Technical BL Coil Investigation and Solution

fmorelli

Lieutenant General
Staff member
Aug 11, 2017
3,748
3,592
0
57
Virginia
Ride
E89 Z4 35i, F10 535d
First off I'd like to thank @SlowE93 and @doublespaces for posting about bimmerlife tuning Motorsport Ignition Upgrade kit issues and solutions. While there had been misfire reports on various forums, we had not experienced them (yet), but were planning to investigate. Misfires at a Sunday track event jumped ahead of the measurement session. Again thanks to those that have been posting their observations, as well as for @Alex@BimmerLifeTuning being responsive to my messages over the past week.

While measurements gave us some clarity on what's actually going on, those measurements also inform some speculation on our part as to why misfires are experienced by some, not by others, and why for some the car runs fine, and then begins to misfire.

With that said, I'm going to post what Barry Battle and I found yesterday evening, with measurements and photos, along with what we found to be the solution for us. We ended up working on @MDORPHN 's 1M which is One Lap of America bound, instead of my Z4, which I'll fix later this week. I hope this is helpful to others. I'll make multiple posts on this thread, as follow-up. I hope the details are helpful to other BL coil owners.

Filippo

1M.jpg
 
Last edited:

fmorelli

Lieutenant General
Staff member
Aug 11, 2017
3,748
3,592
0
57
Virginia
Ride
E89 Z4 35i, F10 535d
I'm going to post a lot of photos, so rather than having the reader paw through everything to come to the conclusion, I want to start with the conclusion and then provide the supporting information that brought us there. Hopefully this allows for more clarity.

Brief summary:
  1. In stock form, we concluded by our measurements that it is impossible for the BL R8 coils to actually sit on the spark plug terminal.
  2. The reasons cars even run without the coil female terminal connecting to the spark plug male terminal is electricity takes the path of least resistance. So spark jumps an air gap from the coil pack female terminal to the spark plug male terminal.
  3. The current modification we've seen posted improves but does not solve the problem: removing ~1/2" of the metal shielding, along with grinding the side two rubber nubs - improves the coil seating, but by our measurement is not sufficient. While the mod does lower the coil pack in the valve cover shaft, decreasing the air gap for the spark to jump, by measurement the coil pack female terminal still is not attached to the spark plug male terminal.
Solution:
  1. The shield must be cut far up the coil shaft to allow the rubber end to flex, much like stock N54 Eldor coils (more on why later). (UPDATE: Alternatively the shield can be completely removed. Thanks to @Rob09msport for posting #140 on this thread.)
  2. We removed the black rubber sleeve at the coil head (it slips off). The two protruding black rubber nubs were cut flush.
  3. On the red coil head, the red nubs (which fit into the black rubber nubs when the rubber sleeve is on) are ground back to the second ring on coil. This leaves a small, remaining red triangular nub which becomes the depth indicator. Once installed correctly, this red "pointer" will be just proud of the valve cover coil shaft.
  4. The 3 black rubber nubs at the bottom of the coil shaft (terminal end) are cut flush.
  5. The coils must be installed with the electrical connector pointing straight back. Only this configuration allows the coil to sit low enough to fully seat on the plug.
I'll post photos and measurements in the following posts. Here are three summary photos, showing the modified BL R8 coil, along with what the coil looks like when properly seated VERSUS when just modified by grinding the side and cutting 1/2" of the metal sleeve, which we tried along the way of measuring.

Filippo

IMG_20180423_205812.jpg


IMG_20180423_201200.jpg


IMG_20180423_205559.jpg


N54 head cutaway. Note how coil must flex to reach plug:

n54_di.jpg
 
Last edited:

fmorelli

Lieutenant General
Staff member
Aug 11, 2017
3,748
3,592
0
57
Virginia
Ride
E89 Z4 35i, F10 535d
The following details support what we found. Let me say in summary, we were surprised a few times, to find out several facts. First we started by taking many measurements of the BL R8 coil, spark plug, and valve cover shaft. At one point with all the data in front of us, we concluded that the modifications people were making (cutting the steel sleeve 1/2" and grinding off the rubber and red nubs at the head end) probably solved the problem. But upon installing one modified that way, while the coil pack sat lower than unmodified, the measurements told us the coil pack was not actually fully seated on the spark plug.

First off ... @Alex@BimmerLifeTuning had told me that there are various versions of this coil pack - that BL's are genuine, and that people buying just the harness and sourcing other coils (I guess there are copies or variants?) may have issues. Our BL setup is sourced completely from bimmerlife tuning. Here's a photo of one of the coil stamping part numbers:

coil part number.jpg


So we took a series of measurements as we wanted to understand how all components sat relative to one another:

For the coil pack and plug: the length spark plug fully seated in the coil pack. The length of the coil pack with plug installed. The distance from the coil pack rubber female end to the female terminal inside and the corresponding length of the spark plug from male terminal to hexagon (so we know full seating depth of the plug in the coil). The diameter of the steel shielding (at the flair, and at the fat end). The diameter of the 2 rubber nubs at the coil pack head (as well as the diameter of the red nubs underneath the rubber nubs.

