Burned Valves

Tunes67 · Post by **Tunes67** » February 8th, 2005, 8:13 pm

As you have seen from my other posts, I likely have a burned valve in my #3 cylinder on my 94 RS DOHC 1.6L. Apparently Tric has also had this same issue and when I ripped the head off a 95 RS this morning at the junk yard I discovered that it too had a burned valve on the #3 cylinder. There are numerous things that can be causing this. Oil hardening on the valve. Too lean of fuel mixture. Cheap gas, dirty injectors, etc.. Basically I am wondering if any of you MX-3 Gurus have nailed down the reason this is apparently so common on the DOHC 1.6L in the #3 cylinder. Seems just to common to just write it off as "Sh*t happens" I will call the Mazda service department tomorrow and see if they have a clue on this as well and will post back any findings I dig up.

Tunes67

Tunes67 · Post by **Tunes67** » February 9th, 2005, 2:41 pm

Ok.. found out a bit of info.. somewhat still speculation.. but from what the mazda tech and I can deduce.. here is what is causing the burnt valve problem. As we all know.. Mazda recommends that we replace the O2 sensor every 60k miles. What happens is that as the O2 sensor ages, it loses its ability to accurately measure the amount of oxygen that is present in the exaust during normal operation. The result is.. the O2 sensor is reporting a RICH fuel mixture issue to the ECM.. which then compensates by leaning out the fuel mixture via signals to the fuel injectors. Over time.. this will cause the valves to get to hot and eventualy pit and degrade.. resulting in a valve that wont seat & seal properly. This situation can also be enhanced or "sped up" by a malfunctioning EGR valve. While neither I nor the mazda tech can provide definative proof to this scenario.. I can say that the O2 sensor in the 95 RS I disassembled at the junk yard yesterday was the original sensor and the car had 99k miles on it. The O2 Sensor in my car is the original sensor and the car has 121k miles on it. How about you Tric? Are you still running the original O2 sensor in your ride? Or have you gone significantly beyond that 60k mark for the known lifespan of the O2 sensor? Anyway.. From what I have been able to find out, the O2 sensor is the most likely culprit. Granted there are many many different scenarios that can cause valves to burn, but, this problem seems to be typical of engines with higher mileage that are still operating on the original O2 sensor. Also.. for the people that are modding their cars, this probably hasnt been a big issue since they end up rarely running their original O2 sensor due to the aftermarket mods they have made to their exaust systems and or engines themselves. Ok.. I am done rambling now.. hope this helps someone out before then end up needing a valve job.

Tunes67

What · Post by **What** » February 9th, 2005, 7:21 pm

the single wire o2 on my sohc looks pretty original with around 160k miles, and when I hook up my wideband o2, I get very around 14.6:1, .1 off where it should be.

For significant temp change at cruise, the a/f would have to be noticeably out. I'd look to other areas unless all cylinders are burning lean to prove otherwise. Fuel distribution in the manifold shouldn't be that bad.

If your motor sat for a long time, corrosion is an easy winner.

I haven't seen enough of the dohc to make a better guess for you though.

Tunes67 · Post by **Tunes67** » February 9th, 2005, 7:40 pm

Thanks for the reply. Does the SOHC engine use a different O2 sensor or ECM than the DOHC? My O2 sensor is a single wire type as well. Are the valves in the SOHC head bigger than the valves in the DOHC? There are a lot of variables here unfortunately.. But so far the only common denominator I have been able to point a finger at is the O2 sensor (possibly the EGR valve as well) But neither I nor the tech at the mazda dealership can claim any certainty on this. On the positive side.. at least I will have a rebuilt head outta the situation

Btw... there is evidence in both heads that all cylinders are running lean.. just apparently the exaust valves in #3 cylinder seem to be the worst affected.

Tunes67

What · Post by **What** » February 9th, 2005, 10:06 pm

I don't know if o2s are same, but I wouldn't expect a large difference other than being slow to respond. Don't know about valve size or how burnt the ones in my sohc would be if I took off the head.

Either way, hopefully I get to trash the engine soon. Certain vehicles, I've disconnected the o2 and the a/f was still quite adequate for daily driving. Even played around running quite lean on highway drives. Before saying it was o2 sensor, it would be nice to verify that your car was running lean as you say with a proper wb02.

