When you click on links to various merchants on this site and make a purchase, this can result in this site earning a commission. Affiliate programs and affiliations include, but are not limited to, the eBay Partner Network.
Been a member here for a bit but just recently purchased my first rotary! A $1200 2004 GT Grey manual Rx8 project hahaha. Bought the car knowing it was going to be a lot of work, so after a few days of cleaning I am finally getting down to the basics here.
Car starts, will idle for a bit then surge (with me holding it at 2krpm) and stall once it seems like the computer allows it to drop back to ~1000rpm idle. Car definitely runs rich, cleaned maf didn't change anything so far and will be investigating my ignition system in the coming days. Trailing spark plugs look gross when I pulled them just now.
Anyways I figured I am all in on these cars (want an FD Rx7 eventually) so I bought a compression tester from RotaryCompressionTester.com. Decided I may as well get this question out of the way and these are the results. Due to the idling issues I was only able to get the coolant temp up to about 180F per my OBDII scan before it just quit. Testing was at just over 150F (car automatically ran fans after shutoff).
Cranked twice each, using trailing spark plugs negligible differences between cranks.
So first off, these show up as "Good" per the tester paperwork, and "Acceptable" per the threads I've found here. But secondly my starter obviously sucks (battery is new and good), so can I trust these numbers with such a big correction?
Thanks in advance for opinions! I apologize for all the info, I like to note all the variables I can when I ask questions like this.
Last edited by Speedanimal; 10-27-2018 at 02:11 PM.
Thanks a bunch for the responses guys! I checked my cat, looks to be good, and got my BHR ignition set in, should be some fluid changes away from firing it up and hoping for some progress.
Well, disappointment set in today. I apologize for the novel ahead, TL:DR at the bottom.
The last two weeks I installed a BHR ignition kit (figured I'll want the coils for future track duty), did my best to check for a vacuum leak, pulled down the cat, it looked good, and cleaned the IAT/MAF. No noticeable difference from that, car still died if I let it go for a couple of minutes at idle. I was also getting a noticeable surge from roughly 850rpm all the way up to 2k. If I held the throttle at 1500 it would surge to roughly 1800 while doing that. Will get back to jumpy idle.
I bought the car having supposedly sat for a few months, so I put a couple more gallons of new premium in it (there isn't much of the old in there), a little premix and a little sta-bil. Not sure if what I'm experiencing is just junk gas...I'd hope the car would at least stumble but run on a mixture of old and new gas. To this point I'd received two engine codes: P0300 and P0410. The misfire sorta makes sense as I really let it stumble to see if it would catch itself, and with the new coils I wasn't terribly concerned. For P0410 I can hear the vacuum cleaner I mean air pump run at startup, based on my forum reading here I assumed this shouldn't keep my car from running. I also noticed a "thump" at 850-1500rpm, almost like lugging a rotor or flywheel issue that would vibrate the whole car until I got the revs over about 1500. It wasn't constant, just a few to one at a time.
Well today, I got under and pulled the ESS off, it was filthy. Cleaned it and did the reset (thankfully my '04 does the OIl Temp sweep). Didn't notice much of a difference there on another startup, UNTIL I put the clutch in (I had left it out before). Suddenly the car went from surging from roughly 1500-2k on warmup to holding at 1500 and then seemed to smooth out at 850 for a minute. I got excited thinking I had just found a bad neutral switch or something. Got under and checked the switches (what a pain to get to by the way), and both the neutral and brake switch checked out...as did the clutch pedal switch.
So I started it up again and noticed that pushing the clutch in does cause a change, typically by dropping the rpm the car wants to run at but doesn't keep the car from stalling. I had changed the coolant earlier this week as part of a fluid overhaul so I figured I would get it warmed up a bit more to get any air out of the system and then call it a day. Through a few restarts got it to 180F coolant temp and then I let it stumble...car wouldn't start back up. Managed to get it to start after a minute of waiting and get the water up to 210F. Stopped it, then tried to crank it again...no dice.
Out comes the compression tester...thankfully doing this again I was a heck of a lot faster to set it up.
Tested front rotor again twice at coolant temp 185F...got 6.9-7.1 Bar on all faces. A very noticeable drop from just a couple weeks ago...
Tested rear rotor again twice at coolant temp 176F...got the previous 7.5-7.7 Bar from before.
TL:DR: Installed BHR ignition, cleaned MAF/IAT, cleaned and reset ESS, checked Cat, checked neutral/clutch switch, did visual check for vacuum issues. Car still jumps around at idle and eventually stalls hot or cold, difference in rpm if clutch is in or not. Rechecked compression after no hot start, front rotor is failing.
So...my front rotor is actually on the failure line. I feel like what I am experiencing is just as much of a vacuum issue as it is a compression issue though. Any opinions? Please critique my next steps.
