Air Fuel Mixture Logic

Hi All:

I am in the process of fixing all the teething problems of bringing up a new engine for my Federal 82 Vanagon (2.0L type IV engine). I am still the process of trying to get the engine to idle well and set the air fuel mixture. I am looking for some suggestions or comments on my reasoning process.

Current situation:

- the FI system checks out and fuel pressure is to specification.

- valves were readjusted (after they pumped up) and compression is now

110-120 in all cylinders,

Problem: Can't get the car to idle well enough to set the timing using a timing light or the mixture using a O2 sensor.

- To start the engine I have the timing statically set at 7.5 BTDC.

- Idle mixture screw 8 (half turns) open and the CO mixture screw mostly closed. These were the settings from my old engine.

- The car will immediately start but will need the throttle to be held open to continue to run (at about 1500 rpm). If I do not hold the throttle open the engine will die. Vacuum at 1500 rpm is about 15 InHg.

By my reasoning, I am running too rich so I need to give the engine more air. I have tried opening both the idle screw all the way and leaning the mixture at the AFM. This improves thing slightly, but still requires more air (by opening the throttle). If I now vary the timing I can get the engine to idle (poorly), but the timing must be advanced around 16-20 to get this to happen. Vacuum decreases when I do this.

By opening the throttle I am also delivering more gas. So maybe it is not the mixture at all but something to do with the injector spray pattern. I have not checked this since the old engine was removed, but a year ago they gave me a nice conical pattern. They were stored dry as not to gum up.

So what does the community think the initial problem is? Any ideas on how to proceed?

Thanks for the help in advance.

Regards,

--Jeff Strickrott

82 Westy, South Florida
Reply to
Jeff Strickrott
Loading thread data ...

.............You're right about getting more gas with more throttle. I'd turn that idle mixture screw back in to about two turns out and see what happens. It's my experience that eight turns out is way way too lean. If you were too rich at idle, it would load up for a few seconds before it quits. If it's dieing immediately when you try to idle, that's an indication that it's not getting enough fuel, in my opinion.

Reply to
Tim Rogers

Is this still the L-Jetronic FI? If so, the most common problem along these lines is that the seals between the heads and the pushrod tubes have perished from the heat and are now leaking. These will leak oil out when the engine sits, and air in when the engine is running. If they are leaking air in it is VERY hard to get the idle mixture anywhere reasonable. This is because the L-Jet design sucks in crankcase vapors AFTER the AFM so this air doesn't get metered.

The pushrod tubes and seals can be removed and replaced with the engine in the van. The only parts you are likely to need to replace are the 8 head-end seals. The other end doesn't get nearly as hot and those don't give trouble.

The most common sign of this problem is oil drips under each head when the van is parked. If you have seen these, replace these seals unless you already have recently.

-

----------------------------------------------- Jim Adney snipped-for-privacy@vwtype3.org Madison, WI 53711 USA

-----------------------------------------------

Reply to
Jim Adney

Thanks all for your feedback so far.

To recap:

  1. The pressure regulator was connected to the air box (not the throttle body) per the FI diagrams and does regulate as required. If the engine had 15 InHg or more of vacuum at idle then the pressure regulator would deliver fuel at 29 psi, but as the best I can get is
13 InHg of vacuum, the manifold pressure is not enough to completely control the fuel pressure.

  1. All the extraneous hoses (aux air, deceleration and the case breather vent) where removed and blocked to see if they were the cause of the low vacuum. This was not the case as vacuum remained constant at 13 InHg. I did not think they would be the source as everything rubber is new and these components worked fine in the old engine. Plus, the new engine runs the same with the hoses blocked or connected. The air box and manifold connections passed the “WD40 test” and no large vacuum leaks were detected.

  2. Cylinder compression is 135 (engine warm) in all cylinders. Valve lash is set to 0.006” (even though the valves are hydraulic) to help work out any of the air that is in the lifters. Preload will be applied after no more noticeable compression on the pushrod is detected (see
    formatting link
    for procedure).

  1. In order for the engine to idle at 900 rpm, I need to have the timing advanced to 28 BTDC, the idle screw 12 (1/2 turns) open and the CO valve closed. I can run the engine at a higher rpm with these conditions changed (how I ran the engine during cam break-in at 2000 rpm), but I cannot get it to idle in any other situation.

