Jump to content
HybridZ

Rpm signal loss


seattlejester

Recommended Posts

  • Replies 138
  • Created
  • Last Reply

Top Posters In This Topic

Thanks for taking a look.

 

I have a 7mgte with late model COP's. 

 

I've tried to piece together the information from a few sources corolla, mr2, supra. There aren't too many people I've found with my setup. 

 

TPS works, and if it doesn't turn over after a few seconds I can flood clear for a few. I will pull out the plugs today after cranking and see if they are wet.

 

As long as it isn't something glaring, that gives me hope.

Link to comment
Share on other sites

Currently at my car.

 

So I pulled the plugs and they don't appear to be wet.

 

I did inject some fuel using the test mode and using the spark test mode I heard a "woosh" so that makes me think the spark and fuel are working. I must have some setting off by a bit I really can't figure out what it would be though.

Link to comment
Share on other sites

So good news is that we got my car to run!

 

Bad news is only for a second.

 

Either the timing or fuel pulse width or something is off.

 

Friend came over to crank while I tried to time. For some reason the gun is not firing when wrapped around the trigger wire. Maybe because it is a cheap one. Just found out that you put an old spark plug wire and use that in-between the coil on plug and the spark plug to pick up a signal o.O, will have to go back tomorrow and give that a try.

 

We were getting some popping from the intake so we shortened the crank advance down until it started catching more readily.

 

The engine now stumbles with regularity, sometime stalls and with a bit of throttle even tries to run a bit.

 

The two problems now other than going off of mechanical timing are that the coil on plug fuse seems to burn out pretty quickly. It's a 20 amp fuse and I am afraid that if I step it up it may harm something. Right now, the fuse is blowing whenever the car tries to start. 

 

Second, it seems like the plugs are getting fouled. When pulled they come out pretty wet. 

 

I'm pretty sure all of this is connected. Perhaps the plugs getting fouled causes the coils to work harder popping the fuse. 

 

So either less fuel (lower pulsewdith) or more amps (bigger fuse) iI' nto the coils to burn through the carbon. I lowered the pulse width which is why it looks a little funny on the cranking setting. With the TPS at 90% (flood clear turned on) the engine would actually stumble for a bit which indicates that maybe too much fuel is making it in.

 

MSQ below

https://drive.google.com/file/d/0B21hrliT9P7pdlhOUE1PdEh2LVU/edit?usp=sharing

 

I'm kind of going off of a few sources for the settings, so if anyone sees anything wrong please give point it out.

 

Maybe screen shots would be easier. Screen shots of main settings below

ScreenShot2014-03-22at95014PM_zps27bf5c1

ScreenShot2014-03-22at95100PM_zps511f7c7

ScreenShot2014-03-22at95027PM_zps7b4cc2f

ScreenShot2014-03-22at95340PM_zps539a2c3

ScreenShot2014-03-22at95350PM_zpsf344b14

ScreenShot2014-03-22at95045PM_zps95905ed

ScreenShot2014-03-22at95034PM_zps30f8685

Link to comment
Share on other sites

Like AkRev said before, your priming pulse is like off the charts insanely high. I think I might be using the default settings for a SBC and it starts 2ms at 180° and goes to 6ms at 10°F. Change your plugs or do what you can to clean them up (propane torch, etc).

 

Secondly, have you verified your timing at any point along this venture? Like unplug your injectors and crank the engine while MS is set to fixed advance. Your trigger wheel will still give an rpm signal with incorrect wheel settings, while the spark events will be way off. It really does sounds like your timing is wrong.

 

Also, aren't toyota 440cc injectors low impedance? You have your injectors configured with PWM on and the duty cycle set to 100%, which means it's sending full current like you would for high impedance injectors.

 

Once you've checked those things, a data log of it sputtering would be helpful.

Link to comment
Share on other sites

Thanks for the response metro.

 

I'll give the priming pulse an edit and see how it likes it. I had a conservative one earlier, but after reading on a corolla forum, it was suggested to run those high values to help with startup. They went all the way to 40 pulsewidth near freezing I think, and I thought that was way too much so I moved the curve down a bit, but given how rich it seems right now I believe I can try the lower settings as suggested. 2ms at running, 6ms at 10*, filling in the rest with linear steps?

 

Mechanically the crank and head is set at correct timing, and I checked the wheel mark to 8 teeth after the missing tooth at the sensor at TDC. I did try with my timing light to read off of the trigger signal for the first coil, but it wouldn't trigger the light. I just found an article this evening suggesting that you run a regular spark wire from the coil pack to the spark plug and read off of that wire so I will give that a shot tomorrow.

 

Propane torch is indeed what we were using to clean the spark plugs. Currently they have been cleaned so I will adjust the settings before checking cranking timing.

