Purs like a kitten...

[FastWoman]

Cozye, I've also had my reservations about the potentiometer "fix," which was why I was unwilling to consider it until now. However, after studying all the EFI info I can find, after seeing empirically just how much adjustment latitude there is, and especially after reading the short description of the temp enrichment in the link that Zed posted, I'm a lot more comfortable with the solution.

I think even with roughly 2k in series with a thermistor that should level out at 240 ohms in a warm engine, I'll still have cold/warm responsiveness in the system. That's because when the resistance is infinite, gas practically pours from the tail pipe. The roughly 2-3k (??? -- from memory) from a cold engine will now become 4-5k, which will still result in enrichment. I don't know if the enrichment will be too little when the engine is cold, but I'd rather have lean running when cold (with inadequate temp compensation) than lean running when hot (from compromises to the AFM adjustment). If this becomes a problem, there is still another solution. I can insert still another thermistor in the thermostat housing (in a plugged hole), and wire it in series with the existing thermistor. That will raise the resistance right much more in a cold engine. I would still have the potentiometer in series (with roughly 240 ohms lesser resistance) to balance out the mixture when the engine warms up.

What I like about this solution is that the correction to the mix is executed the same way at all RPMs and under all conditions. It seems to be an overall gain control, so to speak.

I'll continue in next post, but I'll get this up on the board, since I see you're online, looking at this thread right this very moment! ;-)

[FastWoman]

I've been thinking about how to tune this system. I suppose if I were racing, I'd want peak efficiency at WOT when running the engine at the peak of the power band. However, I'm not a racer. I'm just a semi-puttery, middle-aged woman who wants a cool car and doesn't want to destroy her engine by running it at the wrong mix. I'm first going to determine which series resistance gives me the fastest cruising speed at a fixed accelerator pedal depression. Then I'm going to see how that same setting translates to WOT conditions. I suspect I'll run right much richer at WOT. I might get a tiny bit more power running a tiny bit leaner, but my preference would be to optimize for cruising and to be "safe" at WOT. I suspect that's how the Nissan/Bosch folks designed the system in the first place.

Finally, I'll see how the engine idles at this setting, and I'll make any needed adjustments to the little AFM screw.

BTW, you should remove your screw and examine it. It has a rubber O-ring near its head, which I think mostly serves to keep the screw from wandering. Mine had crumbled practically to dust. (I honestly thought it was dirt.) I replaced it with a tiny sliver of silicone rubber hose that I carefully cut with a razor knife. The screw now has a bit of "hold" to it, and I'm confident it will stay in place. At this moment I have it set rather arbitrarily at 5 turns CCW from fully closed.

Edited November 14, 2010 by FastWoman

[FastWoman]

BTW, I've also re-thought where to situate the series potentiometer. I think I'll put a little Molex pigtail connector on the ECU harness and make a little 20-turn potentiometer/plug doodad (incased in epoxy?) to plug into it. Then I'll make an extension cord that will allow me to make adjustments from a seated position (passenger or driver) as the car is moving. Once the correct calibrations have been determined, I'll remove the extension cord, save it in my "special tools" box, and have it available for future "while-driving" tune-ups.

[cozye]

Sounds like a plan. Let me know how you fair with it. I do not think you will be too lean when cold though, if anything I think it will be too rich. 2k ohms is quite a bit of change on the water temp circuit, the car will think it's 68 degrees. My car seems to run pretty good when cold, and gets leaner when warmed up.

Have you given any consideration to megasquirt with a wide band o2 setup ? I'm thinking that may be a better long term solution. I may bandaid the car with the same tuning setup as you for the time being and look further into a fuel only MS setup for the future.

Like you, I have no desire to turn this into a race car, but I would prefer it to have a decent tune at all throttle conditions. Having a smooth running car is just so much more enjoyable to drive. I've got quite a bit of experience doing EFI tuning on motorcycles for performance (I have a nasty road racing habit), so I'm familiar with wideband tuning and all the variables that come into play with different speed/throttle postions/ gear selection etc..

