The battery is in the trunk and at the far end of the system. The alternator goes to a large terminal strip on the fender, and is thus pretty close physically to the fuse block for the ECU. I do not know if this makes a difference. Should I maybe bring a dedicated line up from the battery to the EFI fuses? I think this is the way Accel was wired.

Sam

_________________

Ooh... Ok, I can see why your friend thought of that. That does make a difference.

I would not run a wire back to the battery. Yes, that'd work, and might work well; but it's not really "elegant". I mean, it's yet another wire... Hmm, lemme think about it. In some ways it'd be nice to have the charging wire all by itself to the battery, then distribute power out from the battery. Think star configuration, with the battery at the center. Just costs copper (and space).

A quick fix would be a choke on the leads running to the ECU, with some additional capacitance on the ECU power leads. Do you have a GND distribution block also? I'm not sure I would wrap the choke around the GND wire also to the ECU. Then again, the noise doesn't appear to be on the ECU power leads, so what the heck am I muttering about...

Will check in later. My head is shifting towards the workday.

A capacitor on the power line to the ECU should filter out any noise from the alternator.

_________________
Ed
64 Valiant 225 / 904 / 42:1 manual steering / 9" drum brakes

Once again, thanks for thinking about this. I know it is an old sad tale by now. Ed there is a capacitor on the ECU power line now, but maybe not big enough. There is a 3 year old power point show on this subject on line with very little text, but lots of intriguing illustrations.

http://www.bgsoflex.com/megameet2008/me ... gnding.pdf

You will notice one page that states that a Stereo Capacitor will work wonders on an illustration of just what you recommend. However the capacitor appears to be quite large. IN asking around, I could not find anyone who knew what a Stereo Capacitor was.

Back to the choke idea, is there any harm here with a belts and suspenders kind of approach to getting rid of noise? Why not choke and capacitor? And, what spec choke do I need to ask for? Supton, you said you thought radio shack has a clamp on choke. That sounds easy. However, I thought the choke windings had to surround the wire carrying the current.

For anyone taking a look at this PDF of the power point show, you will see that there is an entire section on grounding in which they suggest separating the grounds for full power components from the sensor grounds. Unfortunately there is little text as this was serving to illustrate a lecture that took quite some time. Without accompanying text, I am afraid I was not clear on what parts of the system are power parts, and how to separate the grounds. That seemed to me to violate one of the rules of grounding. Another thing they show is all the sensor grounds going to a common bus wire which grounded to the ECU, , and not going all the way to the ECU separately. Right now I have the sensors grounded to the block with everything else. Insight on this?

Thanks again.
Sam

_________________

A car stereo capacitor (note the brand of the one on the pic, RockFord Fosgate, a well known car audio brand) is just a really high value stiffening capacitor. "Small" ones are 1/2 F (yes, half a farad) - avg ones are 1-2-3F, all the way up to 20+F for the carbon pile types. They are intended to keep voltage to high power car amps steady in demanding situations (low bass notes that take immense power to reproduce). For your purpose a 1/4 to 1/2 F is plenty. Yes, they can get physically big. A 1F may be 3" in diameter and 10" tall.

Check here or here - you can try your local car audio store but be prepared to pay a premium as flashy graphics and gimmicks (some usually come with volt meters built into the top) they are connected with are pricy.

When you say chokes you need to be specific. Ferrite chokes are just hollow magnetic cylinders that surround a wire. They make removable ones (two halves that snap together in a plastic case). You probably have some non-removable ones on electronics in your house. Check their power cords - if you see a cylinder (1/2 - 3/4" diameter, has some heft to it) at the end of the AC cord a few inches from the wall plug... that's it. When you just say "choke" alone that implies a wire wound inductor that goes in series with the power wire. This acts as a low pass filter blocking any higher frequency noise. It will have to be rated at appropriate amperage.

I don't think I ever used a ferrite choke in a car but that's not to say its a bad idea. There are car stereo noise isolators that some put on the main radio feed - this could work for a MS as long as again the current rating is appropriate. I thought they referred to this sort of gizmo in the megamanual or one of the other MS sites/literature. If your just itching to do something to the main MS power line I would put a cap on it and call it a day. Remember to wire it as close as practically possible to the MS.

Look for a local industrial or electrical liquidator in your area.

