CIS woes - car runs lean, measures rich

[cmovic]

Hi All:

Up until this weekend, my car has been running great. It's a 1978 911 SC track car with 3.2l 9.5:1 pistons and a M1 cam. Induction is CIS and I'm running headers and a fairly open exhaust. I have an air/fuel meter on the 1-2-3 side of the exhaust.

The usual suspects for air leaks have been taken care of. The airbox, all rubber parts including the injector o-rings were replaced two years ago when the engine was rebuilt. The injectors, phenolic blocks in the intakes and their o-rings were new about 15 years ago.

On Saturday I started it up and noticed a high idle cold. Once it warmed up the AF meter showed lean to really lean at idle. The mixture richened at higher RPM and driving it on the track showed a pretty normal 12.5:1 at WOT.

Up until now, the cold start behavior has been it runs really rich (9.5:1) at start and leans out to 11:1 in about 90 seconds. Once it's fully warm it idles nicely and I see AF in the 13:1 range. As of Saturday it starts at 11:1 and warms to 13:1 and then keeps climbing into the red range (>17:1) and doesn't want to idle.

A little googling and a chat with a retired CIS expert suggested dirt or something in the WUR is causing high control pressures and causing the lean mixture. So, I consulted the handy "CIS for dummies" thread and headed out to the garage to measure my pressures. And that's where things turn weird. Pressures are low (full data below) suggesting I should have rich mixtures. Hopefully someone out there can help me understand what I'm seeing.

Here are the measurements:
1. Year of engine: 1978
2. US or RoW (Rest of World): US
3. WUR model number: 045
4. Ambient temperature at time of test (in degrees C): 23
5. WUR Resistance (in Ohms): 25.7
6. System Pressure (in bars): 5.0
7. Cold Control Pressure (in bars): 0.6
8. Warm Control Pressure (in bars): 2.95
9. Time delta for Cold -> Warm Control Pressure (in minutes & seconds): 5:00
10. Residual Pressure @ 5 min (in bars): 1.8
11. Residual Pressure @ 15 min (in bars): 1.5
12. Residual Pressure @ 30 min (in bars): 1.3
13. Residual Pressure @ 60 min (in bars): 0.5

I ran the test twice. First time was without charging the battery and with a slight leak at the WUR. Second time was with the battery on a charger and everything sealed up. Numbers are slightly higher across the board for the second run but otherwise consistent.

Please let me know if this makes sense and where I should look next. Thanks!

-Vic

[boyt911sc]

Vic,

What is your WCP with vac applied? Thanks.

Tony

[cmovic]

Tony: Thanks for the quick reply.

Just to confirm, WCP w/vac = run through above steps, then apply vacuum to the WUR after the warm pressures are stable? How much vacuum? I think I saw 1 bar posted somewhere but can't find the reference.

Off to the garage...
-Vic

[cmovic]

And rather than wait for an answer I went downstairs and recorded the following:

Warm up control pressures over time:
Bar Time
--- -----
0.55 0:00
1.0 0:54
1.5 1:30
2.0 2:10
2.5 3:10
3.0 4:51
3.0 6:00 (final WCP)

WCP w/Vacuum (psi - sorry that's what mityvac reads)
PSI WCP (bar)
---- ------------
5 2.9
10 2.7
15 2.55
20 2.4

-Vic

[cmovic]

Hmm, I notice the Spec Manual says: "Connect vacuum pump for test on intake connection of warm-up regulator" and lists higher control pressures. Is it referring to the vacuum port on the side of the WUR? My tests above were applying vacuum to the port on the top.

The port on the side of my WUR is capped off and has been for years.

[boyt911sc]

Vic,

There are different ways you can measure the control fuel pressures of your CIS motor. Decide which one is convenient for you. Both will give you same tests results.

• First method, with the motor running and the FP gauge tester connected.
• Next method, just running the FP with the gauge tester connected and using a hand held vac pump.

It is imperative that you know if your WUR is vacuum assisted or NOT. If it is, where to connect the vacuum line. If your WUR in a nonvac type, you don’t need any vacuum to the WUR.

Tony

[boyt911sc]

Originally Posted by cmovic

Hmm, I notice the Spec Manual says: "Connect vacuum pump for test on intake connection of warm-up regulator" and lists higher control pressures. Is it referring to the vacuum port on the side of the WUR? My tests above were applying vacuum to the port on the top.

The port on the side of my WUR is capped off and has been for years.

Vic,

Your WUR is -045 which is a vacuum assisted WUR. You connect the vacuum line to the side port and NOT to the top port which a vent or atmospheric port. There is a discrepancy in your system pressures test results.

Tony

[cmovic]

Thanks for clarifying. WCP with vacuum is 3.5 bar.

Based on the spec book figures, everything except cold control pressure appears to pass. The low CCP numbers seem to explain the rich air/fuel values I usually observe at startup. This makes me think I should start looking elsewhere for my lean idle problem.