Thanks for the tip on the idle screw. I'll pull it out and take a look. Also, FWIW I pulled my check valve for the fuel pump yesterday and inspected it, tested it with air. It holds air pressure, so I'm pretty sure it would hold fuel pressure. My fuel rail is boiling within 5 minutes of shutting down, and I now have an inline pressure gauge permanently mounted. I found out just how quick it boils by pulling the fuel rail inlet line off to put the gauge on after it had been sitting for 5 minutes only to be greeted with boiling steaming fuel shooting out. The fuel rail only holds pressure for about an hour or so, so I think my FPR might be leaking down.

[FastWoman]

OK, well, I've modded my car. Results are pretty good. Construction details will be in the next posting.

I think the enrichment isn't quite all that it should be when the engine is cold, but it's not far off the mark. Idle is a bit unstable; however, there's now something wrong with my air regulator. It seems to be stuck in the closed (warm) position, so my idle speed isn't elevated on startup. I'm sure when I get that fixed, my cold idle will be much more acceptable.

Anyway, I've had my car out on the road now. It feels about like it did when I loosened the AFM spring. The response seems pretty well balanced. I might have a bit of missing at WOT. Also after I give the engine a good WOT rev to maybe 4000 RPM, the drop in RPM pauses/bumps at about 2500 on the way down, and when I get that little bump, there's an exhaust pop. Do you (or does anyone else) know what happens at this 2500 RPM mark and what it means and/or how to get rid of it? I know the pop means I've got raw fuel in the exhaust, but beyond that obvious point, I'm not sure why it's happening there. I'm thinking that might be where the fuel cut ends on throttle-down.

Other than that, the engine seems to be running about right. The fresh plug I put in #1 reads OK -- a bit of brown, and perhaps a touch of chalky deposit on the electrode. I might be a tad lean.

Tomorrow I'm going to pull all of my plugs, most of which are quite black and sooty from prior issues. I'm thinking they might be a bit fouled and could use a good cleaning. I'll also see if I can tweak my timing a bit. (I still have to fix my timing light.)

All in all, this seems to be a pretty workable solution. I like this approach better than playing with the AFM spring.

[FastWoman]

I promised pics and details, so here they are. I apologize for the washed out photos. I didn't take them to be pretty, but rather to be clear. There's a whole lot of black in these photos!

I decided it would be better to put the inline potentiometer at the end of a pigtail, rather than inside the ECU. The first reason is that I can change out the ECU and still have the same mod. The second reason is that I don't have to go into the kick panel to change the mixture.

I tapped into the EFI harness inside the main connector shell. There's a screw on the end of it, and the shell slides apart I clipped the small black wire in the #13 position (labeled on the inside of the connector) and connected the two pigtail leads to the ends:

Here's the pigtail coming out of the EFI harness, which has been wrapped back up. The connector does not have the potentiometer attached:

Here is the potentiometer assembly plugged into the end of the pigtail. I simply heat-shrinked the 20-turn trim pot to the side of the connector:

The connector has a slot where it can be mounted on a prong. I bolted this prong underneath my dash, just above the hood release and vent control:

... and this is the connector mounted on the prong:

Not shown here are the trim panels. The panel that goes across the bottom of the dash leaves a hole where the air/fuel mix screw remains exposed.

Also not shown is the approx 5 ft extension cord I made for this connection. It allows me to make adjustments from the hood compartment area or the driver's or passenger's seat.

Edited November 15, 2010 by FastWoman

[cozye]

Nice install. Did you put your AFM back to the stock spring setting? What resistance did you end up with on the pot?

I did the same today, but more of a temporary test install. I just setup a trim pot on some barrel connectors and put it inline with the water temp sensor under the hood. I tweaked, drove, tweaked some more, looked a vac, listened to idle, drove, repeat, repeat ,tweaked afm, drove, etc..

