Tripminder Calculations

[Lutrova]

Does anyone know how the tripminders/message centers handle changes to the ECU/fuel injectors? Is the system capable of adjusting to larger injectors and still (somewhat) reliably estimating things like fuel economy? Or is that calculation entirely handled by the tripminder unit? I assume at the very least you'd need to keep some sort of EEC-IV ECU. But if you do something like an HO upgrade, will the tripminder MPG reading still be accurate?

More broadly, does anyone have any insight into how data is collected and processed by the tripminder? My car has a few electrical gremlins, and the digital fuel gauge has never consistently displayed the right fuel level. But at any given time the fuel gauge, distance to empty warning, and regular distance to empty readings will all give wildly different ideas of how much gas is in the tank. I would think they ought to agree, since there's only one signal coming from the fuel sender, even if the tripminder is pulling from the fuel gauge instead. I've also thought the fuel gauge and message center could be faulty, but after swapping out for a junkyard cluster I'm still seeing the same issues. (Although both units could just be failing in the same way.)

[GM_Guy]

That is a good question. The tripminder only has two inputs, fuel flow from the ecu (tripminder pin6) and Speed from the VSS (tripminder pin8) everything else is just power and grounds. Given that, I'd say it gets thrown off as the calculations the tripminder makes is based on lopo 5.0 formulas.

[Lutrova]

Does that mean there are different tripminders for the HO Mk VIIs than for the LoPo Panthers or Mk VIIs?

I guess it all depends on what sort of fuel flow signal the ECU is sending to the tripminder. If it's just an on/off fuel is flowing or it's not, then you'd have to have different tripminders for each engine type. But if the ECU somehow communicates the quantity of fuel flowing, then you could use the same tripminder for all models.

[sick88tbird]

I think fuel flow would HAVE to be calculated in the ECU. Like you guys pointed out, different engines, injectors, gearing… they would have to make a ton of different tripminders if the estimations were crudely done at the tripminder based on speed and I/O fuel flow. But what do I know! Lol

[gadget73]

Its set by the injector parameters in the ECM. The real glitch is that its only accurate when the injectors match the programmed parameters, so the 19# with a Mark VII ECM is only really completely accurate with the specific injectors those cars came with. Other 19# will work but since they aren't quite dead-on identical the fuel mileage number ends up slightly off.

[Lutrova]

It's not as though the fuel economy estimate has ever been that accurate. It seems like the computer in my own car consistently underestimates the MPG, which I think has to do with it overestimating how much gas is used at idle.

From a brief look at my EVTM, it looks like the message center pulls fuel level and speed data straight from the fuel gauge and speedometer. So if one or the other is not reading properly, that would trickle down to affect whatever numbers the tripminder spits out. In my case, the digital fuel gauge continues to read all over the place, which in turn screws up the distance to empty figures. The sender at the tank seems to check out, and I've tried cleaning the instrument panel contacts and swapping out the fuel gauge, which leads me to believe there's a crusty connector between the two that's got a flaky connection. The trouble is figuring out which one it is.

[gadget73]

the distance to empty stuff is based on fuel level from the tank, but I don't think it uses current or average fuel economy. At least it doesn't seems to on the Mark VII or Continental. Fill it up, you get an number that has nothing to do with reality. Its a good 100 miles more than what I can actually get from either car.

The average fuel economy number on both are very accurate though. The VII is the same fuel flow signal that the Towncar relies on, straight out of the ECM. The diesel uses a mechanical lift sensor in one of the injectors and some magic box to translate that signal to what the tripminder wants.

the current and average fuel econ don't rely on the level sender though, thats just speed and fuel flow. Can check the speed input easily, just reset the "average speed" function while driving and when it spits back a number compare that to your actual speed. If it matches, the VSS signal is fine. I don't really know of a way to check the accuracy of the fuel flow signal. There is a self-test mode that will confirm if you have signal but the manuals are basically "verify it does something" . Don't remember the buttons to push on a Towncar but its in the shop manual somewhere.

[Lutrova]

I took a look at the shop manual. I didn't realize how different the message center and tripminders were. Getting a fuel remaining readout would be cool, especially since the message center has that figure as part of the quick reference test. But I guess Ford felt they had put enough novelty in it already compared to the tripminders.

It looks like you can run a quick reference test by repeatedly pressing four buttons at once. Further pinpoint testing, however, requires the special Rotunda tester. Most test fails in the manual just call for replacing the display or module anyway. I once tried disassembling a message center display, but the plastic was so brittle that it grenaded itself as soon as I started removing the first bolt.

That just leaves cleaning up all the connections then.

[Lutrova]

I ran the quick reference test yesterday on my message center. Everything checked out fine except for the fuel level. I'd need the Rotunda tool, I guess, to see whether it's the module or the wire from the fuel gauge that's to blame, but I have a hard time believing the module has really gone bad in the fuel level and nothing else.

