Coil Packs and Dwell Time

[belize1334]

That's awesome.

I don't think it will be hard to determine if the igniters are current-capped. I'll just throw one on a variable voltage source with a 1 Ohm resistor and crank up the voltage. If the current rises over 6 amps then we'll know.

Just to be clear brox, you hooked up a second igniter in parallel but are you still running the intrepid/300m coils in series?

 

[broxma]

Correct.

Essentially there are 6 wires which need to be cut on the main harness. I pulled the friction tape back, cut them one at a time, stripped both ends, and soldered all three wires together. I left the white tach wire alone on the clip from the harness and just doubled over and taped the one on the additional clip. I mounted both ignitors on the back of the manifold in the two bolt holes left from removing the intake brace. This allowed me to sort of face them together and up. I needed this close proximity because I did not extend any of the wires so I only had a short length to work with.

The main power wire is a bit of a bear to solder since it is much larger than the others. I am going to leave it in this condition for a few days and see if I have any problems. If not, I will go back and do a more professional job with the soldering a wiring.

I left the coils alone, still wired parallel. Today will have a better condition to test against the norm from yesterday. Temperatures are expected to be the same and I am going to increase boost to about 21 as I said last night. I think running under the same conditions with a 4 PSI increase in boost is a fair comparison. Steve noticed the spark yesterday when I took him for a ride so I'll have him with me again if possible to get an objective assesment.

/brox

 

[belize1334]

I haven't had a chance to test the igniter yet... I'll get to it later... But I have had a thought about the way in which we wire the coils together.

First, I think that series is the only configuration that should be considered. In parallel their collective resistance drops to 1/4 of the OEM coil which means even two igniters might be under too much load (double OEM). Further, the parallel configuration would have them come to full charge MUCH faster which means they'll spend longer with a high current flow and could potentially overheat.

Second, when we wire them in series, I think we're doing it incorrectly. Granted, it works. But, it's not ideal.

The current configuration calls for power to the pos terminal of coil one. Then the neg of coil one goes to the pos of coil four. Neg of coil 4 goes to ground. When the coils charge current flows through them both in the same direction (pos to neg). Then when they dischage, same thing... The current flows through the secondaries in the same direction (pos down to plug and into head). That means that the current must then go through the chassis and through the wiring and into coil one where it branches to plug one and over to coil four and finally to plug four. This has two effects. Firstly it means that the path to coil four has more resistance and thus less spark intensity. This may or may not be noticeable given the breakdown resistance of the plug. Secondly, it means that all the current crossing the plugs for the ignition even must flow through the chassis and the main wiring harness.

If, by comparison, you look at the schematic for the factory coil, you'll see that it's much simpler. There's only one secondary for both plugs. Current flows down plug one, through the head to plug four, back up to the secondary, and back down to plug one. It's an isolated path that removes the wiring harness entirely.

I propose a new wiring scheme. Instead of going from power to coil one pos and then from coil one neg to coil four pos, I suggest it go from power to coil one neg and then from coil one pos to coil four pos and then coil four neg to gnd. Essentially you wire coil one backwards. What this does is cause it to charge in the reverse direction as it's twin. It then also discharges in the reverse direction. Then the discharge spark flows from plug one, up to coil one pos, over to coil four pos (via the wiring) down to plug four, through the head and back to plug one and up into coil one again. Just like the factory coil this creates an isolated circuit path that doesn't involve the chassis and only incorporates one wire (from one pos to four pos) which can be made heavier gauge than the rest of the wiring.

The only possible drawback to this design is that the coils MIGHT have directionality. But since these are all AC currents by the time the secondaries are energized, it shouldn't make any difference whatever. When I measured the inductance of the coils, for instance, I got the same value both ways... no directionality as far as I can tell.

 

[SouthCaliVR4]

Gotta say, you guys rock! /ubbthreads/images/graemlins/worthy.gif this is the best info I have seen on the COP set ups to date & am following this thread with great interest. Hope you don't mind but I'm going to hold off on building a COP till I can do so using your final results. Thanks for the efforts of all involved /ubbthreads/images/graemlins/applause.gif

 

[belize1334]

Ok, just got in from testing my '90 igniter.

I used two 0.6 Ohm resistors in order to vary the load in the system. The first thing I noticed is that the resistance values change when current is applied. With one resistor (supposedly 0.6 Ohm) current was 13 Amps. With two resistors (supposedly 1.2 Ohm) current was about 7.5 amps. So, clearly they act more like 0.9 Ohm resistors once current is flowing and the resistance increases with current. That's gonna make the results a little vague cause it's hard to say what the resistors are really doing.

