MS1 - hard/NO starts

http://www.diyautotune.com
Spoolin6er
Dreams of owning a 240!
Posts: 95
Joined: Tue Jul 03, 2012 2:30 pm

Post by Spoolin6er » Fri Mar 22, 2013 3:19 pm

Definately looks like I have some issues with clean power here. Doesn't seen to care much about rpm, but build boost and it goes crazy?! Could my map sensor be doing something weird, or is this definitely a issue with the car itself? I'm using the stock power setup to the ecu with a relay to close the stock really. (see the diagram i posted a few posts back - stock the ECU relay is controlled by the ECU itself )

https://docs.google.com/file/d/0B1cQUbx ... sp=sharing[/img]

User avatar
Matt Cramer
KA-T.org Sponsor
KA-T.org Sponsor
Posts: 934
Joined: Mon Sep 15, 2008 5:34 am
Contact:

Post by Matt Cramer » Mon Mar 25, 2013 8:11 am

Are you using low impedance injectors by any chance?
Matt at DIYAutoTune.com - Megasquirt ECUs, fuel injectors, wideband O2 sensor systems, and more
Nissan plug and play engine management now in stock!

Spoolin6er
Dreams of owning a 240!
Posts: 95
Joined: Tue Jul 03, 2012 2:30 pm

Post by Spoolin6er » Mon Mar 25, 2013 9:03 am

Yes, the V3.0 board alows this with out an external flyback board right?

EVO 550's

Spoolin6er
Dreams of owning a 240!
Posts: 95
Joined: Tue Jul 03, 2012 2:30 pm

Post by Spoolin6er » Tue Mar 26, 2013 5:31 am

so i l did some searching on noise from low z injectors - fitted a 47uF cap between the center legs of the flyback TIP's and a 4.7uF cap in the bootloader jumper spot. I did them one at a time and saw noticeable improvements with both mods.

The voltage is pretty solid, it still varies a bit with rpm, but not in spikes as it was doing before. Miss fires have definately been reduced, but are still there.

I was rechecking that the sensor grounds are not grounded while the MS is unplugged and found that they are all fine except the CAS. The CAS ground pin i'm using seems to be also grounded internally? Is this right?? The nissan dizzy plug only has really 3 colors - black, white b/w. All the dizzy wires are white or black so idk for sure that ti'm on the right pin

User avatar
Matt Cramer
KA-T.org Sponsor
KA-T.org Sponsor
Posts: 934
Joined: Mon Sep 15, 2008 5:34 am
Contact:

Post by Matt Cramer » Tue Mar 26, 2013 6:17 am

Spoolin6er wrote:Yes, the V3.0 board alows this with out an external flyback board right?

EVO 550's
Yes, but they will inject a bit more noise into the system and be less forgiving if your grounding isn't perfect.

It may very well be that the Nissan distributor is grounded internally.
Matt at DIYAutoTune.com - Megasquirt ECUs, fuel injectors, wideband O2 sensor systems, and more
Nissan plug and play engine management now in stock!

Spoolin6er
Dreams of owning a 240!
Posts: 95
Joined: Tue Jul 03, 2012 2:30 pm

Post by Spoolin6er » Tue Mar 26, 2013 7:30 am

Soo, what's next? I don't see how the grounding could be any better, three wires strait from MS to one lug on the head.

Spoolin6er
Dreams of owning a 240!
Posts: 95
Joined: Tue Jul 03, 2012 2:30 pm

Post by Spoolin6er » Tue Mar 26, 2013 11:40 pm

confirmed that pin i'm using IS supposed to be grounded even with the distributor plug removed. Now, is it the right pin for the CAS signal ground, IDK, but I would imagine it wouldn't work at all with me using the wrong pin... I would think.

