deddex

iTrader: (2)

Hello,

I'm struggling with a few issues with my second attempt at getting my FC Rx7 running. The car was running very well, the engine was smooth and I was beginning to tune the car in boost. After working with some of the in-boost tuning (wastegate pressure only 10.8psi max) the car began to run rough, and would not idle well. A compression test of the car indicated that there is low compression on one face. In trying to trouble-shoot if this was due to a stuck side seal, or if I will have to pull the engine to find out the cause I found a couple issues. The trailing coils do not appear to be firing when the engine is running on either rotor, based on a timing light that shows signal from both trailing coils. These same coils with the same wiring setup were working when tested a week before. This led me to check the composite logger in tunerstudio, and I believe that i'm getting extra triggering, but I'm not very familiar with reading these types of logs.

Based on the low compression on one rotor face it looks like I'll likely have to pull the engine to find the cause of that issue, but I would like to figure out these other issues if possible before pulling the engine. I will list the vehicle setup below and attach a composite logger log to this post. Any advice on these issues is greatly appreciated, on both the low compression and ecu-related issues.

Thank you,
-Tyler

1987 Turbo II
S4 engine with RA super seals
New S4 housings
all new seals
Turblown EFR8374 turbo kit
Dual EGT probes
4x IGN1-A coils mounted in factory trailing location
New MSD 8.5mm plug wires
DIYAutotune MS3 with MS3X

deddex

iTrader: (2)

I had to truncate the composite logger file to update. It should be attached to this post.

Aaron Cake

iTrader: (1)

Either MegaLogViewer is opening the file wrong or your trigger is something weird. Are you running the stock CAS?

Where does your ignition harness run?

Side seals rarely "stick". In my experience, one face down is more likely due to a damaged corner seal from extreme misfiring.

deddex

iTrader: (2)

I am using the stock cas. It runs through shielded wires to the ECU as per your writeup. The cas wires run from the sensor across the top of the engine to the firewall, then through the passenger side grommet to the ECU, which is mounted behind the passenger seat.

Aaron Cake

iTrader: (1)

I think then MegaLogViewer isn't showing the composite log properly because I'm seeing 3 pulses, a pause, 1 pulse, 3, pause, 3, 1, yadda yadda. Not the pattern from the stock CAS.

If you are seeing extra triggers in the log, have you adjusted the pots on the V3 board correctly? Fully counterclockwise.

The shield needs to be grounded at the ECU end only.

Measure the distance between the CAS wheel and the pickup. It should be about business card thickness. They can be moved a bit closer to enhance the signal.

deddex

iTrader: (2)

Thank you for the response. I have checked the spacing on both of the CAS that I have been swapping between and they both have spacing as thick as a businesscard. The pots on my megasquirt are adjusted fully counterclockwise, I will verify again though that this is the case.

I am attaching a new log, as I can't get the one that I had previously attached to open properly, I had manually truncated it to get around the forum size requirement, and likely messed up the formatting. This log shows what I've been seeing, the primary problem is the multiple triggers that show up in the later section of the log. It appears to have the appropriate pattern at idle, but when I rev the engine it begins to show multiple triggers.

What bothers me most about this is that the car was running quite well then stopped running so well, with no changes to the wiring or etc. I pulled the engine from the car last night and will be tearing it down soon to find and fix the cause of the low compression face. During this time I had thought to remove the shielded wires that are currently on the megasquirt and run two new sets, passing them through a different grommet from all the other wiring, and running them as far from other wiring as possible across the engine bay. Is there a good rule of thumb for distance that they should be away from other power/ground wires?

Thank you for all of the help.
-Tyler

Aaron Cake

iTrader: (1)

Your trigger logs look very good.

Do you have access to a scope? Scope the coil trigger leads to make sure the coils are receiving a signal.

deddex

iTrader: (2)

I'm not very familiar with reading these logs, but what I was able to pick up was that I should see twelve spikes of the blue lines between every spike of the green, which makes sense to me looking at the wheels inside the CAS. My log shows this pattern around idle in most of the logs, but when revving the engine up there are multiple of the green like spikes, much more frequently than I would expect. Those patterns are what have made me concerned that I have an ignition issue. Is this normal behaviour for this type of ignition setup?

Unfortunately I don't have access to a scope.

An update on the engine itself, I pulled and disassembled it, the damage is far more extensive than I had hoped. I'll list the damage below.
-4 broken corner seals
-one sideseal broken in half (I'm guessing this was responsible for the lower compression pulse
-one sideseal with sections broken out lengthwise (the bottom half inside the rotor
-one corner seal with flattened spring (FD springs)
-rotor scarring on all three faces
-Housing has large gouges between the exhaust port and the thick portion of the housing (reuseable with reduced efficiency?)

deddex

iTrader: (2)

I'm attaching an image of one of the log sections I was mentioning, showing multiple of the green triggers. Most of these triggers only have about 5 of the blue triggers between them.

