TR issue/stalling with cold engine

Yes, 0.91V on an unplugged O2 sensor (relative to ground) at warm idle is towards the rich side of things -- should be more like ~0.7V (just slightly richer than stoichiometric) and will still vary some so you've got to do sort of an average watching the voltmeter.

There is a lot of hysteresis in the water thermoswitch switching action -- it opens (nominally) when the temp reaches 63 deg C, but won't close again until it has cooled to 36 deg C (although both of these specified switching temps have probably +/- 5 deg C or even +/- 10 deg C tolerances).

 

Were there plugs covering the holes to the mixture screw? If not, that’s a huge vacuum leak.

 

Try a third relay? Don't think that there could be anything wrong with the harness as the 30-to-87 jumper uses that, too. One difference is, that when you put the jumper in place, you are probably wiggle/moving the wires and relay terminals around a little (to new/different positions) so you might make sure that the wires are well-crimped into their female terminals of the relay socket with good electrical connections. Additionally, making sure that the female terminals in the relay socket are clean and tight on the male spades of the relay is no sin. If the jumper solves the problem, a working relay should solve the problem. I guess the good news is that you found where to focus your attention.

Correct (on the unplugged narrow-band O2 sensor wire relative to ground).

Any better instrumentation to measure the exhaust content is only a plus. I'd target something like 1.5~2% CO so I'll leave that as your internet homework to work out what that corresponds to in 14.X language.

(I typed this message before seeing your next two messages and will just leave it here as it could still be something like a bad crimp between a wire and a female terminal that is intermittent depending on flexing it around some.)

 

Way, way bad! If you don't have the slotted cover screws in place and just have open holes = you'll definitely need to readjust and get something to plug the holes (like a rubber plug or something) -- and maybe readjust the throttle plates and/or air bypass screws. Of course, making the actual mixture adjustment requires having the hole open, but then you close it with a fingertip to then read the exhaust measurement.

An Fchatter in France indicated there was a shop making replacement slotted cover screws (but I neglected to record his/her username):

Image Unavailable, Please Login
Image Unavailable, Please Login

There are also rubber plug things like this that could probably work (or be made to work):

Image Unavailable, Please Login

See this post for the strategy of how the throttle plates and air bypass should be adjusted (when the mixture screw access holes are not wrongly open):
https://www.ferrarichat.com/forum/posts/148819203

See my comments about a bad crimp maybe causing this intermittent behavior in post #31.

(It's getting to the point that I should first ask: "OK, what horrible modifications, or hackery, have been done to your TR" before I respond to anything .)

 

Plug those holes. Then redo the oxygen sensor readings again.

 

No need to apologize -- I was just noting how it's changed over the years. When I got my TR in 2000, they were still new enough that very few had any serious modifications or had even been messed with at all. Not that way anymore . Plug those access holes!

 

Gianni -- On a side issue, since you have the black box apart, could you please give the 10-digit Bosch part number shown on the barometric pressure sensors mounted to the back surface on the outside (or post a picture of their top surface). The Bosch 10-digit part number should be something like 0 280 101 XXX (if some slightly-confusing prior information that I have is correct). TIA.

 

Thanks very much. I had/have both 0 280 101 001 and 0 280 101 002 in my notes, and can't remember why -- if anyone knows the difference, or if there is a difference, please post. (Adding 0280101001 and 0280101002 for searching).
My present guess (just a guess) would be:
Bosch 0 280 101 001 = Ferrari 121738 (shown in the 1987 TR SPC for US and CH C7)
Bosch 0 280 101 002 = Ferrari 134808 (shown in the 1990 TR SPC for US, CH C8, and CAT)
and the change was something that Bosch did for themselves for manufacturing reasons, and they work the same.

Yes, it can be a real challenge to move the mixture screws as little as you need to move it and still actually move it. Also, definitely best to "exercise" the airflow plate mechanism motion a little after making an adjustment by gently revving the throttle a couple of times to make sure it all repeats when it returns to idle, and to make the final tiny adjustment going in the CW direction on the mixture screw as the TR WSM suggests.

Also, yes. It shows that the narrowband O2 sensors have a very steep slope in their response curve. Having the hole open also lets in unfiltered air -- bad! Do what you need to do to close them up, and the "fingertip test" should help you confirm that they are truly closed airtight.

 

0.8V on the unplugged O2 sensor at warm idle is OKish -- better than 0.9V

Probably not, but I wouldn't. Unless you separate the airflow meter from the throttle body you really can't get any loose debris out of the bottom the airflow metering unit, and better to leave the debris there than forcing it thru the engine. If there's some oil blow-by residue in the funnels of the air flow metering flaps or on the moving airflow plates, OK to spray and/or wipe that out.

That would probably work. Many just use a screw with a larger head with a flat underside (clean up the top of the boss a little and use a flat gasket) and seal it at the top of the boss. This price is ridiculous, and on the wrong continent for you, but a Volvo or Porsche or Mercedes-Benz source (which all used Bosch CIS) might have a similar solution or even the stock screw:

https://www.pelicanparts.com/More_Info/91111093401.htm?pn=911-110-934-01-OEM

 

Unplugging the O2 sensors also forces the system to run in open-loop mode (just as putting ground on pin 16 of the injection ECU should do during cold-running) so you will have to retest cold-running with the O2 sensors plugged in. But nothing wrong at all with being in better adjustment. Even if the problem is resolved, you'll never know if it was the adjustment or if just getting in and rooting around to investigate wiggled things enough to fix a bad connection . If you still have cold-running trouble with the O2 sensors plugged in, it shouldn't take rocket science to determine why pin 16 isn't going ground when the water thermoswitch is closed and relay 27 actuated (if that's still the trouble).

 

Very good! Glad that you got it sorted out. It's sort of sad how many TR get terribly molested, and a bit surprising that they still run .

 

So the real problem was no tamper proof plug?

 

Plip
I have the exact same problem with my 87 TR. I have gone thru everything with all of the advice from the experts here. No success. I do want to give you a bit of caution. The mixture down in the cavity must be accessed by removing the cover screw (yours are missing) and then removing another screw. That screw is a small blade and one of mine is rounded out so I can not remove it. A hand impact wrench can not be used because of clearance with the deck lid. I could take the engine out (no, no no) or remove the lid, which I have not done because I do not think the cold idle is that bad. I did increase the air amount on both side and the problem is less now.
So, be careful with that screw and I will be watching with great interest on your progress.
Burnell
Ps- could it be an air leak somewhere in the intake system that closes up when the metal gets warm?