Re: 4000 RPM Limiter

[raf40@oes.amdahl.com (Richard Funnell)]

I figured everyone had gone through this cutout at least once by now.

>      Stalking a new gremlin on my TQQC. It runs quite well until we get to
>      4000revs. Then fuel pump switches off. If I keep foot in it it
>cycles on/
>      off.
>      Checked air temp sensor and it has 20 ohms resistance (within
>      tolerances). The throttle switch also operational. My manual doesn't
>      refer tp any other factors that would cause the fuel pump to kick off.
>      Any ideas?
>
>   There are only three things that can cause the early
>   mac-02 style computers to limit at 4000 rpms.  This is
>   the limp home mode.  It means you have a bad input to the
>   computer.  The three are, 1.) air intake temperature
>   2.) coolant temp (sic?) 3.) idle position switch (?).
>   I don't have the doc';s right here in front of me
>   so I may have it partially wrong.  But you've got some
>   simple little electrical problem that can easily be
>   identified with a DMM.
>

You can add excessive boost to that list.  According to the Service
Training manual, the temp sensor isn't included, but I can vouch that it
can cause the problem.  In addition, I've heard that bad grounds, in
particular the cluster ground near the coil, can cause this.  In fact,
anything in the wiring path to those sensors or the ECU ground can cause
the problem.

>I kinda thought the only 4000RPM issue was the manifold air temp, but...
>
---snip---

 which results is a "signal voltage" that varies from
>roughly 70mv (Yes, 0.070 volts) to around 140mv. Just wiggling the connector
>to the ECU will use up about a third to a half of this signal range!

Strictly speaking, its low current that causes conduction problems, though
the two are obviously related.  I forget the details, but I vaguely recall
that below about 10-15ma, contacts really need gold to be reliable.

---snip---

>SFI!
>I drilled a small hole in the ECU case, and ran a small "Hi Fi" cable with
>gold-plated phono plug/jack into the ECU, and soldered the wire to the ECU
>pins (make sure you get the ground right...),

Do you mean you soldered to the gold tip???  In a recent tech tip, Ned
mentioned using something called Stabilant or some similar name.  Has
anyone tried the stuff?  It sounds as if it would be perfect to prevent
this kind of intermittent contact.  I think he said it was available from
Audi dealers, and was recommended by Audi.

>and the other end I ran into
>the cable harness and soldered to the plug pins,  so now I have nice solid
>low-resistance cable connection for the man air temp sensor that is still
>easy to disconnect when I unplug the ECU. No problems for several months,
>so I think I've finally fixed that bug.
>
>You can fairly easily check this out -- pins 18 and 19 are the man air temp
>input to the ECU, and are on the "top" row of pins with the ECU cover re-
>moved. Hook up a good high-impedance DVM across those two pins, and watch
>the voltage; at 4000RPM full boost it should be somewhere in the low 100mv
>range, well below 150mv.
>
>                                        -RDH

Richard Funnell,
San Jose, California
'83 urQ
'87 560 SL