Thank you for the assistance.

Aaron Cake

iTrader: (1)

That's damage associated with misfiring ignition. Sucks. Been there, except caused by intermittent injector, and sadly had to repeat a year later.

I am very surprised you are seeing bad triggers with an MS3X. Unless there is a problem with the wiring (shield not grounded? Or grounded at both ends?), or the CAS...or maybe the pots are not set correctly (fully counterclockwise).

deddex

iTrader: (2)

2016 update!

Car is back up and running, with some significant changes. The engine was entirely rebuilt to repair the damage from the ignition issues I had last year. I also installed the FFE hall sensor kit to try to avoid the ignition issues I was seeing last year. The car started right up and ran, but still isn't quite as smooth as I would expect and I'm still having the issue with the trailing plugs not firing at all. To recap I am running IGN-1A coils, and they spark when triggered individually. I will look through logs and try to get them posted tomorrow, but if anybody has any suggestions about the trailing coils please let me know.

Thank you,
-Tyler

deddex

iTrader: (2)

Attached are my log files and map. They look pretty good to my untrained eye, with the exception that I am only seeing 34 triggers in between each of the missing tooth triggers, is this normal for a 36-1 wheel? Any input would be appreciated, I am really unsure why the trailing coils do not fire while the vehicle is running but will if triggered manually in test mode.

Thank you,
-Tyler

Aaron Cake

iTrader: (1)

Someone more familiar with individual coils will hopefully chime in but looking at your coil settings on the map, they appear correct. "Coil on plug" and "FD mode".

deddex

iTrader: (2)

Thank you Aaron. I have done more investigation today on the car and am still stumped. The trailing coils are now firing, but the spark is quite weak compared to the leading coils. The spark from the leading coils is consistent, and happens regularly. The spark from the trailing is intermittent with "hiccups" periodically and times with no spark being sent. This seems independent of RPM range. The spark is being visualized through the use of an inductive timing light around the plug wires while the engine is running. I believe this is an accurate test, and have tested external spark plugs to verify the spark being strong/weak. When the timing light is not showing any spark being sent the plug with often have some limited spark, but it is very weak compared to the large purple spark normally visible. Below I will list a little of what I've done, and some questions that I have. I appreciate any and all comments on this issue.

If I fire the coils manually the sparks are the same, and I have swapped the signal and plug wires between trailing and leading coils. In this test the strong spark follows the coil triggered as the leading, and the trailing is weaker.

I have run new dedicated 12v power to the trailing coils, instead of feeding all 4 coils off of the same power wire. They are now split with both leading sharing a power source, and both trailing sharing a power source. This has had no noticeable effect.

I have also changed out to new spark plugs, with no noticeable difference.

My trigger logs look great to my eye, but I am certainly no expert on this.

I have read that some people are adding capacitors to the 12V power supply to the coils, I've seen 1uF and 100uF capacitors being used. I have none on my setup, is this a possible reason for the poor performance of the trailing coils?

Thank you,
-Tyler

Aaron Cake

iTrader: (1)

That's puzzling, but I have never used those coils. Almost always I stick with 2nd gen coils.

Do those coils require more dwell or discharge time? That's the only thing I can think of.

deddex

iTrader: (2)

Thank you for the suggestion.

I verified the dwell and discharge time. The coils specify 3.0ms for nominal dwell, and 2.9ms spark duration (I'm unsure how or if this relates to discharge time). I had the settings in the ecu at 3.0ms dwell and 1.0ms discharge time. I updated the discharge time to 2.9ms based on the listed spark duration for the coils, but it made no noticeable difference. I have reverted to the 1ms until I can verify discharge time for these coils.

I think I will try a few things over the next week. Placing capacitors on the 12V power supplies to the coils, and probably running up a new ground from the relocated battery to verify that the engine is as well grounded as it can be. I will likely try to change the pots in the ecu following the megamanual setup for hall sensors, to see if that has any effect. As of now I have left them fully counterclockwise.

As before all suggestions or questions are welcomed.

Thank you,
-Tyler

elturbonitroso

I have use those coils the with dwell time of 3.0 and the charge of 2.5 ,but for Hall effect you have to set up the potentiometer r56 to give you 2.5v on top of r54 check the manual for hall adjustment of the potentiometer

deddex

iTrader: (2)

Thank you for the responses. I adjusted r56 today to try to achieve the 2.5v on top of r54. I was able to bring it up to 2.379v according to my meter, but adjusting further did not increase the voltage so I backed it back down to just above the required turns for the max voltage.

My composite logs, trigger logs, and tooth logs have not been substantially changed by the modification to R36. At least to my eye.

