Update on EFI mod -- potentiometer on coolant temp sensor circuit

[FastWoman]

Hi all,

As you may recall from my "purs like a kitten" thread, I installed a potentiometer in the coolant temp sensor circuit last fall to richen my fuel/air mixture. I did this because my ECU had apparently drifted in its calibration to a progressively leaner state (shorter injector pulses). Inserting resistance in the CTS circuit results in enrichment proportionally throughout all operating conditions -- or so it would seem.

My car has been running like a top ever since, with vacuum in the 18.5 in Hg vicinity. I did have one glitch recently, in that I decided to drive my car to Richmond. My fuel pump died en route, and I had to be towed back home. I replaced the pump and noticed my engine missing a bit when it got warm. I mistakingly took this to be a sign my engine was about to die from some unknown problem, so I quickly returned home both times this happened. My car has sat until today, as I didn't have time to sift through the problem until now.

The first thing I did was to read my plugs. Happily, all are a nice mocha color, so my mix has been great, more or less, since last fall. I noticed a bit more carbon on 5 and 6, while 1 looked a bit leaner than the rest. However, I think these differences were probably relatively minor.

Then I installed a cheap fuel pressure gauge between the fuel filter and fuel rail, so that I could observe whatever fuel pressure irregularities might occur on the road. The gauge, T-fitting, and brass nipples were incredibly cheap from my local True Value.

So I started my car, observed a peppy fuel pressure response, and let the car warm up. As it got warm, it started missing a bit, but only slightly. I noticed that my engine temperature was a bit on the warm side when this was happening. D'oh! The A/C was running, and it was somewhat hot today. After I turned off the A/C, the temp dropped a bit, and the missing improved but didn't resolve.

Then I checked my engine vacuum. I was getting only about 17 in Hg. I found that I was running a bit leaner than optimal, and I richened the mixture by dialing in more resistance into the CTS circuit. That raised the vacuum to 18.5 after I dropped the idle speed to spec. Revving was good, and missing was no longer a problem.

It might be too soon to tell, but there might be subtle differences between optimal wintertime and summertime adjustments when the CTS circuit is adjusted by adding resistance. It might be that my cooling system is a bit strained by the A/C and summer heat, resulting in a leaning out of my mixture. The difference both in temp and mixture is subtle.

So it seems I over-reacted to a few little misses. I should never have doubted my baby. Anyway I though y'all could use the update on how my EFI modification has performed until now.

[Jeff G 78]

I added a potentiometer to mine a few years ago and it has run great ever since. I REALLY need to open the AFM and adjust the spring, but for now, tricking the water temp circuit has done the job.

[5thhorsemann]

I know it has nothing to do with the subject, but as a coincidence, a friend of mine's son bought a 240zx last night and it overheated on the way home. Long story short I went and looked at the car. Blown head gasket. The PO stuck a potentiometer in the temp gauge circut to dial down the gauge to get the car to sell. Buyer beware.

[FastWoman]

Jeff, I'm leary of adjusting the clockspring in the AFM. That was my first approach to correcting my mix, but it rendered the vane tension too loose. Using this approach, the AFM would have swung wide open with too little air flow, leaving the mixture too lean after that. I think it's important to preserve the aerodynamic properties of the AFM and then to adjust the ECU's response to the AFM data electronically, hence the CTS mod. This mod seems to apply the same pulse width correction across all operating conditions.

Horseman, OUCH! Yeah, different coolant temp sensor!

[Jeff G 78]

Same here FW, but supposedly, it's the "right" way of correcting the mixture. I have the plugs looking perfect right now and I get 25 mpg, so I hate to mess with it.

[grantf]

never heard of a 240zx.

[cygnusx1]

Z motors like to drink. Rich mixtures makes them purr very smoothly. There are fixed resistors inside the ECU, that can be swapped, to taylor ALL of the parameters of the fuel injection system. I posted all the info I have over on HybridZ. I have never tried it myself, but the original author of the information did, and verified it all on a bench with a scope.

