Parallel resistor mod?

Much has been made of the fact that a PC-III can reduce fuel delivery while a TFI can only increase it, and that Nomads with Modified pipes and intakes will actually tend to run too rich under some conditions and need the mix leaned out, while at other conditions they'll need more fuel than stock.

There seems to me to be an obvious solution to this delima: Lean the overall delivery of the system & then use the TFI to just bring it back up only in the ranges where you need to. In a way, by putting huge air filters on the engine this is what I've attempted to do. Sometimes it seems to work great but others not as well.

Anyhow, I've been diddling around with the TFI settings on my bike for some time now and have come to a couple conclusions.

With the intake setup I have, the mixture can be set up well only in cold to moderate temperatures. Once the weather is too hot you can never seem to get the thing lean enough at cruise yet rich enough under acceleration.

I'm about somewhat convinced that this is because of the air temp sensor being too exposed. Perhaps it only gets warm enough when the bike is really hot or sitting in hot traffic.

While adding a resistor in series with this sensor will richen the fuel mixture across all RPM ranges, to add a resistor in parallel with it should lean out the mixture across the board; though I do not know how linear this effect will be as surely neither the sensor value nor the ECU mapping will be linear. The hotter the sensor gets, the lower its resistance will be, and the less the parallel resistor will reduce the resistance of the combination percentage wise.

This makes me think I can use a parallel resistor to lean my overall mixture primarily in cold temps (or when the sensor "thinks" it's cold.) This might allow me to use the TFI to richen mixtures at higher temperatures without over-richening at low temps.

I haven't measured the value of the stock temp sensor yet, so I don't know an appropriate resistor value to try, but obviously it's got to be fairly high--perhaps in the megaohm range--to make only a slight alteration of the resistance the ECU percieves of the sensor.

So far, I don't think I've heard anyone mention this idea before, but maybe it has been.

Anybody?

[blowndodge]

Add a resistor in parallel?

According to the manual the stock air temp sensor has values at the following temps:

5.4 - 6.6 ohms at 32F
2.26 - 2.86 ohms at 68F
0.29 - 0.39 ohms at 176F

Whoa! You scared me there for a minute.

Those values should all be in kilo-ohms.

I had to look this up myself, but I just couldn't imagine putting a 800 ohm resistor in line with something only 4 or 5 ohms and having the ECU not freak out tha the reading was hundreds of times what it should be.

Anyhow the output evidently goes from 290 ohms to 6600 ohms. I wish there were more values listed so I could plot a curve that makes sense. I may just use a variable pot of 0~1 megaohm and experiment.

Come to think of it, it seems there was a chart flaoting around where someone mapped this output. Maybe on Gadget's. I'll have to track that down.

[blowndodge]

forgot the K value sorry.

anyway Cad, why not turn down the TPS to a real low value; ie, lean? Just for giggles I turned my down to the minimum 4.29 and it ran like sh*t!!! so lean it wouldn't take throttle. then you could start adding fuel with your TFI without cutting wires. The TPS adjusted down should lean out all values.

Cutting wires holds no fear for me. I've wired lots of cars and bikes and trailers before. Stereos and computers and other stuff too. When I installed my TFI, I cut the wires in the ECU harness so I could solder in pigtails. That way I can unplug the TFI if I want, and didn't have to use the cheapo Scotch-loc type connecters Dobeck gives you with the TFI.

BUT: I do like the idea of a TPS adjustment & will probably do so. I'd like to find a chart of it's resistance/response first. I did find the one for the ATS on Gadget's site. The thing is, there's no way to change that setting while you are riding. With the pot on the ATS method, I could.

Not having a dyno, I gotta tune by the seat of my pants.

[blowndodge]

The problem I see is that there is a resistance value for a temperature of 176 F. That tends to make me believe an open air cleaner will never provide true Linear adjustment since the last time I checked there isn't a place on earth that warm. Except were cactusjack lives!LOL.

that air temp sensor needs to read those "heated" engine temps to work as designed. I'm now thinking I can increase my 1k ohm to 1.5K and see the results since I don't have the open right side air cleaner most of you have.