My proposed next steps:
-Check front O2 sensor on top of header, is there any check other than to just replace it? I couldn't find it with my Torque App.
-Pull intake manifold to check hidden vacuum connections, I don't have a smoke machine, will have to research other methods that don't rely on carb cleaner
-Replace MAF with new one as test
-Check grounds (checked what I could see but didn't look how many there are)
-Start reorganizing garage for a rebuild...
6.9 bar is borderline, but it's not failing yet.. I wouldn't expect the type of issues you're having from that. But, you never know.
If this was my car my next checks, before pulling anything apart, would be to see what the LTFT and STFT are. They should be 0-ish at warm idle. If they're not, at least we know you have a vacuum leak. If they are, then either the computer hasn't learned the trims yet, or there is no leak and your issue is elsewhere. I'd live the O2 sensor alone, nothing here points to it being an issue and the car is smart enough to detect O2 failures.
Then, you say the clutch safety switch checks out, but how did you test that? Actuating the clutch makes a difference, so, I'm thinking, there is something more to that.
Thanks for the help Loki. I should now have my Torque app configured to grab my fuel trims. I'll get that data logged and see if I can gather anything from it. I plan to log MAF, A/F, Throttle, RPM, Temp, Fuel Trims, Fuel Pres (if I can pull that), and will try again to grab something O2 related.
I also think it is time to replace the junk starter...will be ordering a 2.0kw one.
On the safety switch. I unclipped it from the pedal and tested it with a multimeter (at the sensor, not elsewhere) by pushing the plunger. I had continuity one way and open the other, so I assumed that was acceptable.
If you are implying that I should be looking at the connection elsewhere let me know. I have not done as much digging on here for that switch as I have for the neutral.
Last edited by Speedanimal; 11-11-2018 at 09:13 PM.
Might be a dumb question but here goes. Why would there be soo much difference in readings in such a short time. I am asking as I just bought a rx-8 yesterday and am about to pull the trigger on this exact compression tester and want to confirm that it is reliable. OP's results kind of made me worry a bit. So just want to confirm before dropping 300 on a tool
Might be a dumb question but here goes. Why would there be soo much difference in readings in such a short time. I am asking as I just bought a rx-8 yesterday and am about to pull the trigger on this exact compression tester and want to confirm that it is reliable. OP's results kind of made me worry a bit. So just want to confirm before dropping 300 on a tool
Temperature difference between the test, I think. Last week I let my car idle a lot while flushing the coolant. My fuel trims started to climb and lost some vaccum. I had to drive the car hard to fix my trims and get back some vaccum. I think that idling the 8 a lot will cause carbon build up and may make a seal stick a bit.
I don't where the the reading changed from temperature. The rear rotor tested differently from the front, but that's expected. That said in my experience comprrcompr readings can be inconsistent. My car once tested at 80psi at 250 rpm, failing by a wide margin. These days, 4 years later, it tests at 110. Same engine
Might be a dumb question but here goes. Why would there be soo much difference in readings in such a short time. I am asking as I just bought a rx-8 yesterday and am about to pull the trigger on this exact compression tester and want to confirm that it is reliable. OP's results kind of made me worry a bit. So just want to confirm before dropping 300 on a tool
My assumption was a combination of carbon and me getting the car warmer this second time.
However, I'm not entirely convinced that it is bad. The stock starter is pretty worthless and I bet with a 2.0kw starter it would have turned over no problem. It wanted to, it was close, but just wouldn't catch trying to crank at like 175rpm.
It has gotten a bit cold down here in Tejas. Between work and well will haven't made much time in the garage since Sunday.
Super slow progress on this. But slight update with another quick set of questions, thank you everyone so far for your help!
Found out that the gas smell in the garage I was getting when putting the car in start (and not cranking the motor) was the fuel pump priming and giving off a slow drip at the notorious quick disconnect connection under the back seat. I have not been able to determine if it is the pump or the elbow disconnect that is the true cause of the leak, but considering how yellowed and old the pump looks and how I have now disconnected the connector...I figure I will replace both. I have no desire to continue troubleshooting the idle issue until this leak is resolved. Call me paranoid...but yeah I'm a little paranoid.
Fuel pump is on the way (along with 2kw starter) but my question is what is the P/N for this elbow quick disconnect at the pump? From my searching on here I have not been able to find it, don't see it in the fuel section of foxed.ca, and Mazdaswag etc doesn't have it even in their diagrams. There are a few posts that mention piecing together a brass elbow fitting and then using a hose clamp to attach to the pump with a new section of fuel hose. Is this the preferred way to replace that elbow, by removing the quick disconnect portion altogether? Note, my car still needs to have the fuel recall done but my understanding is that the recall only gets me a new fuel pump mount ring, not a pump or elbow.