They may have some large value capacitors for a dollar or two. (25volts or higher)

_________________
Ed
64 Valiant 225 / 904 / 42:1 manual steering / 9" drum brakes

Here goes my dummy border showing up again. Would the starting capacitor for a large electric motor work? Those usually have more than two leads. I was thinking that 2 of those leads might have the traditional capacitor function. I have a couple of large failed ones off our compressor at work. I'm thinking maybe the failed part is irrelevant to what I need here. Is there a way to check the capacitance capability of a capacitor. I am also thinking maybe they were not both bad, even though we replaced two of them off the same motor.

Does anybody know what a stereo capacitor is?

Ed I will look locally for an electronic wholesale/liquidator.

Sam

_________________

Stereo capacitor is just a really large value capacitor. You should not need one, but it won't hurt. Personally, I'd drop the wire gauge before going to such a large capacitor. What size capacitor do you have currently? what are its markings? A motor starting capacitor again won't hurt, but IIRC they are not particular large in value. The high voltage rating is of no use either. But more capacitors won't hurt in this situation, they'll just take up space.

Clamp on choke does surround the wire. Clamp on chokes of this sort come in two (or more) pieces. Install one half on the wire, then snap the other over it. Forms a circle (or square) of ferrite around the wire. You can increase the choking dramatically by first wrapping a few turns of the wire around one half of the choke, then snapping the other half on.

I'm not convinced that the noise is necessarily on the ECU power lines. It's possible for that to be clean, and for your VR signal to have noise coupled onto it by a nearby noise source. You may have better luck with the choke on the VR signal instead--if that is what is bouncing around, that is. Or it could couple on elsewhere.

Alright, I dug through the whole thread, trying to come up to speed. The problem being the VR signal somehow being off/noisy/bad. The VR signal is coming from the distributor, to an HEI module, where it gets squared up and sent to the MS controller, which then "talks" the HEI module to tell it to fire the coil.

If I am reading this correctly, the 4 pin HEI modules will fire the coil when it gets a VR signal, while the 7 pin version has two separate sections, separated by the car ECU. yes/no?

I wonder if getting a second module would be of use. Use one for the VR trigger only, and the other for the coil firing only. If there is any sort of noise inside the unit (doubtful), going to separate units would help. If it's external coupling around the module, onto the various wires, going to two modules would help get some distance between the two.

Something that is in the back of my mind is the fact that spark voltage goes up under boost. Obviously the coil is strong enough to light off the plugs. But, if it is ignition related noise, well, under boost the noise would get worse. Have you tried a tighter gapped plug? I realize it's firing fine, the test would be to see if lowers EMI from the ignition event. Also: does your AM radio receive well? Strong electrical noises tend to take out AM reception first.

Another thought: if you think it could be altenator noise, you could disconnect the altenator.

Have you swapped out the distributor? If the reluctor could "walk" around, it could mess with the VR trigger. You could be dealing some sort of mechanical resonance, that happens around your 3,500rpm. [When does the oil pressure relief valve kick in? wonder if anything changing loading on the camshaft might cause play in the timing... stretching pretty far on this idea...]

Max actually recommended disconnecting the alternator for a test.

The idea of two modules is intriguing. I think your interpretation of the logic is right on. I am not sure how I would wire two separate ones. One of the wires is 12+so it seems both of them would need this power.

The spark gap is 22 now. Pretty small. I will check the AM radio. I think it has never worked well. I have the same wires on my MIata and the AM radio does not work well in that car either. Maybe I need to get wires that have less interference. These are the wires that Dan always recommends. Magnecor.

The Miata seems to have good filtering, as I seem to have no problem with EMI effecting the ECU or drivability.

Tomorrow, I will give you a better summary of things I think might be areas of concern, such as how the ECU gets its power, etc. One of the things that has always been a complicating factor is that none of the Mega Squirt documentation I have seen include the standard relay box in the diagrams. So all of their discussion of fixes for this kind of thing show how you would wire filters and such directly to the ECU. None of my sensors, all grounds, and power circuits all flow through the relay box. Consequently, I am not sure how to make these recommended fixes with the relay box interface. I would need to dig further into the wiring between the two. Supton, I appreciate your thinking about this. I think our other members have gotten pretty much over done on this topic. So it is nice to have a new perspective and new mind thinking on it. Thanks.

Sam

_________________

I applied a ferrite clamp-on choke to the alternator feed with noticeable improvement. I ran it easily up to 4000 RPM today with no misses. I did not push it higher, and I was only up to 5 lbs of boost, which is not WOT. But, this is the first time I have been able to do that since going to ignition control. Part of this improvement is likely the result of the timing improvements I have been making. Almost every cell in the map except the high boost areas have more timing than before. Sometimes by 10-15 degrees.