User avatar
Matt Cramer
KA-T.org Sponsor
KA-T.org Sponsor
Posts: 934
Joined: Mon Sep 15, 2008 5:34 am
Contact:

Post by Matt Cramer » Wed Mar 27, 2013 6:15 am

I would try an injector resistor pack; they can cut down on noise a bit over PWM current limiting.
Matt at DIYAutoTune.com - Megasquirt ECUs, fuel injectors, wideband O2 sensor systems, and more
Nissan plug and play engine management now in stock!

Spoolin6er
Dreams of owning a 240!
Posts: 95
Joined: Tue Jul 03, 2012 2:30 pm

Post by Spoolin6er » Thu May 30, 2013 10:32 am

So I finally fixed my miss fires while driving by using the VR conditioner circuit. This allowed me to use R52 to tune out some noise.

But i'm back to square one - having starting issues. Once it starts it runs great, but starting is very hit or miss. Once again, jumping the car off with another car helps tremendously. I also noticed that push starting it is very effective and starts it every time.

I have found that bumping the starter repeatedly will actually get it to start. Like, bumping it quickly enough to where it is engaging the flywheel while its still moving a bit. (I realize this is terrible for it, but it seems to work)

I also changed starters to an older oem starter, it got worse.

Battery voltage according to MT drops to about 10-10.5v while cranking, but i'm not getting any resets or anything.

Using the tooth logger in TS, I turned down R56 untill I stared to see noise. This should have reduced the trigger voltage to ensure that the trigger voltage can be reached even while the electrical system was under heavy load. This seemed to have no effect.

I put a scope on the input to the VR conditioner, and I am getting a clean square wave even while cranking. The output looked perfect as well. I am getting the correct 12-1 pattern. The scope is showing an amplitude of about 1.3v (1v while cranking) for the input, and 330mv ish for the output (for running and cranking) Is that about right?? I'm using the XG1 jumper as a ground reference.

I am lost as to how my vr out signal could look fine and yet I have a tooth log that is all messed up?? :?

User avatar
Matt Cramer
KA-T.org Sponsor
KA-T.org Sponsor
Posts: 934
Joined: Mon Sep 15, 2008 5:34 am
Contact:

Post by Matt Cramer » Fri May 31, 2013 6:58 am

Do you have a data log you can post of the hard starting?
Matt at DIYAutoTune.com - Megasquirt ECUs, fuel injectors, wideband O2 sensor systems, and more
Nissan plug and play engine management now in stock!

Spoolin6er
Dreams of owning a 240!
Posts: 95
Joined: Tue Jul 03, 2012 2:30 pm

Post by Spoolin6er » Thu Jun 06, 2013 9:42 am

sorry for the delay, was in the middle of moving

datalog:
https://docs.google.com/file/d/0B1cQUbx ... sp=sharing

User avatar
Matt Cramer
KA-T.org Sponsor
KA-T.org Sponsor
Posts: 934
Joined: Mon Sep 15, 2008 5:34 am
Contact:

Post by Matt Cramer » Mon Jun 10, 2013 9:28 am

The way the RPM input signal appears to be freezing implies it's not getting a good read at low RPM. Can you make a tooth log of what it's seeing at that point?
Matt at DIYAutoTune.com - Megasquirt ECUs, fuel injectors, wideband O2 sensor systems, and more
Nissan plug and play engine management now in stock!

Spoolin6er
Dreams of owning a 240!
Posts: 95
Joined: Tue Jul 03, 2012 2:30 pm

Post by Spoolin6er » Mon Jun 10, 2013 11:50 am

I'll try to get a new log up, but it shows very long tooth pulses with no pattern.

The pictures below are from when I was still using the opto conditioner, but the logger results are almost identical. Even when its running, and running well, the tooth log is showing the wrong pattern. Putting a scope on the VR conditioner output gives a clear 12-1 square wave though :?

Image

Image

User avatar
Matt Cramer
KA-T.org Sponsor
KA-T.org Sponsor
Posts: 934
Joined: Mon Sep 15, 2008 5:34 am
Contact:

Post by Matt Cramer » Tue Jun 11, 2013 7:41 am

Do you have a log of it running well on the wrong pattern?
Matt at DIYAutoTune.com - Megasquirt ECUs, fuel injectors, wideband O2 sensor systems, and more
Nissan plug and play engine management now in stock!

