|
|
Lehanover@aol.com wrote:
In a message dated 5/1/2011 1:28:10 P.M. Paraguay Standard Time, echristley@att.net writes:
I threw my back out again last week, and the drugs have my sleeping
pattern all out of whack. Last night, I spent
another 6 hours researching what other people have done. One guy
reported that he started loosing performance with 2*
of split at speed. It was a boosted engine on a dyno. The problem
I have with the RX-7 forum is that they are all guys
looking for ridiculous performance running street engines on a
dyno. You have to read between the lines on all of it.
He didn't go any further than 2*, because of the performance drop
off, but that fits perfectly with what you're saying
about detonation.
I'm going to go with what I've got now, but I'm seriously
reconsidering recasting the mount for the trailing VR sensor
to take out split as a near future project. There is more to fear
from detonation that a little more low end emissions.
> > That problem firing a double ended coil is the gap size on the
plugs. Use
> plugs you can change the gap on, and set the gap at .010" to .015".
> > I used those even with an MSD and never had the problem. Remember
the
> rotor is hauling the mixture past the plugs like a freight train.
> AR2592 Autolite work fine. Same heat range as NGK -10s. But usually
> $7.00 a box of 4 at Autozone. The open tip may get better cooling
than
> the NGKs
Ok, Lynn. That deserves some type of reward for being a brain dead
simple solution to the root cause of the problem.
If you're not getting a spark because you've set up to much
resistance to the current flow, cut back on the resistance.
I learned a bit more about this in the research last night. You
have a couple of very authoritative guys giving very
scientific and clear explanations of why using a split coil won't
work. You have another set of guys going, "Really?
I've been driving that setup for the past 6 months." The
differentiation seems to once again be boost. So I'm
confident that I can run the single coil set using the wasted spark
on the trailing, and I can just swap out the plugs
if it looks like I start missing when the engine is under load.
The ultimate goal, though, is to have two totally independent
ignition sources. The MegaSquirt guys have worked out the
circuitry to run the stock coils or even COP. The problem I have
with that is that it puts ignition control back in the
little homemade box. I like the idea of having the two highly
resilient, time-proven ignition systems that will keep
right on clicking if my fancy little computer goes tits up. I'll
loose control of the advance, but I can be confident
that the limp-home mode of 25* will drag me to crash site.
And that leads me to my next question. Operationally, what would be
the pros/cons of turning off the trailing during
cruise flight? Will it allow deposit to build up and ensure the
plugs are fouled if I ever do REALLY need them? Or is
it just putting a needless load on my generator in order to dump an
incremental amount of extra heat under the cowl?
The wasted spark thing only works on the leading plugs. One or the other trailing housing will have an apex seal between the plugs and will light the next dose of mixture. So leading only for lost spark.
Running without spark on the trailings just collects oil in the cavity and ruins the plugs . Been there, done that. Driver never noticed.
OK. Won't be doing that. Not much point in carrying the system if it won't be there if/when I need it.
Because of "A" apex seal between the plugs the trailing spark must be distributed, either mechanically with a distributor or crank angle sensor or a system with two pickups 180 apart and one reluctor.
Then the lost spark leadings can be triggered from another system with one pickup and two reluctors. Because it needs to fire twice per revolution.
Please tell me if I'm wrong, but I'm pretty sure that doesn't apply to what I have laid out. Traditionally, the "wasted" spark has gone to a cylinder that is on an exhaust stroke. I have the spark being wasted in the same cylinder (and I use that term loosely) and at the same time as the primary spark. As long as I stay away from igniting after TDC, the trailing spark won't get into the subsequent chamber. My initial design is essentially a single coil firing two plugs in the same chamber simultaneously.
The follow on design uses two VRs spaced 180*. On a dual rotor, the chambers fire at 180*. It uses two coil packs, each with two coils, for a total of eight plug wires. To make this trick work, I have to do two things. First, I have to ground one end of each of the four coils. At that point, all the energy from an ignition event will be dropped across a single plug, which is a nice bonus. The second thing I have to do is reverse the plug wiring. The leading plug on rotor one gets the wire from the first coil in the first pack. The trailing plug on the first rotor gets the second coil from the second pack. The EDIS modules don't know which chamber they are firing into, they're just igniting every 180* alternating between coils.
If I space the VR sensors exactly 180* apart, I will have no split. Slide the second VR around and I can make any split I want. Problem is, the split is across all RPMs, because I have no way to control the advance individually. In fact, I had been thinking that I would have no control over the trailing advance at all. I was wrong. The EDIS controller sends out a "PIP" signal that the ECU uses as a tach signal. The ECU responds with a "SAW" signal that is a square wave that the controller uses to determine the target advance. The SAW signal is not asynchronous from the PIP, but there is some timing leeway. I found out last night that I can take the PIP signal from one and then send the SAW signal to both controllers. I just have to be careful not to have a negative split. That is, if I take the PIP from the leading controller, then I have to make sure that the trailing is at least 180*. Else, it will get a SAW before it sends its PIP, and that area of operation has not been explored. The uncertainty lies in that I might be able to get away with 179*, because there is an indeterminate delay between the ECU getting a PIP and calculating a SAW, but it'll be simpler to set thing up to remove this uncertainty.
This same concept would work for a 3-rotor, except that you would use EDIS-6 modules, and space the second VR pickup 120* from the first. A 4-rotor would use EDIS-8 modules, and the second VR would be 90* from the first.
4 days ago the wife and I were waiting for a light to change at a 6
lane highway in our pristine Dodge Minivan. A moron in a 6,000 pound
Trooper II hit is from behind at cruising speed. A witness said he
had his head down and was busy *TEXTING............... .He fired us
into rush hour traffic to the curb lane on the other side, to the
stunned surprise of dozens of speeding drivers. Both seats failed.
Seatbacks laid down flat. Both went to the hospital. He has minimum
legal insurance. I hired a lawyer. grand Nephew of Woody Hayes. I am
going to try for the salvage. I can fix anything. That light had
been green for them for a long time and they were up to speed. Why
we were not "T" boned with 6 chances, escapes me.** *
I used to build totals, and this will just be another one. So My
wife can get a newer van and I can drive this one in a few months.
Still dizzy myself. No sleep yet. When I close my eyes it plays over
again.
* *Lynn E. Hanover
OUCH!!
|
|