Exhaust sputtering

[CriticalB]

Last time I posted here I was looking for a Delorean, and about 3 months ago I purchased 3455, Stainless Bullet's twin turbo. So far my weekend projects have been manageable and I've found a lot of good advice here in the forums. I've replaced the passenger side window motor, rebuilt the mode switch, replaced the alternator and cleaned up some electrical connections, fixed the dome light and engine compartment light switch, recharged the AC, and replaced the injectors and plugs. She was running great after updating the alternator, until this Saturday when I noticed a line to the vacuum advance was unplugged. I plugged it back in and noticed the idle was a little smoother and the boost gauge (which hasn't moved since I bought it) now reads 20 psi of vacuum and actually responds while driving. I also noticed an intermittent sputter from the passenger side exhaust. I disconnected the vacuum again and the sputter went away. With the vacuum connected, it seems there's no sputter when the engine is cold and it comes back when it's warm.

I've been reading up on possible causes, and came across the dollar bill test. I tried it and sure enough the sputter corresponds with a sucking from the exhaust. I'm guessing this means a sticky exhaust valve at worst. I pulled the plugs, they look clean and gaps are good. Just to rule them out, I ordered some new plug wires.

I'm looking for suggestions or advice if I'm barking up the wrong tree. I'll post a video tonight. I should add that I hear the sputter at idle, but not while throttled. It's most pronounced when throttling down and when turning off the car...you can see for yourselves in the video.

Worth knowing - there are no mufflers, and the O2 sensor is not installed, so any other idle issues may be related to that. It looks like it was cut out at some point and the hole in the bung was patched. I ziptied it to the frame until I can get it repaired since it was hanging loose...I guess while I'm asking questions, is that a problem or should I just disconnect it altogether? I have an AF gauge installed as well, but I'm not sure if it's accurate...it has never moved from rich.

[CriticalB]

Here's the video:

https://youtu.be/jUMYVjIwESg


https://youtu.be/jUMYVjIwESg

[David T]

Sounds like one cylinder is not working right. Use an infra-red temperature gauge and see which one is colder than the others. Could be a spark plug, valve adjustment, etc. Might want to do a compression test, might be a blown piston. Make sure the boost is set to under 6#. You should get the O2 sensor going and it will be quieter if you get some mufflers on it. You should also pull the outlet pipes on each turbo and see if there is any oil inside.

[Bitsyncmaster]

With forced induction, you probably need less ignition advance. Connecting the vacuum advance is giving you more total advance but in a stock PRV that should not affect idle advance since the idle switch shuts of the vacuum advance.

I would check your ignition timing with a timing light.

[FABombjoy]

Passenger side sputter may correspond to driver side vacuum issue. I can't tell exactly but also a bit of a misfire sound even at idle?

I can hear a whistle like a vacuum leak. Did you disconnect any pipes when changing plugs/injectors? The island pipes are all rough cast and a candidate for vac/boost leaks.

As to blown pistons - pull the oil cap while its running and make sure it doesn't look like Thomas the Tank engine. I doubt you have a blown piston if connecting/disconnecting a vacuum line has a direct effect on the symptoms.

[CriticalB]

Sounds like one cylinder is not working right. Use an infra-red temperature gauge and see which one is colder than the others. Could be a spark plug, valve adjustment, etc. Might want to do a compression test, might be a blown piston. Make sure the boost is set to under 6#. You should get the O2 sensor going and it will be quieter if you get some mufflers on it. You should also pull the outlet pipes on each turbo and see if there is any oil inside.

I'll warm it up tonight and check the temperatures on the exhaust manifold. I ordered a compression gauge on Amazon and will check that tomorrow when it comes in. I've been debating mufflers...it's not obnoxiously loud, but I've heard the wrong mufflers will cancel out any benefits of the turbos so I planned to spend some time researching that later this summer. I put a new O2 sensor on and dropped it in the hole where the old one was cut out just to see if it made a difference on the AF meter or in how the engine ran...it didn't. The hole just had a wide clamp around it that I removed for that little experiment. Afterwards I put some muffler tape on it for a seal and put the clamp back over that. As for oil in the turbos - I'll check it but what would that indicate?