I hooked up the igniter by grounding pin 3, hooking pin 1 to the resistor and then hooking the other end of the resistor to 12V. I then touched pin 2 to power. With 2 resistors (~ 1.8 Ohm resistance) current was about 6 amps. With 1 resistor (~ 0.9 Ohm resistance) current was about 9 Amps.

This suggests that the igniter is NOT current capped, but does have it's own internal resistance which varies strongly with voltage. Assuming that the resistors both acted as 0.9 Ohm the whole time (decent approximation) then the igniter has a resistance of ~0.5 Ohms at 9 amps and ~0.2 Ohms at 6 amps.

So, what does this mean. Well, it means that as the current rises the resistance over the ignite grows compared to the resistance in the coil. That will slow the charging and will also lower the peak current and thus lower the peak power. By putting two igniters in parallel, they split the current. So, where the resistance would have grown to 0.5 Ohms by 9 amps, now each one will still be down at 0.2 Ohms (or less) up until 12 amps.

To quantify it, lets assume that the coils are better than my resistor and that they maintain their 0.8 Ohm rating. At 12 volts, the total resistance is now the sum of 0.8 Ohm plus the resistance of the igniter. Lets suppose that the final current is about 9 amps. 12V/9amps = 4/3 Ohms = 0.8 Ohm + 0.53 Ohm. That's right on spec because we said that the igniter had a resistance of 0.5 Ohm at 9 amps. Now let's put two in parallel. They split the current so each has a resistance of less than 0.2 Ohms and, summed in reciprocal, their net resistance is 0.1 Ohm. At 13 Amps, the total resistance is 12V/13amps = ~ 0.9 Ohm = 0.8 + 0.1. That's right on spec again. 12 amps split between two igniters is 6 amps each where they both have a resistance of 0.2 and combine to give resistance of 0.1 Ohm.

So... the moral of the story. If you change NOTHING about the coils. But you put a second igniter in parallel, the theoretical maximum current will be as much as 50% higher than with a single igniter and each igniter will carry less current than it would alone which will increase it's lifespan. Further, the decrease in total resistance also decreases the necessary charge time so the coils will rise to a current which is closer to the peak theoretical value in the time alotted. This amounts to a further improvement in current at the time of discharge.

Now, this affect is independent of whether you're on OEM coils or COP so long as the resistance is still 0.8 Ohms. Recalling that the intrepid coils have 3/4 the inductance and thus 3/4 the stored energy at a given current, we can estimate the final stored energy in the intrepid coils with dual igniters as at least (3/4)*(1.5) times the OEM energy. That is, dual igniters with intrepid coils wired in series gives at least 25% more stored energy than OEM coils - probably more considering that they charge faster...

 

[prove_it]

mmmmmuuuaahhahahahahahahhaha

Insert more evil laugh while rubbing hands together here

THIS is why this forum kicks tooners butts, we actually have educated people here!!!

 

[belize1334]

BALLLLSSSS

I was out test driving... got the boost up to about 20 psi on my 14b on 91 octane with no knock (which is why I love my new IC) and all of the sudden... NOTHING. Car went COMPLETELY dead.

Turns out that my COP wiring had got pinned between the head and that little noise reduction box and rubbed through to the valve cover where it shorted the ignition straight to ground and fried the 30A ignition fuse. Of course the simple answer is to throw the OEM coil back on to get home but wouldn't you know it I loaned my plug wires to Ryan for the weekend and he's out of town. I hitched a ride home and wrapped some tape around the short and moved the noise box to the top side of the plate to avoid recurrence. Now I've got to get a ride back to my car so I can bring her home before dark and go autocross in the morning...

BUT... until the wire shorted through she was running SWEET.

Anyway... not to detract from the academic flavor of the thread... but I HAD to vent.

 

[524of1000]

Yeah, something similar happened to Larry earlier this week...or was it last week... Anywho, I didn't get to ride with him today as I was visiting the in laws out of town. Since my spouse is my primary income, I do what I can to keep her pleased(Larry is quite adept with this concept as well). I'M assuming all went to plan in the new trial run today? A whole new beast? /ubbthreads/images/graemlins/devil.gif

 

[prove_it]

haha, that's funny.

So I'm a bit foggy as to the wiring on this project. Did you guys ultimately wire the transistors in parallel to the coils wired in series then?
I think I'm following correctly. If you could draw up a crude picture on MSpaint that would be awesome, as I am a visual person more than a word kinda guy. I think I have a spare 91+ transistor laying around. Would it be possible to use dual 91+ transistors, or will a 90 and a 91+ transistor be needed?

