maticulus

This should prove helpful to a lot of people running boosted applications.

Engine:
3900 60 degree V6 stroked to 4.2 litres, 11:1 compression, T-67 turbo, 93 octane fuel. It's not in an LS car but it was designed with some characteristics of the LS1 in the heads and shares the same bore, so it sports a set of Forged LS1 pistons.

The first 10,000 miles were plagued with intermittent flat wideopen throttle runs that immediately improved by backing off the throttle and reapplying gradually.

I thought it was an AE/PE problem as eventually the black powder of a rich condition would coat the rear bumper. During the Winter it was not a problem for some reason but in the Summer at times full throttle at about 55 mph would be flat with little more than just an increase in noise.

One particular run into boost as Winter approached resulted in a pop out the exhaust. I suspected ignition trouble and swapped the entire coil and module assembly for a backup. The pop disappeared but was replaced with what was clearly a sustained misfire at about the same point in boost until I let off the pedal. I reinstalled the previous coil and module assembly and the pop returned in place of the misfire.

After an engine inspection I found several exhaust manifold bolts loose. I tightened them and the car ran like a champ for about a month and the pop returned but this time the exhaust manifold bolts were tight. The popping or backfire was consistent at about 140 Kpa which finally revealed to me the flat wideopen throttle problem had been spark blowout all along.

The reason it took so long to identify is because I've boosted several engines and never encountered it, however this was the first time I defied common practice and increased compression for boost.

I closed the Bosch Super Plus copper plug gap down to .030 and problem solved up to 10 psi & about 60 deg inlet temps. Colder inlet temps around 55 deg and below caused the backfire to return. Coolant temps would also be lower ~163 deg.

I hot wired my primary coils by having the harness plug, power a relay that provided a direct feed from the alternator and increased the coil voltage by a full volt + over the harness voltage. I also replaced the stock GM coils with MSD coils (MSD wires were already in place). I was told they wouldn't help by a member of the forum representing my car, but because I got them for about the same price as OE I decided why not.

I reinstalled my hardly used NGK Irridium TR6 plugs with .022 gap in place of the Bosch. The first test drive with 44 deg inlet temps (I have an air water intercooler), hot wired coil packs with direct feed to module from alternator and MSD coils resulted in backfire at 140 Kpa.

Disappointed with the MSD coils, TR6 plugs and suspecting the TR6 plugs, I put the Bosch copper plugs back in gapped to .022 and pushed through to just over 13 psi where the wastegate held it.

Boost was pretty much trouble free with the copper plugs but things were a bit more complicated than that. From the start with such a high compression motor and the clutter of tables in GM OBDI code that no one knows as much about as needed, I had chronic occasional decel related knock retard that I tried to combat by manipulating the base coolant temp spark advance table, since the DFCO tables didn't seem to help.

I have a spark advance history table setup in Tunerpro and noticed that something was reducing my commanded spark advance in combination with the rise in coolant temp. I discovered it when reviewing the area of spark blowout and noticed that when boost was trouble free, spark advance was lower than what was commanded in the main spark table.

It also accounted for why the blowout problem seemed to be sensitive to ambient temps being most problematic with cold weather where the commanded advance was at work.

I mentioned the unexplained spark retard to a GM code pro, Robertissar, who informed me that by the way, the base coolant vs spark table in the 8F code mask was BACKWARDS. In other words, when I reduced 30 Kpa to combat decel knock retard, I was actually reducing timing at 100 Kpa and above not to mention the trouble that resulted from the cold start changes I made to the table.

Once that table was corrected the tune improved and the spark blowout was predictable and manageable. In order to consistently keep it under control, I reduced my spark advance by about 2 deg at the Kpa and coolant temp it occurred.

Theory states that cold air temps and wide spark gaps increase resistance so we close the plug gap to combat both. I've since discovered through reading that narrow plug gaps generate a spark with more current and wide gaps generate a spark with more voltage.

That would explain why high voltage aftermarket coils are often useless in resolving spark blowout trouble and why many say their OE coils performed better or as good as aftermarket coils. Generally speaking due to the nature and laws of physics, the OE coil producing less voltage than the aftermarket coil, is probably providing more current in exchange which is what we need to combat blowout.

The higher current produces more heat to help light off the resistant mixture. To drive the point home, look at the spec on a ni-cad battery and its conventional equivalent. The ni-cad battery provides greater current and the conventional battery produces a higher voltage.

It may seem trivial but if you do a side by side comparison with a flashlight you will easily see that the conventional batteries produce a much brighter light than the ni-cad batteries and therefore would be best for a flashlight. On the other hand in an application that needs torque like an electric motor for example, the ni-cad battery will be the better performer.

So, what we would need is the equivalent of a second set of coils in parallel to the first for more current instead of a symbolic second set in series producing more voltage. If you need 12 volts and 500 milli amps minimum to jump a plug gap, 14 volts and 400 milli amps is probably not going to get you there.