 

Yes, stock they are low impedance injectors. I was misinformed and told a resistor box would be required for my install so I went ahead and installed the factory one (did not realize that megasquirt can run low impedance) so I am running high impedance settings with the resistor box installed. Once things get running I may play around with low impedance settings.

 

So current plan of action:

Rig up spark plug wire to coil on plug to get timing light

Adjust cranking pulse width quite a ways downward

Crank

Confirm timing

Adjust timing offset

Start datalog

Crank

Edited by seattlejester
Link to comment
Share on other sites

 

The two problems now other than going off of mechanical timing are that the coil on plug fuse seems to burn out pretty quickly. It's a 20 amp fuse and I am afraid that if I step it up it may harm something. Right now, the fuse is blowing whenever the car tries to start. 

 

 

I'm pretty sure all of this is connected. Perhaps the plugs getting fouled causes the coils to work harder popping the fuse. 

 

You should be able to create sparks all day long without harming or overheating anything.  That's what happens when the engine is running.  And it's actually "easier" for the coil to discharge across a fouled plug than a clean gap.   Blowing a 20 amp fuse consistently is not good.  There's something not right.  Might be that your "dwell" or coil primary "on" time is too long.  The coil primary circuits aren't getting broken to discharge the coils, and the long dwell is causing your fuse to blow.  Or you have a short on the coil ground side that isn't allowing a complete circuit break.  I would double-check the coil circuits and settings.

 

Just a some impartial observations.  I don't have any idea what specific MS circuits or settings would be involved.

Link to comment
Share on other sites

Actually it was suggested to run it up to 10ms on cranking, but I can turn that value down a bit, I don't personally think the longer duration helps any after a certain point.

 

The fuse is blowing after the engine leaves cranking mode (when the engine catches) which should be down to 3ms dwell with 1ms spark duration, turning that value down more does make me worry about idle quality. I've checked continuity with the ground wire and the coil plugs and didn't find any shorts. If there were shorts I would expect it to short out when I start cranking not when the engine tries to catch.

 

Should have some news tomorrow as long as work doesn't run late. 

 

Thanks for taking a look guys.

Link to comment
Share on other sites

Good news!

 

th_F16E41A8-A1FD-4A69-B898-936B4617669B-

 

Turning down the pulse width definitely helped, the engine would barely stumble with clear mode on meaning there was minimal excess fuel.

 

I uprated the fuse to a 30 amp for the coils, added two quarts of oil, turned down cranking dwell and pulse width.

 

I tried timing the motor by making a coil from the tach output on coil one and I was seeing something I don't quite understand. It was actuating the timing light at ~5 btdc (it was set at 6, after confirming I moved it back to 12 btdc) and maybe about 30 or 40 after tdc. It's a wasted spark setup so I would imagine it should be 180 out from the btdc reading, but not that close. I ran a composite log I will upload later in the evening. I have a feeling that the AFR was not reading anything, will have to take a look at my LC-1.

 

My aftermarket oil pressure gauge was showing 0, so I quickly shut the motor off after it got going. I suspect the sensor may be having a hard time grounding to the block with the sealer around the threads. How do you seal something without using thread sealant? Not much space or a surface for an o-ring where the sender sits.

 

Thanks for the help guys.

Link to comment
Share on other sites

If I understand you right, you see spark at 6 deg. BTDC and 40 deg. ATDC.

If you are seeing this, you need to adjust your screws on your board inputs again. http://msextra.com/doc/ms3/ignition.html#vrv357pull

Others can correct me if I am wrong, but if I remember right, megasquirt will read "extra" teeth, and spark early based on the number of teeth it reads. If the  it would count the 35 teeth, it would then wait for the missing tooth on the trigger wheel and restart over again. I had a similar situation when I ran a single coil off of the DIY wheel and had the inputs at full counter clockwise. 

 

As for the oil pressure sensor, I put the sealant at the base of the threads. That way when the thread are screwed in the tip can still make a good electrical connection. You can check resistance with a multi-meter and see if it is a grounding problem or just something is else is wrong with the sensor/wiring.

 

Glad to see you get some progress on it. Good luck!

 

Edit: Just noticed it was a video that you posted. Its hard to tell if it is misfiring, but it kind of sounds like it is. If you can get your timing light under the car and confirm that it is only firing twice (on the 6 btdc and 40 atdc) that might confirm what I was saying.  

Nice set up by the way. Excited to see you post a video of a drive sometime soon. 

Edited by AkRev
Link to comment
Share on other sites

Yes, that is correct. I see at approximately 6 btdc when it was set to 6 btdc, and then a spark further along past the mark intermittently.

