Testarossa CIS issues...cont'd

If you are measuring 0 volts on the red wire with the engine running, that is definitely a bad sign, and the next place to check is the protection relay in the triangular black box over the rear RH wheel well (the protection relay is the thing that supplies the +12V to run the "E" part of KE-Jet). If you made the measurement at the red wire without the engine running, make sure you had the key "on" (in addition to having the jumper in place). I had made an error with a jpeg saying "key off" -- here is the corrected jpeg for doing the red wire test with the engine not running.
Image Unavailable, Please Login
This thread (post #14) has the information for testing what is, or isn't, happening at the terminals of the protection relay if you do have no +12V on the red wire (and, of course, first check if the 10A fuse is OK or blown):
http://ferrarichat.com/forum/showthread.php?t=179893
Image Unavailable, Please Login

 

Yes, that is the correct logic for the US TR "water thermoswitch" -- cold = closed and warm = open

I don't think so, but the schematic shows, when warm, the TTS should be:

W -to- G = open
G (which connects to the BL, white/blue, wire) -to- ground = a finite resistance (so 60 ohms seems reasonable)
W (which connects to the RV, red/green, wire) -to- ground = open

although could be that they made a mistake and swapped G with W (or you did )

Can you explain how you are measuring this? According to the Diagnosis sheet, when warm (i.e., the NTC sensor has a low resistance), the voltage across each EHA coil should be something like ~200 mV, not ~8V (the resistance of the EHA coil is ~20 ohms so when the voltage across the coil is 200mV this gives a EHA coil current of ~10 mA -- which is in the right ballpark). Or are you measuring from one side of each EHA coil to ground?

 

Franck -- First, let me warn you that if you short the two wires going to the EHA coil together, or you short one of them to ground, it is very possible that you will blow up the corresponding injection ECU (I blew one of mine up myself -- and even though the fuse on the protection relay blew, it wasn't fast enough to prevent damage to the injection ECU -- and it cost me $1K to get a used one).

Yes, you should really be measuring the voltage across the EHA coil when plugged in (and not to ground) -- i.e., the volt meter is in parallel to the EHA coil winding. Also, you do realize that to measure the current, the ammeter needs to be in series (not parallel) with the coil? This takes using some male and female (4mm diameter IIRC) bullet connectors to build the necessary jumper wiring to get the ammeter in series with the EHA coil -- are you doing this? Would you like me to prepare a figure describing the set-up I used to measure the EHA coil current?

 

OK -- during your next EHA current measurements also try unplugging the O2 sensors. In that condition (with the engine at warm idle), I believe the EHA currents should default to something like ~7 mA.

 

The best written reference that I've come across for describing KE-Jet with Lambda operation is the TR WSM 330/84 Section D, but you need to make sure you get the whole Section D (some TR WSM don't include the KE-Jet with Lambda stuff) -- pages up to D67 cover TR K-Jet operation (so reading that, and knowing how K-Jet works, is the prerequisite, and covers a lot of the stuff that is common to both K-Jet and KE-Jet -- fuel pumps, injectors, etc.) and then pages D68 to D106 covers the TR KE-Jet with Lambda operation.

Ferrari also produced a separate booklet (part No. 95990856) to cover the later euro TR KE-Jet without Lambda system, but it's just a subset (and sans Lambda) of what's in the TR WSM pages D68 to D106.

The last piece of technical documentation is Diagnosis Sheet No. 6 which gives some testing information (but really requires the Bosch 25-pin connector break-out box to use directly) -- send me a PM with your email address, and I'll email it to you.

 

Welcome to the typical confusion in some of the F documentation

The protection relay does not close with only the key "on" -- the other requirement is that the tachometric relay also be closed, which occurs when the engine is actually running (i.e., the 7-12 coil primary current is switching), or the starter motor is cranking (or you add the jumper wire as described in the jpeg to fool the tachometric relay into thinking the starter motor is cranking). The measurements for both steps 7 and 8 must be made when the starter motor is actually cranking over (or the engine is running, if it will with the injection ECUs unplugged) which closes the tachomteric relay, and subsequently, closes the protection relay. Since you have already measured +12V at the red wire with the engine running, you have already quasi-passed step 8, but doesn't hurt to use pin 2 as the ground to make sure the ground signal is reaching the ECU too. In fact, with the engine running, it is a good idea to measure the voltage on pin 2 relative to the engine cylinder head -- ideally this should be 0V DC, and if it is more than ~0.1V DC this would be bad for the Lambda control portion of the ECUs because they use the voltage on pin 2 as the "ground" reference to compare the signal coming from the O2 sensor. For example, if the voltage on pin 2 has "floated" up to ~0.4 VDC because of a bad crimp in the wiring harness, the Lambda circuit will add extra fuel to get more voltage out of the O2 sensor -- i.e., the Lambda circuit would (wrongly) be trying to get 0.9V average (rich) out of the O2 sensor in its quest to keep the average voltage difference between the O2 sensor and pin 2 to 0.5 V DC. (In fact, my own TR had this flaw so I added extra ground wires in the injection ECU connectors directly from the pin 2 wires to a good ground.)

