PorKen

It seems obvious, but why didn't I put a variable resistor on the Temp II (coolant temp sensor), instead of installing an AFPR?

Currently I'm running 2.86Bar/41.5PSI (stock is 2.5Bar/36.25PSI), with my 3" single exhaust, cam advance, etc.

Could I have added more resistance to the Temp II for the same effect?

Perhaps two adjustable levels, one for closed loop, and one triggered by the upper throttle switch?

2V4V

PorKen,

I wouldn't recommend messing with Temp II beyond what you've already done. But...

Can you do it? Sure. As an analog controller, L-Jet uses that as one of the pulsewidth modifiers. It will not (as least as far as I've tested) see it as 'outside probable range' and discard it in favor of an inboard "map" - that tech is just not there. However, the problem is how do you predict and manage the results? IF you have an o-scope (or something else to measure pulsewidth) then I'd say give it a shot - but I think you'll find it a difficult variable to manage via Temp II.

Gotta run or I'd write more.

Greg

PorKen

2V4V,

Thanks for your input.

My intention was to have a variable resistor, but fixed at one value. (IE. not variable while running, like a piggy-back controller might give.) As far as I know, the coolant temperature resistance affects the fuel injector pulse width directly, at all times.

Now that I think about it, I don't see that the adjustable fuel pressure regulator is necessary on a stock engine. Perhaps if the horsepower and/or rpm limit were raised past the point that the longest pulse-width would not be long enough, but I've only raised the levels of my US 4.5 up to the stock US 4.7.

I already have a mixture meter installed, I guess I'll just have to put back the 2.5Bar regulators and try it.

2V4V

Yup, Temp II is always there as far as I've measured anyway. Even open loop with the full throttle microswitch activated.

Honestly, I've only spent time on looking at certain parameters that affect my current projects - there's some analog I/O that becomes irrelevant with the stuff I'm working with - it's handled in different ways.

You will not "max out" L-Jet until you really build/rev the weewee out of the engine - My dyno numbers prove (and Kibort's too) that US -Spec L-Jet - NO mods- can do at least ~275 HP crank (250+ RWHP) with decent A/F ratios. Certainly more, but that's a really safe conservative NA number.

With some minor mods, there's blower HP in the system. At some point, you need bigger injectors, and the flappy has to go, but analog (hell, CIS) is fully functional for all sorts of HP. Crap, people got pretty amazing HP out of *carbs*...

Greg

PorKen

~400Ω was all it needed.

Stock 2.5Bar regulators with 400Ω resistance added to the Temp II sensor gives the same ratio at WOT, ~12.5:1, as 2.86Bar.

Closed loop still works fine, too.

Being pessimistic, I put in a 15 turn, 10KΩ variable resistor. A 1KΩ would give finer adjustments.

PorKen

Cold start, traffic driving, BWAAAAH, all OK. Even with the cold start delete (here).

Three pots would be best, idle, cruise, and WOT (using the throttle switches to trigger reed relays). The idle could be richer, and WOT leaner (for max HP).

2V4V

Excellent! Glad it worked out. You should be able to find enough data points to determine the curve of Temp II's resistance by temp in the WSM. IIRC, it's in the L-Jet section. It's easy to determine with water and a thermo though if you need more data points to tighten up your curve.

You might want to do some extrapolation for, say, -20F, just to determine a good value for ski trips...

Greg

PorKen

By readjusting the AFM, it's dialed in nicely, at least according to my one-wire mixture meter. I'll have to put in on the dyno again to check the high speed mixture.

I going to get a linear potentiometer, and put it in place of the factory speaker fader (reuse the ****), so I can adjust the mixture on the road. (Road trip - low on gas? Trim lean!)

For wiring, I reckon phone cord would work, and not add much resistance, plus have enough wires to put the relay(s) inside the car. I just have to find that thread that had the best ways through the firewall.