Also what would happen if one transistor was used to control two companion cylinder coils, and another transistor to control the other companion coils? Is it possible?

 

[belize1334]

Basically each Power Transistor just has two "relays" which operate very quickly since they're solid-state design. Each relay is caplable of being triggered and providing a "high-current" ground path for the coils. The factory design uses one channel for 1-4 and the other for 2-3.

Wiring two in parallel just has each relay behave exactly as one alone but they share the current load. As for '90 vs. '91, there's no reason to prefer one to the other. I just figured a '90 and a '91 would be nice since the '90 doesn't have the extra pins for the tacho so the wiring would be cleaner. Brox used two '91 units since that's what he had and he just didn't connect the tacho signal from the second unit. The physical wiring would be painfully simple. You just connect all the pins from you second unit directly to those same pins on the original (unless using a '90 in which case you have to map their function...). Ideally you'd do this with a spare wiring harness so that each unit could be unplugged. One nice thing about doing it this way is that if you unplugged on of them then it's harness would just sit empty but the other unit would still be connected so it'd behave exactly as stock.

Now you COULD rewire the whole thing and have each igniter fire both channels simultaneously, in parallel to each other, for a single coil pair. Then instead of each igniter firing 1-4 and then 2-3, you'd have one igniter fire 1-4 on both channels at once and then the other fire 2-3 on both channels at once. I'd think that this would work just as well except that the igniter might heat up and then cool down since it's doing all it's work and then resting vs. the stock setup where it works half as hard twice as often. Also, in this setup if one igniter failed or were unplugged you'd lose a coil pair. I recommend not doing it this way and wiring them in parallel in the stock configuration.

The coils themselves are being wired just like normal unless you read my post about reversing the polarity of 1 and 2... but that's a separate issue and, at this point, purely speculative.

I'll add a final word of caution. The idea of dual igniters is to allow the coils to pull more current. But, unlike an ARC-2 box or equivalent after-market ignition, they're pulling the current through the factory system. That means there's the potential for burning fuses. Know this going in and don't tell me I didn't warn you.

 

[broxma]

I am almost certain the coils are not directional. They work regardless of which wire is connected to pos/neg.

I followed everything you said about the resistance and the internal resistance of the ignitor.
I had originally considered running the 1-4 set off a single ignitor but again considered the notion that if one ignitor failed you'd be running on two. In the current configuration, as long as one ignitor is working you will have spark. I see no reason to change this setup for now unless some new information comes up which I doubt.

I did the exact same thing with my COP plate a few weeks ago. I threw dozens of codes and am still dealing with the after effects of it.

Here is the report from today.

I turned up the boost to about 23 PSI. Conditions today were actually hotter than yesterday. I was confident that the temps and increase in boost would place the stress on the system to the degree I was looking for. WOT runs in 3rd gear from 4500 to 8500 showed no stutter or hiccup in spark. Power delivery was smooth. Power had noticeably increased even with just a few extra PSI. I brought a navigator with me again who tips the scales around 250 or so, about 100 pounds more than Steve and I didn't notice him. His comment to me was that my car at 23 PSI had more balls than my Evo which ran 27 PSI and 300 WHP . This is considering the fact that my break in on 379 didn't go as planned and my car is only putting out compression numbers of 130/110/125/110 at 9-1 compression. Numbers should be closer to 175.

The car is obnoxiously fast at this point, not just quick. I have been logging my boost/rpm and noticed a drop in RPM to achieve initial boost. The car feels more responsive at low, getting out of the subdivision speeds, even with the very low compression numbers. Jason had noticed that it appeared/sounded as if my tires were spinning at high rpm, 6500+ on throttle, not initial. This is with 235/17's, a locked center and LSD rear.

Tomorrow I am doing the timing belt on my wife's Beetle and turning the boost up to its final setting which I tentatively figured at 28 PSI for the street. Today's test was a huge success. From a near failure point at 17 PSI to cooking tires at speed at 23 with a goal tomorrow of 28. All in a few days.

A side note. Three days ago, I was looking into buying a new ignition system because of the problems I was having at 17 PSI. I had already picked out the 600.00 system I was going to get. Seems I may have solved the problem with a part I had laying on my bench.

/brox

 

Assuming this is the magic bullet for running COP (and its sounding pretty good so far), with two power transistors in parallel, how would you ever know if one of them failed? It seems as though a "smoothness" may be lost, i.e; spark blow out at high boost/high rpms, but it seems reasonable that one may look to other causes as the issue, and not look at one of the two PT's.

