| Author | Message | ||
| JRV (Jrvall)
Intermediate Member Username: Jrvall Post Number: 1934 Registered: 11-2002 |
Jumpiness is rather normal with an accurate reading machine. The difference between .75 & .85 is irrelivant to running condition. >>225-235ppm HC's << Based on the above 200 HC's #, I'd speculate on plugs nearing the end of their best efficiency (or wrong heat range or type), plug wires nearing the end of their effective life span or dirty injectors, or a combination of all or some of the above. There are other possibilities, however I consider them to be less likely, because even a very slight miss (certianly unnoticible by ear) will push the HC's from 150 to 225. Keep in mind that 150 ppm HC is considered 'perfect' and you appear to be on the threshold of getting there. Personally, I'd forget about duty cycle and concentrate on finding a wire out of spec or an injector with a bad spray pattern or a drip. HTH's | ||
| James Christian (Jimc)
Junior Member Username: Jimc Post Number: 57 Registered: 7-2002 |
JRV:DGS: OK. Here are the measurements made Sunday. After much time fiddling around, I ended up back in the same place as before- but using the analyzer instead of duty cycle to find the set point. Using a FLUKE 87 (True RMS), here are the voltages on the frequency valve: Before Adjustment (From Cold)And After Adjustment (From Cold): 6.85 VAC OL (At start-up and for the first few minutes) 6.60-6.88 VAC CL (After About 5-Min once regulation begins) New EGA Measurements: O-L, Hot, O2 Sensor Unplugged, Idle: .75% - .80% CO (A bit jumpy here, don't know if it's the car or analyzer) 225ppm on HC's Plugging in the O2 sensor yielded a bit higher values: .85% CO 225-235ppm HC's This was rather frustrating because the values were not as steady as I would like. I'm not sure if this was the car or the analyzer, but I did us e the appropriate "sniffer" tube on the front of the cat and used a silicon tube to attach to the analyzer so I feel we were getting a good reading. Unless you guys can see something here that I'm not, I guess the next thing to do is take a look at the fuel pressure throughout the warm-up phase of operation. I can monitor ambient air temp and pressure very accurately and report. I've also started to think about valves as well. It's been about 5 years since the last valve adjust. Since I'm approaching the 30k, it might be time. I also may have the opportunity to make some measurments on another 328 to see how close we are. JC | ||
| James Christian (Jimc)
Junior Member Username: Jimc Post Number: 56 Registered: 7-2002 |
"FV disconnected, open loop..." Just making sure you guys were paying attention  Oops... I meant when the 02 sensor was disconnected, forcing the FV to operate at 50% duty cycle. Well, enough said. I guess it's my turn to go out and take care of business. I'll make the adjustment and let you guys know what happends. This has been a very beneficial exchange. Thank You! | ||
| DGS (Dgs)
Junior Member Username: Dgs Post Number: 166 Registered: 5-2003 |
>> don't get ahead of them. ;-) << Hey, that was the simplified version of the message. When an engineer says, "it's simple, really", RUN. Me theory. You practice. Gotcha.  Speaking of theory: It's the one second or so period of the O2 sensor's oscillation that limits the response time of an add-on lambda loop like the 328's. A digital EFI system will sample parts of the O2 sensor curve and apply that to the map for every injection cycle, but the CIS-lambda system can only use the FV to vary one of four pressures which determine the fuel flow rate to the CIS injectors. When the O2 sensor gets old, you can feel the car surging minutely at a constant mid throttle as the mixture varies with an O2 sensor whose cycle period has gone well over 1 sec. | ||
| JRV (Jrvall)
Intermediate Member Username: Jrvall Post Number: 1931 Registered: 11-2002 |
>> The lambda loop is only one part of the system. If your system pressure or control pressure regulator or the springs in the fuel distributor are off, you probably won't be able to get good results setting the base mixture. << DGS, don't get ahead of them. ;-) Just let everyone absorb and try the first portions of the correct procedures. Then if issues are still present we can take it from there. Jim, And here is the anwser to the question. The speed of the electrics. the reason you were able to get it close using Duty cycle is because the O2 Sen reads exhaust gas (somewhat) and compensates (tries too) for corrections, the reason you couldn't get it spot on is because the speed of electric signals was too slow back then, (Thousants of a Second)to work in real time until the advent of the Daul O2 Sen System, Motronic 5.2 (Billionths of a Sec). >>So, as long as one is close with the mixture, adjusting the mixture screw slightly one way or the other (when closed loop) will have little to no effect on emissions at idle.