For the valve cover plug shaft: the depth of the shaft from the black plastic top, to the spark plug seat at the head. The ID of the shaft at the top. We also took measurements of the shaft diameter, internally, using sockets of known diameter and marking depth on an extension.

Let me try and walk through some of the measurements.

30772555_10160248990800335_225422935_o.jpg



Let's start with the BL R8 coil head. The rubber centering nubs measure at nearly 38mm. Since the rubber nubs are meant to compress, I took a compressed measurement using the digital caliper flat area and pressing firmly with two hands - this yielded 36.3mm compressed. The ID of the valve cover shaft is 31mm. This explains why folks were grinding off the rubber nubs and underlying red nubs. The rubber sleeve, with the nibs removed, measures just around 31mm - a good fit to the valve cover shaft.

IMG_20180423_182213.jpg


Next we measured the valve cover spark plug shaft depth. We simply used a factory spark plug socket on an extension, marking it with a Sharpie at depth. FYI the spark plug socket OD is 17.8mm. Note as an aside that the original factory Eldor coils don't have a shielding sleeve (just a rubber shaft) - in the photos below one can see that BMW installed the shielding sleeve in the valve cover (note the hole goes from black plastic to metal). The resulting depth was 158.5mm.

VC depth 1.jpg


VC depth 2.jpg


From there we wanted to understand exactly how far down the coil pack should sit, so we transferred the 158.5mm shaft depth to the BL R8 coil with spark plug fully seated. This was first done with the rubber sleeve still installed on the coil pack. Once we have this, we know without a doubt exactly where the coil pack should sit relative to the valve cover shaft entrance.

coil depth 1.jpg


coil depth 2.jpg


coil depth 3.jpg


I'll continue in the next post.

Filippo
 
Last edited:

fmorelli

Lieutenant General
Staff member
Aug 11, 2017
3,748
3,592
0
57
Virginia
Ride
E89 Z4 35i, F10 535d
So we now knew by the numbers why people were grinding off the rubber and underlying red nubs at the head. We wanted to figure out what was going on in the shaft. Knowing the coil pack shielding flares to 19.5mm at the end, we found a socket that was very close in diameter (19.6mm) and ran it down the valve cover shaft. Remember we have a mark on the extension for full depth (and the new socket was the same depth as the original, narrower socket). Running the socket down the shaft, we discovered that the socket bottomed out on the side of the valve cover shaft. So this confirmed why people were making the second modification - cutting off ~1/2" of coil shielding at the flare end, removing the flare.

spark plug depth socket.jpg


At this point we felt that the modifications people were making seem to address the issues, so we went about modifying one plug per this post. Given we had taken some things apart, and with the benefit of measurements, we decided to do one thing differently - we had the rubber sleeve off, and decided to cut the rubber off with snips, and grind the red support on the head only up to where it slides down into the valve cover shaft. The remaining piece of red nub would give us a red triangular pointer which would indicate the coil is fully seated.

To pull the rubber cover off, the easiest way we found is to slide a small flat-blade screwdriver under the metal sleeve and vigorously lift the rubber nub off the red nub - the red plastic is what prevents the rubber from sliding down (along with general friction on the shaft). Lift one side down, then the other. I also used a bit of lubricant which made the rubber far easier to slide off and reinstall.

Red nub 1.jpg


Red nub 2.jpg


Rubber nub cut flush.jpg


So now we were ready to install the plug and see where we are. Upon installation, to our surprise, we were not all the way down. Something else was getting in the way, and it was not clear to us what the issue was. We went back and looked at our measurements and concluded that the lower rubber nubs seem to be interfering. So we removed the lower 3 rubber nubs.

lower nub 1.jpg


lower nub 2.jpg


With those removed, the BL R8 coil seem to go down a bit, but not enough. We were just miffed at this point.

rubber nubs removed.jpg


Filippo
 
Last edited:

fmorelli

Lieutenant General
Staff member
Aug 11, 2017
3,748
3,592
0
57
Virginia
Ride
E89 Z4 35i, F10 535d
So we were almost there, but miffed. We just could not figure out what was interfering with coil seat depth. We went back and double checked our measurements, in case we were off. No such luck. As Barry went to measure the shaft depth again, he observed that the shaft is not perpendicular to the spark plug seat, evident by the orientation of the installed plug relative to the shaft. We had not seen this before, because we had just pulled the plug and made measurements. Only with the plug installed can one observe this. I hope the photo shows this; but if one looks at their N54 head one will clearly see this.

shaft off center.jpg


So this factor was wreaking havoc on our perfect-world measurements. We now realized that the factory Eldor coils just have the flexible rubber shaft, and that BMW put the shielding in the valve cover. The factory coils can flex. The BL R8 coils have a steel shield that makes them rigid. Given we have shielding in the shaft already, we could cut the shield off higher up, allowing the BL R8 coils to flex as well. (UPDATE: please note one can completely remove the shield instead of cutting it. Thanks to @Rob09msport for post #140 showing the shield removal)

Shield cut 1.jpg


Shield cut 2.jpg


Finally ... the 3 rubber nubs at the bottom ... they have to be cut truly flush, or they interfere. We used a sharp razor knife to cut them flush. Snips were not quite flush and caused issues.