Vic

[cmovic]

I've started on the "looking elsewhere" journey. Reading the Bentley manual last night I decided the next place to look is the AAV and the AAR. To test this, I disconnected the Y hose (sorry if that's not the right name) from the AAR outlet, taped over the AAR outlet and plugged the hose with a cork. I think this removes both devices from the equation because no bypass air should flow.

I started the car and have the same behavior (immediate high idle and AF values in the 13:1) range. If my hypothesis above is correct, this means I probably have a leak downstream of the AAV.

And, I have a suspect: the Y-pipe. In its past life, the prior airbox was only supported by the center back bracket and the Y-pipe rested on the fiberglass shroud and wore a hole in the bottom of it. This saga is documented here. I patched it with JB Weld but I suspect that may have cracked and now it's leaking again.

I'm headed back out to the garage to try to feel around to see if I can confirm this theory. Open to ideas for further testing or places to look.

Vic

[silverlock]

Spray some starting fluid (carefully, avoiding catching clothing in the fan) on the secondary air pipe when it's idling and see if it surges. Same for the CIS boot clamps.

Or, less lazy: remove the CIS boot, pull the pipe, and inspect it properly. Submerge the secondary air pipe and check for leaks (bubbles). Also a good chance to inspect the boot for cracks, leaks and ensure all hose clamps are tight.

[boyt911sc]

Vic,

Use a smoke generator and connect it to your CIS. You need just 1~2 psi. of pressurized air with smoke to locate and identify the hard to find air leak source/s. You could build one or buy one at Amazon.

Tony

[PeteKz]

Agree with Tony. I think it's most likely an air leak from somewhere. That is typical for a lean and stumbling idle (during high vacuum) that smooths out with more load (less vacuum). If you already patched the Y pipe once, you know your way around CIS. Smoke test is probably the fastest way to find it.

[cmovic]

I was thinking smoke tests were next. Thanks for the pointers. The starting fluid seemed to cause surges but it was a little hard to tell. A friend has a smoke test setup that I'll try to borrow next week.

In the mean time, I pulled the Y-pipe and it doesn't hold vacuum. That's good news but for grins, I plugged the Y-pipe connection to the cold start valve and the hole in the boot, basically removing the AAV and AAR from the system. Startup shows the same behavior (high idle, high AF values) so there's probably another leak. I don't think there are any obvious candidates than the airbox so that's not great.

Time to start smoking...

Vic

[PeteKz]

Let us know what you find.

[cmovic]

First attempts with smoke suggest a cracked or blown box but are inconclusive. I injected the smoke in the large port at the back of the throttle body and appear to have smoke coming out the bottom of the airbox. I couldn’t catch it while holding he smoke nozzle in one hand and the video (phone) in the other. I’ll try again tomorrow when I can recruit a second set of hands or attach the tube directly to the throttle body.

It seems like I have two datapoints that it’s the airbox (#1 being the AAV, AAR, and Y-pipes are bypassed so not a variable). Given that and the fact I have an oil leak on the front of the engine (probably bad sender), I may accelerate my fall/winter plans to drop my engine and attend to both of these. If I do that it will be hard to keep my eyes off EFI or carb setups…

-Vic

[cmovic]

Forgive me Father (Tony), for I have sinned. The timing right now is terrible to rush into my fall engine drop plans. Instead of playing with the stinky smoke machine* I tried richening up my idle mixture. It took a lot (1.5 turns) but I've got a hot 13.5:1 A/F ratio and the car starts and runs pretty well. My idle speed valve is all the way closed and I'm still idling at 1500 but I think that's fine to get through one more two day track event. I know this is the wrong way to fix the problem but it puts me in a much better place to be able to get home in September, drop the oil, drop the engine, fix leaks and pull the airbox for a closer look at what's going on. I'm hoping I can JB Weld any patches but if I need yet another box, there will be time to shop for one and not have to rush into a big purchase.

So, stay tuned, and I'll be back in the fall. Thank you everyone for your input. I learned a lot more about CIS this week.

-Vic

*Pro tip - buy pure mineral spirits, hot baby oil has a nasty smell you can't get rid of

[PeteKz]

If you had to turn the CO screw 1.5 turns, you DEFINITELY have a big air leak. Normally, turning that screw 1/8 of a turn makes a significant difference in AFR.

You gotta do what ya gotta do to get through the summer, I get it. However, if you have any time to look more closely, you might be able to patch it up with duct tape and/or silicone sealer and get through the summer that way.

[cmovic]

Yeah I agree this is a big leak and my fix is suboptimal. I'll try taking another look when I get back but so far cameras, mirrors and an endoscope have yielded few clues about the leak. I have a theory that it's near the right side mounting point. A mechanic was rocking the box back and forth during a "tech inspection" and I have a solid mount to the #6 runner in that corner. But that's just a wild guess. I've learned the lesson that CIS requires science and isn't kind to hunches and parts swapping.

Vic