Now then, when I mentioned to Emsvitil that I could not use the AM radio at all with the Magnacore wire, in either the Miata or the Dart, he became suspicions of the spark plug wires as part of the problem. He was recommending maybe trying a set of stock wires. I may do this. I do not have the old wires. I would have to buy a new set. If anybody has a good recommendation for a good set of radio suppression spark plug wires, that would be welcome. Otherwise I will likely go to NAPA and buy their stock stuff. I know there are some trick sets out there with angled boots and such. Any advice is welcomed.

I plan on putting a clamp-on ferrite choke on the 12+ feed for MS also. They come in packs of two, and I bought 2 large, and two small. The alternator feed required the larger one.

I know this seems like slow going, but I have made the mistake in the past of moving too fast, and changing too many things at a time. Refining the tuning has improved things as well. So, not perfect yet, but really getting better and better.

Sam

_________________

I went out this evening, and applied the two small ferrite chokes to both the switched and unswitched 12+ feed to the Mega Squirt relay box. The results were very positive. I started it up, and it idles better. I can tell for sure. I did not take it out and drive it, but things are better.

The main thing is, once I had these chokes clamped around the feeds, I recognized them. As Pierre pointed out earlier, you see these things on the direct current input end of the converters for you lap tops and pocket PC's. I always wondered what the lump was at the end of the cord feeding the laptop. Now I know what it is, and why. Even though the transformer changes the alternating current to DC, there is the potential for a disruptive wave in the DC, and the choke is there to smooth it out. Now that potential is being dealt with by the chokes in my charging system and into the ECU just like the lap top has.

So why have I never heard of this simple, extremely common device before this challenge of filtering the ECU signal arose?

Supton, would not putting a ferrite choke on the VR signal defeat it or mess with its shape? The VR signal is supposed to be a sine wave, or AC. My understanding is these chokes somehow or other etenuate the AC wave. Is that an over simplification of the concept?

Sam

_________________

In reading on the internet, (how it works and Wikapedia) I am getting two different impressions of what they do.

1. One impression is that they keep any RF signal picked up by the wire from getting past.

2.The other is that they keep an RF signal from being broadcast by the wire, and thus silencing its noise to the environment.

So if number one is right, then placing them on input wires close to the device is the trick. If number two is right, the you could put large ones on the spark plug wires near the plugs.

Any insight? These things are cheap, and seem to, at the very least, do no harm. Any thoughts on my inquiry about their effect on the VR signal as Supton suggested.

Sam

_________________

Ferrites can do both. You can think of the property as being an RF block, if you will--it will block in both directions.

That is interesting; I would not have thought there was that much hash coming out of the altenator. Guess I'm wrong. Ferrites to the rescue!

As for the ferrites on the VR signal: they likely will do no harm to the VR signal. Distributor is running at half the engine speed, but at 6 pulses per, so you have in reality 3x engine speed for a signal. 6k rpm x 3 / 60sec/min means about 300Hz. The ferrites cause an AC impedance to be formed with the wire--think of insertion of a frequency-dependant resistor. If you were to use lots of ferrites, sure, you could have an impact. Maybe. Without attempting to figure out the impedance of the VR signal, and the impedance of the choke, though, I'll go out on a limb and say your emperical method is perfectly fine here.

As for "never hearing" before, obviously you're not in ham radio. Us guys have a nasty tendancy to use those things on practically everything in our radio shack...

Oh: the convertor used on your laptop isn't quite as straightforward as you think. Generally speaking, those little (but powerful!) supplies do the following: they rectify AC into DC (with lots of ripple), 170 to 340V. Then use a very small transformer for isolation and to step down the voltage to something much more modest--17 to 20V, give or take. But that transformer is not running at 50/60Hz but rather 100kHz or upwards of 1MHz. At those frequencies ferrites really do start to make a difference on the harmonic content, helping to choke off the emissions and allowing you to hopefully still use your AM radio.

I've never heard of using a ferrite on a spark plug. Can't hurt to try. But I'm going to guess it won't do much. Ferrites work off the current flow in the wire (it's a magnetic thing, not voltage--capacitors "care about" voltage, while inductors "care about" current). Spark plugs use very little current. I haven't seen a good plot of a spark event (you know, one with timebase shown) so I really can't be sure that the ferrite won't interfere with the rise time of the voltage. But I do think resistor core wires will go much further than ferrites on the wires.

It was a ham radio guy who told me about these>

SAm

_________________