Finally, crappy phone photo. Yes, I am aware there is a lot to improve on, but I am having a good time with it.
Super slow progress on this. But slight update with another quick set of questions, thank you everyone so far for your help!
Found out that the gas smell in the garage I was getting when putting the car in start (and not cranking the motor) was the fuel pump priming and giving off a slow drip at the notorious quick disconnect connection under the back seat. I have not been able to determine if it is the pump or the elbow disconnect that is the true cause of the leak, but considering how yellowed and old the pump looks and how I have now disconnected the connector...I figure I will replace both. I have no desire to continue troubleshooting the idle issue until this leak is resolved. Call me paranoid...but yeah I'm a little paranoid.
Fuel pump is on the way (along with 2kw starter) but my question is what is the P/N for this elbow quick disconnect at the pump? From my searching on here I have not been able to find it, don't see it in the fuel section of foxed.ca, and Mazdaswag etc doesn't have it even in their diagrams. There are a few posts that mention piecing together a brass elbow fitting and then using a hose clamp to attach to the pump with a new section of fuel hose. Is this the preferred way to replace that elbow, by removing the quick disconnect portion altogether? Note, my car still needs to have the fuel recall done but my understanding is that the recall only gets me a new fuel pump mount ring, not a pump or elbow.
Finally, crappy phone photo. Yes, I am aware there is a lot to improve on, but I am having a good time with it.
I had the fuel pump recall done recently. When they did it, they broke something on it. I think it might be the same tube. They replaced my fuel pump. I had to go without my car for a few days.
If you ordered a standard replacement for the fuel pump, it should come as a complete assembly with pump, filter, housing, etc. If you ordered an aftermarket fuel pump, youre just getting the pump.
Im willing to bet that the outlet on the pump housing is deformed and a large cause of the leak, hold a straight edge to it and verify. Its possibly cracked, but more than likely it has deformed and bent over time.
i ordered a replacement fuel pump back in 2014 from advance auto parts. I cant remember the brand, but when i opened the box it was the actual authentic mazda fuel pump assembly with the genuine mazda product sticker on it and everything.
Finally took the car around the neighborhood for the first time. Still getting an issue where the car doesn't want to stay idling. Essentially coming up to a stop sign if I put in the clutch and release the gas to start braking the car will die. Will start right back up, and proceed to do the same thing at the next stop if I take load off of it.
I tried (for a second time) to run the car with the Maf unplugged today, this time it actually fired up and seemed like it would idle fine at roughly 1200, just super rich obviously. Plugged Maf back in and received the previous surging of roughly 300-500 rpm from wherever I held it, where putting the clutch in will kill the car as it settles idle back to 800 and then trails off. So puzzled there as it wouldn't fire up the last time I tried that. The throttle body also "chirps" like crazy with the MAF in.
On Sheet 1 you have four graphs:
RPM | MAF/Throttle Position
STFT | Voltage
Apologies in advance, export to google drive didn't turn out great. Can I correlate what I have here to anything? To me it looks like both the MAF and Throttle Position follow the rpm accordingly. But am not sure about the STFT, note that the car probably doesn't have a drive cycle on it since I started troubleshooting. If I need to get more time and more data on it to get a clearer picture I can do that, should get that on Thursday.
Since the last time I have:
-Replaced Fuel Pump to fix leak
-Installed 2.0Kw Starter (not really as troubleshooting)
-Troubleshooted All Vacuum Solenoids and Hoses, only Vfad failed, left in, replaced a couple cracked hoses, and vacuum tested lines, and actuators with a vacuum pump per the forum diys
-Re-cleaned ground under airbox, and at both fender wells.
The data log looks ok to me, for a car that's on itsefirst drive cycle.
You know it's kinda normal for it need to learn to idle and not stall after you disconnect the battery. One drive cycle should do it though.
Have you reset the ESS profile yet? Key to ON, press brake pedal 20 times within 8 seconds or just keep tapping it until the oil pressure gauge does a full sweep. That's all I can think of at the moment.
The O2 sensor voltage won't tell you much, the front O2 is a wideband that signals in current. I don't think it's even up to temperature in your log, so not in closed loop.
How old is the gas in the car? May need some freshening up
The data log looks ok to me, for a car that's on itsefirst drive cycle.
You know it's kinda normal for it need to learn to idle and not stall after you disconnect the battery. One drive cycle should do it though.
Have you reset the ESS profile yet? Key to ON, press brake pedal 20 times within 8 seconds or just keep tapping it until the oil pressure gauge does a full sweep. That's all I can think of at the moment.
The O2 sensor voltage won't tell you much, the front O2 is a wideband that signals in current. I don't think it's even up to temperature in your log, so not in closed loop.
How old is the gas in the car? May need some freshening up
Thank you as always for the help Loki!