 

That makes sense that the extra light is from detection of "extra teeth." So I would adjust the noise threshold (center pot) until it stopped reading the noise at the ATDC point.

 

The log should clear things up, I want to say it looked pretty noisy till it was running for a bit. I wish I uploaded it last night. Will have to wait till I get home.

 

Just a bit at the tip makes sense for sealing the sending unit. I was thinking maybe the rubber compound that goes between the threads may be better, but as long as it is grounding and not leaking it shouldn't be a problem. I'll have to test continuity and then maybe pull the sending unit to reseal it.

 

Thanks, almost there I hope.

 

Edit: Yes sorry should have mentioned it is a video. It definitely doesn't sound clean, I need to wait till I can get a helper to give me a hand while I try to time the car, if it is picking up extra teeth and injecting fuel or running spark inconsistently the misfires wouldn't be surprising at all. The wires still really need some addressing and maybe some discreet split tube covering, it's a mess right now :(.

Edited by seattlejester
Link to comment
Share on other sites

msq:

https://drive.google.com/file/d/0B21hrliT9P7pTWp3ZTFueWJ1VEE/edit?usp=sharing

 

datalogs:

stumbling

https://drive.google.com/file/d/0B21hrliT9P7pT2g3cmtsRU9hYjQ/edit?usp=sharing

running for a few seconds

https://drive.google.com/file/d/0B21hrliT9P7pQXhUd0xhR0RGNEU/edit?usp=sharing

 

Looks to me that it isn't counting the teeth correctly till the end, so definitely need some fine tuning there.

Edited by seattlejester
Link to comment
Share on other sites

If you Zip your msq and catalogs, they can be uploaded directly.

 

Reading all these trigger setups from MS2 makes me glad I went with MS3. My setup was simple once the V3 board was setup for the optical sensors. Of course the headache was back when I was helping James setup the MS3 for the RB CAS. Once I got that setup done and stable, the L28 was easy. I still am going to move over to the crank wheel soon with almost three times the resolution and a stable crank signal. I think there is too much slop with cam based wheels due to gear slop or chain/belt play.

 

I am glad to see that this tune is starting to come together. Keep it up and don't get discouraged.

Link to comment
Share on other sites

Thanks, I am really hoping no more leaks spring up or other gremlins pop out.

 

That makes sense, I was wondering why the forum wouldn't let me upload the raw files.

 

I am glad to move away from the cam sensor as well. I was reading up to 10* of difference reported at higher rpm on a miata.

Link to comment
Share on other sites

Looks like there are a lot of sync loss at low rpm. It looks like you have the sync loss gauge on your dashboard. Keep adjusting your inputs and gap until you get that worked out.

Did you adjust the inputs after the video but before the datalog on post #34?

I ask because it seems like it would be reading the teeth well enough to not be firing at the 40 atdc.

Link to comment
Share on other sites

Yes, actually quite a bit down there it looks like. Either I need to remount the sensor so that it is even closer, or adjust the pots to increase the voltages threshold and lower the noise threshold. This may seem really simple and I apologize, but to adjust the pots, do you put them back to 0 each time? As in go counter clockwise 5 turns to 0 it out before making an adjustment? Data log is from the same day, I just didn't get a chance to upload it till yesterday.

 

From what I can see, it is not picking up a good teeth read till the very end of the running data log, which makes me believe that you are on the money regarding the reading of the extra teeth. Also to note (if I am reading it correctly) is that the AFR stays at a perfectly consistent 14.7ish and does not seem to fluctuate at all, so that may need some revisiting.

Link to comment
Share on other sites

What is your current gap between your wheel and hall sensor?

They recommend a 1.5mm as the max gap. With my VR sensor I ended up mounting it with a .020 inch gap, or about half a millimeter. (just a note, make sure your wheel is truly centered before you set your hall sensor to it...... ask me how I know this....)

 

As for the input pots, you can handle that however you want. I would offer that you go full counterclockwise and then set it to your base. After you get a log from base then make your adjustments to get to where you need it without returning to 0 every time. From what I am researching, most with a hall sensor have just left it at the base, and adjusted gap to make it work.  

 

Have you had a chance to chase down the lack of oil pressure reading yet?

 

Good Luck!

Link to comment
Share on other sites

Oh I cringe to ask lol.

 

Wheel really is centered pretty well. I left a bit of a gap just in case the mount for the hall sensor flexes a bit. I will measure the gap but I want to say maybe 1mm between the tooth edge and the sensor.

 

I haven't gone back since Monday, hopefully some time will be put in this weekend. Definitely need to rig up an exhaust of some sort, cabin gets awfully fumey with the exhaust dumping next to the transmission at the moment.

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.


×
×
  • Create New...