If you have a problem with spark blowout, it can be in the tune, also I have read it and experienced it for myself; copper plugs are the best application for boost and the little V shaped groove in the ground strap of the Bosch Super Plus plug is said to be beneficial as well although I don't recall the benefit I read about.

In my opinion as well as the consensus from what I've read, installing aftermarket coils to combat spark blowout is a waste of your money, especially considering the quality of the products are often questionable. My engine requires 3 coils and although brand new, not one of my MSD coils which were manufactured in China are identical to eachother, how's that for confidence in quality.

If you install aftermarket coils and it cures a blowout problem, I'd be more inclined to suspect you had weak OE coils to begin with.

Hope someone finds this useful as I had many a headache before getting to the bottom of this problem, not to mention having to endure the insult of not being able to pull away from a Ford Ranger as a result of it.

maticulus

My spark blowout problems continued on and off intermittently and were the result of:

Too much water being injected.
Weak OE coil.
New MSD coils used without the complete MSD outfit.

all resolved, and the latest to date, New original AC Delco coils purchased at a too good to be true price on ebay.

The misfire problem was fixed and the car running fine. I installed a brand new set of OE coils from ebay on my boosted V6 to make sure there were no ignition problems and took the car to the dyno.

The motor was breaking up by 4100 rpm on eacy of three pulls at 13 psi where before the coil change it was running clean through 5000 rpm and as high as 18 psi.

I thought it might be the low mileage plugs gapped at .025. I replaced them with no improvement. Finally I looked at the new coils and found cracks in the bottom of the casing on 2 of the 3 in use.

Suspecting they might be counterfeit, I looked at the box and found that not only were they made in China, they were also intended for cars driven in Central America and the Caribbean, no mention of the US which probably explains why the casing looks different than the Factory coils that were on the motor.

Some of the vendors are also calling aftermarket parts OE although they carry an aftermarket name and price.

I reinstalled the factory coils and the misfire cleared up immediately.

The cracks are at the top of the yellow marks and the factory coil is on the right. Note what the box reads.

oange ss

iTrader: (50)

good info

stevieturbo

This will sound rude.

But as the post is very long and seems to be lots of waffle in the thread....what is the actual point ?

it isnt an LS engine, it isnt using LS coils, and in general seems to have little or no relevance to anything LS whether boosted or not ?

Which rarely have spark blowout problems because the cols and setup are good in the first place

Trucker82

If you have a problem with spark blowout, it can be in the tune, also I have read it and experienced it for myself; copper plugs are the best application for boost and the little V shaped groove in the ground strap of the Bosch Super Plus plug is said to be beneficial as well although I don't recall the benefit I read about.

Hay Maticulous,

I have a 4.8 twin turbo with breakup issues around 4000, and am having problems trying to find the right heat range for the Bosch Super Copper Plus. Do you know the part number? I really appreciate the help! Also I have the D585 Delco coils on order and a new GM ls2 harness.

Thanks man!

TrendSetter

my evo randomly started missing under heavy power after about a year of no problems. it had a toyota coil on plug conversion which most evos use as an upgrade making plenty of power. I put new plugs in and it still had a miss, tried 022 gap to no avail. I messed with the dwell times and that helped but it didnt go away. I finally just built a harness to hot wire some d585 coils directly powered from the battery and now it runs great.
it seems like the brute force approach of super hot coils is the easiest solution. ive never had any ignition problems on my tt4.8 with all stock ignition.

kingtal0n

iTrader: (1)

Current is voltage over resistance (literally V/R = current), so if you increase voltage, and leave resistance alone, the current goes up also. If you set resistance to = 1OHM then current = voltage, so 14volts would be 14amps at 1 ohm of resistance.

Sometimes a cell (battery) will have it's own limitations that are imposed, causing a "lower voltage" to seem like "more current" but if there any these subtle traits it has to be that the internal resistance of the cell or somewhere along the line, R is lower. Colder wires have lower R, I would reason.

corvet786c

iTrader: (6)

mine are gapped to .020

1970camaroRS

iTrader: (8)

I thought these were the same people.

Dian

so was arf monitored and what did it do?

pantera_efi

iTrader: (1)

Hi ALL, reading the above posts states a problem with a "fix" NOT EXPLAINED.

First,"rule out" the Chit/Chat.
A good coil is the IGN-1A, a 25AMP coil, NOT to be commparied to the GM 10 AMP coil.
Next, the "wire" a coil Batt + to the ALT/Battery Buss with a HIGHER measured voltage is fine.

The coil "charge time" does equal supply voltage/impedance value in time, the time required to "fill" the coil flux.
When the coil is full of electrons, the winding becomes a DEAD SHORT.
THUS it would be easier to change the coil dwell value BEFORE a wiring change.