Step 7 is confirming that the tach signal from the ignition ECU is reaching the injection ECUs -- if your tach is working, this is a good sign that this signal is present, and the only way it would be reaching the tach, but not reaching the injection ECUs is if there is a wiring problem.

Your TR being able to start and run quasi-OK is an indication that the manual mixture tweaks are set towards the rich side. During normal warm idle, the EHA is only causing a fairly small amount of extra fuel to be added so tweaking the manual mixture screws a bit rich can easily compensate. It's very much like the 308QV K-Jet with Lambda where, if the protection relay is dead, as long as the mixture tweak is reset at warm idle, it doesn't run too bad. One difference though is that the KE-Jet also uses the EHA (which would be dead) to provide enrichment during throttle blips; whereas, the 308QV uses the WUR, so 308QV K-Jet with Lambda get along with a dead protection relay better (if they have been retweaked when the protection relay is dead).

Good Hunting!

 

steve, where is pin 2 in this description? I don't have the dignostic sheets.

 

"pin 2" is pin 2 of the 25-pin injection ECU connectors (either one). However, without the "factory signal breakout box", it isn't easy to access pin 2 with everything plugged in. IIRC, this ground signal also goes into the black triangular box to one of the relays so it can be more easily measured there (but I'm away from my TR documentation at the moment). Tonight, I'll post a Dropbox link to Diagnosis. Sh. No. 6. Do you have a copy of the 509/88 US TR wiring diagram book?

IIRC, Jagbuff eventually traced the source of the problem that he was chasing here to a bad crimp at the connector pins for one of the barometric capsules (no signal from the barometric capsule = looks like super-high altitude to the injection ECU = the injection ECU drops the EHA current to zero trying to lean things out for such a high altitude condition).

 

ok Thanks! I'm tracking. I don't have current on either EHA. So I was wondering what's going on. Was thinking a ground could be bad.

 

so I checked my barometric capsules. V is 7.5 for both and looks like V1 is 5.5 for both. The third wire I got 0V for both. Looking at the chart a V1/V is around .73 which corresponds to an elevation of 7500 feet. I am in detroit. Could both these capsules be busted? or did I do something wrong?

 

I didn't mean to imply that you should necessarily suspect your barometric capsules. Rather just that darn near any small thing can go wrong, and it can take a lot of effort to find it. This link should get you the (corrected) 4-page Diagnosis Sheet No. 6:

https://www.dropbox.com/s/gm81j3cq3048qe1/TR%20Diagnosis%20Sheet%20N%206%20US%20KE-Jet%20Injection.pdf?dl=0

which is really a series of tests (starting with a cold engine) where, if something is out-of-bounds, it needs to be fixed for sure, and, if all is OK, the system should be working (but even that doesn't guarantee that there can't be some other hard-to-detect Gremlin still mucking things up -- especially inside the ECUs).

With regard to the V1/V issue -- you'll have to compare the pins used for the V1/V test to what you did - for example, if V1 is measured in a way that would give the difference of the two voltage measurements that you report (i.e., V1 = 7.5V-5.5V = 2V) you'd get a much more reasonable result for Detroit (and my guess is that, since the two are the same, the trouble is in the measurement and not the barometric capsules themselves - but please let us know).

 

OP states a stumble at 3k as you accelerate-correct?
You are pulling values from the stem both hot, AND cold(closed loop and open loop), correct?

You are testing this KE car as if it were a Mercedes M117! Engines car!

Late TRa have a fairly common problem which can be eliminated with you answering a couple of questions-this assumes that most of your static data info ias sll within factory(Bosch) specifications?...

If yo disconnect the O2 sensors from the harnesses-both of them-and monitor the HC/CO numbers: what happens to your numbers once the cat hits 450degreesC?
Also, when HOT, you set L and R fuel towers to-if memery serves me correctly, 1.25-1.50%
THEN you plug them in-when car shifts to closed loop what happens to both O2 values AND port readings for HC/CO
Do the skyrocket?
If so, ck for bad ground in the lambda wiring loom-the drivability problems you describe will present if the ground is poor or bad
This is found on the rt side, inner set of runner studs on the 2nd or 3rd stud....there is a "loop" wiring end as well as 1-2 others on this stud. There will be "white" crusty detritus in this hint...
Stemless wire wheel it all clean, put a copper electrical coating paste on threads and washers and retorque...if ground fixes this, your sky high readings at port and O2 harnesses will drop like a stone....

This is more common that you might think-say this at the dealer many times...
It's well worth checking this ground circuit as it's a known issue in LATE TRs....

 

I would say that most common problems with the CIS are from letting the car sit unused. The fuel gets old and gums up the injectors, and the metering slits in the FD.

I would start by draining the tank, and filling it with fresh fuel. Then pull the injectors, and soak them in carb cleaner overnight. Make sure the poppet valves are working, and that the filter screens are not clogged. Also clean the spark plugs, as an improper fuel mixture will foul the plugs. I use the propane torch method for cleaning spark plugs.

Run the car at least once a week.