Dies at idle, stutters above 2000 rpm, progressively worse

I think I have a vacuum leak somewhere.
84 GSL
Almost all stock (updated radiator and radio)

Shorter story:
Noticed some stuttering about 2 months ago around 3000-3500 rpm.
Became more noticeable each week.
Idle became erratic jumping up and down, eventually stalling.
Engine starts fine with choke engaged.
Now engine dies when choke disengage when normal coolant temp reached.
Last driving I had to pull out the choke when stopped to hold the RPM to 1200 – 1800.

I am now suspecting a vacuum leak at or near the carb.

More details:
Initially the issue seemed to be ambient heat related, worse at 88 degrees than 78.

I eliminated the igniters as a cause. I have 2 in use and 2 spares and 1 dead. The 2 in use and the 2 spares all tested good using the FSM test procedure. (I keep the bad one as reference.) I cleaned off the 2 in use and applied fresh thermal paste, reinstalled but made no difference. I tried the other pair and still no difference.

Earlier this year I replace the plugs, rotor, distributor cap and fuel filter, checked the timing and adjusted idle mixture and speed. It ran without issue. I rechecked all but the idle mixture, idle speed and timing, and found no new issues.

Not yet completed:
Replace aftermarket fuel filter with one from Atkins Rotary which will not collapse from the lower hold down.
Properly check the coil resistance. Trailing at .5 - .6 and leading at .7 - .8 with the engine warmed up but not running. Should be 1.3 ohms. Both coils were replaced at the same time some years ago.
Have not check the fuel pump pressure or output.
Have not jumped the fuel pump cut relay to take it out of the system.

Any input/suggestions will be most welcomed.

 

i believe your instincts are correct and a vacuum leak is probably at play.

 

I was curious last night and pulled off the air cleaner. The front of the carb and the carb insulator base were way cleaner and dry when I rebuilt the engine 16k miles (10 years) ago. There is no odor of gas or oil, but my sense of smell is not what it once was. However the last times I drove it and parked in the garage I thought I did smell gas. The gunk on the front of the carb starts at the gasket above the float window. The carb insulator base has some gunk in the middle section and a slight gunk cover at the rear. I did check for a possible coolant leak where the radiator hose connects to the water pump, but the thin layer of dirt on the metal connector was undisturbed.

Is what I'm seeing in the images connected to my main problem, or just one more problem?

I'll be checking for vacuum leaks at this point.




was undisturbed.

Is what I'm seeing in the images connected to my main problem, or just one more problem?

 

is it related? my guess is probably not. if you're not cleaning things somewhat regularly, then i think that's just what 10 years looks like. when was the last time you gave the carb a thorough cleaning? 16K miles in 10 years ... so it doesn't see much constant use. that can lead to some problems like this.

i would focus on tracking down the vacuum leak and/or whatever carb issue you may or may not have. if it's not idling without the choke, then clearly something is up with your intake.

 

A few things to check;

1) When was the last time you changed spark plugs? They're good for about 10-12k miles, and then will degrade slightly over time so little that you'll barely notice, but putting new plugs in and then it feels like you gained a bunch of HP. Also, same for ignition wires.
2) Cap & Rotor fairly new? They don't last forever and any tune up should include new ones. While they're out, remove the spark insulator plate and take a close look at the vacuum advance mechanism. Does it look like it's moving? It should have clean metal where the arm rotates the plate. Also check the vacuum line to the vacuum diaphragm for the vacuum advance - if this is broken, it could cause stuttering at certain engine speeds and load / vacuum levels.
3) Shutter Valve? You'll have to search on that here, but I seem to remember it may cause similar symptoms.

That should give you a few other ideas,

 

New plugs, rotor, distributor cap earlier this year. Yes, it ran better.
Ignition wires are maybe 9 years old, or about 14K miles. Coils about the same ager.

Thanks for the vacuum advance. I'll look into it.
The shutte valve is on my list, but I'll start there. At least that part is still available.

I knew this would not be a quick fix. However I did find a vacuum line goint to the #1 antiafterburn value not really connected. (red arrow in photo) It slid easy off, push it back in tight, but no help with the idle.
I also checked the hose from the valve to the intake manifold. About 19 years back the car stalled in traffic. I got it out of the way, popped the hood and looked around. the hose was not attched to anything at the top. I saw where it should be and pushed it back on. Problem solved. Yesterday I took the hose (blue arrow) completely off. It was not the formed Mazda part (N201-13-756), just a generic hose which wanted to uncurl, which finally explained why it orginally popped off. I've order a new Mazda part. Dad had an Atkins remanufactured engine ($1500) installed in 1997. That is probably when the hose was replaced.

Thanks again for the suggestions.

 

Update:

I installed the correct hose from #1 anti-afterburn to the intake manifold. It initially seemed to help when the car was stationary, but not driving. It started OK with the choke out. Started to stumble at idle so I held it about 2000-2500 rpm for about 3 more minutes. The idle was then stable. Taking the rpm to 3500-4000 and letting it drop multiple times the idle returned to stationary. I stopped and let the engine cool down. Late started again with the choke out. When the choke released this time, no stumble. Repeated going up to 3500-4000 and dropping down to stable idle. Left the car idle for some time and every time I checked the idle was stable. Later the same day I ran an errand and the original problem returned. Next day a not good non-driving test. It started OK, but ran rough and died at idle.

