blown engine post successful dyno tune, ideas on what might have happened?
 

Blew my stock internals 13B yesterday during the first highway run immediately after having it successfully tuned on the dyno. At a bit of a loss as to what might have happened. Specifically regarding what would have changed between many dyno pulls then instant street blow up. Looking for any suggestions that might point me in the right direction. :scratch: I'll give a rundown of the situation and what I know:

Background: Engine was built 5-6 yrs ago, basically stock, maybe 30k miles on it. Engine was running great. Showed great vacuum at idle. No indication of any issues. It has been running on a street tuned pfc for for 2.5 yrs now, 12k miles, at 13lbs boost. I routinely do wot pulls on it in all gears, 1 track event, no issues. Car has fmic, wi, supra pump, 550/1300s. Fuel filter has about 12k miles on it.

Yesterday I swapped the PFC for an Adaptronic. Got it up and running with a very reputable Adaptronic/rotary tuner. I had also just installed an Innovate LC2 wideband which at the time we thought was calibrated/working. On the way to dyno, we attempted to do auto tune which failed miserably. AFRs showed lean if I added extra throttle. Manually made adjustments to the map on the way to the dyno, took it very easy on the drive, had it running reasonably well on the way over.

On the drive over, noticed that AFRs were lean in the boost areas no matter how much fuel was added to the map indicating something was up with the fuel system. Car hadn't had the recommended fuel wiring mods (learned about these post blowup :icon_tdow).

No idea if the car had always been running lean with the PFC or if this was only since adding the Adaptronic. I tend to believe that in 2.5yrs of lots of wot pulls and one track event it would have blow up if it was always running lean.

Told the dyno shop, an rx7 specialist, about the lean situation. They made a jumper for the high speed fuel relay, basically causing the pump to always run in high/battery voltage mode. In hindsight this doesn't make much sense as a fix since the car would switch to high voltage mode on it's own w/o the jumper when needed assuming it's all working properly.

Found out the Innovate was giving bad readings, sensor malfunction, who knows. Probably explains why auto tune was a mess. Also means, what I said above about running lean might be garbage info. Swapped to the shops wide band.

Here's where it gets weird. We did many pulls on the dyno while tuning the car, maybe 10 full pulls. The engine was flawless. Put down 340hp/300tq on a dynojet. AFRs kept conservative under load, around 10-11. Everything looked good.

Got the car off the dyno, turned it off. 15 min later headed off to the highway. Did a couple pulls, 3rd-4th gear. Felt reasonably strong, but not the kick I would have expected from an additional 80 or so whp from what it had before. On the next pull, engine lost power and that was it. Car won't start. Sounds like a blow seal when you try to restart it.

Didn't have the laptop hooked up on the highway drive as its battery had died. Innovate was probably still giving wack readings anyway so that wouldn't have helped diagnose, but definitely seems like the car went very lean and blew a seal.

So between the dyno runs and the street pulls what could have changed/gone wrong?

For sure I need to do the fuel electrical rewire, but I have a hard to believing that was the issue since it ran fine on the dyno. We left the fuel pump high voltage jumper on so it was definitely getting the same voltage as it was on the dyno.

The car was tuned to make a lot more power than it did with the pfc tune, so it's definitely possible that an issue that had exist previously only reared it's head under the added stress.

Pulled the fuel pump, nothing is melted, shorted, sock looks fine.

When testing voltage at the pump, we see high voltage with the jumper in and ignition in (more like 11v). Strangely, w/o the jumper, when turning on ignition we don't see any voltage. We should see the pump prime as I understand. We didn't take the key out of the ignition completely between turning to on. I'm not sure if this is necessary to cause the pump to re-prime.

Any thoughts/suggestions would be appreciated. :scratch:

Was the gas tank low ?

You said you're running adaptronic with an innovate wideband, but the innovate was bad so you tuned with the shop wideband. When you buttoned it back up and got on the road, did you plug the innovate back into (or was it still plugged into) the adaptronic?

The adaptronic references the wideband, so if it was bad, it could have hosed up the tune, no?

Good question. Forgot to say, fuel level was just under half so fuel slosh shouldn't have been an issue.

ptrhahn, the adaptronic was running open loop so it wouldn't have been using feedback from the wideband to adjust the map.

Closed loop feedback is generally not used above a few psi( set this way bu default and how I set it too).