YES to the Copper Temperature Resistance change, higher = greater resistance.

THIS IS WHY I state it is important to "throttle" the Coil Dwell.
Example : The engine "cruise" dwell = 2ms with engine "boost" dwell = 4ms.
This will keep the coil winding cooler with greater output when required.

Lance

stevieturbo

Which would have a greater impact on coil winding temp ?

The fact they are bolted to a hot engine, or the current etc itself ?

roastin240

iTrader: (13)

The current inside of the coil for sure. If you bump the average dwell value above what the coil can operate at continuously, then you will have a failure. I have ran pretty high dwell settings at high load/rpm but its only quick spurts and does fine. The 1 time I doubled dwell to around 4.5ms while cruising, it fried the coils (this was 10 years ago when I thought more=better).

That being said, on a work project I ran 2 MSD DIS-4 systems on a constant speed generator I threw custom 13:1 CR pistons BBC platform engine that ran on natural gas and at equivalence ratios of 2 and higher. The ignition system was needed in order to fire the mixture (which was also preheated significantly to help as well). Before the ignition system, it would not run. But the engine runs at 1800rpm, so i get the full 12 spark events per cycle over 20 crank angle degrees and it claims 115mJ per spark event or something like that. So...12*115mJ per cycle = 1380 mJ total introduced to the chamber per cycle.

Another project 2 of our Electrical Engineers worked on that I provided engine level and controls support on was a variable capacitive discharge ignition system that could discharge between 50mJ-200mJ at a speed of 2000Hz with a power consumption less than 120W (<10 amps). It had to be fully controllable at all times. Goal was to spark ignite heavy fuel engines. The few sparks we made were closer to 1000mJ and blew the ground straps off the plugs in 1 spark event ha ha ha. We did high speed visualization in a pressure chamber I designed.

Sorry for the rant. Thought maybe some would be interested in those projects. I had fun.

Hey Pantera - do you have the appropriate ignition coil settings to be used in Tuner Studio for those coils you talk highly of? When I say appropriate, I mean long life durability with the ability to maintain a 1200-1400 CHP LS setup on e85 and 30psi? I am interested if you have any info you can send me. Preferably some sort of quantitative bench testing for power, energy, and speed, or a high speed spark visualization of a d585 vs. these coils at the same dwell setting. Lets see what you have!

stevieturbo

Up til a few months ago I had to fire/charge 2 coils at a time...never had any issues with any runtime or mileage despite this at 4.5ms

At low rpm's, there is still far more off time than on, ie duty is very low despite the same charge time.

At higher rpm's, obviously duty will increase for same dwell. With mine being run as wasted type spark ( didnt have enough ign outputs at the time for one coil per output ) say 6000rpm, would be like running the coils at 12000rpm.

Still no issues at 4.5ms dwell either with 514's or now the IGN's

I am back to full sequential operation now with a different ecu but I've just left dwell around 4.5ms at lower loads and 5.5ms at higher.

Even at 4.5ms around idle, the coils are only pulling about 7-8A max. Still havent tested them at higher dwells though.

pantera_efi

iTrader: (1)

Hi Steve, you observation is valid.
A coil is rated to 175C with the IGBT rated at 125C though in real life the same IBGT family (same dye) when tested to 175C will operate with current capacity degraded.
The under hood temperature of 100C is often a maximum read.
Thus the coil brackets are a heat sink at times.

Hi Roastin, I sell a LOT of my IGN-8s AND IGN-1A coils to CPS in Paris, Texas.
Yes Natural Gas, 1800 RPM, BBC engines for Oil Field pumps.
Your 12 spark CDI math ?
I would NOT trust the MSD reports due to supplied current to the CDI. (RMS)
The NEWEST Woodward Ignition Systems use their version of the IGN-1A coil WITH ION sensing.
They DROPPED the CDI program.

The term Arc Duration is MOST IMPORTANT.
What would be better ONE LONG SPARK for the 20 degrees of crankshaft rotation OR 12 start/stop/start/stop sparks ?

Your coil question = YES, I would need to know which coils, the cylinder pressure, AND the SPG ?

Your research into the Ignition System is a GREAT "share" for LS-1 Tech members.
I ask YOU to share you knowledge about the Spark Plug Gap AND the effect it has on ARC Duration.
Thus your recommended SPG/RPM (crankshaft rotation degrees/time)

Steve, the rotation of the crankshaft at 6000 RPM is 10ms.
The MAX IGN-1A dwell is 8.8ms at 14 volts.
The IGN-1A coil was measured by Mototron with a required specification of 102mj and a coil dwell of 2.03ms using the standard SAE "pico" load.

My self and Jala (Jarie Lane) has measured the IGN-1A at 280mj with the 8.8 ms dwell.

Does this help ?

Lance