Even a slight increase in backpressure in that cylinder, retarded ignition timing, improper fuel mixture (too rich or too lean), small spots of carbon on the valve, sulphate deposits, alloy selection and valve construction, etc. etc. lots of possible 'chicken and egg' scenarios on high mileage motors. I personally think Mazda set timing too far back, and runs too rich WOT. If you are handy and have nerd friends, you could determine the cause of failure and likely which caused which with proper metallographic examination. I had a friend in charge of failure analysis and accident reconstruction that took advantage of company equipment to do failure analysis on a valve from his nitrous swift. He was simply wondering if the valve failed before the spark plug tip.

I say get some guys to put their cars on wideband and go from there. I'd give more specific options, but I'm just beginning to wade in the mazda parts pool.

Tric · Post by **Tric** » February 9th, 2005, 11:54 pm

I still using the same O2 that came with the car!! it has over 500,000km on it!! but I know its running right because everything is a nice tanned color and my spark plugs are not burning off faster then normal!! A set of normal NGK plugs should in normal operation the plug gap increases about 0.001in (0.025mm) for every 4000km of driving! with a nice tan color! if your running lean that gap happens alot faster!

mitmaks · Post by **mitmaks** » February 10th, 2005, 12:32 am

I went to http://www.schucks.com and looked up o2 sensor for gs and theres one said "recommended change interval 100K" so i guess o2's should last about that much

Tunes67 · Post by **Tunes67** » February 10th, 2005, 2:12 am

Thanks again for your thoughts guys.. I do appreciate it

Tric - See all my plugs are showing a nice tan as well with normal gap as well.. yet still a burnt valve on #3.

What - Thats interesting news on the retarded ignition timing at WOT. I typically run 4-6 degrees of extra advance on my cars to increase performance and mileage.. sounds like when I get done with the head job I wont have too much problem doing this. I always run premium gas as well to avoid detonation. Will have to see how it goes..

Mitmaks - I looked up a O2 sensor at that site for the RS I have, they have a couple of different ones ranging from 60k to 100k.. but I was told by the mazda tech that the factory O2 sensor is supposed to be changed out every 60k miles - this usually occurs with those cars that people actually stick to the dealer maintenance schedules. I do plan to upgrade to a better O2 sensor this time around - heh.. if all else fails throw money at the problem LOL

Likely I wont ever find a definative end all answer to this.. and now that I have a second head to build for a spare.. ehh.. doesnt really matter.. more of a curiosity than anything else.. though it bugs me that there can actually be a burned valve issue or even a compression issue and not get any kind of diagnostic or CEL codes from the car at all.

Again.. thanks for everyones input on this

Tunes67

Yoda · Post by **Yoda** » February 13th, 2005, 10:46 am

The intake manifold on DOHC in the MX-3 could be part of the problem. The runner of the manifold are cast like an equal lenght 4-1 header in reverse. If there is a slight error in the casting it could cause a lean condition in one cylinder at certain rpms. The earlier B6 DOHC couldn't be used in N. America because they didn't meet EPA emission spec after '92. The change in the MX-3 was that the engine became leaner burning than previous engines. This combined with a intake runner that leans out one cylinder may be the cause of your problem. My guess is that is the engine was operating above a certain rpm more often that what Mazda's engineers had assumed in their calutalions you might get a blow torch effect only at high rpm's which would still leave normal burn indication on the plugs. The SOHC has a similar issue but on the first cylinder caused by the intake manifold being too large at that end of the manifold. This causes a drop in air flow into the runner for the first cylinder on the SOHC but this can be easily fixed on this engine with epoxy filler where on the DOHC would mean complex flow matching of the runners. O2 sensor do tend to degrade after about 60000km (40000miles) and go downhill from that. Single wire O2 sensor tend to be very narrow band meaning very slow to react and inaccurate when not at proper operating temp. I've converted over to a 4 wire on my SOHC just for this reason.

Tunes67 wrote:Thanks again for your thoughts guys.. I do appreciate it

Tric - See all my plugs are showing a nice tan as well with normal gap as well.. yet still a burnt valve on #3.