Generally there is more load and lower temps on the street so this is also a factor that needs to be checked after dyno tuning. A lot more airflow over intercooler vs fans, and dyno rooms usually get hot. Lastly not uncommon to see more boost pressure on the street.

take a look at your plugs. of course you will probably find one of the rotors lean but it would help in the diagnostics process to know if both rotors or one rotor was lean.

i do recommend to anyone running an aftermarket ECU that they log fuel pressure,

Howard

Lower temps on the street meaning more fuel would need to be added to prevent it from running lean? It's possible but can't say for sure. On the dyno, temps seemed to be around 30C with the water running. With the pfc previously I noticed it a bit lower at times but generally not massively different.

Ya checking the plugs to see if both rotors went lean should provide some indication as to what happened. I'll report back once the engine is out.

As for the fuel relay jumper, is my understanding of that correct? It wouldn't have provided any "fix" for the fuel issue? As I understand, the only thing it would fix would be if the high voltage trigger or relay wasn't working for some reason. With 13lbs on the pfc would I have been able to get enough fuel if the high voltage mode wasn't working? The factor that the jumper was still in makes that irrelevant anyway. Anything else that we could cause fuel voltage to drop suddenly on the street?

You are correct that forcing the resistor into high speed mode does nothing.One needs to add a larger 12V supply to the fuel pump from the battery. One should always be logging fuel pressure, esp since its so easy with the Adaptronic. If you have a wire in model and are not running an OMP you can wire the sensor directly to the ECU. If you are using a PNP you need an innovate signal conditioner(SSI).

Its hard to say exactly what happened to your car without anyone here being there....

Ya, totally understand, hard to diagnose wo being there. I was mostly wondering if there were any obvious gotchas that I was missing given that we weren't able to richen up the map on the way to the dyno, then it ran perfect on the dyno, then went lean again on the street and blew up. I'm at a loss.

I have the pnp model. Will definitely add the fuel pressure sensor. I assume you can add ignition cut if fuel pressure drops below a set threshold?

First thing that immediately stands out to me is your injector setup. Are the 1300s bored out 850s? If that's the case there is a good chance one stuck. They have an extremely high failure rate. Best case scenario they stick open and flood your motor. Worst case they stick shut under boost and you blow the motor. I've had over a dozen sets fail on me through the years to the point where I will not recommend them, install them, or tune a car with them.

No matter how thorough you are on the dyno, you can't 100% replicate real world conditions including, load, airflow, temperature, etc. I like to street tune for drivability, then finish up the high speed stuff on the dyno. I always verify the tune on the street afterwards. A tailpipe wideband like most dynos have isn't accurate enough to tune off of in my opinion. I only use it as a second reference. And if the dyno doesn't have load control, it is basically useless for tuning.

Sucks about the motor. I would look into your fuel setup and your spark plugs to get an idea of what happened. And definitely fix the wideband.

Hey Ihor,
Actually, funny you ask... this is actually the engine you built up for me 6 years ago and did all the mod work to (fmic, pfc street tune, injectors, wi, etc) about 3 years ago now :) Just to be clear to people reading this, the engine has been flawless since the time irp did all the work on it. If you recall, this is the black fd my dad used to own that I've since taken off of him. Given that, you'd probably know best about the injectors, but I'm pretty sure they're proper 1300s. Have you ever had one of those stick?

Definitely a huge fail on our part not having the laptop hooked up when we first hit the street. In hindsight that was a really stupid (and expensive) move. Did a couple pulls before it let go. Could have let off after seeing something wasn't right on the first pull.

On the dyno we had the shop's wideband plugged into a 2nd bund on the downpipe. The readings should be 100% accurate. I believe the dyno has load control. Many many rx7s tuned on it.

I switched away from the 1300s a few years back because of the issues the had. If I recall correctly there wasn't anything else out there at the time to fit a stock setup that needed to pass emissions. I would suspect the tune was off due to the wideband but still send the injectors out to test. Any reason you switched from the power fc? There really isn't a point for a stock turbo car.

Definitely a huge fail on our part not having the laptop hooked up when we first hit the street.

No way. Sounds to me like you got a complete dyno tune and had the go ahead to drive it without a list of fix this fix that.

Ripping around the block directly after the tune is exactly how the last dyno tuner I went to verifies his tune and the customer vouches their satisfaction with it.

Got the car off the dyno, turned it off. 15 min later headed off to the highway. Did a couple pulls, 3rd-4th gear. Felt reasonably strong, but not the kick I would have expected from an additional 80 or so whp from what it had before. On the next pull, engine lost power and that was it. Car won't start. Sounds like a blow seal when you try to restart it.


To me it sounds like they did not re-plug in the MAP sensor or un-tee it (so it is open to leak) from when they had been logging boost on the dyno.

