Protection Relay Conversion and Fuel Distributor Starting Alignment

(Starting a thread from a PM because I need to post an attachment)

Steve:

I received the newer relay and moved the wiring around in the socket and the FV is now buzzing away as it is supposed to. Many thanks.

I am afraid I need to trouble you further. As I mentioned earlier, my car was heavily "compensated" by the previous owner/mechanic in terms of mixture and idle adjustment. I am having a rough time finding a starting point in terms of the mixture and idle setting to keep the car running so I can get a good reading and start tweaking the mixture/idle settings.

I am also not sure what is going on with my duty-cycle meter. IT seems to indicate between 12-14%. Not sure if it the inverse since I beleive the FV valve is negative slope.

Do you know how to set a good starting point to keep the car running and to begin the mixture tuning ?

Thx, again

Oleh

 

Oleh -- You should get a copy of the Mondial8/QV WSM 281/83 and read Chapter D. I can't recall where, but others have put this Mondial WSM up on websites as a .pdf file (if someone knows where Oleh can download this Mondial WSM file please post where).

This jpeg describes the "rough" starting alignement of the mixture screw given on page D29:
Image Unavailable, Please Login

Can't comment on your duty cycle measurement seeming strange.

 

Yes (but only if you trust your duty cycle measurement )

If the duty cycle does seem "wrong", check if the oil temp switch and the water temp switch are open (they both should be when warm).

 

To the AG (light blue/yellow) wire of the frequency valve (while it is plugged in and working).

Oleh -- You might try this too. If you are now measuring the "dwell" between the PN (beige/black) wire and the AG (light blue/yellow) wire of the FV, try measuring between the AG wire (the "+" lead of the dwell meter) and ground (the "-" lead of the dwell meter). Mathematically, they should give the same result, but perhaps your meter needs the reference ("-") signal to be ground (rather than the +12V of the PN wire) -- just a thought...

 

Didn't you use the simple test with a light bulb and the battery that Paul recommended here to determine cathode vs anode:

http://ferrarichat.com/forum/showpost.php?p=138962539&postcount=6

If it was in backwards, I think that the biggest flaw it would cause is to prevent the cold start injector from firing during cold cranking. You could just unplug it during warm idle to see if it makes any difference in the duty cycle measurement (but I'd be very surprised if it did).

 

Obviously, the mixture is not quite right because it should run OKish even with the O2 sensor unplugged (although it is a clear sign that the FV duty cycle is changing when you unplug the O2 sensor).

One thing you might try is unplugging the safety switch (this way the FV will keep running even if the engine stops running) and then measure the FV duty cycle when the engine is running with the O2 sensor plugged in, and then unplug the O2 sensor and see if the FV duty cycle changes (it should if the engine dies). If your measurement detects no change in the FV duty cycle when the engine dies, then I still wouldn't trust your FV duty cycle measurements.

Alternatively, you could try arbitrarily adding a little richness (1/32~1/16 of a turn CW on the mixture screw) and see if that keeps the engine running when the O2 sensor is unplugged.

 

Try as described in post #7.

 

I can't accept this answer yet Just as you get ~6 VAC when measuring between the two wires on the FV when it is running, if you measure the AC voltage between the AG (light blue/yellow) wire of the FV and ground you should also get ~6 VAC when the FV is running (and a finite dwell value when using a dwell meter). Can you please measure the AC voltage between the AG wire and ground when the FV is running to confirm that that does measure ~6 VAC?

Unlikely IMO, as those two relays are used to control the air injection system and the cold start injector functions, but maybe you should ensure that your air injection system is really "off" as, if it isn't, (during warm-running) the O2 sensor will "tell" the injection ECU to (wrongly) add more fuel by increasing the FV duty cycle. Not sure the easiset way to do this on your model, but (see page 73 in your 307/84 OM) what you want to do is physically block the air line that runs from the airbox to the cut-off valve "A" and onto the check valves "B" -- i.e., you want to ensure that no air is allowed to enter the check valves (which is what the cut-off valve should be doing when it is "off").

 

I like that AC voltage measurement result much better than "not geting any signal", but am still doubtful of your corresponding dwell measurements. The US 328 technical documents (which also uses a K-Jet with Lambda system) gives the following relationship between dwell and AC voltage when measuring between the two terminals of the FV:

6.4 VAC = 50 % duty cycle

7.4 VAC = 60% duty cycle

7.6 VAC = 65% duty cycle

Mathematically, this makes sense for a rectangular wave switching between 0V and 12V, so I have to conclude either:

1. the waveform is flawed, and is not a rectangular wave shape (I guess you'd need an oscilloscope to confirm/deny), or

2. something is still wrong with your dwell meter device(s)/method -- are you sure that your dwell meter is reading directly in duty cycle? If you are on an 8 cyl scale (i.e., the device thinks you have an octagonal cam lobe shape), and the dwell meter is reading 22.5 degrees that would be roughly a 50% duty cycle.

Feel free to give a shout if you need something else, but I don't have any better idea right now.

 

Dwell meters designed to work specifically on a points based ignition system should be hooked up with with the "+" lead of the dwell meter on the AG wire and the "-" lead of the dwell meter on ground to be used in the exact same manner. Doesn't mean that you couldn't get a meaningful measurement using the two terminals on the the FV, but this might give the complemetary (open) reading rather than the closed reading (i.e., "dwell" is defined as the degrees of cam lobe rotation during which the points are closed).

It doesn't matter what cylinder scale you use if you are measuring dwell as long as you know how to translate from degrees to duty cycle -- i.e.,

on the 4 cyl scale: 45 degress = 50% duty cycle

on the 6 cyl scale: 30 deg = 50% duty cycle

on the 8 cyl scale: 22.5 deg = 50% duty cycle

Hopefully, you new meter will have a direct "duty cycle" output so you can avoid this confusing translation

 

Aaack! I hate it when we start to get near here

I still don't like that the engine behavior changes (it stalls) when you unplug the O2 sensor, but you don't detect any change at all in the duty cycle at the FV. AFAIK, the only way the injection ECU can change how the engine runs is by changing the duty cycle of the FV.

Have you tried measuring the voltage on the O2 sensor single wire when plugged in and the engine at warm idle? If the voltage on this wire is wandering between 0.1V DC (lean) and 0.9V DC (rich), that would show that the FV duty cycle must be changing.

Have you tried just adding some richness by tweaking the mixture screw to see if you can get it to stay running at warm idle with the O2 sensor unplugged?

Also, how is your cold-running immediately after cold start-up? -- does that idle OK?

I would suggest that you confirm/deny the health of the oil temperature switch before pursuing a different injection ECU. See fig 68 on page 77 of your 307/84 OM. This oil thermoswitch should be closed when cold (below 15 deg C) and open when warm (above 25 deg C). You can just manually unplug it to simulate an "open"/warm condition if it seem wrongly closed when warm.