Help with yogurt cup test please...

[rossiz]

been driving for about 300 miles on the new head/cam and have been trying to hold my judgment until things wear in, but she's just not at 100%.

what's been done:

• re-built head
  
• set valve lash at install, then hot adjust after an hour or so of run time
  
• set timing to 35 degrees max with no vacuum advance (seems to get best result so far)
  
• all new injectors - nos oem
  
• new CSV
  
• all new sensors at the thermostat housing
  
• checked/cleaned all bullet connectors from sensors
  
• checked/cleaned ground connections
  
• all new connectors at all injectors + cold start
  
• manifold cleaned out, new gaskets
  
• egr system removed, manifold sealed off
  
• 6-1 header
  
• k&n cold air intake
  
• i've adjusted afm by moving the white gear to where it hits the best idle and best vacuum
  

symptoms:

• smells like it's running very rich
  
• stumbles at wot up around 3500
  
• plugs read rich/sooty on 3, 4, 5, 6, but 1&2 seem fine if not almost lean
  
• vacuum is down around 14 in hg at idle
  
• when the small vacuum plug is pulled to put the gauge on (letting air into manifold) the idle goes up (indicates rich?)
  

tried to do the yogurt test yesterday but couldn't seem to get any useable info - here's what i did:

• pulled afm and plugged afm-to-tb hose w/plastic cup
  
• pulled breather hose to top of valve cover off and blew into it with thumb over valve cover hose inlet
  
• kept my pressure gauge on the small vacuum nipple at end of manifold and watched leakdown
  
• pulled vacuum lines from everywhere i could, one at a time while watching the leakdown for differences, couldn't find any
  

the break booster is rock solid.

the leakdown happens really fast - couple of seconds. couldn't hear any hissing or air escaping, wouldn't it just be going around whichever intake valve was open and into the cylinders?

am i doing this wrong?

seems the only thing i was able to accomplish was putting a horrible dent in my hood when i closed it on a tool i had left on the radiator crossmember :stupid: i am not happy...

[siteunseen]

I put my valve cover on upside down once, thermostat side is straighter IIRC. Could it be the EGR tunnel sucking in air somehow?

[Zed Head]

I would reconsider using the AFM to adjust mixture. When you change the spring tension you change the shape of the enrichment curve, not just the end-to-end magnitude. In other words, for example, if you adjust it richer, it will be more rich at low RPM than at high. You could actually turn the spring far enough so that there's no pressure on the vane and it just bounces around until it makes contact with the spring again. If you adjust it leaner, ?, I'm not sure exactly where it ends up. I think this is right, although I can't back it up with equations.

It's much simpler to use the potentiometer on the coolant temp. circuit. It's quicker and moves everything up and down equally. If you put the knob inside you can tune on the fly. Also, the idle mixture and ignition timing are very different than they are off-idle. You don't have vacuum advance, but you're still on the idle enrichment circuit of the ECU program, assuming that your TPS is working right. As you add throttle, you lose idle enrichment. And, the ECU is programmed for vacuum advance, so you're off-spec. there also. At low RPM your timing is very retarded, from what the ECU expects.

That's the thing about the stock EFI system. It's designed for all of the pieces to work a certain way. There's not much leeway for change.

On the cup test - don't forget that there are certain places in the engine's cycle where the intake and exhaust valves are both open (overlap), where it will be impossible to create a seal. If everything's right and it won't hold, another ten degrees of rotation might seal it.

[S30Driver]

When you did your AFM adjustment ... was the area the wiper travels across in good shape?

Did it appear to be un-tampered with from the previous owner - seals intact?

About how far, how many teeth was your adjustment?

[FastWoman]

I'm sorry to hear about your dent! If it's any consolation, I have one just like it, so I feel your pain.