<< Well sort of but not really. Yes the O2 sensor will try to compensate, but it's electrics are not fast enough to work in real time as explained above. Also the main setting is the Static Setting at the FD Screw, with the O2 Sensor a mere Helper Device, not an actual adjuster/compensator. HTH's ps: like DGS & I have stated, leave the FV alone, as it affects Fuel Pressure (somewhat)which affects System Pressure Which Affects Overall Mixture Across the entire running range. The system is Hydraulic with Electrical enhancements. Please keep in mind that any changes in FP affect Distrib Plate Operation, which in turn affects mixture at corresponding rpms. | ||
| DGS (Dgs)
Junior Member Username: Dgs Post Number: 164 Registered: 5-2003 |
You don't disconnect the freq valve for the open loop test. The frequency valve sets one part of the pressure controls, the mixture screw, control pressure regulator, and AFM set another part. They work simultaneously, so you have to open the lambda loop to set the other part at a known level. The original CIS was a "hydraulically controlled" system, using varying fuel pressures as the control "circuits". CIS with lambda is a hybrid (e.g. bastardized) mix of CIS and electrical controls. Disconnecting the O2 connector outboard of the AFM will force the Bosch lambda system into open loop. You then set the base mixture. With the O2 connector plugged in, the lambda system varies the frequency valve duty cycle based on sensor parameters. From service bulletin 80-29: Water temp < 57degC, Oil temp < 25degC : Voltage across frequency valve: 7.4vac +/- .2 :: 60% duty cycle (Open loop) (F105C040 Engines) Water temp <57degc,> 25degC : Voltage across frequency valve: 6.4vac +/- .2 :: 50% duty cycle (Open loop) Water temp > 57degC, Oil temp > 25degC, Cat temp > 300degC : 5.3-6.3vac +/- 0.15 :: Closed loop (variable duty cycle) WOT enrichment (from throttle switch) : 7.4vac (60% duty cycle) Since the oil temp and WOT switches are parallel, cold and WOT looks the same to the lambda system. The lambda loop is only one part of the system. If your system pressure or control pressure regulator or the springs in the fuel distributor are off, you probably won't be able to get good results setting the base mixture. Now that I think about it, mixing fuel pressure and electrical controls in the same injection system sounds ... almost as crazy as using gasoline to cool the electric fuel pump motor. | ||
| James Christian (Jimc)
Junior Member Username: Jimc Post Number: 55 Registered: 7-2002 |
JRV, DGS: Based on what I know of feedback systems, it sounds like once the mixture is set (FV disconnected, open loop), the fuel system adjusts the frequency valve around the point it chooses- based on the signal coming back from the 02 sensor (once the FV is reconnected.) So, as long as one is close with the mixture, adjusting the mixture screw slightly one way or the other (when closed loop) will have little to no effect on emissions at idle. But when one heads down hill and off the throttle, the loop is not working becuase it's all the way lean. The frequency valve is pretty much trying to shut the fuel down and then the mixture one dials into the mixture screw becomes more critical as this setting sets the relationship between the air coming in and the fuel opening in the fuel distributer when the frequency valve can no longer keep up. So, it's critical to have the engine running well in order for the fuel injection management system to keep the car where it's supposed to be. Is this close or just a misunderstanding of the system? | ||
| JRV (Jrvall)
Intermediate Member Username: Jrvall Post Number: 1928 Registered: 11-2002 |
Jim, Depending on the car and condition it should take between 1-2 hrs to complete the full range of items discussed here. The exhuast grill under the car must be removed to access the before cat test port, and the drilling of the FD plug if necc. The rest is testing/adjusting, double checking. The Smell is the unburned HC's. Once the HC is at or below 150ppm (pre-cat) and below 50ppm at the tail pipes the smell will deminish considerably. | ||
| Jim E (Jimpo1)
Intermediate Member Username: Jimpo1 Post Number: 2062 Registered: 7-2001 |
Ok, how much should I pay somebody to do the below for me? I'm NOT a mechanic, I have the same sulphur smell problem, and I have no testing equipment. | ||
| James Christian (Jimc)
Junior Member Username: Jimc Post Number: 54 Registered: 7-2002 |
JRV: Thank You. I'll try it and follow up later! Jim Christian | ||
| JRV (Jrvall)
Intermediate Member Username: Jrvall Post Number: 1926 Registered: 11-2002 |
