MSII Extra serial 321. Have LC-1 Noise issues and settings in error help!

[OhBilly]

Mike,

I doubt the battery could be causing you much fuss so I wouldn't put too much effort into that one. However, there are two things you should do/check: (1) Check your plug gap and set it to about .025 to .028". Any more gap than that and you'll end up with misfires, especially at load, and those misfires can cause resets via pin 6 on the CPU. (2) Solder a 0.1 uF (100 nF) cap across pin 2 and pin 6 of the CPU. Pin 6 is your reset pin and voltage fluctuations there will easily cause a reset. In my research, it seems nearly everybody has to install that cap to get rid of resets, especially under load. The cap will hold pin 6 at a much more steady 5V and make the system substantially more stable.

[motomanmike]

I'll have to email Matt on that one or maybe he will chime in.  I'm running this 3.57 board.  I sold the 3.0 I built to a friend for his Supra a little while ago but i'm pretty sure i've got one of those caps from the kit still. Wish I hadn't now, they are so much easier to work on.  I'm half scared to touch this 3.57 pcb.  You are talking about pin 2 and pin 6 of the 40 pin socket U1?

Edited January 4, 2013 by motomanmike

[OhBilly]

Yes, that's the one. This thread here is pretty informative as to what's going on:

 

http://www.msextra.com/forums/viewtopic.php?t=27704

[motomanmike]

I looked I don't have any so it looks like i'll hit the bins at radio shack this week. If they don't have them i'll have to order from digikey i guess. I"m assuming this same procedure will work with the MSII boards?

Edited January 5, 2013 by motomanmike

[motomanmike]

Its been little bit of a successful morning. I hooked my wideband back up and routed it nicely. Soldered a different LED on the status of the LC1. I got the car started and long and behold it stayed online for awhile. So I drove it. It didn't run the best but I was able to run a little around the school parking lot across from my house with the live VE analyze and get it driving. It wasn't as jittery. So I drove it down the road about 3 miles and back carefully watching the wideband and driving easy. Afr's were never above 15.5. Then the car wouldn't idle right and kept wanting to cut off. I got it idling halfway right and let it run about 5 minutes. AFR's in the mid 12's. Then i started watching the realtime display and i kept getting small voltage spikes. My wideband isn't hooked to anything MS other than the signal wire. I ran a separate relay to power it. I shut the car off. Then tried restarting it and now I can't get it started. I get a little putter here and there trying to do the flood clear with the TPS and rolling off of it it would sputter and try but wouldn't catch and stay fired up. In this time i've pulled all six plugs twice and run them on the wire wheel and re installed them. Something in the priming pulse, cranking pulse, and timing i think are affecting the startup. Its progress. It stayed online a good hour today just running and trying to tune a little. I've just got to get the startup nice and snappy and I think i'll be heading in the right direction. Its odd to me things quieted down some with the wideband hooked back up. Its very strange because that's all I've done is hook the wideband back up. Voodoo i tell ya. There is Voodoo in that black box. I'm going to let everything cool down. Let the battery charge a bit and try it later tonight so see how it acts in the warmup enrichment parts of the starting cycle. I'm happy with what i've gotten done though. I was able to keep it online much longer than I have in a LONG time so its progress.

Edited January 6, 2013 by motomanmike

[motomanmike]

I walked out there a little bit ago. Same thing. It will crank and then barely sputter. Won't fire up.  I'm pretty sure the plugs are fouled so i'll get a set or two this week and see what happens.  If that doesn't work i'll reflash the firmware and see if that clears things up.

[OhBilly]

Mike, I have found that turning the priming pulse completely off is the best way to go until you get the car starting reliably and idling reliably over all of your expected ambient temps. Otherwise you will be spraying fuel every time you cycle the key off and back on, and that can quickly add up to being too much fuel in the cylinders. I have mine starting very well and I still haven't turned the priming pulse back on. It takes maybe a few extra seconds of cranking but it's no big deal.

 

Your cranking advance should be about 10 degrees BTDC. Cranking threshold should be about 300 RPM. Assuming you are using stock 270cc injectors, your required fuel should be about 11.3ms. Your cold cranking PW should be about 88% of that, or 9.9ms. Your hot cranking PW should be about 23% of that, or 2.6ms. For all your other temperatures, just do a linear interpolation between the hot and cold and that should get you very close, at least close enough to get it to fire up.

 

For afterstart enrichments, these numbers should be close (from cold to hot):

70, 64, 58, 51, 44, 37, 30, 22, 15, 7

 

For warmup enrichments, these should also be close (from cold to hot):

165, 155, 145, 136, 128, 122, 115, 110, 104, 100

 

It may help if you post up your latest .msq and perhaps a datalog.

 

Good luck.