I would check your ignition timing with a timing light.

Also ordered this from Amazon. If nothing else, owning a DeLorean has been a great excuse to fill out my workbench.


Passenger side sputter may correspond to driver side vacuum issue. I can't tell exactly but also a bit of a misfire sound even at idle?

That misfire sound is what I was referring to as sputtering and is what has me concerned. It corresponds to the exhaust sucking air in (each sputter pulls the bill toward the exhaust) However, it's way less frequent on a cold start. I just cranked it when i got home from work and it might sputter once every 10 seconds. When warm, like in the video, it does it as much as 1-2 times per second.


I can hear a whistle like a vacuum leak. Did you disconnect any pipes when changing plugs/injectors? The island pipes are all rough cast and a candidate for vac/boost leaks.

I didn't have to disconnect any pipes for the plugs and injectors. I'll inspect them, though. I hear what you're talking about in the video, but I wonder if it's not the mic picking up something that's not there. I've never noticed it in person and it doesn't seem pronounced when I listen now.


As to blown pistons - pull the oil cap while its running and make sure it doesn't look like Thomas the Tank engine.

Confirmed - no creepy train faces staring back at me :)

[FABombjoy]

Confirmed - no creepy train faces staring back at me :)
Haha, that's a good sign :D

If you google symptoms of "the dollar bill test" too much you might become a hypochondriac.

I think what is happening is that you have one (or more) misfires which cause gaps in the exhaust pressure waves. Then during valve overlap on an intake stroke there is exhaust reversion due to the lower than normal pressure in the exhaust. The whistling could be caused by these oddly timed pressure waves across the turbine wheel or something, that much I'm not sure about.

I would pursue spark & fuel tests before becoming concerned about valvetrain problems.

Oil in the intake would indicate failed carbon seals in the turbos. You would see evidence on the plugs (blackened/tarry ring around the ground electrode) and the inside of the intake manifold will appear bronzed. If you were misfiring and had oil I would expect to see oil in your exhaust pipes by now (oil mixed with water on cold start)

[CriticalB]

Well, ruled out spark plug wires. I got brave and decided to follow the guide to remove the whole manifold to access them. While I was in there I popped in a new distributor rotor that the previous owner never got around to installing. Took around 4 hrs and I feel like I learned a lot. Cranked right up afterwards, but idled rough and did some idle hunting. Turned it off, checked the hoses and cranked it again and this time it found its idle quickly and held steady. I could swear I heard a steady ticking from the passenger side of the engine that I didn't notice before, wondered if a plug is firing now but not combusting?

I'd have liked to spent a little more time investigating but it's late and I thought it best to let the neighbors get some sleep. The compression gauge arrived today and the timing light will be in tomorrow so I'll do some more testing tomorrow.

[DMC-81]

I'm not familiar with turbos, but aside from what's been mentioned so far, how are your exhaust gaskets and valve adjustment?

[CriticalB]

The timing light came in today and I had chance to try it out. With the vacuum advance disconnected and plugged, the timing mark lines up dead on with the 13 degree mark. It goes out of whack and the idle changes when the vacuum is plugged up and I now I think I see why.

If you recall my first post, this started when i found a loose vacuum connection and reconnected it. Now that I've studied the vacuum routing in my car vs what it's supposed to be according to the engine cover sticker, it looks like one of the PO's made some modifications. I'll try to describe how it's currently mapped, but it may make more sense if I post a diagram later.

There's a 4 port connector in the back near the CPR. One port runs to the boost gauge, one to the firewall facing side of the CPR (the outlet on the other side of the CPR is plugged), one port goes to the intake manifold. So the connection between the CPR and delay valve doesn't exist on my car. I don't even see a delay valve. The last port, which i originally found unplugged, goes to the vacuum advance.