Spoolin6er
Dreams of owning a 240!
Posts: 95
Joined: Tue Jul 03, 2012 2:30 pm

Post by Spoolin6er » Tue Jun 11, 2013 12:00 pm

so i guess its not running as well I had thought/hoped.
It feels pretty smooth, the the log shows a tach drop out at around 163 sec. It pops at about this rpm pretty often, but i always assumed it was the rev limiter.

https://docs.google.com/file/d/0B1cQUbx ... sp=sharing

Below is a random tooth log screen cap during a pull (some where around 4000rpm)
Image

Spoolin6er
Dreams of owning a 240!
Posts: 95
Joined: Tue Jul 03, 2012 2:30 pm

Post by Spoolin6er » Mon Jun 17, 2013 6:00 pm

thoughts?

User avatar
Matt Cramer
KA-T.org Sponsor
KA-T.org Sponsor
Posts: 934
Joined: Mon Sep 15, 2008 5:34 am
Contact:

Post by Matt Cramer » Tue Jun 18, 2013 7:54 am

Other than the one RPM dropout, the one thing that jumped out at me is that the acceleration enrichment is triggering too easily. Try increasing the threshold value.
Matt at DIYAutoTune.com - Megasquirt ECUs, fuel injectors, wideband O2 sensor systems, and more
Nissan plug and play engine management now in stock!

Spoolin6er
Dreams of owning a 240!
Posts: 95
Joined: Tue Jul 03, 2012 2:30 pm

Post by Spoolin6er » Tue Jun 18, 2013 9:29 am

ok, yeah I see what you mean. I honestly haven't made it that far into the tune because first and foremost it hates starting.

I really need to get a handle on this, its killing my starter and battery to have so many starting attempts. I did find a new method of starting it though. I really dont understand why I'm having so many issues with the tach signal.

The scope shows a clear 12-1 signal but it is showing an amplitude of about 1.3v (1v while cranking) for the input, and 330mv ish for the output (for running and cranking) Is that about right, sounds very low?? I'm using the XG1 jumper as a ground reference.

neverlift
Knows Some Stuff About 240's!
Posts: 263
Joined: Mon Apr 03, 2006 11:35 am
Location: fla

Post by neverlift » Tue Jun 18, 2013 9:48 am

what are your cranking settings I found a lower ignition timing cranking made for smoother starts, also make sure the skip pulse is set to about 2(I cant get by with 1 but starts right up every time without kickback on 2)
Image

User avatar
Matt Cramer
KA-T.org Sponsor
KA-T.org Sponsor
Posts: 934
Joined: Mon Sep 15, 2008 5:34 am
Contact:

Post by Matt Cramer » Wed Jun 19, 2013 6:22 am

That is way too low. What voltages do you see if you test the board on a Stim?
Matt at DIYAutoTune.com - Megasquirt ECUs, fuel injectors, wideband O2 sensor systems, and more
Nissan plug and play engine management now in stock!

Spoolin6er
Dreams of owning a 240!
Posts: 95
Joined: Tue Jul 03, 2012 2:30 pm

Post by Spoolin6er » Wed Jun 19, 2013 2:24 pm

I'll have to get back to you on that, my car and stim are 600 mi apart...

Spoolin6er
Dreams of owning a 240!
Posts: 95
Joined: Tue Jul 03, 2012 2:30 pm

Re: MS1 - hard/NO starts

Post by Spoolin6er » Mon Nov 11, 2013 7:32 pm

So, its been quite a while, but I've finally been ab;e to start working on this project again.

I'd like to test it on the stim again, but my stim is only a v2 and doesn't simulate a toothed wheel like the Jim stim. (borrowed a JimStim previously, but moved since then) Can I mod my v2 stim to also simulate a 24-2 tooth optical wheel?