[motomanmike]

Awesome information. I'll turn the priming pulse off. Makes perfect sense. Cranking is 10 BTDC and I think cranking dwell is 6 i think. Also my cranking rpm is at 200 which could be a big part of the problem, its probably getting into the VE table during cranking and dumping way too much fuel. If I get off work early enough this week i'll change the plugs and adjust my MSQ.

 

I reverted back to a restore point yesterday of when the car would fire right up so it makes me feel even more so that my plugs are fouled. Attached is my current tune. VE'S have not been tuned so it may still look very rough. Also it appears my AFR targets are way off. I'll get to that point very soon i'm hoping. That is the fun part i'm looking forward to.

 

3.2.4 FIRMWARE on MSII 3.57 box for now

current tune.zip

Edited January 7, 2013 by motomanmike

[OhBilly]

I took a look at your tune and I notice a few things that might be causing you some trouble.

 

1. You have max running dwell set at 1.5ms, but I think most people have it set to about 3.0ms. My running dwell is set at 3.0ms and it seems to work fine (280ZXT coil).

2. A few of your cranking PW percentages are a little goofy. I'd try using the default values (cold to hot): 326, 301, 276, 251, 226, 201, 176, 151, 126, 101

3. Your ASE and WUE have some weird spots in them, I'd use the numbers I posted in my last post above.

4. I'd definitely turn off your priming pulses. You've got some pretty big numbers in there which will likely flood your engine pretty quickly with just a few off/on cycles of your ignition key.

5. Increase cranking RPM to 300 (or even 400) like you mentioned. I used to have my cranking RPM set at 200 and it gave me a fair bit of trouble (it would only start after my battery started to get drained down and the cranking RPM started dropping).

Edited January 7, 2013 by OhBilly

[motomanmike]

I will change everything you recommended.  I had my dwell at 3.0.  I was trying to curb some of the noise by lowering it down until the car started misfiring and then up by .5 as i've read to do but it could be part of the issue i'm having.  Just not enough time for the coil to charge right. My spark plugs wires show in route for delivery today. I doubt i'll have time tonight to do anything due to a work meeting after work (go figure).  Hopefully i'll have an update by wednesday.  Thanks again!! 

[motomanmike]

Thanks for all the advice on the priming pulses OhBilly. I set all priming pulses to 0. I had no idea how much it actually affected the fuel pump itself.  I really thought it was how much pulse the injectors gave when you turned the key on, however it greatly affects the strength and amount the fuel pump runs.My pump sounds normal now, I can't even hear it when the car is running now.  I put my new Magnecor wires and put in some new plugs. Set dwell to 3.0. Cranked about 1.5 seconds.  It didn't fire up immediately so i popped a vacuum line off and it fired right up.  I let the warmup cycle run at idle until she warmed up. Then put the vacuum line back on and she's back up and running pretty good.  I took it down the street and ran a decent length data log.  I've still got alot of tuning to do but its back.  So awesome.  I really think the priming pulse was key to it starting up.  The wires can't hurt either. I'm still getting some pretty nasty voltage spikes but with the new plug wires its not affecting the pulse widths nearly as much so the car doesn't feel as jittery.  It does somewhat but i'm pretty sure its mostly the tune itself being so rough because there is only a few spots on the datalog that you can see the voltage spike affect the pulse width.

 

  Attached is the longest data log i've been able to run since this all started. I got one reset as I rolled the key back to shut her down at the end.  Its not the cleanest but I did hit 7.5lbs of boost and the car drove really well considering what I had expected. AFR's stayed well in the safe range under boost and driving, running rich which is fine by me at the moment.  The wideband is what it is for now.  I'll keep the delay at 1/3 a second on the controller and make sure I watch the lambda delay using the VE analyze. Later on i'll move it further back in the exhaust system.  I'm very happy.  Now to install all new parts from the flywheel back.  I've got a T5 trans, ACT clutch, eibach springs, kyb struts, poly bushings, R200 rear, mustache bar and some silver powdercoated Techno Toy Tuning rear lower control arms to all install. Should hold the power from this set up fine which i'm not looking for much more. I'm hoping to have it buttoned up by the time Ocean City cruisin gets here so I can show the muscle cars what HybridZ is all about.

 

Thanks again everyone.

getting there.zip

Edited January 9, 2013 by motomanmike

[OhBilly]

Congratulations on getting it running again. Glad I could be of some help but I wouldn't attribute your fuel pump working better to getting rid of your priming pulses. The priming pulses do exactly what you thought they did, which is spray fuel when you turn the key on. They shouldn't have any effect whatsoever on how the car behaves once it's running. Did you do anything else differently that could have possibly caused the fuel pump to work better?

 

I took a quick glance at your datalog and didn't notice anything too out of whack. You are definitely running rich the majority of the time, looks like AFRs in the 10 range typically. You are getting AFR spikes on throttle lift but that is probably due to your decel fuel cut settings, and are probably normal. Since you are running rich all over the board it would probably be prudent to drop your required fuel value a little bit to get you closer to a more nominal AFR before you start tuning. Otherwise you are going to end up with quite large changes to your VE bins and it will take a lot of time to get the whole fuel map smoothed out. Since your VE map is already pretty lumpy it may be beneficial to go back to a smoother map before you start.