With these final modifications, the BL R8 plugs now by measurement sit completely and securely on the spark plug, at full depth.

coil full depth.jpg


I'll end up with a post with a few installation notes.

Filippo
 
Last edited:

fmorelli

Lieutenant General
Staff member
Aug 11, 2017
3,748
3,592
0
57
Virginia
Ride
E89 Z4 35i, F10 535d
Last post on this modification. Note the first post shows a summary of what needs to happen.

When installing the modified BL R8 coils, we found they need to be oriented front to back: that is, with the coil connector facing straight back. This is beaused, installed at an angle, the coil head hits the passenger-side wiring harness black housing. There is room for the wiring harness to plug in - you can choose to plug first, then install, or one can carefully slide the plugs in place once the coil pack is installed. This is true for the first 5 cylinders. The 6th has a different set of rules.

The 6th cylinder coil pack interferes with the plastic wiring harness enclosure that runs down the exhaust side of the motor. The installation requires a bit of finesse, but one can get full-depth installation without causing issues. For the 6th cylinder coil pack install:
  1. Remove the passenger side strut tower reinforcement bar.
  2. Plug the harness into the coil pack.
  3. Set the coil pack down in the valve cover shaft.
  4. Use a large head screwdriver and note the wiring harness black plastic enclosure can be wiggled left to right.
  5. Press the harness enclosure away from the shaft and walk the coil pack down onto the hole. Good lighting really helps to see what is going on. The side ribs on the coil pack want to catch the black plastic wiring enclosure. Wiggle the screwdriver to help the coil pack ribs clear the plastic.
  6. I could not get good pressure on the top of the coil pack for the final 3mm or so of seating (this is where that red index marker is helpful to know exactly how deep the coil pack is seated. I used the handle end of a plastic dead-blow hammer on top the coil pack, and then hit the top of the hammer head with my palm. This gave me space to push down and seat the coil pack on the plug. Do this only for the last few millimeters of the install, when the coil pack has cleared the black plastic wiring enclosure.
Three other tidbits:
  • We originally thought everything installed fine, because we heard a "click" when installing the plugs. We believe that click was actually a noise made by the shielding in the valve cover when the rubber caught it. We don't know that for sure, but what we do know for sure is the click wasn't the coil pack terminal attaching to the spark plug.
  • The fact that these coils could run motors without misfires while the coil and spark plug terminals were not connected to one another is impressive I suppose. Path of least resistance, I suppose.
  • From our experience prior to modifications, when we incurred misfires after not having them for 4 track events, then "reseated" the coil packs which caused the misfire to leave, likely just brought the gap from coil pack terminal to spark plug terminal close enough to overcome the path of least resistance.
Hopefully this information is relatively accurate and helpful to others. Again thanks to all that have posted on this issue, along with your solutions and suggestions - they were helpful. We hope that we don't see any more misfires.

Thanks again to Barry Battle @ 3DMMotorsport for all the assistance and expertise, and @MDORPHN for letting us hack up his BL kit before hacking up my Z4 set :).

Filippo

IMG_20180423_205812.jpg
 
Last edited:

Torgus

Brigadier General
Nov 6, 2016
2,671
2,194
0
Boston
Ride
ACF 6466 E92 + METH
Great thread! Very informative.

Amazing they worked better than stock when not even physically connected.
 
  • Like
Reactions: fmorelli

MDORPHN

Corporal
Jan 28, 2018
196
162
0
Ride
BMW 1M
Great thread! Very informative.

Amazing they worked better than stock when not even physically connected.

Agree. And now that the coil is properly seated on the plug, I suspect I can run a gap greater than the .028" that I had been running.

Neil
 
Last edited:
  • Agree
Reactions: Torgus

wuX2

Lurker
Mar 21, 2018
18
5
0
Bay Area, CA
Ride
2008 135i
Great work. Was there any followup from bimmerlife after your findings? Curious to hear what they had to say about it.
 

STE92

Corporal
Mar 3, 2017
224
83
0
Ride
E92
+2

Second or third thread, looks like they sticking to user buying cheap coils story.
 

JBacon335

Corporal
Nov 7, 2016
227
107
0
40
Toms River, NJ
Ride
07 335i Sedan, 1988 Mustang GT
I'm actually amazed at how my car runs so much better than the factory coils when apparently these aren't even fully on.
 
  • Like
Reactions: Snowman