Mod part Two

To facilitate ease of access, it's fairly easy to drill a couple holes in the plastic floor board as shown below.

Mount the ECU on the floor board, and then look at the placement of the test points from all four sides, as straight on as possible, and make a mark.

Remove the ECU, and using a straight edge draw a line between each mark. Use a post it note with Xs to more easily check the alignment with the test points.

Drill a ⅝" hole at each X.

Now you can access the test points with out having to dismount the entire ECU from the floor board - all you need to do is pull back the carpet and viola.

To keep dust out, I placed this plastic door over the holes. The plastic is just cur from some store blister packaging, and it is simply taped into place on the top edge, making a hinge.

Cheers

Andy

 

.



Dude was a smart mofo

 

Dude was a smart mofo

was???

 

Thanks Andy. Great instructions. I have an 81 380 SL I just rebuilt the top end on (did not pull heads) and need to dial it in. My volt meter with duty cycle is on order.

 

I remember reading this and being impressed with the work that was done. I later replaced my WUR with an electronic WUR that could be programed for better performance and have a FAR wideband EGT and gauge. The problem with the frequency valve is it is working against the K jet system. It is constantly trying to keep the fuel air ratio mixture constant while vacuum to the WUR is enriching or leaning it based of vacuum. When you accelerate you want the fuel system to enrich the mixture for best performance but the Lambda system is trying to lean it out. It does go to constant 60/40 at idle and full throttle which is helpful. I just have my frequency valve set at a constant 60/40 so it no longer interferes and the car runs more like a pre-80 model. I have the fuel air ratio at about 13.8 while at idle or constant speed.
These cars run better at 13.8 - 14 instead of 14.7 (stoichiometric) which is the only correct reading from a narrow band EGT that controls the Lambda system.

 

In 1985 the pressure step switch was added to the lambda circuit, this puts the lambda system into the warmup enrichment mode (same as oil temp <16C) when the vacuum drops during accelleration. I think I have seen this was sometimes retrofitted to earlier cars, perhaps to help solve the problem you describe.

 

Andy,
You saved my bacon today. Thanks to your mod I was able to get proper readings. Even with the scope on the x11 the picture was so convoluted that I couldn’t decide on direction. Only after connecting to the clean signal directly on the board thing started to click.
Thank you again.
Mark

 

Hi Guys, I did the mod to my ECU as well. But unfortunately the readings on my DMM show 0% all the time. I know from the exhaust smell that my 380SL (1982) is running to rich. After the car had been adjusted by CO measurement at the exhaust, I had to enrich the mixture again because the car stalled sometimes on red lights at idle… so now I wanted to adjust it properly by measuring the duty cycle. Do you have any tips for me? Do I have a broken ECU unit here? I tried leaning the mixture slowly. I thought maybe it is running so rich the ecu is sending maximum lean signal to the valve (0%). But still I read 0% duty cycle. I double checked that I used the right PINS 16 and 17 for the mod. I also checked every other parameter like compression, ignition timing, all the relevant system pressures of the K-JET. The car is running fine with the rich mixture. It’s just a bit lazy on revving and it’s a bit smelly… so I am pretty sure it has to be lambda that’s causing the stall issue…. Looking forward to hear from you

 

If everything else is working, you have three parameters (on an '82) that govern fuel mixture when at operating temperature: Lambda (O2) sensor, the ECU and the frequency valve. One of those 3 is not doing it's job. I have a similar problem (richness at idle / lean at 2,000 rpm) and have had a marked improvement by grounding the O2 sensor with a ground strap to the body of the car. This has greatly reduced the richness at idle and improved acceleration because I can dial in more fuel without it stinking up the air. My next course of action is a new Oxygen Sensor because the 10,000mi one I have as probably been encrusted by rich combustion....

Couple more questions: can you hear the frequency valve buzzing away? Why did you mod the ECU? ...Cheap Chinese Oscilloscopes do their job just fine reading pins 2-3 on the diagnostic plug.

 

I soldered a wire to pin 17 and connected to the positive on my DMM set to duty cycle and I get around 50% at idle. When I reverse polarity as instructed in the directions above the meter reads OL.

 

There's been some discussion here about the removal of catalytic converters that results in the O2 sensor not getting hot enough to signal the ECU to perform correctly. Has anyone wired in a modern 4 wire (self-heating plus signal ground) O2 sensor? How do euro models do it?