My two questions are:

  1. Why is the timing so advanced?
  2. Why is the engine vacuum so low?

For background on the engine:

  1. It has new AMC head with new valves.
  2. New Mahle pistons and cylinders.
  3. New lifters and stock hydraulic grind cam.
  4. Every seal/gasket or anything else rubber has been replaced (including the seals on the throttle body).
  5. New O-rings on the PVC vale (so it does not leak).
  6. A good OEM oil pump, no scoring on the case or gears.
  7. No leaks at any seals/gaskets.
  8. New oil cooler.
  9. All my aux air valves are reconditioned (valve functions smoothly, no rust or dirt). All other FI components check out to specification. I have verified my cam/crank timing and it is in correct alignment. The distributor drive gear is where it should be. I am using the stock OEM vacuum/mechanical advance distributor. The injectors do not leak and have a “fairly ” conical spray pattern. I say fairly as I have never seen an ideal cone from an injector so far. I have two good (never modified, good resistors paths) AFM’s and I get the same response from both. And yes I have connected the plug wires to the right locations. I am running a Pertronix Ignition system, so the point gap is not an issue.

What I will do next:

  1. To solve the criticism of having the fuel pressure regulator providing too much gas I can just connect it to a vacuum source to get
29 psi and see how the engine performs at idle.

  1. I will perform a leak down test to see if anything unexpected is leaking (manifold, air box, heads, etc.). But other than these simple steps I am at a loss as to what to try next.

Can anyone explain why the timing must be so advanced? What would you try next?

TIA Regards,

--Jeff Strickrott

Reply to
Jeff Strickrott

I cant help you, but pop over to:

formatting link
and locate Ray Greenwood in one of the forums(either T 4rum or injection forum) he is very helpfull and is extremely knowledgeable regarding the Bosch FI systems...

J.

Reply to
BergRace

....................How did you do this?

.............If there really are no leaks in the induction system (with FI, this includes the case) and your compression readings are OK, cam timing is a likely culprit for low vacuum and compensating with more ignition advance would point toward 'late' cam timing.........maybe?

..................As for running too rich at idle, try disconnecting the plug on the cold start injector (temporarily) and see whether that makes any difference.

...............Did you build this engine? Is the pulley correct for it? I'd verify true top dead center for cyl. no.1 to be sure that your ignition timing is where you think it is.

Reply to
Tim Rogers

Thanks all so far for the suggestions.

I verified my crank/cam alignment by seeing when #3 exhaust valve just closes and #3 intake just starts to opens. I compared this to my TDC mark and the pulley mark/timing scale. I had determined TDC when the engine was out by using the interference method (wooden chopstick in the spark plug whole) measuring when #1 cylinder was at TDC. From my calibrated eyeball this is about 0.5-1 degree off of where I have marked TDC on the flywheel.

Compression was 135 psi warm last week when I tested this, and every gasket/hose is new. Know of a good way to test for leaks on the complete intake/exhaust system?

I have plugged up most of the extraneous hoses with no change in vacuum. So I would like to test the complete system. The Bosch L-Jetronic manual suggest looking for leaks with compressed air and soap. By sealing the exhaust and intake do you think it would be possible to either pull a vacuum or pressure the case?

I am worried about pressurizing the case as I don't want to force oil out the crankshaft seals.

No difference. I had tested all the injectors last year right before we decided to get the engine rebuilt. None leaked at that time. My test on two (#1 &#2) showed that they were still working fine. Is it likely that the others would dry out and start leaking?

No, I did not build this one. This is from Fast German. My attempt at rebuilding is still on my engine stand in process, but we ran out of time and needed a new engine soon. However, before putting this one in I went over it as much as one can without actually splitting the case and pulling the heads.

If anyone wants to hear my opinions on Fast German they can pmail me.

Regards,

--Jeff Strickrott

82 Westy, South Florida
Reply to
Jeff Strickrott

..............Don't pressurize the case. If you haven't already done this, disconnect the crankcase breather hose from the induction system. Leave it open but securely plug or block where it was connected (I've never worked on a type4 but it shouldn't matter with this test). Start it up and measure the vacuum to see whether it's any different. It should run perfectly OK since the crankcase fumes are only routed to the intake to reduce emissions. I'm not suggesting that you leave the breather hose detached indefinitely, just temporarily inorder to see if there's a problem. A crankcase leak or excessive blowby might be causing your low vacuum and would make it run lean (I'm still skeptical about it running too rich.......the system usually has to have some kind of sensor or ecu problem for that..........too lean is much more likely............the crankcase hose is downstream from the airflow meter and would cause problems if there was too much flow of air to the induction system).

