Temp sensor resistor

[chaseincats]

Hi all

I was doing some reading and found that a lot of folks have been adding a potentiometer in line with the temp sensor so the efi doesnt keep leaning out the mixture according to temperature.

I was reading that the efi runs the most stable with the car at 175 which means keeping the resistor reading at around 300 ohms.  If I were to add that and set the 1k resistor to about half, wouldn't that mean it is constantly richening the sensor's reading by 500 ohms?  What I'm getting at is my car will run at around 185 idling but 170 on the highway.  If I'm looking for the engine to be reading a constant 175, the potentiometer trick would just tell the computer it is constantly running cooler regardless of rpm and wouldn't really do what I'm looking for, correct?

The best writeup I found at AtlanticZ here.

 

Any ideas?

-chase

Edited March 31, 2023 by chaseincats
Added additional info

[Zed Head]

Your post seems to have flipped the cause and effect relationship.  Coolant temperature affects sensor resistance.  But sensor resistance does not affect coolant temperature.

[chaseincats]

correct, sorry if it didnt read that way - this is tricking the ecu telling it that the car is running at 175 when it isnt to modify/stabilize the fuel mixture the ecu provides

[Zed Head]

It's just a way to richen the mixture across the whole temperature range, and air flow range.  It's a constant.

Edited - I'm not sure that the actual curve is as flat at the end as Nissan shows it.  Maybe it is though and with added resistance the ECU never sees a fully warmed up engine.

The whole curve will get shifted to the right.

Edited March 31, 2023 by Zed Head

[chaseincats]

what would the difference be then between adding this resistor vs messing with the AFM?  Would it garner the same result then?

[Zed Head]

The AFM vane is controlled by a spring.  So you have to get in to how the force curve changes as you loosen or tighten the spring.  For example, if you loosen the spring, the vane will move faster at low air flow, than at high air flow.  I think, my memory of spring rates and how to work with them is fuzzy.  And I don't know if a clock spring (which is what you'd call the AFM spring, I believe) is a variable rate spring or constant.  It's probably out on the internet somewhere.  Maybe even in an old paper book. Plus there is some preload.  Some things can just be logic'ed out though.  The ECU might behave as it would when the AFM is maxed out at mid-range air flow rates if the spring is loosened enough to let the vane move to its limit, even though the air flow rate is still changing, because the spring is not hindering vane movement as much.  So you'd end up rich at low air flow and lean at high air flow.  Because the AFM signal to the ECU has stopped changing. 

Or, if you tighten the spring the AFM vane might not move at all at low air flow rates.  So you'd be lean at low air flow for a certain spread of air flow rate, then still lean as more air passed by but did not move the vane.  Basically you change the slope of the curve. 

The resistor slides the whole curve over, AFM adjustment changes the slope of the curve and can create artificial limits.  The ECU and AFM and injectors and sensors are all finely balanced to work together.

[chaseincats]

1 hour ago, Zed Head said:

The AFM vane is controlled by a spring.  So you have to get in to how the force curve changes as you loosen or tighten the spring.  For example, if you loosen the spring, the vane will move faster at low air flow, than at high air flow.  I think, my memory of spring rates and how to work with them is fuzzy.  And I don't know if a clock spring (which is what you'd call the AFM spring, I believe) is a variable rate spring or constant.  It's probably out on the internet somewhere.  Maybe even in an old paper book. Plus there is some preload.  Some things can just be logic'ed out though.  The ECU might behave as it would when the AFM is maxed out at mid-range air flow rates if the spring is loosened enough to let the vane move to its limit, even though the air flow rate is still changing, because the spring is not hindering vane movement as much.  So you'd end up rich at low air flow and lean at high air flow.  Because the AFM signal to the ECU has stopped changing. 

Or, if you tighten the spring the AFM vane might not move at all at low air flow rates.  So you'd be lean at low air flow for a certain spread of air flow rate, then still lean as more air passed by but did not move the vane.  Basically you change the slope of the curve. 

The resistor slides the whole curve over, AFM adjustment changes the slope of the curve and can create artificial limits.  The ECU and AFM and injectors and sensors are all finely balanced to work together.

So in that case, it sounds like you set the baseline with the potentiometer on the temp sensor and fine tune it with the afm?

[Zed Head]

That's not where I end up.  My general feeling is "don't mess with the AFM".

[Captain Obvious]

Well one difference between adding a resistor to the temp sensor circuit and messing with the AFM is you can only make it richer with the resistor, not leaner. By messing with the AFM, you can (in theory) do either.

So out of curiosity, Is there a problem here you're trying to fix? Or is this an academic discussion only?

And...

10 hours ago, chaseincats said:

I was reading that the efi runs the most stable with the car at 175 which means keeping the resistor reading at around 300 ohms.

Where did you read that? 

[cgsheen1]

The EFI does depend on the coolant temperature sensor for correct fueling. As the coolant temp goes UP, the resistance DROPS. The FSM shows a table for expected temp vs. resistance values in the Engine Fuel section (page 53). Why would you want to ADD resistance to the signal using a variable resistor? Else, why would you remove the coolant temp sensor (thermistor) by replacing it with a manually controlled variable resistor? The aged wire and connectors in your engine harness may already be skewing signal with additional resistance. Additional resistance which would fool the ECU into thinking the engine was running cooler that it actually is.

I think a more productive route would be to measure the resistance at the sensor, and then at the ECU - to see how much additional resistance wiring and connectors might be adding. If the skew is significant, a new wire and connector from the sensor to the ECU might beneficial. In my years with this early EFI and ECCS in S30's and S130's, I've found 90% of the problems to be in wiring and connectors. I've always advised: Clean both sides of your connectors, then do it again, Ya... then do it again. and don't forget the ECU...

Most people ignore me and just go straight for the AFM. When you know everything there is to know about that Bosch L-Jetronic AFM, go ahead and bust it open. Trouble is, if you did know everything there is to know about that AFM, you'd never actually do that...

[Zed Head]

42 minutes ago, cgsheen1 said:

Why would you want to ADD resistance to the signal using a variable resistor?

Because it has been shown to work, very well.  There is much discussion around the EFI Z based forums about why old EFI Z's very often have lean condition intake popping problems that are then cured by adding resistance to the coolant temperature circuit.  Fuel quality, electronic component drift, etc. have been proposed as potential reasons that the system doesn't work the way it was designed to.  But, whatever the cause is, the added resistance is the cure.  

There aren't many reasonable aging reasons for resistance to drop in the coolant temperature circuit.  Maybe something in the ECU changes.  Who knows.  But cleaning connections in the coolant sensor circuit will only lower resistance, making the air-fuel ratio leaner.

I know that you moved on to the ECCS system and other more developed Nissan EFI control systems long ago.  The primitive 280Z EFI system has its own idiosyncrasies.  The 280Z EFI has a fever and the only cure is more resistance.

[kickstand80]

If you have popping going on then you have something really out of spec. If everything is working as it should, there will be no lean popping. 

I have considered removing my cylinder head temp sensor and using a variable potentiometer like this:

https://zcardepot.com/products/variable-resistor-for-fuel-injection-mod-280z-280zx

That way you go rich to start the car and lean it out as it warms up.

If you have a Air/fuel ratio meter you could nail it down perfect while cruzing