There is no solenoid valve. I know what to look for and it's simply not there. The electrical connector is hanging loose and there is a loose vacuum hose that routes under the manifold, so I assume it goes to the thermal control valve. It was plugged with a small bolt, so I assume it was removed intentionally? Since this is a turbo setup, I don't know if I should try to get this back to the factory routing. I do have the installation guides that came with the turbos, so I'm going to comb through them tonight and see if I find anything that requires modification to this part of the vacuum system other than a port to the boost gauge. If not, I'll order a vacuum solenoid and start putting things back to spec.

There is constant vacuum in this setup, I imagine because there's no vacuum solenoid. As long as that's the case, should I just leave the vacuum advance unplugged? what would be the impact of doing that?

Lastly, i've checked when cold, warm, under throttle, etc. and there never seems to be vacuum to the 'lower' hoses in the diagram (connected to the throttle edge tapping and charcoal canister.) Does that make sense? Is it a miracle this car is even running?

[CriticalB]

I do have the installation guides that came with the turbos, so I'm going to comb through them tonight and see if I find anything that requires modification to this part of the vacuum system other than a port to the boost gauge.

Read the docs, apparently this is how it's supposed to be set up. So the vacuum advance is constantly getting vacuum from the intake manifold port. However with everything connected this way, the idle has changed since I changed the plug wires yesterday.

Engine cranks and idles around 775 RPM, as it always has, but now after throttling it returns to right around 1000 RPM and idles there until next time the car is restarted. Don't know what I could have done to cause that to start out of nowhere...

[FABombjoy]

Ok, that's a lot of mods :D

Distributor:
Sounds like it's set up for manifold vacuum instead of ported vacuum. Island has you make that mod but BAE didn't. I'm not sure that it's particularly valuable. Our cars cut off the vac advance with the solenoid to (I believe) help with idle emissions. You can run it connected to manifold but it may have a tendency to high idle. I'm honestly not sure what the other effects could be.

Volvo used the throttle tap instead of our goofy solenoid arrangement to accomplish vacuum cutoff to the vac advance canister at idle. DO NOT DO THAT ON AND ISLAND OR BAE TURBO CAR! Draw-through turbo setups will send the throttle edge back in to vacuum when under boost, which will cause the ignition to advance under boost.


Control Pressure Regulator


There's a 4 port connector in the back near the CPR. One port runs to the boost gauge, one to the firewall facing side of the CPR (the outlet on the other side of the CPR is plugged), one port goes to the intake manifold. So the connection between the CPR and delay valve doesn't exist on my car.

Sounds like an attempt to add fuel enrichment under boost. I'm not sure it's going to work very well, honestly. In my experience you need the factory 2 hose setup to operate correctly when cold. The setup you are describing would cause poor cold running and who knows what kinds of AFR under boost. I would think that the C/O adjustment has been thrown way off to compensate.



Lastly, i've checked when cold, warm, under throttle, etc. and there never seems to be vacuum to the 'lower' hoses in the diagram (connected to the throttle edge tapping and charcoal canister.) Does that make sense? Is it a miracle this car is even running?
That should only see vacuum when the throttle begins to open

If you want to return the CPR & distributor setups to stock I have the delay valve, solenoid, and many other leftover K-jet parts.

Are the hobbs switch(es) in place? Any other deviations from the Island manual?

[CriticalB]

Quick update, since I last posted I've been through a couple of tanks with Techron added and the sputter seems to have almost vanished. Not sure if there's a cause/effect relationship there, but thought it was worth noting. It could have been the new plug wires or rotor. In all that driving, I did note the idle seems to be getting progressively higher. On a cold start it would hunt, and after warming up it would settle anywhere between 1k - 2k. I began to suspect something was up with the pipe of agony, but since I just put all new stainless fuel lines I was hesitant to undo them all and check. I finally bit the bullet today and pulled the fuel manifold out and found that, sure enough, the O ring behind the bronze cap was rotted and must have broken up when I took it out previously. I put in a new ring, greased it and put everything back together and the idle is back to ~775.


Ok, that's a lot of mods :D

If you want to return the CPR & distributor setups to stock I have the delay valve, solenoid, and many other leftover K-jet parts.

