Is there any way to keep the trailing plugs firing down to about 500 rpm?

I have my idle set about 850 rpm, occasionally it will drop down after clutch release and go below 800, then the trailing quit firing and the idle gets rough and runs down around 700 rpm. I short blip on the throttle gets it back up and smooths out the idle.

Seems like this would give the closed loop idle stuff a hard time.

TIA,

Scottt

 

the trailing doesn't affect the idle at all... if a blip fixes it, there's probably something else causing it. To confirm, you can turn off trailing (make sure to unplug power from the trailing ignitor as well), and see if the behavior is the same.

In fact, it's already well-known that trailing doesn't fire below 800... It's done that way by design because firing below about 600 rpms would cause us to need to rewrite some code, and it's not trivial. However, once it's firing, it shouldn't turn back off unless you drop below 700 rpms....

The best I could do is make it come on at 700 and turn back off at 600... below that, I'd have to add code to the low-speed-spark section to make trailing fire.

Tofuball and I haven't seen this problem at all... or rather, we were using leading only for a long time, at least until I fixed trailing, and we saw the same behavior where a quick blip would fix the problem... trailing was off the whole time since it didn't work yet.

Most likely you're getting to a spot in the map where your table is a little rich or lean causing the engine to sputter and run a little rough until you stablize it with the pedal. Idle recovery can be one of the harder things to tune, even after you get the rest of your table tuned.

To avoid the revs dropping that low in the first place, you may want to change your fuel-cut settings a bit... specifically, change the time before it comes on to 1 second, and increase the revs where fuel comes back on a bit; if you're using 1200, change it to 1300 for example. That'll keep the engine from falling on its face so hard.

Anyway, let me know how it goes.

 

Well I tried running on leading only by turning off trailing spark. It idled very rough and giving it a little gas produced backfires every time.

Not sure why, looks like I have some checking to do also. I'm pretty sure above 0 psi I'm hearing some light pings. I currently have the max boost set to 5 psi so I need to stay off the pedal till I get this all resolved.

I'm gonna check the timing again and take a look at the plugs.

Scott

 

I wonder if your leading plugs are in good shape.... tofuball and I ran leading only on his car for thousands of miles with no problems.... it's well known that trailing doesn't produce much more power... the main reason it's even there (according to some mazda rotary design docs that are posted somewhere in the megasquirt section) is for fuel economy.

If you're hearing light pings above 0psi, try retarding the ignition a little bit in your ignition map... maybe 2 or 3 degrees.... also try running slightly more split if you're not running any there...

I'll try to see what I can do to get trailing to fire even on crank... it won't be at the programmed split though, it'll just be at the same time as the leading until the 800/700 rpm cut-off point. I'd only have to do a lot of rewriting I think if firing trailing using the split table below that rpm.

 

Don't bother yet. I'm also not sure firing at the same time would be good--Normally there is 15 deg of split at idle.

My guess is one of the plugs is toast. I'll know more tomorrow.

Scott

 

firing at the same time shouldn't ever hurt anything... renns runs zero split all the time with no ill effects on his EDISed megasquirted rx7.

I'm also thinking that one or both of your leading plugs is toast. That can definitely cause knock.

 

Well the leadings don't look too bad. The rear leading one appears to be "too clean", almost like the metal is shiny and the electrode has been cleaned off.

Next, I'm gonna check compression then check the trailing plugs for grins.

Ken, if I hook up the timing light to the front trailing and check it against the -20 red mark on the pulley, assuming it matches--then there is no way the trailing signals could be crossed--right?

 

All plugs look normal based upon checking condition against the 3 sets I have from prior use.

Compression test is good. No changes from previous 2 years.

Timing looks good, checked all four plugs and they appear to be firing at the right time. The rear trailing is kinda hard to know since there is no mark for it, but no marks are visible so that's a good sign--at least it's on the correct side of the pulley.

Fuel pressure looks good.

Ken, for the wheel decoder, I have questions about two settings.

1) Enable hi-res dwell == on | off ?

2) Wheel decoder routine == 024s9 | 025 ?

Just installed a new LC-1 sensor today so I'll be double checking my AFR to make sure I'm getting the right fuel.

TIA,

Scott

 

I use hi-res on, and wheel decoder 025 for rx7s... and the way I check to see if the trailing plugs are in the right place is to take the cover off the CAS and set the split to 0.... then flash a timing light at the CAS, for the front rotor, you should be getting a flash between your trigger A tooth and return A tooth... for rotor 2, it should be between your trigger B tooth and return B tooth.

As for your problem, I still think something's wrong if leading-only doesn't work right... there are plenty of people who have run leading only without problems.

 

Ok I tried ignition without trailing on just to see what would happen.

Front rotor had good leading spark and no trailing spark--just like it should be.

Rear rotor had intermittant leading spark and steady trailing spark. I have no idle how this could be possible. Tach seems to read out in the car but is off by a good deal.

I thought both leading sparks worked off the same coil--maybe one of the wires is bad--but, it seems to work OK with the trailing spark activated in MS.

Any ideas?