To make sure I understand after doing some research after posting yesterday. The STFT readings of -20% are acceptable as the drive cycle hasn't been complete? I assume a good reading here after a full drive cycle or once LTFTs are set is somewhere between +/-5?
Also, I've read from some other threads here that our O2 sensor on the header is an AFR wideband so it doesn't output a voltage signal. Is the correct value to investigate for this on a scanner simply the Measured AFR? My assumption is this shouldn't osccilate like a narrowband O2 and simply spit out roughly 14.7.
Just looking to have my data logging setup correctly for a drive later this week. I have reset the ESS so I was sorta expecting this. Just wanted to be sure. Gas is mostly fresh, or at least stabilized. Could still have some pretty old stuff in there from when I got the car though.
I got these for my compression test and it seems to misfire a little is it time for a rebuild or does some one have any clues. Ive changed sparkplugs, coils, the aftermarket sparkplug wire was too short so i have the ones on for right now. Any ideas would be great for my next step.
Took me a while (as usual) but got a temp tag for the car and took it for a drive last night. I should preface by saying that i got the car around the neighborhood for real a few days back and after a few loops it stopped dying when I pulled up to a stop sign, but was definitely running rich.
Switched that test to an actual decent road last night, speed limit is 45. Worked as high as 4th gear in most cases. Since Google Drive did something weird to my spreadsheet last time, should be attached as an .xlsx now. Put about 25 miles on it.
Observations:
- Car still dies at a stop sometimes, and it seems to be linked to the clutch pedal. Putting the clutch pedal in will usually cause it to drop down to as much as 500 rpm before stalling. It will idle with a surge if I leave it in neutral with the clutch out. It doesn't do it, every time, and does start right back up afterwards.
- Still very rich, LTFT seems to hover between -10 to -18% which leads me to my next observation.
- Can't find a reading for the upstream Wideband O2 sensor. I was able to find the Downstream (which unfortunately didn't get logged with Torque Pro but displayed between 0.6-0.8 V the majority of the time ) but no O2 option would display anything other than 0 for the Wideband. Can Torque find this sensor? I suspect if that isn't reporting then the fuel trims would be crap no? However, Torque's "emissions check" displayed the O2 Sensor as "complete". Which it said was "incomplete" the other day.
- No hesitation or anything that seemed bothersome to me when pulling up to 6k+ rpm or more.(I started logging after giving it the beans for a bit hoping to clean things out).
- Car threw two codes, P0850 pending, and P0410 (I didn't hear the vacuum cleaner when I started the car so no surprise there haha).
Today I got back under to check the Neutral Switch a second time and it again tested good, open when out of neutral, and 150-200 Ohms when in neutral. I pulled what I believe is the clutch position switch? the red one closer to you when sitting. It reads open at everywhere until almost fully extended at which case it is closed (continuity). I will need to see if that is what I should expect (haven't searched for it yet). My understanding is the other switch is the clutch start safety which I assume would only keep me from starting the car if it failed?
Anyways, am I on the right track to suspect the Wideband O2 to be giving me that rich condition? Still working on the surging and stop stall.
The wideband reads current, not voltage, so possible you won't see a raw value for it.
Just read AFR, that is its output. Commanded and actual AFR should be roughly the same. If they're different, then the ECU isn't able to reach the commanded AFR even with fuel trims, so somehow you're getting too much fuel.
In general the ECU is trying to lean things out with - 17 trim
It's possible you have a problem in a single rotor, so the shared O2 sensor can't make heads or tails of the alternating rich and lean pulses.
Loki thanks again for the help, I really appreciate your time!
I added the AFR Commanded to my AFR tab, the trend is the same (it spikes with the Measured) but definitely doesn't veer from 14.7 like Measured does. I'm sure you already recognized that but I had to visualize it. Attached for anyone wanting to see it.
So it sounds to me like you suggest running down the fuel system as my next step? Will do my research there and see what I find.
So when you let off the throttle, the engine stops injecting fuel and measured AFR goes to 20 while commanded stays at 14ish. This is normal.
Do you still the stock intake and if so (or even if not) does it have the 2 straightener screens before the MAF?
I'm not convinced it's the fuel system, but by all means, feel free to explore. Maybe a leaky injector.
I am running the stock airbox, with clean filter and both screens. I haven't noticed any damage on them.
The Maf's readings should be charted in the attachment above, I didn't notice anything weird. I can convert the numbers to g/sec if that helps, currently in cfm.
Past pulling injectors and checking the neutral switch's signal at the Ecu's connector I'm kinda stuck on where to go if this doesn't solve my issue. And I'm not optimistic I will discover anything from the fuel system but we shall see.
10-27-2018, 02:01 PM
Hello! Can I trust these Compression Numbers?