2V4V

PorKen,

At some point, you might want to go wideband. The Innovate unit is (relatively) inexpensive and quite accurate. And you'll have a lot more (and more accurate) data to work with.

You might want to also consider a log pot (audio pot). I'm not sure the linear is gonna give you the range you need. But it might. Just a thought.

Greg

PorKen

2V4V,

That Innovate! data logger ($350) would be very helpful.

It appears I only need a 0-1K range (maybe 1.5K). As I understand it, a logarithmic potentiometer goes up faster as you turn it (because humans can't differentiate volume levels well at higher dBs). I'd rather have more turns, with more precision over the whole range.

PorKen

I found this cool counting **** to go with a 1K, 10 turn, linear pot:



There is a built in lock, and the measured resistance is actually within a few ohms to the dial so it will be easy to return to the best setting if changed during driving.

I am going to run 4 conductor phone wire (22 gauge, solid) through an extra hole in the grommet where the hood release cable runs through to wire it.
...

While driving around today I noticed the car is more responsive with the Temp II tweak versus having higher fuel pressure.

2V4V

Porken,

It is advised not to use solid wire in any automotive app. Especially since you will be traversing a pretty movable field (ie, hooking to the engine). Solid wire simply cannot flex like stranded, and will work harden and fail. But, if it's a temporary install, you should be able to get away with it for a while. The other concern would be the insulation life in the engine compartment - phone cord just ain't up for that kinda abuse for long.

(BTW- Not trying to pee on your Wheaties. Just spreading the BTDT's of others.)

Personally, I'd recommend 18 or 20 ga. stranded. At least TXL insulation. Better if you have extra money to spend.

Aren't those just the most cool *****? I have a couple in my electronics box, just waiting for a suitable project.

Sounds like it's going well. Maybe keep an extra ~6K resistor in the car in case it really gets cold - just a thought.

Greg

PorKen

2V4V,

Thanks for the tip. I will have to find better wires for the permanent install.

Perhaps I was unclear, I'm not replacing the Temp II sensor, just adding some resistance to richen the mixture enough to compensate for the more open exhaust, etc. It should be 'set and forget', ie: the Temp I and II and the rest of the L-Jet system should operate normally in different weather conditions without any additional adjustment.

However, since I'm a tweaker (no, not a meth head ), I want the option to change the mix 'in flight', on road trips for example, using the mixture meter as feedback. More torque or running hot/richen, easy running cruise/lean it.

I've found that even while in the L-Jet O2 sensor loop you can adjust the rich peak and speed of the lean/rich swing with fuel pressure or this mod. The 'computer' is still trying to maintain a balance according to the O2 sensor voltage, but the actual mixture over time can be adjusted richer or leaner.

BTW: Any idea what the reference voltage is for the temp sensors? Would it be the same as the throttle switches (9v)? I ask because if for some reason I wanted to lean the Temp II sensor, I'll need to go postive.

2V4V

Although the sensors are 2-wire, the second wire just grounds out to the block (I'm pretty sure Temp II goes to the cam tower ground point as does most of the ECU stuff). I'll look at a diagram later just to confirm which chassis ground is referenced.

I supect (knowing the Germans) they wanted to use a consistent ground point to avoid the ground differentials that can occur when you use a "one wire" style sensor and just ground via the sensor wherever that happens to be in the block.

Or they just got a good deal from Bosch.

Temp I is grounded in the ECU unit because it it located in the AFM. (L-Jet only).

Greg

PorKen

Thanks, but I knew the sensors themselves go to ground (right cam tower).

What I'd like to know is what the internal ECU voltage referenced against the sensors. If I knew that, I could add a slight voltage to the Temp II output, in order to lean the mix.
...

For anyone interested, I scanned the '80 L-Jet descriptive service bulletin a while back, but only just now uploaded it (I believe it's also on the Morehouse CD, right-click, 'Save Target As...', 2.3MB zipped jpegs): LJet_AFC.zip