What a waste of a day off....

I knew it was going to be one of those days when my laptop wouldn't communicate with the Link ECU, luckly I brought a back up unit that I normally use, it's just that the 1st laptop I had was way, way faster...(normally)

Mark D (McMark) showed up and helped stomp on the gas while I played with the laptop, we started @ 1000rpm and went up in 500rpm increments...strange feeling with the pedal @ WOT in 4th gear and the motor would not go over the set load...

I kept hearing a misfire and then all of a sudden under load it got real bad, we changed out the distributor cap since the original one had some viewing holes cut into it for timing....it just got worse...then I noticed the air intake elbow was a little loose so I fixed that...I was hopeing that was it...nope still really bad misfires coming out of the exhaust....finally I load a new MAP setting and the car seemed to be idling better, then all of a sudden the car dies..."hey Mark, did you turn it off???"

"nope"

"oh $hit"

I get out the voltmeter, I have voltage like I should, no apperant spark....

So, it's one of the following I think
a bad MSD box, a bad optical sensor in the dizzy or the Link is messed up....I have a check list of items to go over before sending the Link back for testing....

I think i want to replace the optical sensor within the dizzy with a hall effect unit, any leads or ideas?

McMark dyno's his car, I'll let him tell you about it...he had better luck than me

Dang Mike,

very sorry to gear that BS happening.

If you plan for a weekend I might be able to help out in the future.

Cool that Mark could be there fo rmoral support if nothing else.


Hey, you know what's up better than most. It'll get done. No problemo'.

Couldn't make it today, sorry. Way too much work.

I think you should just use the Mallory Unilite straight for ignition, and remove the CDI box completely. Do fuel only, and get the car running really well. THEN, use an 009 or something like that (I have several if you want one), gut it, lock it, and do ignition experiments with that AFTER you get the fuel system working 100%. You can swap the Mallory in and out as needed until the tunable ignition works and is reliable. CDI boxes are just another thing to break, and unless you NEED it (which you shouldn't for a near stock 1.8), is just going to be an unnecessary distraction.

Yeah, it was a major bummer for Mike. We started to Jerry Rig something and the dyno operator told us we'd be better off just leaving it and coming back. He was right. Better to fix it and come back later than to burn time fixing it in the room. I hope it wasn't rude of me, but I jumped my car on the machine and bought up the extra hour. I was able to do some idle tuning and get the full throttle tuned in. It was nice, but frustrating not to be able to tune the full table. He said it would take two more hours to do that. Next time, I don't think I'll be there. I'll just drop the car off and let them do it. But I am making 86 hp and I think 99 ft/lbs.

Of course, I had my share of bad news as well. Right now my car is sitting on the side of the road just outside of Napa. I did a little side of the road trouble shooting and the fuel pump is blowing the fuse. So I'll tow it home and check that out.

It's been a real good news/bad news day.

Some time's this isnot what we need but it is a good thing that you found the problem on the dyno and not on the street any time spent on a dyno is good time .When you look back you will not do the same thing twice on your next trip to the dyno you will have your car set right for the day

bizarrely enough - I could really use some dyno time right now! Bummer that it didn't go well. Sorry to hear that. Sorry to hear about Mark's car too.

we have a "runner"......problem fixed...sorta...

broken electrode on the rotor......

The original electrode was too skinny and giving me problems with spark scatter, so I rigged up a new one. It had no issues with the original distributor cap...

Once we installed the new cap, the car really ran terrible until it just wouldn't run any more....I "think" this could be the reason

No new rotor, I fixed the broken one to at least get the car off the trailer and into the garage....I still have a slight miss that I have to hunt down.

James, I think I will switch it back to running only the fuel with the Link and tune it that way for now.

That rotor being so wide can definately add some weird things to the timing.... It can also throw the rotor out of balance...

I tried that once but had spark scatter so bad at 5,000 RPm that the engine acted like the valves floated..Swapped the rotor back and the problem was solved.

That sucks. I wonder why there was a problem.

QUOTE (McMark @ Nov 1 2005, 07:44 PM)

That sucks. I wonder why there was a problem.

