread......
P&P Tuning
420.5whp / 359.8wtq
I must agree 100% with Mark. you guys have been beeting your heads so hard for so long over cracking the eco ecu that when an answer is dropped infront of you, you deny it. quit thinking of the idea as TPS and MAP! STOP! think of it as voltages. it makes total sense. OL and CL can be changed on comand with HPT, as well as when it turns on and off. you can control when PE starts too. all of the other variables involving spark can be disregarded/changed to reflect.
Well, it's not so much negativity as constructive criticism. No one wants you to not try; we just don't want you to be overly optimistic about the possible results.
One of the key aspects of muddling with MAP sensor inputs and swapping different BAR MAP sensors is that at key-on, the PCM takes a barometric pressue reading. This is a scalar for all fuel delivery once the engine starts. If one has put the at-rest voltage in a different region, the barometric pressure reading will be highly skewed, and over time, the PCM will rebel with odd issues and wandering fuel mixtures. As we cannot control this feature, we can't effectively overcome this.
It is for these reasons that HPT provides different OS for the cars they've released 2- and 3-bar capabilities for. It's not just as easy as mapping the voltages. Oh, if it only were, lol!
Bill Hahn Jr.
Hahn RaceCraft
World's Quickest and Fastest Street J-Bodies
Turbocharging GM FWD's since 1988
www.turbosystem.com
Michael Richardson wrote:I must agree 100% with Mark. you guys have been beeting your heads so hard for so long over cracking the eco ecu that when an answer is dropped infront of you, you deny it. quit thinking of the idea as TPS and MAP! STOP! think of it as voltages. it makes total sense. OL and CL can be changed on comand with HPT, as well as when it turns on and off. you can control when PE starts too. all of the other variables involving spark can be disregarded/changed to reflect.
lmao, really? how do you do that? I've been tuning with HPT on this ECM for years and never could do that.
OL/CL has two variables... TPS and delay. when TPS threshold is met, the delay rpm engages. until the RPM raises above that value, and WOT continues into the following gear, the ECM is still in closed loop.
and if you give the ECM a MAP sensor voltage output instead of TPS, yea it can sorta kinda work, but I think you'll find the ECM going into open loop way too often. you'd most likely be able to solve this by changing the TPS value for OL operation, but like i said before...
the ve tables are exponentially scaled. so who's to say that from 4.5V to 5V isn't considered 50%-100% throttle, or, that 2.5V to 5V the ECM does nothing different except interpolate between the two values (which leaves a lot of flexibility for interpolation.. something not really that great when it comes to a VE map).
until someone can answer those questions, I'm not convinced. Can it be done? I want to say yes, but the wary side of me points to no. And it won't be a simple solution either, alot of tables will have to be altered in illogical ways in order for the MAP sensor to work as a TPS sensor.
I'd leave the stock MAP sensor in place, and use a seperate MAP sensor (in whatever bar rating you want for your desired boost) and wire that to the TPS wires.
Quote:
'm just curious how the ECM would react with the open loop/ closed loop switch over. It uses TPS as the primary threshold to engage open loop.
Good point. It should still switch from closed to open loop. Ok, lets say that the ECU is set to switch at say 40% or greater TPS. Assuming the Voltage curve for a stock TPS sensor is linear - 0-5 Volts is 0-100% TPS, the ECU is going to want to see a TPS value of 2 volts or greater to switch. So, the ECU is looking for 2 volts on this TPS pin. 2 volts or greater (on a 2bar) is going to be about anything from ~5 inches of vacuum and up. The ECU isnt going to be blind as far as TPS in a way. As you push the throttle, the MAP sensor is going to act just like a TPS in a way that voltage is going to go up based on the throttle opening and manifold pressure going up.
Quote:
you could force open loop all the time, and just run off the VE map, but you also run into issues with the multiple VE maps..
It should still switch, so we shouldnt have to force open loop.
Quote:
but again, i'm not sure how they'll react without seeing ~0 volts on the TPS sensor to engage them.
The ECU will be "seeing" TPS activity that is similar and directly related the the actual throttle position.
Quote:
for instance, the high rpm VE map goes from 50% throttle directly to 100% in the next row.
Yeah, I wish it was more linear. I am spoiled by Hondata - you can add rows, add columns, and change the scalars.
Quote:
Bottom line is, its not going to work...
Pretty good chance it may not. The real bottom line is nobody has tried it... yet.
Quote:
The software just doesn't go deep enough.
It really doesn't... I'm sure if there was more demand for this, HPtuners would invest the time and provide a solution.
This is just something im curious about, and would like to try. I'm not going to count on it working. The real plan seems to be moving towards something like your PortFueler setup.
