Printable View
Hello all,
im having an issue with my TPS% reading 30% at idle.
It also reads 204% at pedal to the medal.
heres my info:
-AMC 304 V8
-using 5.7L throttle body with 65lb/hr injectors
-1227747
-$42
-IAC is 18 steps
-TPS voltage is .57V
-idle 800 RPM
-open loop idle AFR 13.9
-elevation 5000'
-using 42.xdf from tunerpro's website.
I pulled the power to the system overnight, but nothing changed. Is there something I'm missing?
IIRC, to read TPS% correctly with a 7747, there is a datastream value that needs changed.
TPS voltage seems fine, so nothing odd there. the TPS% value shouldn't be capable of displaying above 100% though.
TPS percentage is not a native part of the '7747 datastream, and ADX values are usually a fudged range that doesn't always work out to be accurate.
The only TPS value that is in the '7747 datastream is TPS voltage.
IIRC Mark had mentioned that there was a change that could be done in the bin to send out a proper TPS percentage value but it still wasn't 100%correct, just closer than trying to manipulate the TPS voltage value.
great info! I figured it was simply a calc based on TPS volts, but it didn't make sense how it could be so wrong.Quote:
Originally Posted by Six_Shooter
so knowing this, how should I tune stuff dependent on TPS%? for example, I'm tuning my PE circuit now, Is it okay to tune PE by offsetting MIN TPS% for PE by 30%? So I'd set PE to 100% instead of 70% for a true 70%.
Ignore TPS % in the ADX, it's not valid.
I can't give you any solid answer on how to tune parameters that use TPS %, other than feel and trial and error. You can guestimate what the TPS percentage would be, based on volts, and the pedal under your foot.
Alright. This is what I'll do. Thanks guys!Quote:
Originally Posted by Six_Shooter
There should be no need to guess or do manual calculations. Try editing the adx to get what you want. In the editor you find it under the Values. Basically, the TPS Volts and TPS % should be the same on the General tab except for the upper range and alarm (5 vs 100) but the formula on the Conversion tab would change to scale it differently. I believe the raw data goes from 0-255 for 0V to 5V. Once you open the adx editor and take a look it should be fairly clear how you can edit it.
Or, try the file from this site and see if it works to give you what you want,
http://www.gearhead-efi.com/Fuel-Inj...2&d=1375131735
There's one huge flaw with that idea, TPS voltage never goes to 0, even at idle, so any conversion done on the TPS voltage is going to be incorrect. Even adding an offset does not work properly. I tried for a good long time to get an accurate TPS % on a '7747, but never got it close enough to be useful.Quote:
Originally Posted by lionelhutz
I'll try this tonight! Thanks for the pointers.
Does that PCM have an internal %TPS value? If yes, how do the 0% and 100% values get set? For example, does a routine use closed throttle voltage to set 0%?
At any rate, I wouldn't call it a huge issue. That number will be closer than guessing based on your foot position....
That's the thing the internal calc stays internal and It's not fed to The ALDL datastream. Trust me guessing will be As accurate as trying to calculate based an incorrect formula from TPS voltage. Although If you go back and read what I wrote, you'll see that I suggested oobserving the TPS voltage along with knowing where your foot is.Quote:
Originally Posted by lionelhutz
Another part of the problem It's that you might be able get what seems like an accurate range on one vehicle probably won't be on another. The '7747 does not auto calibrate the TPS at start up like newer ECM/PCM do.
Observing the TPS voltage (which is already a scaling of the raw value) and figuring out what %TPS it is or just scaling the raw value to show as percentage do the same damn thing.....
No, no it's not.Quote:
Originally Posted by lionelhutz
The TPS voltage != TPS %
The TPS rests at between .4 and .56 volts IIRC , and at WOT will range between 4.5 and 5 volts, both of which are a pretty large range of discrepancy.
.4 volts != 0.
4.5 volts != 100%, when 5 volts can also be WOT.
There is a vast difference between TPS voltage and TPS % as reported by newer ECMs, and is not based on the same ADX value.
As I have already stated the '7747 does not report TPS % through the ALDL datastream, it only reports voltage. I believe there is a modification that can be done to the bin to enable the internal TPS % scaling to be reported on the ALDL datastream, but it is NOT something that happens in OEM form.
Newer ECMs, such as the '7730 (and all codes that run in it) will auto calibrate the TPS voltage to 0 at start up even though the TPS voltage is not at zero, typically, like the '7747 between .4 and .55 volts, though I have not seen below .48 volts personally (at least not without a TPS error being reported). The '7747 does not do this.