Please take a look at that.

So you're saying it might have the bored out 850s on it? I'll have to check once it's apart. The duty on them with the new tune would definitely have been much higher than anything they've seen before. Is it possible that could have caused one to stick closed?

The wide band was hooked up to the downpipe. Why would it be off?

Reason for switching to the Adaptronic... newer ecu, something new to play with, get it ready bit by bit for a future efr :) Mostly because I was going to get the car tuned on the dyno and figured I might as well swap the ecu at that time.

BLUE TII, Ya I agree with you. Had the green light, but I guess for this unfortunate situation, taking it easy at first and double checking some logs might have avoided an expensive bill.

Interesting thought regarding the MAP sensor. I'll ask the shop to take a look tomorrow. We did have a T on it during the dyno runs. hmm

a malfunction at the MAP sensor would be very close to the top of my list.

there are two hose connections that will break your motor if they disconnect:

manifold to fuel pressure regulator

manifold to MAP sensor

both should have double tie wraps at both ends

if your MAP sensor hose came loose in boost your motor would think it was NA and fuel would be greatly reduced as well as timing would be jacked.

similarly if the hose to your fuel pressure regulator were to disconnect you would go way lean.

given the shop did work w the MAP sensor hosing...

sorry for your loss and hope you are up and running soon w a solid understanding as to what happened.

Howard

Most dyno widebands go in the tailpipe for ease of use. If you had it in the downpipe that is way better. Are you using the stock or the onboard map sensor?

Ya, it's looking like a potential map sensor issue is on the top of my list now. Will ask the shop to report back on it.

Wideband in the exhaust seems like a terrible way to tune unless maybe you're running wo the cat. This car still has a hiflow cat on it.

We used the 4bar sensor built into the adaptronic ecu. My hose routine was pretty indirect though. I wonder if a potential lag in the map readings could have caused the issue, although it didn't seem to be an issue on the dyno. The routine goes from the manifold, to the drivers side quarter panel, into the driver's footwell then over to the stock ecu location.

I believe there is a length requirement for the map sensor and that routing would exceed that by a long shot. Sounds like it wraps the entire car almost.

ya, doesn't seem very ideal. There's a predictive table in the Adaptronic configuration to deal with the reading latency. Anyone configured the 3bar GM MAP with the adaptronic? I have that sitting in the engine bay currently, but in Andy's (Adaptronic) videos he seems to indicate that calibrating it properly can be a pain. He was recommending that people use the stock MAP or the built in Adaptronic MAP.

Where is your pressure reference for the map sensor?

On the back side, towards the firewall, by the throttle body/start of the uim. Right by the stock map sensor location. I was told this location gives the most accurate readings.

Yes thats fine. Personally I've never had an issue running the hose into the car to ecus that have an onboard map sensor.

Ya, don't think that was the issue. Would have had to have been something more catastrophic like the hose coming unhooked or the T adaptor they used for the dyno readings not being removed properly.

Even tbough you say your wideband is off what were the readings

Most of the stuff I've read about the innovative wideband is that after 500hrs it loses the accuracy you would want. The tests I've seen show .5 AFR swing. So if you tune for 11 it could be 11.5 or 10.5 AFR. Actual. Most consumer grade widebands are just rebranded Bosch LSU 4.2 widebands.

It's also what I currently have in my car just monitoring with a gauge. I am going to be upgrading mine to a production grade NTK which will be much more accurate; even after 500hrs. It's long since been used by tuners and was difficult to find and very expensive until just recently. It can be found under the AFR500 name plate.

It's probably not likely to be the problem that caused your motor to blow as you were not using it. But there isn't much point of upgrading an ECU if it's being given less than accurate feedback readings to tune by.

Good luck and sorry to hear about your motor. I have felt that pain too many times.

"Most consumer grade widebands are just rebranded Bosch LSU 4.2 widebands.

It's also what I currently have in my car just monitoring with a gauge. I am going to be upgrading mine to a production grade NTK which will be much more accurate; even after 500hrs. It's long since been used by tuners and was difficult to find and very expensive until just recently. It can be found under the AFR500 name plate."


i one thousand percent agree w the above and use NTK widebands. the Bosch 4.2 need re-calibrating very often and doesn't read well in the 10 handle.

as per Zoom Zoom the NTK setup is currently available as AFR500.