When I did this test, I didn't measure leakdown with a gauge. Heck, I'm not sure I can even blow hard enough to register much on a gauge, even having played oboe in high school. You should be able to feel what's going on just with YOU. There's a certain rate at which you can blow air into your engine, depending on leakage. You can get a pretty good feel for that. And there's a certain rate at which pressure will leak down to the point that you can blow more air in. You can feel/estimate that too. You can compare this leakage rate with a known leakage rate through a small vacuum fitting, e.g. the HVAC control line that attaches right next to the brake booster line. I found that my vacuum leaks, in combination, were FAR less substantial than leakage through that open fitting, and I was satisfied with that.

If your leak-down occurs through your valves, it would have to be either from open intake to open exhaust or through the intake and out the rings. Either would be abnormal. You could try turning your engine a few degrees to see if that seals things up. I have to mention here that I've never tried or heard about a "yogurt cup test" on a fresh engine. Perhaps it needs to be broken in before it will be tight. But honestly, if you assembled everything well, the engine should be pretty tight.

You are not likely to hear any hissing, unless perhaps you apply some air pressure from a compressor (I'd suggest low pressure, maybe 10 psi). You might also try recruiting a smoker to help with the test. You can blow smoke into the engine and look for smoke billowing out somewhere it shouldn't.

Backing a bit, and this is important, your plugs don't all read the same. Until the read the same, you might get a lot of mixed information from your engine. Furthermore, if your cylinders are doing different things, there are no settings that will optimize your engine's performance. That's surely why your max vacuum is 14 in Hg, and not at least 18 or 19. Thinking off the top of my head, and not necessarily in the order I'd test, I'm wondering...

Do your injectors have the same flow rate? Being NOS, they might have come from different manufacturing lots with different flow characteristics. Usually parts from the same lot will be pretty much the same, so this isn't as likely an issue with newly manufactured parts. It seems unlikely this would result in wide variance from rich to lean. To test, try swapping your #2 and #3 injector plugs and see if the #2 and #3 plug readings reverse.

I just checked the schematic for '78, remembering that the injectors are batch fired in two banks of 2 and 4. I don't know why the banks aren't split 3 and 3, but they're not. However, in the '78, the banks are 1/2/3/4 and 5/6. So that doesn't account for the differences between 1/2 and 3/4/5/6.

Have you replaced your ignition components? You might have some weak ignition issues resulting in incomplete combustion in 3/4/5/6.

You might re-check your valve lash. Maybe adjustment differences between 1/2 and 3/4/5/6 could result in performance differences that would result in incomplete combustion in 3/4/5/6.

Is your compression even across all cylinders? This would tell you a lot, including possibly some clues about the high (?) yogurt cup leak rate.

Once you've got your cylinders doing the same thing, then worry about tuning the rest of the engine.

For starters, I've never set timing the way you've done it. By the book, you would disconnect and plug the vacuum line to the vacuum advance. Then you'd time the engine to 10 deg BTDC. I think most of us find our engines like a few deg more advance. Mine is set to 13 deg BTDC. BTW, check the breaker plate in your distributor to ensure that it rotates freely without binding. That's a common problem with our OEM distributors.

I would recommend first adjusting your AFM to the correct spring tension per the Atlantic Z article. This time you'll use an aluminum beer can, instead of a plastic yogurt cup. This will ensure that your AFM pegs out at the correct air flow rate. Beyond that, fuel delivery keeps increasing per RPM. Anyway, I try to stick with the design, and that's what I've decided for my own car. If you richen the mix by loosening the clock spring, then the AFM might peg out prematurely, such that there's no enrichment for higher RPMs, hence leaning out.

After you've adjusted your clock spring to spec, then see whether your engine is running rich or lean. My guess is lean, due to what I'm going to call "typical '78 ECU drift," and you would fix that by adding a potentiometer in the CTS circuit. That will get you by until Lenny finally releases his much awaited Hellfire ECU.