I found that if I put the AFM back to stock settings, and dialed up about 1100-1200 ohms on the pot, it ran smooth but lacked torque and felt a bit flat. So I richened up the AFM a bit again, not as much as before and dialed the pot back. I'm at 4 teeth rich on the AFM now, and right about 800 ohms on the pot. This is pretty close for me. It's rich enough that the lean miss fire is gone, and not too rich that it feels soggy. I may tweak it a bit more but this is close enough for me to give it some drive time to get more familiar before making any more changes. I've also got to get the suspension freshened up, bushings etc,, I've got so much rattle, bounce, and shimmy going on that it makes it hard to feel the subtle stuff on the motor at speed. I've got new tires on it, and I think they might have one not perfectly balanced but I'm sure they will blame it on my lack of strut dampening and rattling ball joint so I'll fix that stuff first.

As far as popping on deceleration, do you still have your BCDD valve on the throttle body, and is it operational?

[FastWoman]

Thanks, Cozye Yes, I used the stock spring setting and ended up with a resistance of approx 2.6 k.

I think I'm about where you are on the tuning. I'm pretty close and will be just driving it a while to get a feel for it. I'll periodically do plug readings, calculate gas mileage, etc.

BCDD: Yes, I still have it. I'm pretty sure it's operational, but I suppose that needs a re-check.

G'nite!

[cozye]

Thanks, Cozye Yes, I used the stock spring setting and ended up with a resistance of approx 2.6 k.

I think I'm about where you are on the tuning. I'm pretty close and will be just driving it a while to get a feel for it. I'll periodically do plug readings, calculate gas mileage, etc.

BCDD: Yes, I still have it. I'm pretty sure it's operational, but I suppose that needs a re-check.

G'nite!

Check the BCDD. That exhaust pop on decel is a pretty common thing if you BCDD is not operational. You can do a "field" adjustment of the BCDD, but disconnecting the 12v wire to the solenoid to simulate going faster than 10mph, and reving the motor. Turn the screw on the bottom until it has a hard time idling down, then back it out slowly until it idles down gradually (for me this was approx 1/2 turn).

If it doesn't do this, then the BCDD is probably bad. They are NLA, and it's hard to find a working one so you might be stuck with it if it's bad. A lot of people just eliminate it and plug the hole in the TB.

[FastWoman]

Thanks yet again! I'll try that.

[FastWoman]

Oh, and you had asked me whether I had considered MegaSquirt. I did think about it, but I didn't put too much time into studying it. I was discouraged by the fact it's incompatible (I think) with vane-type, closed-loop systems. I didn't want to get into a major redesign of the system. I think more and more of these EFI systems are going to be going south, and maybe there will be enough people wanting to stick with EFI (rather than converting to carb) that there will eventually be a turn-key retrofit for our systems (whether vane-type or something more modern). When such a system surfaces, I might well make the switch.

[cozye]

Oh, and you had asked me whether I had considered MegaSquirt. I did think about it, but I didn't put too much time into studying it. I was discouraged by the fact it's incompatible (I think) with vane-type, closed-loop systems. I didn't want to get into a major redesign of the system. I think more and more of these EFI systems are going to be going south, and maybe there will be enough people wanting to stick with EFI (rather than converting to carb) that there will eventually be a turn-key retrofit for our systems (whether vane-type or something more modern). When such a system surfaces, I might well make the switch.

Yeah, I do not think it is that difficult though. Just get rid of the AFM all together and all you need is an IAT and MAP sensor. From my initial research it looks to be a much more modern type EFI, will reuse everything except the AFM, and can be configured for closed loop with a wide band o2. I really like the idea of a wideband. I've used them on my motorcycles for tuning on the race track and it's extremely reliable and easy to use. Personally I think I would do fuel only with it, not spark. I'm going to go this route I think, but will stick with my temporary work around for a while and work on other repairs. (suspension, paint, etc..). One of the main things I don't like about the current EFI setup, with work around is that the changes affect every rpm under every throttle setting. A system like MS you could configure for a particular air fuel ratio on a much more granular scale, and have it calculate the changes necessary to hit your target. Much more efficient.