 

On my 82 380 I soldered wires to pins 2, 16, and 17 so I can watch O2 sensor and FV duty cycle at the same time while I'm driving. At idle the O2 reads around .5V and the FV is at 50% but while driving the O2 sensor reads .88V and the FV is at 88% which tells me the FV is trying to lean it out but the O2 sensor says it's not getting the job done. My next move is to replace the distributor cap and rotor since both are in bad shape. I'm thinking I'm getting incomplete combustion. I hooked the sensor of my timing light to each of the spark plug wires and I get a very erratic strobe.

 

On my 82 380 I soldered wires to pins 2, 16, and 17 so I can watch O2 sensor and FV duty cycle at the same time while I'm driving. At idle the O2 reads around .5V and the FV is at 50% but while driving the O2 sensor reads .88V and the FV is at 88% which tells me the FV is trying to lean it out but the O2 sensor says it's not getting the job done. My next move is to replace the distribu tor cap and rotor since both are in bad shape. I'm thinking I'm getting incomplete combustion. I hooked the sensor of my timing light to each of the spark plug wires and I get a very erratic strobe.

When the FV is on (open) it makes the mixture richer, off (closed) it makes it leaner. On the scope you can see when it is on or off. I assume the duty cycle measurement is to be interpreted as higher duty as being on more, making the mixture richer.

 

FV can only do so much. It can only add more or lessen the amount it is adding. If the mixture from the Fuel Distributor and pressure control from WUR are not already producing a lean mixture it's just going to make it worse.

 

I've tried leaning the mixure out more but it gets to the point where the idle is rough and it's hard to start. Wouldn't poor combustion cause this symptom? I lean it to the point where the O2 sensor is happy but with incomplete combustion it runs rough. I have not tested fuel pressures yet. I've read that best practice is to disable the frequency valve when setting AFM. Get it best you can and when the FV is reenabled it should fine tune it for you (assuming that part of the system is working right). I would use O2 sensor voltage while setting the AFM with the FV disabled and error on the side of lean. I also need to try adding a grounding strap to the O2 sesnor to see if that changes things.

 

I'm getting 50% at idle with the o2 sensor around .5V but since the controller has a bias voltage of .5 I'm thinking it ignores the O2 voltage at idle and just sets it at 50% (as long as it sees the throttle valve switch closed or oil temp switch closed, which is another thing I'm monitoring (pin 6) since I've checked it before and sometimes it only closes after multiple cycles of the throttle). I do know with the K-E the WUR and FV are eliminated. Volkswagon and Porsche used the same K-Jetronic and they fed pin 17 to the diagnostic plug so you can read it there with a DMM. Mercedes opted to require the Bosch tester to perform that function at the diagnostic plug. BTW, I bought a NOS Lamda controller and plugged it in and the car runs the same. I haven't opened it up to probe pins yet. I'm hopeing I can use my current controller with all the wires hanging off it to troubleshoot and then swap in the new controller when it's right. May not be able to count on them performing the same, though.

 

The system is active at idle. It will go into a 50% fixed cycle if the readings from the O2 sensor are outside of a predetermined range (such as immediately after cold startup), or if the sensor is disconnected. The oil temp switch when closed puts it into a fixed 60/40 (60 on, 40 off) cycle.

Here you can see it from my meter, idle rpm.

 

If it was mine and I had to have the FV work to pass inspection. I would disable the FV then look at the voltage from the O2 sensor. If was still rich I would start looking elsewhere to fix the over rich problem since it would be impossible to fix with the FV. If it was lean with it might be able to help but since the O2 sensor is only correct at one point you don't know how lean or rich it is.

 

Only thing I would add is Full throttle also puts it into fixed 60/40. Where I have mine permanently set. FV stays functional but removes it from interfering with the mixture. I understand needing this if you state requires inspection. It just fights the enrichment from the original design for performance except at full throttle.

 

I am not required to get a state inspection of any kind. My ignition parts ar edue in next week so I'm not going to do anything further until I get those installed. Then I will adjust the AFM with the FV disabled while reading the O2 sensor. Will report back with the results. Appreciate it!

 

Installed new cap and rotor and made no difference. Looks like my problem is with the fuel distributor. Took the car out for a good drive and when I got back I pulled all the spark plugs. Numbers 1 and 5 were fouled with soot and the rest were white. Pulled the injectors at 5 and 6 to compare and 5 is spraying out 3-4 times as much as 6. Swapped the injectors (5=6) and port 5 is still way over producing vs 6. And this is without pressing down on the air sensor plate. I was expecting no spray without pressing down on the air sensor. I read fuel pressures of 4.7 bar at 5 and 4.5 bar at 6. Shutting off control pressure did not seem to affect it nor did adjusting the air mix screw (180 one way and then the other). Do I need a new distributor?