[rossiz]

thanks all for the great posts - some really good things for me to chase down. i will approach in a methodical, systematic way to cross items off the list one at a time.

i will re-do the yogurt-cup and beer-can tests and spin the engine to get a closed-intake condition

the PO replaced the ecu a few years back so i'm doubting it has drifted. he also replaced the afm, but i haven't done the calibration thing yet - looks fun

as far as the injectors go, i didn't realize they were in separate groups - i just thought they all went back to one set of leads since they all fire at once?? so i might not have them on the right injectors because the PO replaced all the plugs with loooooong leads and ace hardware crimp connectors - i re-did them as they were all over the place but may have switched some around... is this an issue? is there a way to figure out which one is supposed to go where?

to be honest, i've been so upset by the dent i put in the hood that i haven't had the heart to open it back up. just been driving it and and quietly cussing. fortunately the curve of the hood is such that i can't see it from the driver's seat - otherwise i'm not sure i could handle it. to think this hood lasted 36 years without so much as a scratch and i managed to do that to it through sheer carelessness...

ANYWAY, i've got a weekend planned with my son and so i'll probably jump back on it next week.

stay tuned...

[240260280z]

The dent could be the cause of the lumpy idle..... sorry, I could not help myself.

[Zed Head]

I fixed a dent in my hood using a block of 2x4 as a backer and a finishing hammer. Mine was pretty small though. The metal on these cars is pretty think and malleable. You might be surprised.

On the banks of 4 and 2, I think that the original EFI might have been designed for a four cylinder engine and when they added two more cylinders they just added one more circuit to control the two new ones. The power supply is two leads I believe, one is crimped to four resistors in a package, and the other is ganged in to two. If you look down behind the clutch master cylinder you'll see the two packages (each big rectangular metal piece actually has a set of round ceramic resistors inside). Maybe you have a bad crimp. It's a tight fit but the resistor pack can be removed for testing, or tested in place. There's a test in the FSM, page EF-56 for 1976. 78's the same.

It's been discussed also, that the transistors in the ECU might control them in batches of two and four. I think that superlen or Steve J or Captain Obvious might know for sure. So if you had one transistor go bad it might only affect two injectors or four. The wiring diagram does show 5 and 6 as together but who knows what they did in the ECU.

Injectors are built to a spec. so I can't imagine that the flow rates would be far off. Especially for an EFI system that doesn't have any self-trimming functions.

For what it's worth, my cylinder #1 plug has always been much cleaner and whiter than the others. Even with a different engine. I've wondered about it. Maybe cylinder 1 gets more air or runs hotter.

Good luck, hope it's something simple.

Forgot to say also, the injector harness runs to the middle of the intake so if somebody flipped it when putting it back on, then 1 and 2 would be where 5 and 6 used to be. Just a thought.

Edited August 14, 2014 by Zed Head

[240260280z]

1 leaner is often said to be due to intake runner and plenum design. I have not seen data to verify.

[Captain Obvious]

Despite the independent looking circuits, all six injectors are fired at the same time. Always.

However, there are two output transistors in the ECU and they divide the task in two. One of the transistors does 1, 2, and 3 and the other transistor does 4, 5, and 6.

The DROPPING resistors do it 4-2, but the ECU does 3-3.

But, since the control signals to both transistors are tied together all the injectors are told to open at the same time.

Edited August 15, 2014 by Captain Obvious

[Zed Head]

I think that we were looking for low voltage or bad control, leading to slower/lower opening times, and a lean mix. Bad connections, high resistance, bad transistors, etc. Something that ties 1 and 2 together.

Of course, it might just be an artifact of something else going on. Since the engine is running, I would do the old ear to screwdriver test. It's a Nissan standard. Quality of the click is the key.

[sscanf]

If all else seems right, check your BCDD - this is where I found a leak on my 1976 Z when I resorted to the yogurt cup test. I could hear it though. Blow hard, find someone with good ears. The condition of my BCDD diaphragms were so bad that I'm surprised that I don't hear about the same problem more often here.

It sounds more like you have a vacuum leak but some other diagnostics worth trying (to help eliminate these systems as being part of your problem):

Try running with the TPS disconnected and observe any differences in performance/gas mileage.

Also maybe try disconnecting the CSV after you start.

I would add the temp sensor pot as well (even though you have a newish ECM - who knows exactly how these things are reconditioned).