Page 4 of 6

Re: 2004 Mazda MX5 1.8 VVT #50

Posted: Thu May 02, 2019 10:27 pm
by kb1gtt
That's some good news. Glad to hear it's moving along.

Re: 2004 Mazda MX5 1.8 VVT #50

Posted: Sat May 04, 2019 10:47 am
by Crazy Striker
I started the car fine but it was idling horribly rich.
I may revert back to NB2 base settings since I no longer have issues running the engine.
I notice a lot of what can be kickbacks during cranking. The starter motor suddenly stops and the car resets itself.
I'm also chasing engine noises. Car started to develop some strange noises after all those failed cranking attempts. I don't know if the engine mounts may have suffered or something like that.
I'm also struggling with VVT assembly (possibly) related noise

Gwendal

Re: 2004 Mazda MX5 1.8 VVT #50

Posted: Thu May 09, 2019 3:17 am
by AndreyB
We are (hopefully) here if/once you need help :) No promises obviously.

Image

Re: 2004 Mazda MX5 1.8 VVT #50

Posted: Thu May 09, 2019 6:43 pm
by Crazy Striker
I was pretty busy these times so I had to leave the project for a little.
I worked on it this afternoon and found the source of the loss of synchro. When desoldering filter cap on OPAMP I burned a pad that connected 10k resistor to the OPAMP input.
Just soldered a solid jumper to it to solve this all and RusEFI catches the pattern way quicker than before.
I couldn't get the car to start but was pretty close to with the NB2 preset ( except for pattern inversion needed with OPAMP inputs).

I know you're here for us and I'm pretty thankful for that ;) That's one one the things that made me chose RusEFI over other alternatives

Gwendal

Re: 2004 Mazda MX5 1.8 VVT #50

Posted: Thu May 09, 2019 9:17 pm
by Crazy Striker
I'm also working on a more definitive approach for wire interfacing using a PCB to link the 64 pin space on Frankenso to the 72 Pins NB2 connector and my auxiliary harnesses to avoid the kind of issues i've ran into before (snapped wires and such).
I did the layout and now linking the two together.
Keep you updated

Gwendal

Re: 2004 Mazda MX5 1.8 VVT #50

Posted: Fri May 10, 2019 7:59 am
by Crazy Striker
I saw on another thread that the warm-up fuel multiplier switched from % to a multiplier value.
Should I account for it in my settings ?
Last time the car ran it ran very very rich (like 8-9 AFR)

Re: 2004 Mazda MX5 1.8 VVT #50

Posted: Sat May 11, 2019 12:58 am
by AndreyB
Crazy Striker wrote:
Fri May 10, 2019 7:59 am
I saw on another thread that the warm-up fuel multiplier switched from % to a multiplier value.
Should I account for it in my settings ?
Only the display label got switched, nothing has changes in the logic.

Re: 2004 Mazda MX5 1.8 VVT #50

Posted: Fri May 24, 2019 6:26 pm
by Crazy Striker
Hi everyone !
I'm back and still unable to start the car. Crank deciding is now perfect, I definitely have spark and fuel but I still have this strange behavior. Either I spit loud bangs from the exhaust either I have kind of kickback where the engine stops and the electricity goes down. I was able to get a few near catches but nothing wonderful.
I've played with IAC, timing and fuel but still no luck...
I'm not desperate as I know it works but I have to wait before trials because battery doesn't like that much being drawn like that for long...

Cheers

Gwendal

Re: 2004 Mazda MX5 1.8 VVT #50

Posted: Fri May 24, 2019 8:01 pm
by MgFoster
Gwendal,

Congrats on getting it running. I've gone through many of the same issues and it is so rewarding when it finally works! I wish I would have been more active over the last couple months and I probably would have been able to help you on a couple of your previous issues.

Heres my thread from where I had many of the same Crank and Cam signal issues. https://rusefi.com/forum/viewtopic.php?f=3&t=1462

So I have been driving my car for roughly 5 months now on rusefi and have been battling stubborn starting issues the whole time!!! Ive set off some car alarms in my last apartment complex from backfiring while cranking :lol:

I had a breakthrough 2 nights ago that is pushing me in the right direction.

The cranking IAC multiplier table. This is a MULTIPLIER and it was defaulted all to zero for me!!!! This meant that the IAC was closed no matter what I changed the setting to in the cranking settings window. Ive set it all to 100 for now so I can mess around with it.
Screen Shot 2019-05-24 at 3.53.09 PM.png
Screen Shot 2019-05-24 at 3.53.09 PM.png (202.17 KiB) Viewed 22490 times

Another note, kickback is usually the cause of ignition being too far advanced, igniting the charge way too soon. I realized the ignition advance table I was running had ignition timing as high as 16 deg in the highlighted cells here.
Screen Shot 2019-05-24 at 3.56.40 PM.png
Screen Shot 2019-05-24 at 3.56.40 PM.png (182.05 KiB) Viewed 22490 times
I suspect that when the engine crosses the cranking rpm threshold, 500 rpm in my case, it jumps to this 16 deg timing value and that is when the kickback happens. I am experimenting with lowering these values with good results.