Howard

I've noticed this as well as I've had to install more than one wideband when I had the innovate setup. My innovate setup always had quirks. I had to use heat sinks. Sometime when I got heavy on the throttle, it would lose signal occasionally. I also got tired of doing the calibration. Ever since I installed an AEM unit, I've had zero problems. Plus you don't have to calibrate them amd I was able to remove the heatsink.

fuel pulsation causing lean spots?

I know this is a bit of a derail, but if you go with a different wideband controller than Innovate but wanted to log egt's or fuel pressure also, what are the options? Only thing I can figure is use the Innovate analog module to log the 0-5v output from another brand wideband and convert back to AFR inside the adaptronic. This seems like a step back however since a serial digital signal should be more accurate communication to an ecu than a sensor reading converted to digital inside the wideband, then converted to analog and back to digital for the ecm to read. Am I wrong?

You ask: What Happen?
Rotary Murphy's Law
Open the engine, find what failed and then find the possible causes from there.
Common sense dictates, it easier than diagnosing prior to that, with so many variables.

It's funny because I had the AEM UEGO and sold it to get the innovative unit for the plug n play feature of the adaptronic. I never had a problem with that thing, I'm now considering selling the Innovative for another...

What is the timing at 5500-8k rpm at full boost, WOT? What is the leading/trailing split value?

Are you positive there isn't an exhaust leak pre-wideband?

Compare the timing from the PFC vs the new tune on the Adaptronic. Are there any noticeable differences?


Finally,

I read WI on the list. When does your water injection begin, and how much is sprayed. Was the car tuned with the water injection spraying?


https://cimg5.ibsrv.net/gimg/www.rx7...e9cfc9dae8.jpg


This is an rx7 basemap, post a pic of yours. Unless you don't want to share your tuners info.

Im assuming this car is making around 12-13psi to achieve those horsepower numbers. It just seems like regardless of the brand or calibration, if its anywhere between 10.5AFR to 12.5AFR, it can't possibly be that catastrophic to blow the engine from running slightly lean. Still sounds like a timing issue, or MAP reading as mentioned earlier

https://cimg4.ibsrv.net/gimg/www.rx7...a16dc88d6b.jpg




Is this the model you're referring too?

The model that was so popular with tuners was the NTK Powerdex. It was discontinued and replaced with something different. The only way to get the NTK Powderdex is to buy the AFR500 which was created by the original Powerdex manufacturer. You can read more details on Ballenger Motorsports website.
If you decide to buy the AFR500 be sure to upgrade to the Production Grade NTK sensor. They package it with the Bosch 4.2 LSU in the base package. $50 upgrade gets you the NTK sensor you need with the AFR500 controller to get the best wideband for the money on the market right now according to many who are better informed than myself.

I purchased mine with the optional shorter 7 foot harness vs the standard 13 foot as mine will be "in car".
Attachment 614919
Attachment 614920

Beautiful! !!!!

Updates?

Hey guys, sorry for the hiatus. Lots of great comments here. Appreciate the help from the community. The car is almost back together now w/ some new goodies. If you're going to throw money at it, you might as well throw a lot of money at it... I guess...??

Upon pulling the engine apart, it was apparent that both rotors blew an apex seal. That narrows things down a bit. Clearly a no fuel issue rather than something going bad with a particular injector. I probably should have checked this sooner, but it wasn't until we tried to get the car started with the rebuilt engine that we noticed that the fuel pump was still not getting voltage. Eventually traced this to a bad EGI relay. Fuse was ok, but relay was dead.

So couple questions:

1) The EGI relay suddenly going bad can cause the engine to self destruct? That seems a bit crazy (by design). No EGI power means no fuel pump which is obviously a big problem, but there's also the EGI fuse and fuel pump fuse that could potentially go out. I can't imagine if any of those go out it means goodbye engine. Is the ignition somehow connected to this circuit such that if the fuel pump loses power ignition stops firing? Seems like that might save the engine. So I still don't understand why I lost the engine.

2) What can cause the EGI relay to suddenly fail? Just bad luck? The fuel pump relay was jumpered into high voltage mode at the dyno. Could this have somehow impacted it? I still question why this was even necessary because even with the jumper it still wouldn't get around any voltage drop issue in the stock circuit wiring. Only seems like it would help if the fuel pump high voltage relay died, which it hadn't. As a reminder, the car had many full range pulls on the dyno, afrs looked great. Only when taking it on to the highway did it let go on maybe the 2nd pull.

I'm rewiring the fuel pump and installing a new EGI relay, but still not fully confident that we've solved the issue w/o knowing exactly what went wrong.