Originally Posted by Blowndodge "Darksider"

...that air temp sensor needs to read those "heated" engine temps to work as designed....

OK, but we want it work better than designed, right? I'm thinking that "as designed" means they ping when it's hot.

Right now I'm running rich at low temps and not rich enough at high temps. If the sensor doesn't get hot enough, the resistance won't go low enough. By adding a resistor in parallel I can get it to go lower (the ECU will think the air coming in is hotter than it really is) but the problem is to reduce the fuel when it's cold out so when I add it back in with the TFI I can add enough to prevent pinging without adding too much when the engine/weather is cold.

Right now, after looking at the chart and some quick calcs, I can see that a megaohm was way off. I think I want to start at about 10k~12k ohms.

10k in parallel will give me 282 ohms when the thermistor should be showing 290 and about 3976 when it should be showing 6600.

From the chart, the ECU won't notice much difference at temps over 90^oF but at 32^oF the ECU will think it's actually about 45 ^oF. That will cut way down on the ECU's enrichment at low temps, but cut it less and less as temps climb, still cutting it slightly in the 100^oF range. I'm hoping this will allow me to crank the TFI up to restore the low temp mix while boosting the high temp mix more than before.

[blowndodge]

Where are your TFI settings to be running rich at low engine temps?

This is why I initially thought the water temp sensor might be a good place to put a 330 ohm in series at. Looking at the values in water temp when the fan kicks on at 212F the sensor will be reading about 160F. At 212F there is almost no resistance left.

By not running rich enough at higher temps are you saying you getting pinging with the TFI installed? Wouldn't that be a function of the #2 pot? Assuming your pinging during roll on and not steady cruise? I know what you want to do, I'm just trying to figure out the simpliest path!? If there is one.

Right now, about 1:30,2:30,1:30,8:30, but I've diddled the settings all over the place. Even with the TFI completely off it's a little rich when cold at idle but off-idle isn't rich enough then.

If I adjust the #2 pot up to reduce pinging under serious acceleration it's too rich under light acceleration. It's not pinging at a steady cruise, and of course pings less when it cools off somewhat (like hitting the open road after riding in stop & go traffic at 100^oF.)

Anyhow, I'm hoping since the TFI adds fuel on top of what the ECU wants to send, that I'll be reducing the acceleration fuel curve by making the engine think it's warm when it's cold, but about the same when it's hot. Then I can crank that yellow pot back up and it won't over boost the fuel at cold temps and low loads.

[bobzinger]

Maybe someone sells temp sensors and we can get one with different reading to achieve what we want. Or maybe one that's not linear so it adds more fuel when it's hot.

The stock one's not very linear. It drops like a stone from 32^oF to 50^oF, then more slowly from 50^oF to 80^oF, then levels out a lot from there on up.

You can see the chart here: http://www.gadgetjq.com/ping_fix.htm but it's in centigrade. I had to convert.

[blowndodge]

It's designed to not be linear on purpose. As ambient air heats up the amount of burnable oxygen molecules is reduces and it's not linear. For example and I'm not going to go into the real math of the equation: Lets say the air temp goes from 70 to 80F and the amount of air molecules is reduced by X- ? (some unknown number at the moment) When the air goes from 80 to 90F the function is not X-2X (or two times the unknown number. It's more a exponential function of X. Gradually reducing at a every decreasing level for every degree increase in temps. Clear as mud?

Yup. And I have no desire to try and make it linear. I'm just going to push the whole curve down proportionally. It will drop about 30% at the cold end, but only 1/2% at the hot end. That might be a tad much on the cold end, but I'll see as soon as I get the bike going & tweak the TFI again, exactly what happens.

[blowndodge]

I wonder with your TFI off why your rich at start up?