I'm guessing that the contacts in the old cap had been worn down some or that the tolerance of the new cap was on the minus side???

Sorry to hear about your "issues" at dyno time - my beater Volvo drove me nuts with a simalar problem.
I would have guessed yet another bad MSD box - I gave up on them after the 3rd one died and MSD lost 2 of them during repairs. My old boss worked on the old Shadow UOP can-am cars and had a huge box (75-100) of dead MSD boxes from the race cars.
Do you want that industrial PC for your mill???
Steve

what causes the MSD boxes to die?

b

QUOTE

CDI boxes are just another thing to break, and unless you NEED it (which you shouldn't for a near stock 1.8), is just going to be an unnecessary distraction.

The MSD box is needed with the Link ECU, it's basicaly a relay/triggering device for the coil.

Link [5vdc sq wave output] > MSD box > coil

I don't "need" the multifiring of the box...it just that the Link cannot handle the flyback from the coil and I believe it uses the MSD for that.

Man, I swear the only luck you have is bad luck. You know reading your stuff is what got me going down the MS road. I guess it's time for me to come clean. This is probably happening cause my wife spends every night sticking pins in a little doll that look kinda like you.

Dave

Yea, bad luck tends to kick me in the @ss once in a while...the problems I've had with this FI and with the MS are mostly owner induced due to not reading the instructions or carelessness

On a more positive note, the owner/operator of the chassis dyno was impressed with the software/interface and mentioned it was one of the best he has seen...

I called the shop and left a message later that night to let him know what the problem was and that I think I still have the timing screwed up (right before the car died I "thought" I read 23° advance with the timing light on the fan at idle) The owner called back yesterday and voluntered to drive to my house to help me with the timing issue free of charge.

Also, Ben M (airsix) sent me a link to an Ignition Coil Driver which should be able to replace the MSD box completely..it's pretty amazing what such a small part can do.

ignition coil driver

This part is discussed endlessly on the MSEFI forums. Lots of circuit diagrams and info on using it.

Yeah - I think the MSv3 board can run a set of those for Coil-on-plug.

Tasty. I might have to upgrade

Mike,
You had the link firing the coil before? I thought you were running the link fuel-only. If the link was firing the coil, where was the ECU getting it's timing signal from? Was it getting it from a pickup mounted in the distributor (like a pertronix or optical pickup?). If you were feeding a timing signal to the link via the distributor, is your distributor locked? The link would need to get it's timing signal at the same crank-angle regardless of rpm/load. An unlocked dizzy will change the angle at which the ECU get's it's signal depending on rpm/load and the ECU will calculate the wrong crank-possition. I know you know this - I'm currious how you were feeding the crank signal to the Link though.

-Ben M.

IIRC mike has a Mallory Unilite (optical pickup) and he locked out all advance on it.

mike - why did you have to mod the rotor tip?

QUOTE (Aaron Cox @ Nov 3 2005, 10:13 AM)

mike - why did you have to mod the rotor tip?

Helloooo, because he's Mike.


Don't be so hard on yourself Mike (yes, that's right, I said "hard on"). They don't call it a learning curve for nothing. How many times did Benjamin Franklin fail to make the light bulb?

QUOTE (airsix @ Nov 3 2005, 10:11 AM)

Mike, You had the link firing the coil before? I thought you were running the link fuel-only. If the link was firing the coil, where was the ECU getting it's timing signal from? Was it getting it from a pickup mounted in the distributor (like a pertronix or optical pickup?). If you were feeding a timing signal to the link via the distributor, is your distributor locked? The link would need to get it's timing signal at the same crank-angle regardless of rpm/load. An unlocked dizzy will change the angle at which the ECU get's it's signal depending on rpm/load and the ECU will calculate the wrong crank-possition. I know you know this - I'm currious how you were feeding the crank signal to the Link though. -Ben M.

optical sensor in Unilite(locked)>Link>MSD*>Coil

the Link needs to see the signal 10° BTDC, I modified the rotor as a half-@ss attempt of changing the relationship between the rotor electrode and the trigger windows of the rotor...I need to make them indepentant of each other

*MSD acts as an ignitor or coil driver, to me it's just an on/off switch to dump the voltage on the coil

aaron, I'm going to get rid of the optical sensor and replace it with a Hall effect unit, if it works as planned, I'll be selling the unilite sensor (sorta cheap )

Fiid,

I've been reading more on the MSv3, pretty cool, I should waited...oh well....