 

It's hard to say, I don't quite understand what you're saying for starters:

Are you saying you turned off trailing, and then did the checks? With trailing off the front rotor has good leading and no trailing, but the rear rotor has intermittant leading and steady trailing? Trailing shouldn't be firing at all in that case.... it sounds like it's getting +5 volts continuously on the coil select wire, and getting its normal igt signal...

I think you might have 2 problems: your plug wire could be bad, or that post on your coil could be bad.....

The other problem (assuming I understood your description correctly) makes me think you have at least 1 crossed wire...

You are correct that both of the leading plugs fire at the same time from the same ignition coil, which is making me think that either the plug, wire, or that post on the coil could be bad.

 

Yes that's right. The intermittent seems to also have a pattern, it's not random.

Yea I'll have to double check the wiring. But explain how with trailing set ON in MS it seems to spark all the plugs correctly--then I disable trailing and the rear leading starts to act up? If the wire, post or plug was bad then wouldn't it also be bad when I have trailing on? Perhaps for some reason the coil isn't charging good enough to spark both plugs at the same time--is this possible?

Ken, have you had a chance to disable trailing and put a timing light on the trailing plugs? I'm just wondering--I wouldn't have done it unless I had these problems.

Crap I don't understand how it could work so good with it on then not work with trailing disabled. This sounds like it will be tough to figure out.

 

Scott,

The leading coil has to fire both plugs in order to complete the secondary circuit, so I'd guess you've got a partly fouled plug, or poor coil output. You might try swapping the plugs front to rear, or if drop in a new set if you have them handy. Do you have a scope, or even an old dwell meter? Try checking to see if dwell at idle changes while toggling trailing ON/OFF. Who knows, maybe there's a code issue. In rapid development mode, sometimes older stuff gets broken while new features are being added.

BTW, did you or anyone else ever test a stock FC ecu setup to see what the actual coil dwell signals looked like?

As for simultaneous firing of leading/trailing, that's what Mazda recommends in their race prep manual, for n/a cars at least. I ran for one season in that mode, as I was using a single crank-fired VR sensor. I've since switched over to a CAS-mounted 72-2 wheel triggering the two stock VR sensors at a slight stagger, to give a fixed split of about 5 degrees or so. Both methods worked just fine, with no ignition issues from startup to 7.5k.

Roger.

 

No I haven't checked the dwell, sounds like a good idea though. I've got a meter just need to figure out how to make it work for a rotary.

Somewhere in my readings on tuning, I saw it was not good to run zero split on a turbo engine--I'll have to see if I can find it. I knew for NA's it was OK.

Thanks for the advice Roger.

 

Found it

 

that just means don't run zero split on boost,... (if you're using my split table, you're running zero split on boost).

It should still be safe to run zero split off boost, especially since if/when I fix the code for cranking, you'll be running zero split on crank anyway. (probably only on crank).

 

If you run no split, you will have to run less total timing under boost but it can be made to work fine. You can run more total advance with a wider timing split. This will make more power.

N/a is a little different. I have a very backwards philosophy on timing that I've only ever seen one other person try but megasquirt can't do it. Once I get the RX-8 code, I'll be able to test the idea out better as I can make that work the way I need it to (with a little creative wiring) due to it's lack of wasted spark. The trailing spark dropout might make the idea unfeasible though unless I raise my idle up to 1000 rpm or so.

 

The MS2 should be able to do negative split when I get the time to make it work with the rotary. I've gotten a hold of the MS2 code that supports a wheel decoder... so I'm going to clean it up, make it faster (right now it won't handle over ~8k rpms) and get it working right for the rotary. However, I have 2 other software projects I want to finish before this one (the rx8 code, and the msns-extra wheel decoder that actually uses all the teeth on the wheel... at least for timing, if not for timing and dwell).

All of my software projects come after I get my first batch of plug 'n play units out the door. I've got all the megasquirt units built for that, and I'm working on the daughter cards and board mods right now... Last will be the adapters...

 

Guys,

I went out and retested the ignition setup with a timing light and rpm/dwell meter. Somewhere--somehow there is something messed up. I'm still not convinced at all that it has to do with a plug/wire/coil issue. This same setup is firing fine with the stock ECU and electrically working with MS when trailing is running. If you still think it's a plug/wire/coil--please explain why and how.

Here's what I got:

-----------------------
Rotary Trailing ON
-----------------------

Front Rotor
Leading--Steady on -5 mark, fast blinking
Trailing---Steady on -20 mark, medium blinking

Rear Rotor
Leading--Steady on -5 mark, medium blinking
Trailing---Steady but no mark visible, medium blinking

Dwell
Leading--2.5 deg @ 900 rpm
Trailing---7 deg @ 450 rpm

-----------------------
Rotary Trailing OFF
-----------------------

Front Rotor
Leading--Steady on -5 mark, medium blinking
Trailing---No spark

Rear Rotor
Leading--No Spark--but 1 or 2 occasional
Trailing---Steady but no mark visible, medium blinking

Dwell
Leading--1.0 deg @ 850 rpm
Trailing---3 deg @ 425 rpm

I tried to include a guess at the blinking speed of the timing light and I'm not 100% sure the trailing dwell was hooked up right--guess I should have two dwell readings on the trailing. The dwell readings are from the 8-cylinder scale.