User avatar
Matt Cramer
KA-T.org Sponsor
KA-T.org Sponsor
Posts: 934
Joined: Mon Sep 15, 2008 5:34 am
Contact:

Re: MS1 - hard/NO starts

Post by Matt Cramer » Tue Nov 12, 2013 7:35 am

Spoolin6er wrote:So, its been quite a while, but I've finally been ab;e to start working on this project again.

I'd like to test it on the stim again, but my stim is only a v2 and doesn't simulate a toothed wheel like the Jim stim. (borrowed a JimStim previously, but moved since then) Can I mod my v2 stim to also simulate a 24-2 tooth optical wheel?
You could try this:

http://www.megamanual.com/router/crankwheelpulser.htm
Matt at DIYAutoTune.com - Megasquirt ECUs, fuel injectors, wideband O2 sensor systems, and more
Nissan plug and play engine management now in stock!

Spoolin6er
Dreams of owning a 240!
Posts: 95
Joined: Tue Jul 03, 2012 2:30 pm

Re: MS1 - hard/NO starts

Post by Spoolin6er » Tue Nov 12, 2013 8:04 am

Very cool - I never thought of that. I'll give it a shot this weekend and report back. I may even try to generate a square wave using matlab to better simulate an opto wheel...

Spoolin6er
Dreams of owning a 240!
Posts: 95
Joined: Tue Jul 03, 2012 2:30 pm

Re: MS1 - hard/NO starts

Post by Spoolin6er » Sat Mar 29, 2014 11:10 am

Ok, I'm back at it again. So I replaced the battery, and that has helped starting tremendously! I can't believe my setup is so sensitive to battery voltage?? Either way, doesn't matter to me right now as long as it works. I still get the occasional kick back, but it only does it when its really cold, so i think its again voltage related. Either way, I've still got bigger issues at the moment.

I'm still having tach noise issues in the higher rpm range. During a pull I can watch my in car tach studder as I'm accelerating. You can't feel them too much in the lower gears, but if I'm going a high speed pull under a lot of boost you can really feel/hear it. These slight shudders in the tach signal barely show up in the log - they look more like tiny flat spots in the other wise increasing rpm.

But then I get to 5500-6000 and all hell breaks loose. Tach needle starts bouncing and the engine starts backfiring. For the longest time I thought it was the rev limiter doing weird things, but its set to 6700 and after looking at the logs, the rpm signal is dropping to 2500ish for one time step and then coming back.

I don't have a Jim Stim so I pulled the MS and my distributor out and wired my optical CAS to my early gen stim. I hooked the distributor shaft up to a dremel and went to testing. RPW looked perfect every where except right around 6000, tach signal was intermediately dropping out to around 2000. Looking at the log showed that "spark angle was dropping sometimes when rpm wasn't? Maybe it has a slightly different sample time?

https://www.dropbox.com/s/sr8avgc5xqnf0 ... _bench.png

I hooked up the scope again and found that the lower voltages I reported earlier were due to me setting it up wrong. This time I got a perfect square wave at the "tachselect" pin with a magnitude around 10-11v (which is what my power supply was putting out). And another perfect square wave at the VROUT pin with a magnitude of about 1.3v (i'm using the VR conditioner). I couldn't see any change in the wave form at either location even when the tach was dropping out in MT.

I re-melted all the solder joints to the processor and all the tach input circuits and went back out to the car. I hooked up the scope in car, and saw a nice 14v square wave at tach select and I think 1.5v at VROUT. Reving through the rpms while stationary I din't see any issues with the square wave, although it doesn't seem to trigger the issue as easily free reving (possibly because it stays at an trouble rpms for less time? Could explain why its less noticeable in the lower gears as well?)

I drove it and took a regular and a tooth log and the tooth log seems fine except that there is a gap every so often that is probably my poor netbook trying to catch up. But it happens everywhere so i don't think it is related. Here is a screen shop of the in car log.

https://www.dropbox.com/s/oua5qebjc5nvt ... %20car.png

What is going on here? I have been fighting with this thing for WAY too long. Could it be the processor? Seems like the input to it is good.