>>1. Air bypass (idle adj.) all the way in 2. Adjusted idle stop screw for 850rpm at idle 3. Opened air bypass (idle adj.)to set idle to spec. 1000 rpm<< 1-3 is correct. The next step should be, disconnecting the 02sensor for a pre-cat sniff & set, to .80% Once the static mixture setting is dialed in at the FD screw then the O2 sensor can be plugged back in to see if it's duty cycle is in spec and that it maintains the static setting within the range. 4. O2 sensor sweep reading: 80-800mV sweep (closed loop idle) 5. 02 sensor reading(open loop)idle: 800mV 6. Frequency valve at 50%, open loop, idle 7. Frequency valve sweep (closed loop) 46-54%, at idle. (I used a portable scope for this measurment and adjusted the mixture to get this) | ||
| James Christian (Jimc)
Junior Member Username: Jimc Post Number: 53 Registered: 7-2002 |
JRV: I read your posts a few times with much interest. You are taking a different approach to this than I. I was setting up the duty cycles per the manufacturing recommendations regardless of emissions. It was tricky, but I know I hit it dead on. It would be very interesting to do as you suggest, and find that I simply needed to lean the mixture slightly and have the problem go away! After you review my last message and comment, I will take the car to my friends shop, hook up the analyzer, and try what you have suggested. I don't want to hurt the engine. Although this is a "toy" (read hobby) for me, I don't want to break it. Based on the information I provided, are we close to where we need to be? Thank You! | ||
| James Christian (Jimc)
Junior Member Username: Jimc Post Number: 52 Registered: 7-2002 |
OK! Let's try this route. Thanks for the patience here- I'm not a mechanic. I replaced the injectors and O2 sensor myself, as I know this is recommended if the original injectors are still in place. What you see below I learned from the Bosch book, communication with others, and a friend has an analyzer. Yes, I know where the mystery plug is and have removed it to access the mixture screw. The hole is securely plugged with a small rubber stopper with handle at the moment. It is air-tight. 1. Air bypass (idle adj.) all the way in 2. Adjusted idle stop screw for 850rpm at idle 3. Opened air bypass (idle adj.)to set idle to spec. 1000 rpm 4. O2 sensor sweep reading: 80-800mV sweep (closed loop idle) 5. 02 sensor reading(open loop)idle: 800mV 6. Frequency valve at 50%, open loop, idle 7. Frequency valve sweep (closed loop) 46-54%, at idle. (I used a portable scope for this measurment and adjusted the mixture to get this) Not too sure about these, but here goes... Pre-cat cold: CO= 0.70 CO2= 14.8 HC= 205ppm Pre-cat hot: CO= 0.83 CO2= 13.9 HC= 230ppm Post-cat hot (Idle) CO= 0.07 CO2= 11.1 HC= 57ppm Post-cat hot (2500 RPM) CO= 0.01 CO2= 14.1 HC= 31ppm The car runs great. I've driven two others, and when I "blip" the throttle, it seems to rev-up a tad bit slower than the others... But it might be my imagination. Also, if I blip the throttle too hard, the engine sometimes stumbles (just a bit) before reving up and I hear some quick chatter from a relay under the dash on the passanger side. Just a sputter, click-click, and up it goes. but only when "blipping" the throttle and only sometimes. Does this help? | ||
| JRV (Jrvall)
Intermediate Member Username: Jrvall Post Number: 1917 Registered: 11-2002 |
<<shouldn't>> Uhhh..I was just testing to see if everyone was paying attention...;-) 0.80% Yes if the factory plug is still in it does need to be removed by drilling a small and inserting a small sheet metal screw to pull it out or by prying. To remove the Factory Sealing Cap the Air Cleaner must come off to allow working room. For those conscientious and industrious enough to have gotten to the testing stage, also do a steady state 2500 rpm test/check of CO/HC while you're at it. The CO will likely be .5% ...with 1%..about the max you should see. HC should be in the 120-150 PPM range . Anything over 200 is not that good. | ||
| DGS (Dgs)
Junior Member Username: Dgs Post Number: 160 Registered: 5-2003 |
Ur, two things, JRV: Shouldn't that be 0.8% rather than .08%? I think that plug, on US model 328s, is a pound-in cap that must be pried or drilled out -- unless its been converted previously. (Presumably, Jimc has gotten the plug out if he's been fiddling with the mixture screw.) | ||
| JRV (Jrvall)
Intermediate Member Username: Jrvall Post Number: 1915 Registered: 11-2002 |