 

I think you are probably correct. May need to rebuild or replace Fuel Distributor. Would check the control pressure from the WUR also since it may also need attention. Check the return line also to determine if it does not have blockage. Is this car new to you or was it running ok previously without the possiblitly of someone messing with the mixture control ?

 

@jmartin919

You said "I read fuel pressures of 4.7 bar at 5 and 4.5 bar at 6"

How did you obtain those pressures? Did you connect the fuel pressure tester between FD port #5 and injector #5? And between FD port #6 and injector #6?

It is my undertanding injectors are supposed start opening (around) 3.2 bar.

 

I got the car last November and it wasn't running very well. First indication of a problem was the large piece of carbon stuck on the side of the O2 sensor. I replaced that and then started looking at Lamda system. O2 sensor said rich condition and FV was trying to lean it out. All plugs would be sooty quickly after idling or very short, slow drive. On a longer drive plugs 1 and 5 were sooty and the rest looked ok. After feeling good about ignition I started looking at port 5 on the distributor and comparing that to port 6 and here we are. Oh, and I shimed the WUR to try to lean it up a bit and the control pressure went from 3.1 bar to now 4 bar. Return line to the fuel tank is clear and I measure 750ml of fuel returning in 30 seconds ( a little low and I contribute that to my accumulator not working). As far as reading fuel pressures at the distributor ports, I disconnected the injector line at the distributor, screwed in the fitting and connected the pressure gauge. Yes on the 3.3 opening pressure, hence the reason why I'm getting fuel out of the injectors without even pressing on the air sensor plate. Backing off the air sensor screw ccw 180 makes no difference that I can see. #5 injector puts out 3 tbsp of fuel in 30 seconds and #6 1 tsp. The pressures are different by .2 bar but is that enough for that big of a difference in fuel delivery? I feel like the piston is not seating like it should to close off fuel flow, or at least get the pressure below 3.2 bar. But then why would all the injectors be getting the same amount? If I do go for a rebuilt distributor any recommendations on source?

 

Oh, and I shimed the WUR to try to lean it up a bit and the control pressure went from 3.1 bar to now 4 bar.

I would start by bringing the WUR back where it needs to be as in the K-Jet with Lambda systems in the US that WUR does nothing at all to affect mixture once the car is up to temperature. Then look at rebuilding and balancing the fuel distributor as it does sound like you have an unbalance in delivery. Also is your O2 sensor properly grounded? I recently added a grounding strap to the sensor body and that greatly improved the ECU's ability to meter fuel through the frequency valve.

 

O2 sensor is well grounded. Clamping jumper cables to the body of the sensor and then to engine ground did not affect the output as I went back and forth with it. I do intend to put the WUR back the way it was once I get the distributor sorted out.

 

Most would suggest CISFlowtech also Kjet Specialist in Australia sell rebuilt fuel distributors. Colin Duncan with Kjet Specialist seems to be very responsive to email inquires. Also note their prices are in Australian $ so much better than what appears on the website. I rebuilt mine with a kit about 6 or so years ago so I don't have any experience with buying a rebuilt FD.

 

Opened up my distributor. My problems have primarily been with cylinders 5 and 1 as far as sooty plugs. Just so happens the port for cylinder 5 sits directly over the main fuel inlet and for cylinder 1 it's right next to it. By the looks (and feel) of my diaphram I'm thinking I'm getting leakage into those two ports. The diaphram reminds me of the little diaphrams you find in 2 cycle machines. They get brittle and won't pump or meter properly - and ethanol speeds up the demise. This diaphram is wrinkled and kind of stiff, especially at each of the 8 pockets that it has taken the shape of. If I press on one of those areas it is stiff as a brick. Anyway, think I'll be ordering a kit. Tempted to replace just the diaphram to prove or disprove my theory. Everything else looks excellent to me.

 

Yea you can always just open it up again and do more cleaning and inspecting. Just suggest you leave the fuel injection lines off and push gently on the plate and see if all the ports rise together. You can also see if you mis-aligned one of the valves and you get a fountain. I think I would go ahead and vat it, put new o'rings and inspect the piston openings.

 

I think it was K Jet Specialists where I read about that. Hook all the lines up except injector lines, lean out the air sensor screw to make sure fuel doesn't come out the ports, richen it until it does, then back off 1/4 turn I think. That's the base setting. Put it all together and tweak it when the engine gets up to temp. Can't find the doc now. Correct me if I'm wrong.

 

Yup, that is real close to the CIS Flowtech method also.