Had to go look:
-for a snooty grocery getter SUV? You could do just as well,for WAAAAY less money;
-I bet the electronics would make the 220 look like a Radio Shack 50-1 kit!
Actually hate those things. The area I live in is populated with soccer moms driving $200k SUVs while texting/yakking/not paying attention to anything around them. Drives me nuts. Totally agree about the looks of the Bentaygaā¦
Now if I could find a late-80ās mint Lambo LM002 I could just drive over everyone else.
No info on Ducellier coils? There is no decent info on these anywhere. This is what happens when a race shop builds cars. Whatever is lying around at the moment gets tossed in.
Iām no expert, butā¦ in the case of an ignition coil, doesnāt it just really pivot on whatever the coilās needed resistance is?
You should live here: I think Colorado is the Stupid Useless Vehicle capital of the country.
I think you are referring to a ballasted (1.5 ohm) vs. non-ballasted (3 ohm) coils. Yes, of course you must match a new coil to the resistance of the old one (this car is an HEI type ignition, so it should have a 1.5 ohm resistor). But there are coils that put out 20,000 volts, and coils that put out 60,000 volts. More volts ( lets you increase plug gap)=bigger spark=faster flame front=more torque. There are a few other advantages to higher voltage (it pushes current through the wires faster, allowing less time to leak, more current available for high load situations). So I need to know what Iām starting with, and again the service/parts manuals are no help. And this coil is 27 years old and prone to failure what what Iāve read.
Sorry, meant to say that this car has electronic ignition, not an HEI system (male terminals on distributor), but still should be around 1.5 ohms.
Dr. Solly,
Do you have pictures of, or can describe more about the ignition system. I understand the Zytek role in the XJRS ignition system. Itās just grafted in the middle between the distributor pickup and the standard 4pin HEI GM amplifier in the AB14 Lucas box - then matched to a relatively low impedance coil. The Zytek box then takes the positional input from the distributor - in this case acting both as a cam sensor AND engine speed sensor (via tandem hall effect sensors), modifies the timing, and sends a YES signal down to the amp - which is responsible for providing enough dwell time to charge - but not overcharge - the coil, break the connection at the āYESā moment, and trigger however much spar the coil is capable of. (The coil is somewhat matched to keep from burning out the amp - but itās usually 0.6 Ohms on a modern single coil replacement for the dual coil system). (I should note that on the Marelli car, the amp is only responsible for charging & complying with the trigger signal, --the Marelli box provides a coil dwell time to the dual amps - so a little different).
Yes, a bigger spark is better, but a lot of that is down to the amplifier and how fancy it is + how capable of sending LOTS of columns of electrons into the coil w/o burning out for that BIG - long lasting spark.
This is why racers often swap out the ignition boxes for fancier stuff ā better coil - better coil control - better spark, fancier spark (multispark.) Then thatās the usual inductive discharge boxesā¦ then there are capacitive discharge boxes that dump all their energy into the coil at once usually resulting in one very short but INTENSE spark.
The lower the OHMS of the primary windings, the more amperage / columbs of energy can be applied (to the extent the driving electronics can handle it), and the faster it can charge up. The amount of high voltage induced depends on the speed of change in the magnetic field, the number of windings on the secondary coil, and the strength of the magnetic field. Modern Elec Ing systems donāt need ballast resistors anymore, BUT the electronics do have to handle the ability to charge the coilā¦ especially SUPER low Ohm coils like at 0.2 0.3 Ohms.
~Paul K.
1 Like
Good questions Paul. Iām going to have to pull up the service manual for the ignition and see what I can find. I believe the ignition is very similar, if not identical to what you are describing for the XJRS but Iāll double check.
Does this seem right? Would be nice to know more about the contents of the box attached to the coil:
The base of that cam-driven distributor appears to be very flat- like not much thereā and no extra wires so no clear indication that it contains anything other than a rotor (no pickup). That suggest, like the Marelli XJS V12, the pickups & timing bits are elsewhere - on the crank
~Paul K.
Something else that might be interesting to think about is the energy available for spark output VS input voltage VS RPM. Per the Megasquirt site, a standard coil or at least an LS1 coil (and from what I can find this about avg for coils) is on the order of 4ms to reach full output potential at automotive voltages.
The standard four-cycle engine spark repetition time rate formula is t=120/(N*RPM) where t is seconds and N is number of cylinders
For a 6cyl that means the charging time available between ignition events is as follows:
800 = 25ms
1000 = 20ms
1500 = 13ms
2000 = 10ms
3000 = 6.7ms
4000 = 5ms
5000 = 4ms
6000 = 3.3ms
7000 = 2.9ms
8000 = 2.5ms
9000 = 2.2ms
10,000 = 2ms
11,000 = 1.8ms
As you can see, even for the V6, let alone the V12 the ability to ramp to full charge stars decaying above 6000 RPM (and would start at 3000 RPM for the V12, which is why it had 2 coils initially and a 25thousanths spark gap).
Thatās at least with inductive systems charging the coils at 14.5ish volts (there are charts about this). Coil on Plug, or CD (Capacitive Discharge) systems donāt really have as much problem because in the case of the former - each coil is taxed less often, and the latter - the dump-to-coil voltage can be ridiculously high like 400V to the primaries. The result is more joules to the coil and higher output at any RPM.