James,

I'll be looking more on msefi for info on that driver, on the surface it seems too simple to be able to completly replace the MSD, but if it could, that would be nice....

You're going to pull the Unilite and fit your own Hall-sensor to the Mallory? Why?

Stop re-engineering things that work, dude...

Seriously, run the Link fuel only with the Mallory returned to normal operating mode. AFTER you get it all tweaked and working, use some junk 009 locked to fire the ignition using a Hall sensor, the Link, and the MSD box. You can swap back to the working Mallory quickly, so you can take your time to get the ignition system sorted and still have a usable car. Alternatively, you could leave the 009 unlocked and provide the crank signals from the fan using a method like Ben's, then you won't have the rotor phasing problem anymore, and you can still switch back to the Mallory to run the ignition at any time to restore the car to working order.

Okay, that makes sense. Sounds good. I'm currious about the Link only needing it's signal 10 degrees before TDC. How does it trigger the coil at say 27 degrees before TDC if it doesn't get its input signal until 10 degrees before TDC? Does it count off 323 degrees before it fires (ie. it's always behind one rotation). My ECU get's its input signal at 44 degrees before TDC so it has time to "think" before firing the coil at up to 40 degrees (high vacuum) BTDC.

-Ben M.

ps - Watts, you crack me up.

QUOTE (lapuwali @ Nov 3 2005, 11:16 AM)

You're going to pull the Unilite and fit your own Hall-sensor to the Mallory? Why? Stop re-engineering things that work, dude... Seriously, run the Link fuel only with the Mallory returned to normal operating mode. AFTER you get it all tweaked and working, use some junk 009 locked to fire the ignition using a Hall sensor, the Link, and the MSD box. You can swap back to the working Mallory quickly, so you can take your time to get the ignition system sorted and still have a usable car. Alternatively, you could leave the 009 unlocked and provide the crank signals from the fan using a method like Ben's, then you won't have the rotor phasing problem anymore, and you can still switch back to the Mallory to run the ignition at any time to restore the car to working order.

word. stop reinventing the wheel.

if link sucks for timing, just use it for fuel. get it running and driving....have some fun....and the fuck it up

The Link may use a "next cylinder" method, which is similar to what some GM systems do, and MS has a mode for it. The pulse arrives X degrees before cylinder C, but the ignition is calculating the advance for cylinder C+1. So, the pulse is really 90 + X crank degrees before the plug that's actually being fired, on a four-cylinder engine.

QUOTE (lapuwali @ Nov 3 2005, 11:50 AM)

The Link may use a "next cylinder" method, which is similar to what some GM systems do, and MS has a mode for it. The pulse arrives X degrees before cylinder C, but the ignition is calculating the advance for cylinder C+1. So, the pulse is really 90 + X crank degrees before the plug that's actually being fired, on a four-cylinder engine.

That's cool. I like it. Thanks for explaining.

-Ben M.

QUOTE (lapuwali @ Nov 3 2005, 11:16 AM)

You're going to pull the Unilite and fit your own Hall-sensor to the Mallory? Why? Stop re-engineering things that work, dude...

why?

frustration, stupidity, fear of the unilit failing....I dunno

I believe I have a much better grasp on things right now, I just made some errors that I can fix...I'll just have to write off the money spent on the dyno as a learning experiance not to monkey around with certain components.....it's my own fault for not making sure the timing was dead-nuts on.....on the plus side, I learned about some more neat electrical components that I could use for other things in the near future

James is correct....current cylinder upto 500 rpms,
next cylinder for above 500 rpms.

You can make the v2 megasquirt do everything that the v3 does - you just need a daughterboard and some wiring between the boards.