Btw, here's the wiring diagram. I follow the fuel pump / relay wiring, but get a bit lost with the EGI wiring. I see a line from the EGI relay going over to the ECU, some "solenoid" and "el unit". Maybe the ECU knows to cut ignition if this line goes dark? I'm running the Adaptronic PNP. Hopefully that logic is built into it. Can anyone confirm? It also looks like that line enables triggering on the "A/P relay" which looks to be part of the A/C system.

https://cimg7.ibsrv.net/gimg/www.rx7...681734a25c.png

It doesn't appear that the EGI relay controls the fuel pump relay. The power to close the circuit opening relay is jumpered off BEFORE the EGI relay and then grounded by the ECU. According the the circuit wiring I am unsure how a bad EGI relay could cause the fuel pump not to run, although it would cause the injectors not to fire.

Ya good point. I thought about what I wrote afterwards and it didn't quite make sense. Is it possible that the egi relay failing could pull the trigger line down and also cause the circuit opening relay (provides power to the fuel pump) to trigger off since they're on the same line? You're right in that an egi relay failure should have caused the ignition to fail which I would think would have prevented a lean condition. I'm grasping at straws here...

I think the egi relay malfunction disabled your injectors, causing the lean condition. The coil packs derive their power directly off the ignition switch and then triggered from the ecu. I think your root cause is correct, but the assignable cause is the injectors failing to fire instead of the fuel pump. If the egi relay failed in such a way to draw down the voltage I would think you would have blown the 15a engine fuse, was it blown?

I went over to the shop today to take a look at the car first hand. Trying to debug the car via second hand information wasn't working very well.

So turns out the only issue was, as you said, the 15A engine fuse was blown. The egi relay was fine once the fuse was replaced. I think your theory is sound. Engine fuse went out, therefore injectors didn't fire, engine went lean. What kind of insanity is that? Fuse blows and the engine is toast? How can it be designed like that? Feel like I'm missing something here. :scratch: Why is a 15A fuse needed to just trigger a couple relays anyway?

I have no idea, but I think a better circuit design would tie in another relay to that fuse that, if the fuse blew, would drop power to the coils and the ignitor, thus shutting down the engine and preventing that lean condition. I can only imagine how many engines have been blown because of this very problem. I think I will fix this on my car, I should sketch up a schematic and post it for others to critique. I am no expert on rx7's but I am a service engineer for Cummins by trade so I have to deal with poor circuit design and software on a daily basis.

I am in the process of upgrading my fuel pump, fuel pump wiring including a 40a relay and inline fuse. The stock wiring, relay, fuse etc aren't up to the task of handling all of the amps the bigger pumps are drawing under higher pressures.

I have the parts and am in the process of documenting the install in my build thread if you want to see the process.

I'm using the supra pump. Would the current draw on the pump change as you're using additional fuel? I figured the current draw would be constant at a given voltage. The pump flows at a certain rate at a given voltage. I wouldn't think this would change as the injectors draw more fuel.

But, in general, I agree, it's very likely I over taxed the stock wiring and caused something to pop. Stupid on me for not rewiring the pump beforehand. Expensive lesson to learn. Not to pass the buck as I should have better understood what was going on at the time, but the shop gave me the silly fuel pump relay jumper and I assumed somehow that guaranteed battery voltage to the pump.

I have power to the pump wired directly from the starter positive terminal (goes straight to battery +) now w/ a relay triggered by the Adaptronic turning on. Hopefully that's bullet proof.

I will be selling the "discontinued" Powerdex Wideband kit with NTK sensor soon, if any are interested.

I haven't read through all this but if I was a betting man I'd start with the fuel pump.

Poor electric current may be killing the pump.

Hopefully you have fuel pressure gauge but if you don't get one and keep an eye on it.

Here's a simplified version of the wiring diagram for the fuel pump, with some color added to highlight how certain fuses and relays affect the distribution of power:

https://www.rx7club.com/attachment.p...p;d=1459008183

if you look at the switched side of the EGI relay, it does four things.

1. it powers the injectors, the ecu grounds the injectors
2. it powers the ecu, pin 2b.
3. it powers the air pump relay, if you still have it
4. it powers all the other solenoids on the engine, like the BAC valve, and the solenoids if you still had them

the coil side of the relay gets switched power from the ignition switch, through the 15A engine fuse, and this wire also powers the fuel pump relays and ignition. the actual fuel pump power is a different source.

if the switched side of the EGI relay failed on a stock car, its likely it would just stop running (you can have the car idling and remove the relay and see for yourself). it is possible that in a modded car at full boost something bad might have happened