[motomanmike]

I didn't change a thing as far as the pump is concerned which is really strange.  I noitced my AFR targets were really rich, I had set them in the low 12's originally as a safety precaution. Like I said i've only driven it maybe 200 miles since the swap. I've got a few different maps i'll copy the AFR targets and see how it acts after some datalogging and i'll drop the required fuel down a few notches to lean it out some.  The clutch is fine under normal conditions but as soon as boost hits its slipping.  I'm sure i'll spend alot of time this weekend with it but jees i'm SOOOOO glad my pulse widths are staying pretty solid because before they would spike and the car would buck.  I noticed the decel spike on the wideband and i've never been able to get it not to do it. Just one really quick spike then its back to normal AFR's.  I guess when that throttle plate snaps shut its a quick transition.

 

Billy you've given me a lot of great advice and I really appreciate it.  How are you making out on the track car?

Edited January 9, 2013 by motomanmike

[OhBilly]

The AFR targets on your tune, at least the "current tune" you posted above, look okay so I wouldn't go changing those just yet. I assume you have EGO feedback turned off at the moment? If you don't have it turned off I would turn it off until you get your tune closer to where you want it. You should think of using EGO feedback as what you use to "fine tune" your VE bins. Get them close first, then let the wideband do its thing. Letting the wideband do the tuning when you are this far out will likely give you a VE map that is unpredictable outside of your normal range of driving conditions. Typically you'll be running in a diagonal set of VE bins running from bottom left (low load, low RPM) to upper right (high load, high RPM). The wideband would likely do a pretty good job of tuning within that narrow diagonal band, but not outside of it because you don't spend much time there. But you do spend time outside that narrow band for short periods of time and if the wideband hasn't tuned those areas very well, which it really can't, you'll end up with that jitteryness you talk about.

 

The track car progress is progressing at a snail's pace. Someday... I hope.

[motomanmike]

Very good advice. I have the EGO off but i didn't realize it would actually change your VE table with it on I thought it just did some corrections based of current VE and adjusted pulsewidths towards a target AFR. I didn't realize it would actually change the VE bucket in that part of the table.  I'll leave everything as it is.  I plan to run a good 1/2 hour to 45 minute datalog sometime this week and set the VE analyze on very hard and see what it does to the tables.  Do you use the EGO at all with your set up?

Edited January 9, 2013 by motomanmike

[OhBilly]

I think you're right about simple EGO correction not changing your VE bins, it is Autotune that does that. However, utilizing EGO correction when your tune is so far out is even worse in that case because your VE bins are not changing. MS is constantly trying to get you into your target AFR zone, iterating over and over until it gets there (if it even has enough time to get there). This is a great recipe to create jitteryness!

 

I would advise using a well established VE map, perhaps one of the last ones posted by cygnusx1 in the map sharing thread. His VE maps are smooth and would be a great starting point. After getting the car started and warmed up on a known decent map, I would drive the car without putting much load on it and watch your AFRs, getting a good idea whether you're running rich or lean. Since you would be running on a pretty smooth map, I would then simply adjust your required fuel value up or down accordingly, a little at a time. Re-drive the car and again watch the AFRs to see if you're closer. Keep doing this until you get it as close as you can. Only then would I turn on EGO correction to let MS fine tune things, but I would make sure that I had a known good AFR map loaded beforehand. With that in mind, it might be a good idea to use a VE map and an AFR map that are known to work well together.

 

I currently am not using EGO correction and don't plan to until I get the tune closer first. Unfortunately I don't have much time to work on the car so it will likely be spring before I get started on it again. It runs pretty well just like it is though...

[motomanmike]

Sounds like a plan.  I've got a copy of cygnus's most recent tune.  If I remember though his upper kpa range goes to 220kpa i think.  I'd just have to scale it down to my set up.  I didn't want to waste that many more cells of my table that i'll never see so my upper limits I have set at around 165 i think.  Thanks again.  I am in your boat with time.  Its the hardest thing for me to come by.  I need a break from this mess too, i've kind of been all consumed with it the past few months. 

[OhBilly]

I wouldn't scale his maps to fit your kPa range, I would take his maps and get rid of everything above your max kPa. Then add additional rows as needed to fill in the missing kPa lines that cygnus' tune doesn't have (interpolating your new lines as needed).

 

As just an easy starting point though, I would just use his maps as is and not worry about his big kPa range. Your car should run pretty well regardless.

[motomanmike]

Billy are you using any noise filtering on your tach input signal? 

[OhBilly]

Billy are you using any noise filtering on your tach input signal? 

Nope.