Reply to
Tim Rogers

Hi All:

Thanks so far for all the advice and an update from the trenches on what I have done and where I am.

Timing: I wanted to be really sure about the cam timing, so I degreed the cam to see when the valves open and close. I have included an image at

formatting link
If I use 0.015for when the valves open I get: Intake Opens at 28 BTDC and closes 70ABDC, Exhaust opens 68 BBDC and closes at 18 ATDC. That gives durationbetween 266 and 278 with an overlap of 46 degrees. Remember this wasdone with the engine and the lifter in the car, though I picked up theoutside of the lifter body to avoid errors due to lifter compression,so it is probably not that accurate. Accurate enough I hope toindicate that the valve timing is close to where it should be; thatthe cam is not off one tooth. Vacuum: I took the induction system off and found that one of the injector seals was leaking (the new paint on the intake runners caused a bad seal), and the new air box/throttle body gasket did not seal as well as I though it should. Form my collection of new gaskets I found one that fit the best and used Blue RTV to make sure it did not leak. Before installation the complete induction system would hold a 15 InHg vacuum until the duct tape on the intake manifold leaked (5-6 minutes). During stages of reassembly I was able to pull and hold a vacuum (for 30-40 seconds). When completely installed with the intake boot plugged I could achieve (using the starter to turn the engine over):

· 15-17 InHg with aux air valve, deceleration valve and all vacuum lines connected but with PVC and brake booster plugged, · 13 InHg with PVC connected but brake booster plugged.

Ignition: I have two rebuilt distributors, new ignition coil and Pertronix electronic ignition. Plug wires are new. I do not detect and binding or other sticking with the distributors.

FI:

- Cold start valve works (does not leak),

- Aux air valve works,

- Wire harness and grounds in good shape (buzzed out from ECU connector),

- Cold air timer (real name?) works,

- Injectors: 1 leaks but is within the 2 drips per minute test, all spray conically and put out approximately the same amount of fuel. Now I have read that the spray pattern is supposed to be very fine (misty), but mine appears to not be. So injectors will be added to the things to get list.

- AFM passes the simple test in Bentleys.

- Both temperature sensors read within their respective ranges.

Valve Lash (history): After running the engine for 20 minutes (during the first part of break-in) and 5 miles, I adjusted the valve cold for 0.006” (per Boston Bob’s hydraulic valve procedure at

formatting link
#1 and #3 cylinders reading around 90 psi on compression prompted this. After the adjustment, compression read 110-120 psi cold and 135 psi warm. I figure that as the engine had sat for 1½ months before installation that the lifters had bled down and trapped air. After I run the engine more then I will adjust the preload on the valves.

Status: After all of this work the same symptoms exists:

- By ear the engine runs best with timing between 20-28 BTDC (measured with timing light),

- It is not possible to set timing anywhere near factor spec of 7.5 BTDC.

- I can achieve a bad loping idle (800-850 rpm) with the idle screw all the way open and CO screw all the way closed. Engine runs better with the throttle held slightly open.

- Looking at the plugs they are black with carbon after just a short period of running, indicating that the mixture is too rich or spark too weak. I am running the recommended Bosch plugs of W8CC.

When I contacted the engine builder and told them of my findings, they stated that the vacuum is normal (13 InHg at idle) and they think that my problem is with the AFM; that I need to adjust the spring tensions. I am hesitant to do this as my AFM's (I have one California and one federal version) still have the original silicone rubber sealing them shut. The federal version worked fine on the old engine until I dropped a valve seat. Both AFM's behave exactly the same when installed.

I still don't really understand why a rich mixture would account for the timing needing to be so far advanced? Could it be that the installed cam is not the correct profile for my FI system?

The builder could not tell me what the cam profile should when I called and the only information I have for the stock hydraulic grind is from CB performance (IO 26 BTDC, IC 54 ABDC, EO 60 BBDC, EC 20ATDC, duration 260 and overlap 46). My curves are similar and I hope that the differences are due to errors in measurements.

I have a third federal AFM that needs to be rebuilt (worn resistor tracks) and I will experiment with changing the spring pressure on this one and see if the engine performs any better.

I will get back with more status information on Sunday.

Regards,

--Jeff Strickrott

Reply to
Jeff Strickrott

MotorsForum website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.