This is all find and dandy unless your spark voltage exceeds the amount needed to jump the spark gap to the point where it begins arcing anywhere and everywhere itād like to around distributor bits because it can now reach grounding bits it couldnāt before. So bigger isnāt always better unless balanced by how things are setup.
Re the XJ220 setup vs XJRS, vs Marelli, the XJRS definitely uses the AB14 Lucas box with the GM amp in it. The 4 Pin GM amp inside is based on a mc3334 Motorola chip, and has a fixed dwell control. Normally the GM Amp is triggered by an inductive pickup, which is sort of a āsawā wave. Thereās no dwell information inside it like could be in a square wave. The XJRS uses Hall Effect sensors in the distributor, though, instead of an inductive pickup, so the ECU is seeing a square wave. Whether the ECU triggers the GM amp with a square wave output or saw waveā¦ I donāt knowā¦ BUT I can ASSUME fairly reasonably, that unlike the Marelli computer (square wave output w/ dwell info), the Zytek computer provides NO dwell control information to the GM Amp as it wouldnāt use it.
What does that mean for the XJ220 also using a Zytek ECU? IDKā¦ but it suggest maybe that the XJ220 ECU isnāt sophisticated enough to do dwell mapping based on vehicle voltage levels. While itās true that ing amp and coil need to be basically matched spec wise, NOT having ECM dwell control does make it easier to switch out ignition boxes.
~Paul K
2 Likes
I believe this was the main reason of dual-point distributors: it is an interesting problem, one thatās been visited for many decades.
1 Like
Correct on all your identifications. Pickup is a Hall effect mounted at the crank, the dizzy is empty except for the rotor. The box next to the coil is a Lucas AB14 ignition amplifier. Whatās inside the amplifier box is anyoneās guess. Everything is out for rebuild at this point, so the only thing I can test is resistance, which wonāt tell me much. Despite all the broken down foam in the fuel lines, filters, pumps and injectors, we were still able to drive the car around the parking lot before the teardown. So I assume the ignition system is running properly, but I have never been a Lucas fan (had an E type and an XJ6), and the Ducellier coil seems to have its own problems from what Iām reading. Trying to keep the car original if possible, so Iāll have to wait another 6-7 weeks until itās all back together.
Iāll just have to wait until everything is hooked up again and I can plug in the Zytek diagnostic unit and see what exactly the ecu is doing. I am a bit amazed that a 1994 twin turbo race car has no knock sensor, so no capability to retard timing if detonation occurs. I will be installing standalone knock sensors for each bank along with 2 wideband O2 sensors to a standalone A/F gauge, and manual/electronic boost controller. So Iāll be part of the ecu control system. What a terrifying thought.
2 Likes
Paul- a copy of the 1972 Popular Mechanics? I assume thatās for bathroom reading?
1 Like
For the bit on āhot roddingā ignition systems, ya, buddy!
Somehow, through some unbeknownst-to-me microcircuitry wizardry (@Ray_Livingston, you listening?), thereās likely some method to remove The Good Doctor from the equation, and integrate it into the system.
Huh. My bet, with it being labeled an AB14, is that the contents are just an off the shelf Jag box with the GM Amp inside. Itās not really āLucasā but in name and accessorization only. Itās probably worth cracking the box open (very easy via 4 screws) and taking a photo of the inside to identify the GM module (which I highly suspect is what is there.).
The coil looks to me like a standard 0.6 Ohm latter XJS unit:
If it becomes needed, or of interest, there are modern drop-in-replacment 4pin GM Amp modules from the likes of MSD, Crane, Pertronix, etc that product-description-wise say they can handle higher amperage to the coils AND some of which that have actual built-in RPM limiters or adjustable dwell (via little surface mount pot adjusters.)
~Paul K
Different coils will want different maximum dwell times and so concluding anything from a different coil isnāt really valid. The ECU will probably have a lookup table inside that keeps its charging transistor switched on or off to match the required load.
Megasquirt only gives you all of these settings to play with because it has no idea which engine it will be running, so you wonāt unnecessarily heat a coil with too much dwell.
It is quite common to drop the spark plug gap in boosted applications.
kind regards
Marek
1 Like
CORRECTION- The XJ220 does not have a crank sensor. At the front end of bank A is a round short canister mounted in the same place that the distributor is mounted on b bank. Inside this is a 2 piece (upper and lower) Hall effect sensor, and a trigger wheel doing double duty as both a camshaft and a crankshaft sensor. To that end, there is a round disk with square teeth driven by the exhaust camshaft. The upper end of the Hall sensor counts the teeth to determine crankshaft speed (It has some type of algorithm to get crankshaft speed from the cam, like multiplying by 2). This trigger wheel also has a square cutout in its bottom half. The lower part of the Hall sensor uses this to synchronize with TDC at cylinder 1. The signal is sent to the ECU in a squared off sine wave, and the ECU takes it from there.
1 Like
Quite similar to my Hyundai, which uses a Hall effect sensor on a squared-toothed extension of the intake cam, itself being driven internally, by a small chain from the exhaust cam.
1 Like