 

[broxma]

Well, consider that the system as we have laid out, with a standard ignitor/COP setup is barely working at even modest levels, it is safe to assume that at an extreme level of performance a loss of one of the ignitors would be pronounced. If I had a navigator I would take some video and show you the difference between two runs, one with a single ignitor and one with dual. Realistically, the devices are fairly robust. I have actually never had one fail on me spanning 15 years and probably a dozen cars. I would estimate the likelihood of failure as very low however probably increased due to the new configuration. More importantly, if one did fail, you'd still be moving. It does equate to a redundancy but one with an added benefit.

/brox

 

[broxma]

I wanted to point out a few things since I think this post sort of testifies to the way DSM modding has been done for the last 20 years. To my knowledge, this use of the dual ignitor to increase spark potential in the COP setup has not been done before. This is a fairly amazing piece of information considering the sheer volume of DSM sites and tuners out there. Some of the guys over at EvoM are pretty sharp cookies and we didn't conceive of anything like this in that entire post. Credit needs to go out to Belize1334 here provided that the future testing does back up what appears to be a better system. This same ignition system has been in use for decades now and while the bits and parts have changed, the main system has remained the same. Not a lot of data has been gathered on the actual workings of the system as it has mostly been discussion about functionality, i.e., will system "X" work on Car "Y". Another site will eventually pick up on this and claim it as their own I am sure. GVR4.org is small potatoes in the overall 4G63 community and tends to be focused a lot on restoration due to the age of our cars. It is nice to see a serious, albeit small, tech effort come out of the community and I'm glad to be a part of it.

Now onto updated results.

It is late as George Burns but I got a few hours sleep earlier and have found myself not tired in the least. It is also, unfortunately, 2:00 a.m. here, which means drunks and cops. Regardless of the risks, I set out on the highway to push this testing phase to another level. I have a small issue which is that my boost gauge reads in k/pa, roughly bar, and only goes up to 1.5. I have another few gauges I will install in the near future to get an accurate reading and will probably install the 3 bar map sensor so I can read boost directly from my EvoScan logs. Ignoring this issue, I turned up the boost again.

At what reads out at about 1.9 bar I let off the gas as it is 2:00 a.m. and I don't have anyone to call should something go a foul. The turbo was not at full boost at this point and I expect it would have went well over 2 bar. The real increase in butt dyno power was again noticeable. Power delivery was outstanding. My car has a bit of a rake to the stance which is a good thing because the front lifts up past horizontal during acceleration. Had I dropped the rear more, I may have lost sight of the road. There was no spark break up at an estimated 27-29 PSI high RPM in third gear. I was limiting boost by throttle so was only about 1/2 to 3/4 throttle during the runs. I also did several quick launches in 2nd gear almost to redline, again, throttle limited for boost reasons, again not a single hiccup or stutter.

Now compare.

The car 2 days ago had spark stutter at 17 PSI, gap set at .022. Tonight, 28(ish) PSI, same gap, perfect.

I have changed only one thing between then and now. I may actually pull the plugs and increase the gap to see if I can find a breaking point.

P1 - Adding an additional ignitor in parallel to the factory ignitor should increase available spark at the plug due to electrical magic.
P2 - Brox added an extra ignitor in parallel to his car and his ignition problems have disappeared.
-------
C - Adding an additional ignitor in parallel to the factory ignitor benefited Brox's car.

That is as much of an argument as I can lay out right now. If we can add a few more premises for the argument we can solidify the conclusion. For people who are interested, I have ignitors I can get on the very very cheap. If you consider this modification to the ignition, in order to add valid premise to our argument, I will gladly help make that a reality. I am not looking to make any money off this. I just want others to validate the argument.

I dare not post a schematic of this because it is so simple a solution as to be easier said in words.

If you have two identical ignitors as I did, you simply cut back the wrap on the factory harness going to the ignitor about 7 inches or so. You will be cutting and soldering the three outside wires from each side. The white wire is the tach wire and does not need to be cut. Cut the wire, strip both exposed ends, strip the corresponding wire on the harness of the new ignitor, and slide some heatshrink down the factory harness side wire, Tie the new ignitor connector wire to the side opposite the one you slid the heatshrink on, tie the combined wires with the other, solder, heatshrink, done. Repeat for all six wires. Take the white tach wire from the new connector, fold it over and heatshrink it. Done.

As far as mounting, again, I used the bolt holes for the factory intake manifold brace. I drilled out the hole on the ignitor backplate to fit the 10mm bolt(14mm head), and bolted them in facing up and center.

If anyone would like to add data to this set, shoot me a PM. I can run down and grab several ignitors and harnesses and will probably do so anyway.