94 Civic - 10.82@134
99 Corvette - 11.71@125
DeeJayPanzone wrote:ImPhat0260/Cavattack2000 wrote:John Higgins wrote:
I feel your pain.... Can't wait to see the "My car won't run,whats wrong"/"My sh!t blew up" thread......
Thats the point of trying it with a close to stock motor, Its a lot harder to make a quick mistake and blow up a boosted car then a clost to stock n/a car. Then if it works it can be applied to boosted ones and so forth. I dont see the point in being so negitave about something when people are simply looking for ways to make it eaiser to tune cars and help out the j-body community.
Ok... So many have pointed out reasons why this won't work... Here are a few more... How will PE be engaged? If no TPS is seen, PE enable will not work. Next you all the scanner issues. You won't be able to log anything properly. I can think of more reasons, but John, PJ and Bill have covered most of it. Also, I see your car is an automatic. I hope you realize that there are a sh!t load of TPS based transmission tables. My guess, even IF it were to work/run (which I seriously doubt it will), your car will probably not even move. If it does move, it won't shift properly.....
P&P Tuning
420.5whp / 359.8wtq
Quote:
Still, I would be curious to see how it would react with a 2 bar MAP wired up, just for shts and giggles you know? I think having two rows for fueling in boost is better than none...
Thats honestly the same way I feel about this. Why not try it.
Quote:
I guess tomorrw we are gonna try this on my car, so you will know lol. Im not boosted but at least if it works and we can get the car to run decent and whatnot we will know if it works.
This guy's down for anything lol he's awesome. See you tomorrow.
Quote:
I feel your pain.... Can't wait to see the "My car won't run,whats wrong"/"My sh!t blew up" thread......
From the looks of your sig, it seems you've made 400 horse on the pushrod 2.2 motor. That is truely impressive, and I'm sure people told you it could never be done... but you tried anyway and did it.
We are trying this on a stock NA car. I will put the wideband from either the Civic or Vette, and throw it in Panzones car. We will be cautious with this "test".
Quote:
I must agree 100% with Mark. you guys have been beeting your heads so hard for so long over cracking the eco ecu that when an answer is dropped infront of you, you deny it. quit thinking of the idea as TPS and MAP! STOP! think of it as voltages. it makes total sense. OL and CL can be changed on comand with HPT, as well as when it turns on and off. you can control when PE starts too. all of the other variables involving spark can be disregarded/changed to reflect.
Thanks, it would be cool IF it works.
Quote:
we just don't want you to be overly optimistic about the possible results
Its always wise not to be over confident, I'm not optimistic, just curious.
Quote:
at key-on, the PCM takes a barometric pressue reading...
I'm not as familiar with the ecotec ECU as I should be - the Honda ECU has a barometric pressure sensor on the board that is used for that reading. It sounds like the ecotec uses the factory map sensor underhood. I was planning on leaving the stock map sensor in place and wired how it should be as stock. Was going to try a GM 2bar and/or Honda 1.7 bar... separately wired to the tps plug.
Quote:
lmao, really? how do you do that? I've been tuning with HPT on this ECM for years and never could do that.
One way to force open loop would be to set this tables values all to 304...
Quote:
I'd leave the stock MAP sensor in place, and use a seperate MAP sensor (in whatever bar rating you want for your desired boost) and wire that to the TPS wires.
Yes, I was thinking the same.
94 Civic - 10.82@134
99 Corvette - 11.71@125
Quote:
As you push the throttle, the MAP sensor is going to act just like a TPS in a way that voltage is going to go up based on the throttle opening and manifold pressure going up.
yes, I've tuned before I understand this.
the only problem is, for certain modes to be engaged (such as idle and decel for example) there needs to be a 0 percent TPS input... even at idle or on decel, MAP value doesn't go below whatever idle kpa value is, which from experience is roughly around 20kpa.
so the idle and decel would never be engaged, because part of the conditions that must be met for the ecm goes into decel mode is 0% throttle input.
Quote:
The ECU will be "seeing" TPS activity that is similar and directly related the the actual throttle position.
similar, yes. but again, you lose the 0% inputs which, I think, will cause more problems than you're anticipating, this is what I'm trying to tell you. I work with HPT all the time, trust me.. its going to cause problems.
and you don't have much programmable control over closed loop parameters with the interface HPT gives us. Like Bill mentioned, and Ryan was also hinting at, the depth just isn't there.
its a good idea, but you're going to have to pull some tricks to get it to work right, if at all. Like I said, I
want you to prove me wrong, but without some creative thinking and tuning of the available parameters, you're going to fail.
have you messed with HPT much for these cars? you may be able to get past the lack of 0% input by messing with the DFCO, but how exactly is beyond me.