[FastWoman]

Wow, Dave, fantastic! My next stop will be HybridZ to dig up your postings! It might be nice to set up an ECU with 20-turn trim pots accessible through the case for periodic recalibration.

Jeff, the "right" way is only some other person's opinion. I don't think Nissan ever meant for us to mess with the clock spring, or else it would be in the FSM. To me, messing with the CTS circuit makes much more sense. Better still, I might be able to mess with the base pulse, per the info Cygnus cites!

It was a good day for my Z. This was the first evening it got out since the humbling experience of being towed back from Richmond (poor baby). We met some friends at a nearby pub. They had questions about their VW beetle, parked next to my Z. As I was chatting with them about it, some guy piped up and said, "THIS is the car you should be interested in!" I smiled and said, "Thanks! That's my car." We had a long chat, going down memory lane about his 280ZX from long ago. Then some 20-something kids walked by. One of them circled my car and said, "This car is SO SICK!" Made my day.

Anyway my engine ran like a top again. No more misses.

[cygnusx1]

Let's save some time....

http://forums.hybridz.org/index.php/topic/95316-braaps-l6-efi-induction-advice-and-tips/page__p__896858__hl__%2Bfuel+%2Bcut__fromsearch__1#entry896858

[FastWoman]

Thanks! I've saved a copy to my hard drive in my 280Z reference folder. Great stuff!

Edited May 27, 2011 by FastWoman

[Z train]

Jeff, the "right" way is only some other person's opinion. I don't think Nissan ever meant for us to mess with the clock spring, or else it would be in the FSM. .

Sorry,but playing with the spring is the right way to do it.There are a lot of things not in the FSM.Playing with the spring was SOP back in the day to make them faster,or in some cases-lean them out.As long as you mark the original settings,moving the spring is NOT a big deal.

[FastWoman]

Ztrain, I didn't say messing with the clock spring is a big deal. I've done it -- and then undone it. I also don't see much harm in it if the adjustment is only a few teeth. However, if the "correct" adjustment is around 15 teeth richer (as it was in my case), that introduces a serious issue with the overall response of the AFM.

If you look at the AFM's design, the vane stays almost closed at idle, under some spring tension. It would not make sense to actuate the potentiometer farther than that. Why meter air at lower-than-idle flow rates?

From there, the vane will open up fully with a certain (large) air flow rate that's well within the operating parameters of the engine, and it will continue to add fuel beyond that point based on temp readings and RPM. If the two ends of the potentiometer are to correspond correctly to idle and that larger airflow rate beyond which the AFM readings are ignored, the clock spring has to be at a certain tension. If it's adjusted 15 teeth richer, it will peg out prematurely. No correction will be applied beyond 4000 or 4500 RPM, and the engine will be running very lean in the peak of its power band at WOT. That's definitely not a good thing! :nervous:

By altering the CTS resistance, a proportional correction is applied throughout the AFM's response range and (more importantly) beyond its response range, into the higher RPM ranges.

It's very interesting to read BRAAP's approach in the above-linked thread: "In short, I found the AFM works as originally designed and works INCREDIBLY well, and due to the narrow range that it functions, which is only below 4500 RPM and mostly under cruise conditions, the only tuning or alterations I perform to the AFM is to adjust the spring tension, but only after the WOT tune is set by altering the water temp sensor resistance. " So he basically uses the CTS circuit to tune the mix under heavy flow and sets the clock spring to calibrate to low flow. That's a bit more sophisticated than my approach. (A dyno would help.) However, it's plain to see his primary adjustment is of the CTS circuit, because "this offers a linear AFR change across the entire RPM range under all conditions."

That's basically the point I'm making. BRAAP and I reinvented the same wheel, more or less. I made educated guesses from the info in the FSM, and he did one better by actually reverse engineering the ECU. Our primary approach is essentially the same.

Edited May 27, 2011 by FastWoman