What - Thats interesting news on the retarded ignition timing at WOT. I typically run 4-6 degrees of extra advance on my cars to increase performance and mileage.. sounds like when I get done with the head job I wont have too much problem doing this. I always run premium gas as well to avoid detonation. Will have to see how it goes..

Mitmaks - I looked up a O2 sensor at that site for the RS I have, they have a couple of different ones ranging from 60k to 100k.. but I was told by the mazda tech that the factory O2 sensor is supposed to be changed out every 60k miles - this usually occurs with those cars that people actually stick to the dealer maintenance schedules. I do plan to upgrade to a better O2 sensor this time around - heh.. if all else fails throw money at the problem LOL

Likely I wont ever find a definative end all answer to this.. and now that I have a second head to build for a spare.. ehh.. doesnt really matter.. more of a curiosity than anything else.. though it bugs me that there can actually be a burned valve issue or even a compression issue and not get any kind of diagnostic or CEL codes from the car at all.

Again.. thanks for everyones input on this

Tunes67

Tunes67 · Post by **Tunes67** » February 15th, 2005, 3:35 am

Well as of this monday evening.. we finished the valve job on my head. New exaust valves and we reground the intakes (replaced 2 as they were slightly bent somehow) valve grind.. new seals.. new springs.. etc.. we did find one more oddity about the #3 cylinder.. When doing a valve height check using the spring deck (where the spring seats) as the base line.. we discovered that the #3 spring deck was .020 lower than the other 3 cylinders. Basically this means that the springs for the exaust valves were under LESS tension than the other cylinders. While this is not likely to have caused the burned valve in and of itself.. it could be another contributing factor. Since the second head I yanked out of the junk yard also had # 3 cylinder burned exaust valves.. after we get this head out of the cold tank we plan to measure it as well and see if this is something common to the head during factory machining. This issue was easily corrected with a shim under the springs on #3 cylinder. Will keep you guys posted on this.. and Thanks for the info Yoda

Tunes67

What · Post by **What** » February 15th, 2005, 8:13 pm

Unless the profile on the cam was oddly different, that could make a good difference. The valve sheds a whole lot of heat each time it makes a tight seal with the head.

Tunes67 · Post by **Tunes67** » February 16th, 2005, 11:12 am

Well we finished measuring the spring deck height on all cylinders of the junk yard head and found less than a .006 varience across the board. So apparently I just got lucky with my head being the one that was off. What is kinda weird about this whole thing is the spring tension specs.. the manual shows that the exaust springs are under less tension than the intake springs. This seems a bit odd to me but maybe it was done this way to reduce the stress on the valves that are tortured the most. I cant really say.. but it does seem a bit odd in my mind. Anyway.. I also got my second set of injectors rebuilt & flow tested. All are now showing good spray patterns and a flow rate of 100. Resistance and voltage drops were excellent. With a bit of luck.. The shop will finish assembling and tuning my ride today

Tunes67

Tric · Post by **Tric** » February 17th, 2005, 3:05 am

so how does your ride run after the new install of the new head?? I know it made a world of diffrance for me!!

Tunes67 · Post by **Tunes67** » February 17th, 2005, 3:58 am

I just got her back today.. needs a bit of fine tuning.. (the shop wasnt asked to do the tune up or timing.. etc.. ) But the idle is smooth now and she doesnt seem to be so gutless between 50-60 mph when getting on the freeway. Definately a vast improvement. I also noticed that the trans seems to be shifting smoother from 2-3 and then also from 3-4.. no lunging like it was before. Just need to time it and set the idle.. will get that done tomorrow. There does seem to be a vibration when at curb idle that I can feel in the steering wheel (if I let go of the steering wheel I dont notice it at all) So I may still have a air leak or vaccum leak somewhere, but this is nothing compared to how it was running before hand. Gas mileage has been impacted as well. I took her on a 75 mile test drive and barely touched a full tank. Hoping to see around 400 miles to the tank full. In the end.. I didnt get quite the good deal on the car that I thought I had at the time.. but even with tires and the head job.. I spent a little less than blue book for her and she should last for a long time.. Just want better rims.. better tunes and maybe even a paint job

Tunes67