It'll just be an approximation. You are recommending to watch voltage and guess at the %TPS because just throwing in a calculation to get a direct approximation of the %TPS isn't good enough? What the fuck is the difference?
There is no %TPS. You have to basically make a best guess at it. The easiest way is to throw a calculation into the adx file and get a number. It won't match exactly what that ECM is using internally but it's at least an easy starting point. There, does that suit you?
Now you're throwing in arguments about how newer PCM's work, which have no bearing on using a '7747....
Look, I've used the '7747 quite a bit, I've tried a lot of things, one of which was to get an accurate TPS % value in the datalog. After many, many attempts to get something that was even a good approximation, nothing ever was, not even close. Changing the formula, adding removing offsets, etc. It was pointless in the end.Quote:
Originally Posted by lionelhutz
To understand how TPS % output works, referencing other GM ECMs is most certainly valid, to contrast how things work, to help further explanation, since it seems that you only have experience with newer ECMs that have this ability built in.
Keep it civil, or you will be taking a break from the site, we try to keep things nice here and don't tolerate hostility.
I'm trying to explain things to you nicely, but you don't seem to be able to grasp that it's just not as simple as you believe. I too thought that it would be simple to put a formula in manipulating the TPS voltage, but after attempting it myself (going against what other people had said), I confirmed that it's just not that simple.
So in your recommendation, when the throttle is closed and you get a TPS% value of around 12%, how is that helpful? Ok add an offset, now the rest of the range is off, reading low, because you have to take away what ever it was over at the beginning. So now change the formula, oh wait, now the offset is wrong, readjust that, and the range is off again. So now we look at the other end of the scale, and see that it's not reading 100% when the throttle really is, play around with the formula to get that right all while trying to get the other end to stay the same and hoping that the scaling through the middle actually works as well...
Go and play with it if you want, I won't waste my time any more and will not recommend to anyone else to waste their time trying to set up a TPS % value based on the stock ALDL datastream of the '7747, it's an exercise in futility. If if someone was to get a TPS % value that worked right for that truck at that particular point, then the TPS needs adjustment or replacement, the whole process starts all over again.
The '7747 doesn't auto calibrate the TPS. So, lets assume the '7747 ECM uses %TPS internally.
- What voltage does it use as 0%?
- What voltage does it use as 100%?
- How could you not calculate the %TPS from voltage when you know the PCM never does any calibration on the %TPS?
And the bonus question - how did you know the %TPS you were trying to calculate for a '7747 was wrong when there is no %TPS output in the datastream?
Because I'm smart.Quote:
Originally Posted by lionelhutz
When closed and WOT do not match 0 and 100%, and then when manipulating the formula causing any point in the throttle movement to be off. That auto calibration is what made it possible to have TPS % in the datastream.
I also believe that the tables based on TPS % are just approximations that someone has put in there because it seemed to make sense when they disassembled the the code. There have been many inaccuracies of the $42 code found that have perpetuated to this day, because that's how it was said to be originally.
The long and the short of it is, it's not as easy as you want to believe it is.
Ya, SMRT smart.
I wouldn't be surprized if the '7747 doesn't use %TPS internally at all. So, when you tune a %TPS value you're probably adjusting a value that is using the TPS voltage (the 0-255 counts value).
Getting all wound up because you can never make a display show 0% when let off the pedal or show 100% when you put the pedal to the floor is rather dumb when that value has no correlation to what is being used inside the ECM.
Here's the only way I can think of to test your %TPS meter on a '7747. Set the TCC unlock (or lockup) all at the same %TPS. Then, drive and observe the TCC lockup vs the %TPS meter. I personally couldn't be bothered. The last time I programmed a '7747, the calculated %TPS got the TCC table close enough it works like I expected.
here is a file eagle mark was working on be for he got sick. the 5.2 adx shows 0 % at .50-51 on my truck. if I remember it went to 94% which it would read different on another tps.
thanks for the file. I offset all my TPS% values by 30 in the old XDF to compensate, and when I loaded the one you posted, my TPS% numbers were exactly what I thought they should be instead of 30% high. I feel like all TPS calcs should've been done in volts to negate this discrepancy, but it must've made more sense at the time.Quote:
Originally Posted by ony
thanks guys.
the only % setting I remember messing with was when it was time to enter pe mode which I have a v6 and I try keeping it at 70% throttle and a 2.5 second delay.
This was an incorrect assumption on my part. The values in my tables are real, and don't need to be offset. It is simply the readout from the XDF. Now what you guys were saying makes more sense. Now that I really think about it, the XDF and ADS files are totally separate and of course a readout from one wouldn't affect the other. Haha. Live and learn.