Are the hobbs switch(es) in place? Any other deviations from the Island manual?

If there's a hobbs switch, I don't see it, but I'm not sure exactly what I'd be looking for. I appreciate the offer for the delay valve, etc., but since I think I found the idle problem, I think I'll try to keep it as close to original as possible. Well, original + turbo mods, anyway.

[David T]

If a can of Techron helped it means one or more injectors was not spraying well and is now better. You really should have either catalytic convertors or mufflers installed and the O2 sensor in and working. My advice would be to try to get the turbo install back to an original installation configuration. Many owners did mods to their cars and some of them worked and some didn't. Get the original paperwork for the turbo install and try to get it back to that as a start. Might be worth checking the fuel injectors and the valve adjustments. The Island Turbo install will run the idle higher than 775 because of the way the vacuum is hooked up to the distributor. Now that you have a timing light you should check the vacuum and mechanical advance. If your base timing is 13 it is too much. It should be between 8 to 12 depending on your gas and altitude and how much boost you are using. You really should not go above 5 psi. Use the highest octane gas yo can buy to prevent detonation. In a forced aspiration Delorean stock PRV you won't hear the detonation and it can quickly destroy your pistons.

[FABombjoy]

Island / BAE have a Hobbs pressure switch (or two) installed to force the system to go open loop and add fuel enrichment on boost. The "book" method of installing them is bad as it cut/splices the O2 sensor which adds resistance to an already low voltage signal. You can get the same effect by grounding terminal 11 on the lambda box.

At first I had two adjustable Hobbs switches, one for low boost (triggering the WOT switch) and one for high (triggering terminal 11). After datalogging I found that terminal 11 added far too much fuel - like 10-11 AFRs. I turned it up the activation pressure and left it as an overboost failsafe. Eventually I switched to the UTCIS programmable CPR and dialed in a more progressive and sensible enrichment curve.

If the previous owner set the car up to use the CPR to enrich under boost and you reinstall the O2 sensor, you really still want some way to force the system to go open loop. I did this on my own car by grounding the wide open throttle switch which prevents the lambda box from trying to lean things out under boost. The lambda box will fight for fuel control and can do a remarkable job of leaning things out.

It's difficult to recommend any course of action without any data logs or precise list of fueling mods. There are a lot of ways to accomplish the same task and the Island/BAE methods aren't necessarily any better than what you have running currently.

[David T]

Island / BAE have a Hobbs pressure switch (or two) installed to force the system to go open loop and add fuel enrichment on boost. The "book" method of installing them is bad as it cut/splices the O2 sensor which adds resistance to an already low voltage signal. You can get the same effect by grounding terminal 11 on the lambda box.

At first I had two adjustable Hobbs switches, one for low boost (triggering the WOT switch) and one for high (triggering terminal 11). After datalogging I found that terminal 11 added far too much fuel - like 10-11 AFRs. I turned it up the activation pressure and left it as an overboost failsafe. Eventually I switched to the UTCIS programmable CPR and dialed in a more progressive and sensible enrichment curve.

If the previous owner set the car up to use the CPR to enrich under boost and you reinstall the O2 sensor, you really still want some way to force the system to go open loop. I did this on my own car by grounding the wide open throttle switch which prevents the lambda box from trying to lean things out under boost. The lambda box will fight for fuel control and can do a remarkable job of leaning things out.

It's difficult to recommend any course of action without any data logs or precise list of fueling mods. There are a lot of ways to accomplish the same task and the Island/BAE methods aren't necessarily any better than what you have running currently.

Some owners even tried blasting the cold start valve to keep from running too lean. None of these worked well because none of them were calibrated for the desired result. There are also water/meth injection schemes. The basic problem is boost control. Very rudimentary using straight boost pressure to limit, you guessed it!---- boost pressure! Today, with systems like MSD, you can fine tune all of your control parameters and they don't have to be linear. You can program fail-safe features and if they fail you can have other back-up features to prevent detonation.