The modification to R36 has had no noticeable effect on my spark issue. I am still having weak spark on the trailing coils (spark B and C) with consistent spark on the leading coils (Spark A and D). Again this is determined by seeing the trailing coils throw a visibly reduced spark when grounded against the housing along with the inability to consistently trigger an induction based timing light. While these coils and trigger wires will reliably produce strong spark when manually instructed to spark via tunerstudio. Swapping the coils and wires has the effect following the trigger wire rather than the coils or plug wires.

Another issue I have noted and would like clarification on is the correct setup of these spark wires in an application such as mine. Below I will show 2 tables, the first table is found on a reputable ECU suppliers how to page as well as in the MSextra manual for the MS3Base/v3.57 manual. The second table is found in the msextra manual for the MS3X/v3.57 manual. I have tried both wiring configurations and find that they both produce a similar idle, but I have the same issues with spark relating to the same trigger wires. Any input on this would be appreciated.

I plan to add capacitors to the power supply wires for the coils as mentioned previously, and continue to troubleshoot this to the best of my abilities. Again I welcome any suggestions or comments.

Thank you,
-Tyler

Table 1(MS3Base/v3.57 and how to website):




Table 2(MS3X/v3.57 manual):

muythaibxr

What firmware are you running?

More recent firmware revs no longer have "RX8 mode"

For FC or FD with a wasted leading you'd use wasted spark and rotary stroke with 2 rotors, and set trailing to FC if it's the old style coils, and FD if it's FD style. For something like the old RX8 mode, you'd just set Rotary stroke and COP and be done with it.

Ken

deddex

iTrader: (2)

I was running an older firmware version up until last night. I updated to the 1.4.1 firmware from a much older version. Rookie mistake there. After that update and some rewiring of my coils to match the new configuration (Table 2 above) the engine fired right up. and seems to be running quite well, my egts are much closer to each other now so overall that looks promising.

That being said I am still experiencing the weak/intermittent spark issue associated with Spark B and Spark C. I did not have long to investigate the issue last night, but I can still see "hiccups" in the spark from the two coils fired by those triggers. I am still unsure of the cause of this. I would appreciate any further suggestions if anybody has something to add.

Thank you all for the suggestions and responses,
-Tyler Healy

muythaibxr

Sounds like either a wiring issue or an issue with the outputs on the MS.

deddex

iTrader: (2)

Update time.

The spark issue seems to be mostly resolved with the firmware update. Initially after updating the MS firmware the spark issue seemed to have minimal effect. The next day with no other changes the spark issue seems to be fixed. I plan to monitor this frequently over the coming weeks, but for the last 4 days or so I have seen very few dropped sparks on any of the outputs. There are some very minor drops, but they are infrequent and inconsistent.

I have moved forward in tuning and have hit a few snags. I was initially able to tune idle and achieved a stable idle ~900 rpm with minimal trouble. After beginning to tune in low load areas I ran into a problem with the idle oscillating that I am trying to fix. The RPM and AFR are both oscillating similarly at idle. I'm going to remove the BAC valve (which has not been active to this point) to eliminate it as a potential cause for this idle oscillation as there is some indication it is not behaving properly. Any suggestions of other things to investigate would be welcomed.

Thank you all for your helpful responses with this.
-Tyler

muythaibxr

For the oscillation, check ignition timing and VE and make sure that they're as flat as you can get them in the idle area. What AFR and timing are you idling at? In my experience the 13B is pretty sensitive to both at idle.

With stock ports on an S4 and S5 NA, and on my RX8, the engine likes to idle around -5 degrees (5ATDC)... on the S4/S5, they also liked to idle pretty rich, like between 12.5:1 and 13.2:1. The rx8 idles dead-on stoich, if not slightly lean (14.7:1-15:1).

Ken

deddex

iTrader: (2)

I started with the base map on Aarons site. The AFR's I have tried are between 12.5 and 13.4. It was idling well at 13.0, before the oscillations started. For timing it looks to be running at about 18 degrees. I have not started modifying the timing, but I will try your suggested levels tomorrow after I get the temporary BAC blockoff on. The engine is an S4 turbo 2 with a "medium" street port. What I don't understand is why the car idled well, but then began this oscillating idle with no changes to the maps.

Thank you for the input,
-Tyler

deddex

iTrader: (2)

Quick update to this. I haven't had much time to work on the car this week, but I did get to fiddle with it for an hour or so. I removed the BAC valve and that reduced the idle oscillations, they were just as frequent, just with a smaller range. After that I followed the advice by muythaibxr and adjusted the timing and AFR. I was able to achieve a very smooth idle by moving to 5°BTDC for timing and an AFR of 12.6. This produced the smoothest idle, and lowest Kpa (~44kpa). moving the timing either direction was detrimental. I will have to see if this remains consistent over the coming days, but this is promising.

The change in timing was pretty dramatic from my starting point of 18°.

Thank you yet again for all of the help with this project.

-Tyler