I have also installed injector resistors and got rid of the PWM and a different distributor with the same 24-2 wheel.

User avatar
Matt Cramer
KA-T.org Sponsor
KA-T.org Sponsor
Posts: 934
Joined: Mon Sep 15, 2008 5:34 am
Contact:

Re: MS1 - hard/NO starts

Post by Matt Cramer » Tue Apr 01, 2014 6:38 am

Can you capture the high RPM break up in a tooth log and post it here? This will let me play back the signal you're getting.
Matt at DIYAutoTune.com - Megasquirt ECUs, fuel injectors, wideband O2 sensor systems, and more
Nissan plug and play engine management now in stock!

Spoolin6er
Dreams of owning a 240!
Posts: 95
Joined: Tue Jul 03, 2012 2:30 pm

Re: MS1 - hard/NO starts

Post by Spoolin6er » Wed Apr 02, 2014 6:22 am

The following google drive folder has 3 tooth logs and 3 normal datalogs: (all tooth logs at 750ms)
https://drive.google.com/folderview?id= ... sp=sharing

Interestingly, the two pulls I logged on 4/1 didn't capture the extreme tach spike that I posted a picture of previously. The engine DID break up at high rpm however, tach needle jumping around and all. I also kind of pushed through the break up and it started making power again but I think the run that really showed that didn't make it into a log. You can tell where its breaking up by looking at the AFR's, they go really lean where it starts misfiring. (I'm using an AEM all in gauge type wide band - pretty sure its linear off the top of my head)

I tried to keep those two logs short so they are easier to look at. The third log is much longer and the start of the data log is a good bit before the tooth log. The tooth log also ends well before the datalog. I believe the tooth log starts around 100 sec into the datalog. This log has the huge spike I posted the other day.

Also, not sure if its related, but my MAT signal is looking noisy. Its a fairly new GM open element sensor (2 wire grounded to MS).

User avatar
Matt Cramer
KA-T.org Sponsor
KA-T.org Sponsor
Posts: 934
Joined: Mon Sep 15, 2008 5:34 am
Contact:

Re: MS1 - hard/NO starts

Post by Matt Cramer » Thu Apr 03, 2014 6:59 am

You're seeing an extra pulse in the middle of the missing tooth. Do you have false trigger protection enabled? Might want to change that, and be sure the distributor is grounded properly.
Matt at DIYAutoTune.com - Megasquirt ECUs, fuel injectors, wideband O2 sensor systems, and more
Nissan plug and play engine management now in stock!

Spoolin6er
Dreams of owning a 240!
Posts: 95
Joined: Tue Jul 03, 2012 2:30 pm

Re: MS1 - hard/NO starts

Post by Spoolin6er » Thu Apr 03, 2014 7:54 am

I checked and false trigger protection is enabled.

The CAS ground goes directly to the MS. The MS is grounded to the head.

Can you explain these extra triggers you're seeing? I count 10 short tooth pulses followed by a long pulse (tooth 11 and 12). Is that incorrect? Maybe I'm just missing something. What do you use to play back the logs? I was plotting them in excel using a bar graph like the one seen in TS.

Thanks for the help.

User avatar
Matt Cramer
KA-T.org Sponsor
KA-T.org Sponsor
Posts: 934
Joined: Mon Sep 15, 2008 5:34 am
Contact:

Re: MS1 - hard/NO starts

Post by Matt Cramer » Mon Apr 07, 2014 8:11 am

I played it back in MegaLogViewer. Usually you see the pattern you describe - but I'm seeing a few points where the long bar is absent.

You might want to experiment with turning false trigger protection off.
Matt at DIYAutoTune.com - Megasquirt ECUs, fuel injectors, wideband O2 sensor systems, and more
Nissan plug and play engine management now in stock!

Post Reply