/brox

 

[prove_it]

It's sunday morning here and haven't checked yet but I think I have some spares laying around. I should be able to check around to it today. I have a mild setup and I am curious if I can see any difference.

Quick question: are NA and turbo transistors the same?

 

[belize1334]

'90 Transistor reads J122 and '91+ reads J722T. From the FSM it appears that the n/t of the same years used the same ignition design but I can't confirm that they have the same part numbers.

 

[broxma]

They are the same part. I have pulled a few from the yard in the past.

/brox

 

I know that this thread has focused primarily on COP. However, do you believe one would also be justified in doing this on a stock coiled car?
P.S. Thanks to Roger B. Scott and Broxma.

 

[toybreaker]

This is some nice work, gentlemen!

Quoting broxma:


I mounted both ignitors on the back of the manifold in the two bolt holes left from removing the intake brace. This allowed me to sort of face them together and up. I needed this close proximity because I did not extend any of the wires so I only had a short length to work with.



The ignitors make some pretty substantial amounts of heat during operation, so while this might work ok for testing, I would try and mount the ignitors individually in the factory configuration with each of their backs against their own heat sink.

Some heat transfer/dilectric grease applied between them and their respective heat sink surfaces would also increase their ability to transfer their operating heat out of the units.

It will increase their lives exponentially

(if you have friends in the car stereo biz, see if you can score an old audio amplifier heatsink. They make bad ass mounting plates for projects like this)


Quoting Broxma:
The main power wire is a bit of a bear to solder since it is much larger than the others. I am going to leave it in this condition for a few days and see if I have any problems. If not, I will go back and do a more professional job with the soldering a wiring.

I left the coils alone, still wired parallel.



Quoting Belize1334:
I'll add a final word of caution. The idea of dual igniters is to allow the coils to pull more current. But, unlike an ARC-2 box or equivalent after-market ignition, they're pulling the current through the factory system. That means there's the potential for burning fuses. Know this going in and don't tell me I didn't warn you.



I was thinking the same thing...


Please note the coils are fed directly from the ignition switch down the same wire that feeds the fuel pump leg of the mpi relay.

Increasing the current carried by the ignition switch may result in a significant shortening of it's service life.

At the very least, it will result in some voltage drop out to the coils as well as the fuel pump.

A seperate dedicated relay used to feed your new coil circuit, located out in the engine compartment would probably be the mostest bestest way to work the supply side.

If your fuel pump is already re-wired, you have some breathing room, so you *might* be okay, but I would re-wire things as follows;

Coil Power
Install a dedicated coil power supply relay out in the engine compartment.
Feed it fused power from the battery down a quality ten gauge (or eight awg if youre feeling froggy)
Switch the relay on using the original black/white wire that used to power the coils.
(an interupt could be put on the ground leg of the relay as a hidden security feature...)
power supply side done

For the connections out to the coils, I would make a jumper harness that plugged into the original harness and then into the twin ignitor multiple coil harness, (leaving the option to go back to stock if the whiz bang shitarree lost it's desire to throw a spark)



It "appears" to my untrained eye that the ecu is just tickling one leg of a switching transistor in the igniter case when it wants to fire a coil.

It's this transistor located in the ignitor that will switch and carry the current that energises the coils. This means the ecu will only see the load that it takes to switch a transistor. The actual coil ground will run thru the legs of the ignitor transistor and out to ground on pin 3 (black wire) from the ignitor unit, not back thru the ecu.

I must say I don't completely understand how much loading the ecu "sees" when it triggers the transistor in the ignitor to fire the coils, but I do know it looks like we're preparing to double that load. To an uneducated mook like me it seems like we should be sure the ground traces from the ignitor triggers are up to the extra loading, and the actuall switching components are up to the task... long term ...



Hopefully, Steve P or Keydiver will chime in and educate us on the subject. They know these ecu's from the board up.

It would suck to pop the ecu /ubbthreads/images/graemlins/bawling.gif


At any rate, I would probably optimise/upgrade the ground path out of the ignitors, as they will see higher loadings.





Roger, I'll get you a 90 ignitor pigtail out on Monday. If you need anything else, please don't hesitate to ask.

Good luck with the project, gentlemen, and please keep this thread updated.

I like to see people try new things! /ubbthreads/images/graemlins/smile.gif

 

[prove_it]

Toybreaker:

You mention build a jumper harness. I was also thinking the same thing too. Question is does anyone know of a male connector similar the transistor side that could be used? This would allow a much easier and quicker install and provide simple plug and play for those who do not wish to hack their harnesses. That's what made the COP systems more appealing too.

Also I think you bring up a good point about the ignition switch currents and fuel pump issues.