The connector to the module also looks pretty clean, and the fuel gauge jumps around every couple minutes, even with the car stationary. So really, I believe the true culprit is probably a bad connection somewhere between the fuel level sender and the gauge.

There's supposedly a connection behind the dash, but the EVTM isn't very specific as to where, and there's a lot of wires back there. It's going to take some sleuthing to find it.

Also, regarding the fuel economy figure: the shop manual suggests that a bad fuel flow signal from the ECU would result in an overly high MPG estimate, as I suppose the message center wouldn't receive all the fuel data. My distance traveled reading seems to be correct, so I have no idea how my car is underestimating the average fuel economy.

[gadget73]

Check at the tank, its the most likely spot to have a crappy contact since its under the car. Its on the front side of the tank, no need to drop it or anything like that.

No fuel flow signal will give you like 150 mpg.

[Lutrova]

Back when I was still living in Tacoma, I asked my mechanic to take a look at the fuel gauge issue. He only got as far as ruling out the sender or its connection, which I've personally confirmed more recently. It's a very clean contact.

Is there some way to verify connectors with a multimeter while you're working on cleaning them?

[sly]

Backprobe the connector. Poke the probes along side the wire on the back side of the connector inside the rubber around the wires. Then you can check the connections. Typically if the connector at the sender is good, next thing to check grounds... under the dash for the gauge, chassis ground... then the connectors to the cluster and the gauge itself in the cluster. Might also try pulling the antislosh module out of the cluster and see if it works then. If it does, the antislosh was causing the issue. IIRC, the antislosh is parallel to the guage, not in-line... but if it's in-line, you may need a jumper to bypass it instead of pulling it out. Not sure exactly how that would be accomplished though.

[Lutrova]

I've never heard of an anti-slosh module, nor can I find anything about it in the shop manual. Maybe it's integrated into the digital fuel gauge assembly? If so, I doubt I can get at it without obliterating the plastic housing holding it together.

If I understand correctly, the fuel wire is sending resistance to the gauge, which in my case is fluctuating and causing an erratic reading. So if I'm back probing a connection, am I looking for changes in resistance from one side of the connector to the other? Or am I probing each connector against a ground?

[friskyfrankie]

I know the anti-slosh module is a separate board on the analog IP's but not sure on the digital ones. Maybe it is part of the IP?

[sly]

It's usually a small board wedged into the side of the cluster. No clue about the digital dashes though. But I've seen all of those issues on analog dashes, so I figured I'd throw that out there for consideration.

[Lutrova]

Taking a second look at the shop manual, there is a brief statement on how the digital fuel gauge has anti-slosh mechanics built into the module, which probably means a failure there wouldn't be serviceable.

Pivoting slightly, another issue I've encountered with the Message Center is its display flickering off completely, or the text changing to some nonsense or a single, super bright character. The incorrect text seems to me like a bad connection in one of the wires between the logic module and display, but the display turning off entirely could be a power supply issue. Or it could still be one of the logic-display wires, but the power supply is interesting because of an oddity in the wiring diagram.

The other two digital gauges are supplied by power only in Accessory or Run, but the Message Center components have constant power. What's strange is that the same circuit that supplies Message Center power also appears to run to the engine compartment light. I'm currently running a junkyard-find light, which has to be hit just right to turn on or off, which doesn't seem like intended behavior. The shop manual appears to be silent on its operation. Does anyone know what's supposed to trigger that light on or off? Is it opening the hood? And if so, how is the light aware that it happened?

[kishy]

Just straightening some uncertainties from above:

The boxy box (to 89, non-digital) anti-slosh module is a dedicated little box somewhere in the vicinity of the glovebox. It is often but not always orange.

The 90 analog cluster (which despite its aesthetic differences is fundamentally the same thing across Ford-Mercury-Lincoln) introduced the board which plugs into the cluster itself.

I was not sure how the digidash Lincolns did it; never had occasion to figure it out.

Hood lights use a mercury switch (or some comparable assembly); as the light assembly tips backwards the switch closes and the light turns on. You should find that the wire going to it is always hot.

[gadget73]

I'd be very unsurprised if the message center module doesn't just need an overhaul. Old electronics that live in a bad environment just don't hold together forever. I've had a lot of issues with buggy BS that turned out to be capacitors leaking goo on the board.

[Lutrova]

I can definitely see that being the case. As I look more into it, there just aren't that many wires connecting everything, and so far the amount of corrosion on the contacts has been minimal.

Have you found bad/leaking capacitors to be the most common form of failure on old car electronics? I don't think I can pull apart the display boards and have them survive, but the logic module is in a metal case, so it might be possible to disassemble and clean up. And the size of the circuits on these '80s boards look like they can't be too tricky to solder.

[sly]

Most of the old stuff is through hole... so yeah... usually easy.