>>Some comments: I'm convinced that there is no way to adjust the mixture on this engine.<< Oh Contraire. The mixture 'must' be periodically adjusted to compensate for wear, dirt, age, condition etc. The Factory notion that these Fuel Distributors never needed adjustment was total baloney. Beside the Fuel Distrib Inlet Plate is a plug that must be removed then access to the mixture screw is obtained. The procedure is: 1- Unplug Exhaust Gas Test Port Before the Cat. 2- Test CO & HC with O2 Sensor Connected 3- Test CO & HC with O2 Sensor Disconnected 4-Unscrew Mixture Protection Screw, insert 3mm Allen and turn clockwise to Richen, Counter Clockwise to Lean mixture, set to .08% CO, plug hole, recheck, blip throttle a few times to stabilize...recheck and reset till CO stays right at .08% CO at idle speed (adjustment hole plugged) then when idle is exactly 1000 & CO .08%, replug O2 Sensor and recheck. The mechanical adjustment is supposed to keep CO in range with minor help from the O2 sensor. The O2 sensor is not supposed to do all the work. HTH's | ||
| j scott leonard (Jscott)
Member Username: Jscott Post Number: 503 Registered: 1-2002 |
So that is what I smelled last year! May cost you 5 points this year. | ||
| DGS (Dgs)
Junior Member Username: Dgs Post Number: 159 Registered: 5-2003 |
The mixture setscrew should be adjusted with the Lambda loop unplugged. There's a plug outboard of the air filter (under the lip there) that, when unplugged, should disable Lambda. You then set the "open loop" mixture to the correct CO reading, then plug the connector back in and let the Lambda loop "tweak" the mixture for the driving conditions. But if the base mixture is off, there's only so much the Lambda loop can do. 50% duty cycle at the freq valve may indicate that the Lambda computer is in "open loop". The loop only kicks in when the coolent, oil, and cat are warmed up, and the throttle is below WOT. If the coolent or oil temp sensors, throttle microswitch, or cat thermocouple aren't working right, you'll never leave open loop, so the 50% on the freq valve might not translate to the correct CO reading at the exhaust. You need to check it with a gas tester. If you have a 'scope, you might look at the output of the Lambda sensor. This should oscillate a bit over about 1 second. If that period goes much longer than 1 sec, your O2 sensor might need replacing. | ||
| TomD (Tifosi)
Advanced Member Username: Tifosi Post Number: 4084 Registered: 9-2001 |
david is right, have that done and I bet JRV can help | ||
| David Feinberg (Fastradio2)
Junior Member Username: Fastradio2 Post Number: 248 Registered: 4-2002 |
Jim, I'm sure other will chime in on your concerns...but here's my 2 cents worth. Simply, for the moment...stop throwing money at your car, and have the CO and HC levels checked by someone with an exhaust gas analyzer. Excessive sulfur smell tends to indicate an overly rich condition. This can be cause by several conditions, such as, but not limited to: Base mixture level set incorrectly... Poor injector spray patterns... This is a fairly simple electro-echical system...but, without knowing specifically what's coming out the exhaust, you're just shooting in the dark... Regards, David | ||
| James Christian (Jimc)
New member Username: Jimc Post Number: 50 Registered: 7-2002 |
Hi Everyone, Here is the situation, which I've been dealing with for 4-years now. Not a problem, simply a nuisance. The car is a cherry '89 GTS. 28K miles, excellent running condition. I've owned the car since 16k miles. We have quite a hill between home and town. At the bottom of the hill, there is a stop sign. When my exhaust catches up to me, it's sulfur in odor. So, I replaced the cat with another and same problem. I've monitored the frequency valve with both scopes and DVM's, and the duty cycle is 50% at idle. The engine leans downhill and richens under full throttle acceleration- as is normal. Everything else works fine, and sorry folks, no FCCC (Ferrari Carpet Cruise Control!) I've replaced plugs, plug wires, fast idle valve, cold start injector, oxygen sensor, etc. I even tried replacing the windshield wipers- but that didn't help either! Go figure. I was certain that would work. I even went so far as purchasing a book on Bosch mechanical fuel injection systems, and all appears to be working just fine. Some comments: I'm convinced that there is no way to adjust the mixture on this engine. The computer will set the duty cycle to achieve stoichometric mixture (through the frequency valve) regardless of mixture screw setting. Of course, the danger is that the computer can only control the duty cycle to a certain extent so if one where to adjust the mixture screw too far one way or the other, the computer would not be able to correct for a too lean or two rich mixture. So Team, I challenge you to help figure this one out. The only way I've been able to resolve the issue is to run without a cat. I know this is better, but it's simply the thought of knowing something is up and not being able to tackle it! Will the computer connection tell us anything about timing that we can't find out otherwise? Has anyone heard of a Jovatron (sp?) and is the device used to monitor the timing of the engine through this port. Ugh. Please help. Jim |