The real low dwell on the leading with rotary trailing OFF may explain why only one plug is firing. No wonder it runs rough when trailing goes off below 800 rpm--it's running on the trailing plug only.

[RANT]Gotta admit I'm getting tired of dealing with these problems--first ignition and now my ve table seems to be off by a good 10%--I spent a good month getting that thing tuned in and now it's for crap.[/RANT] Man I'm real proud to be running this MS but I need to get thru this soon or I'll put the stocker in for a while and cool down.

The good news I got autotune working with a 2nd o2 sensor setup--whoo hoo.

Three questions I have:

1) Can anyone deduce from the symtpoms what is wrong?

2) Is it possible some kind of cross talk is occuring?

3) Is there another software build I can try--is 029G ready?

TIA,

Scott

 

This is just a guess. There's a lot I don't understand yet.

Fouled rear leading plug. Some of your symptoms sound a lot like what I had, which was that the front plug wire would fire for the rear rotor but not for the front rotor. Yours seems to be the reverse, with some additional considerations. When the trailing plugs are not helping you get NO spark from the rear leading. When they are helping you get only the front spark.

I put some chalk on the front edge of my crank pulley 180* off from TDC so I could also see the rear firing location. In my situation my front leading plug lit up the chalk but did not light up the standard timing marks. The rear plug lit up both. Add the chalk/tape/paint to the pulley and see if you're getting a rear leading spark when you're supposed to.

I took the leading plugs out, pulled the ECU fuse, and cranked until no more fuel came out, then put fresh plugs in. Problem solved.

Good luck,

 

This is not a megasquirt problem, you have a problem with the coil, plug, or wire. If leading on the front rotor is firing fine, and not on the rear rotor, then there is a problem with that coil in particular, or the wire, or the spark plug. If the megasquirt was at fault, you wouldn't have spark at all on leading. They are both supposed to fire at the same rate. Because one is firing and the other one is firing less often, that also tells me this isn't a megasquirt problem... the megasquirt can't tell one to fire more often than the other. The leading plugs should fire at the exact same rate no matter what...

Also, if you turned off trailing in software, you should make sure that the two LEDs that were set up for trailing aren't blinking. they should both be very bright... if you set up those LED's for spark output, but turned off trailing, the results are going to be undefined.... mainly because if you were to turn off the megasquirt and turn it back on with those settings, you'd just get a fuel pump that turns on and off (error condition). To "really" turn off trailing, you've got to turn it off in software, change those two LEDs to something other than ignition, and unplug power from your trailing ignitor/coil assembly.

Your fuel map is most likely off due to the problem you're having with spark. If not all the fuel is burning (and it looks like that's the case) your fuel map will be off. Also, if your spark advance or split are different from the stock advance or split, that will also cause you to need to retune. That is totally expected.

I suggest that you switch the plug wires around (just reverse them on one end or the other) and retest. If the problem doesn't move, switch them around on the other end, and see if the problem moves... if it still doesn't move then you know it's not the wires... in that case I'd suspect the plug itself... I'd take Eagle7's advice in that case... if that still doesn't solve the problem, then you know the coil/ignitor aren't working right.

I understand that you're getting frustrated, but like I said, this isn't a megasquirt problem... this is a problem with supporting hardware.... I think you probably flooded the engine somewhere along the line, and only the rear rotor didn't clear out. IF the stock ECU doesn't exhibit this behavior, the only reason I can come up with is that maybe it commands more dwell... which is causing it to fire anyway even with the flooding. If that's the case, you can tell the megasquirt to dwell for 3.5 or 4 ms and see if it helps, but honestly even if it does, you should still follow Eagle7's advice.

Finally, I would suggest turning off trailing and unplugging power from it until you get this problem solved. Once you get the engine running on leading-only, and it runs nicely, then turn trailing back on. I strongly recommend to everyone that they do this... it helps them get started and make sure all the essential components are working correctly before adding hardware and software complexity by wiring in and turning on trailing. I'm sure I've said it before on this forum as well.

 

OK, that may be it. I did not turn off the LED settings or disconnect the trailing coils.

I just pulled the plugs last weekend and they looked fine. I'll keep going on this some more, hopefully I can figure it out.

Scott

p.s. And I did get the alternating fuel pump so that must be it.

 

^^^^ that did it. Turned off FC trailing, disabled the led spark outputs, and disconnected the trailing coil--all works great--runs real smooth. Took it up to 5 psi boost and no pings at all--before I would here mini-pings at about 0 psi.

Running on leading only works fine.

Now to figure out what's going on with the trailing.

Oh yea, undefined operation is bad

 

OK, what kind of split were you running at 0 psi? Advance?

 

At 0 psi and rpm > 3800 the advance is 25 deg with a split of about 12.5 deg.

I think I was getting a bad trailing spark--now just to figure out why. Are the LED/coil outputs sensitive to emf interference? I.e. Can the wires easily crosstalk?