Testing results:

The idle issue is not related to ambient temperature. It did not hold idle from 70 to 87 degrees.
All 6 solenoid valves on the rats-nest tested OK per FSM procedures.
Distributor:
---Both ignitors test good.
---Leading vacuum advance OK. Plate moves with vacuum.
---Trailing vacuum advance does not hold vacuum and plate only moves with external force.
---Plugged the hose from trailing vacuum control solenoid valve and the idle problem persists.
Shutter/Coasting valve tests OK per FSM. Also applied vacuum to Coasting valve and no leaks.
No. 2 Anti-Afterburn valve tests OK.
---About 20 years ago I replaced the No. 2 Anti-Afterburn valve for a similar problem. I plugged the solenoid and hose to the valve to take it out of the equation. The problem persists.
It was not possible to test the No. 1 Anti-Afterburn valve as I could not remove the hose from the air pump.
Dashpot works as specified.
No other vacuum leaks detected. 10 years ago I replaced all the rats-nest’s vacuum hoses.

Questions:
Why would it idle as it should on two starts, but not before or after?
Can a vacuum leak be intermittent? I’m beginning to think it is something is stuck/clogged, except for two good tests.
Could the issue be with the carb? The last rebuild was 1997 at 128k miles. (47k miles ago.)

Not done:
Fuel pump test.

Upcoming:
Ordered new coils and ignition wires. Last replaced August 2002 at 133k miles (current 175k miles). Both coils out of spec. The highest reading I could get was: Leading 1.1, Trailing 1.2 ohms. They should be 1.35, 10% plus/minus. (Plugs, distributor cap and rotor all replaced earlier this year.)

(Image: top was the installed hose when I got the car 21 years ago, the bottom is the Mazda hose now installed.)

 

After capping all the vacuum lines on the rats-nest, replacing the trailing vacuum advance, replacing the hose on the botton of the purge valve going to the carb, and testing for vaccum, nothing helped the idle.
Switching to the right side, checking for vacuum on the different valves, I found the #2 Choke Diaphragm did not hold vacuum. Capping off its vacum line solved the idle issue.

On a test drive the car did stumble, it felt like a momentary loss of power. In neutral it was smooth to 5k and quick back down to idle.
The clutch engages very close to the floor. Could I be experiencing slipping?

Any and all suggestions/replies are appreciated.

 

Ensure the clutch master cylinder is full, and slave cylinder and the line going between them don't have any leaks. You can also adjust the height at which the clutch pedal engages.

 

You were correct. #2 Choke diaphragm. Thank you.

 

The rubber hose is leaking at the slave connection. The fluid in the master always turns dark.
I've ordered the parts to replace the master, hard line, and the slave. I already have a stainless steel flex line.

Thanks for your help.

 

I'm recalling that clutch masters these days might have a difference in the thread depth vs the factory part. The part still works but you might need to stack some washers to take up the slack. I forget exactly how but I do recall having to use a couple.

 

Thank you.
I will try and match the new with the one in the car, but I may have already replaced the master in 2004.

 

I'm glad you are replacing all three items in the system Tim, twenty-seven years ago before I knew any better, and before I found this RX7 forum I had a leaking master cylinder so I replaced it. Two months later I had to change the leaking slave cylinder, and one month after that the line. Toruki is right about the master cylinder studs, I changed my system last year, and the and the master cylinder I bought didn't have threads all the down the studs and I needed to add a couple of washers to get it to tighten down on the firewall.

 

Check to see if the threads on the clutch master now are longer than the new one, if they are swap them. I just replaced my original with an aftermarket unit and didn't learn about this until after I had already thrown out the original. Worse case is you will add a couple of washers to both threads from the inside then add the nut and tighten. No biggie but baffling if you didn't know about it!

 

Just check the master, and my memory. This will be the second time to replace the master and yes, there are additional washers installed. I may have the original in a box somewhere, but probably not at this point.

Thanks for the suggestion.

 

And the hard line. It was cheap. Now just waiting for the parts.

 

Where did you get the hard line at? I replaced everything else except the hard line and if it's cheap like you say I may as well do that too.

 

Only for 84-85 and newer Rx7's.
https://www.atkinsrotary.com/Flywhee...-Clutch-Pipes/

 

Got an 85 so that works, thanks.

 

Success!
Vacuum leak found. Used part purchased and installed.
Clutch flex-hose leaking. New slave, flex-hose, hard line and master installed. (That left nut on the master was not fun.)

Thanks everyone for your suggestions and help.

 

And it doesn't get any easier as you get older Tim, that why the verifying the threads before installation is important, Nothing worse that getting the left nut on, and tightened down, just to find out you need washers to get it tight and you didn't use any.
:-(

 

I've broken the ear off of the clutch master because of this. Lesson learned on checking now.

 

Tales of the clutch left nut.
After over a dozen attempts to get it started where it would slip out of my fingers, on the last attempt it fell and disappeared. Not behind the front carpet, not under the floor mat, not even hiding behind the left wall panel. Took a stud out of the old master and the one nut and off to the auto parts stores. (O'Riely was out, Advance had a pack of 4.)

Again trying, and still failing.
I took a 2' length of 1' blue painting tape. Taped one end over the top of nut and folded the rest into a handle I could hold on to. a few conterclockwise/clockwise tries and it grabed onto to the stud. After a few turns on, pulled off the tape an the rest was easy.

 

Yeah some of those nuts are hard to reach and screw on, glad to hear you finally got it on there!