I would use two coil drivers and run wasted spark. It would only need two coil drivers, and an edis coilpack - which you already have right? Then you get rid of the dizzy altogether (all solid state).

Can the link do direct spark with two coil drivers?

QUOTE (fiid @ Nov 3 2005, 04:45 PM)

Can the link do direct spark with two coil drivers?

Yes it can....I'll get this version running perfect, no need to give up or tackle another possible problem

hmmmm...I just got an e-mail from Link, I asked for the wiring diagram for wasted spark (you only get the diagram for what you originally configured the Link for, I told them dizzy so those are the diagrams I got, however the box will do more with just a click of the mouse)


anyways....both the home office of Link in New Zealand and their US rep recommend staying with the distributor saying that for the most part that I will not achieve gains by going with COPs or conventional wasted spark.

I just have the fear of the MSD box going south on me...luckly I have an spare unit, not that I want to carry it around

mike - hijack... how do you TIME a MSD style igintion? will it read 4x as high? (4 sparks vs 1) {below 3000 rpm}

QUOTE (Aaron Cox @ Nov 3 2005, 05:19 PM)

mike - hijack... how do you TIME a MSD style igintion? will it read 4x as high? (4 sparks vs 1) {below 3000 rpm}

good question, I wondered if the multiple spark had an impact on the timing light...hopefully James will reply and give us some of his knowledge on the subject

QUOTE (Mueller @ Nov 3 2005, 05:27 PM)

QUOTE (Aaron Cox @ Nov 3 2005, 05:19 PM) mike - hijack... how do you TIME a MSD style igintion? will it read 4x as high? (4 sparks vs 1) {below 3000 rpm} good question, I wondered if the multiple spark had an impact on the timing light...hopefully James will reply and give us some of his knowledge on the subject

oh bummer - i was hoping you knew...

James???

logic would say that when timed over 3k, it SHOULDNT be an issue, due to returning back to a SINGLE spark setup....
but might make timing initial advance at idle more fun (XXX / 4) lol

Multi-spark boxes usually confuse adjustable timing lights, just as wasted spark systems do (the knob will no longer be accurate in terms of degrees of advance). However, most multi-spark systems are done by 3000rpm or thereabouts, so if you follow the usual rule and actually time for max advance, you're alright. Some of the newer whizzo timing lights have adjustments for multiple sparks per rev (wasted spark), though I don't know if any of them (yet) adjust for multiple sparks per event.

For non-adjustable timing lights, you'll get a flash per spark, and one of them will happen at the "right" time (the actual point of advance), so you'll still see the mark. Your eye is slow enough to respond that the multiple flashes shouldn't hurt, although the mark may appear to be more "ghostly" than usual.

So, for timing with a multi-spark box on a 914, mark the fan for max advance as well as TDC, and use a non-adjustable light, or an adjustable one set to 0.

rock on james....

so, an adjustable one set at max advance, ABOVE 3k, would not work?

It would work only if your particular CDI/MSD box switches to single sequential spark above 3k. AFAIK, most MSD boxes switch to standard 1 spark for each cylinder above 3k and so an adjustable timing light would work okay. But as I mentioned, anyone who is curious about this should check the documentation for their particular brand and model.

QUOTE

mike - hijack... how do you TIME a MSD style igintion? will it read 4x as high? (4 sparks vs 1) {below 3000 rpm}

always check total advance, granted what happens regarding ignition advance between 1000 rpm and 7000rpm does matter, but total advance matters most, on distributor based systems. Checking timing wit h MSD or Crane multi-spark boxes is the same as with any other system.

Another note to add is many aftermarket EFI systems rely on a crank reference. usually this is Cyl 1 TDC, however you cant expect the system to fire Cyl 1, 24 or so degrees before TDC if the ECU sees the TDC event at TDC?. (wondering if this make sense)

So you tell the ECU how many degrees to offset the Cyl 1 spark event. To do this you need to know the static timing.. ie how many degrees the system makes with no correction from the ECU

I have lots more to add to this but I can hardly keep my eyes open now LOL. Work get in the way way to often