Ive attached the Current Tune in case you would like to look at it. I suggest placing an IAC gauge on your gauges page and monitor it under cranking to see its actual position.

There are SO many things to get right, it really gives you an appreciation for how much effort mazda engine calibrators put into the OEM ecu!

Good Luck, I hope some of this is helpful.

Matthew

Re: 2004 Mazda MX5 1.8 VVT #50

Posted: Sat May 25, 2019 10:55 am
by Crazy Striker
Hi !
Thank you very much for you very complete response, I will try to look at it as soon as my battery get charged again.
All those attempts are polling hard on the alternator and I can hear it whining when I deplete the battery

Thank you for the tune also !

Gwendal

Re: 2004 Mazda MX5 1.8 VVT #50

Posted: Sat May 25, 2019 12:04 pm
by Crazy Striker
For the crank and cam inputs, are you using the opamps and not he MAX chip? I don't knoq if I have to invert the signal with this configuration. I removed the capacitors to avoid getting the signal filtered.

Thanks again

Gwendal

Re: 2004 Mazda MX5 1.8 VVT #50

Posted: Sat May 25, 2019 12:44 pm
by AndreyB
MgFoster wrote:
Fri May 24, 2019 8:01 pm
The cranking IAC multiplier table. This is a MULTIPLIER and it was defaulted all to zero for me!!!! This meant that the IAC was closed no matter what I changed the setting to in the cranking settings window. Ive set it all to 100 for now so I can mess around with it.
Thank you for the usability issue report - this is now https://github.com/rusefi/rusefi/issues/802
I hope to fix this today

Re: 2004 Mazda MX5 1.8 VVT #50

Posted: Sat May 25, 2019 4:02 pm
by MgFoster
As far as the crank and cam inputs, I dont remember exactly, and my ecu is tucked up under the dash again, but i did find this pic of how we wired in the crank signal. I dont remember what happened to the cam signal.
IMG_2755.JPG
IMG_2755.JPG (3.5 MiB) Viewed 22456 times


As far as getting your trigger right and what not, heres my settings:
Screen Shot 2019-05-25 at 11.48.38 AM.png
Screen Shot 2019-05-25 at 11.48.38 AM.png (92.81 KiB) Viewed 22456 times
Remember to set vvt offset properly :
russian wrote:
Sat Jan 26, 2019 2:11 am
vvtOffset is the angle between cam sensor and VVT zero position.

With VVT actuator disconnected and VVT in fully closed position, you adjust vvtOffset to see zero VVT position gauge.

It's a big mystery to me why is this angle so different between the few engines I've tries this on, I do not have an explanation for that.

Step 1: get a good vvtOffset to get a good zero with electrical plug disconnected.

Step 2: confirm you have a diode similar to IAC setup

Step 3: try aux pid0 p=i=d=0 and different offset values to confirm that you can move your VVT actuator at all.

Step 4: help us understand what exactly is going on, maybe shoot a video and upload to youtube?
The trigger angle offset needs to be set with a timing light. Set fixed timing at 10 deg and adjust it the offset until you read 10 deg on the harmonic balancer.


One more note for starting, use fixed dwell. This is what all the megasquirt guys use.
Screen Shot 2019-05-25 at 11.50.10 AM.png
Screen Shot 2019-05-25 at 11.50.10 AM.png (107.59 KiB) Viewed 22456 times

Re: 2004 Mazda MX5 1.8 VVT #50

Posted: Sun May 26, 2019 12:44 pm
by Crazy Striker
I stole your tune and set the car to 10° fixed timing and it runs ! Very rough and very very rich but it runs !
I also used a timing might to check the timing and the mark was spot on 10° 😎 or maybe it was the wrong mark but it wouldn't be too difficult to set. The engine was cold by the way.
Sorry if I haven't made a video 😅
Now that I know what was wrong I can go on and start tuning.
MgFoster, I suppose you have the blue RX8 injectors or some EV14 regarding your settings. I changed the size prior starting.
Thank you very much you made my day 😁😁

Gwendal

Re: 2004 Mazda MX5 1.8 VVT #50

Posted: Sun May 26, 2019 2:35 pm
by MgFoster
Gwendal, thats awesome!

Notes on checking timing:
- Make sure you're on the 10 deg BTDC mark and not 10 deg after.

-It shouldn't run THAT rough on 10* fixed timing. I suggest making sure there is no timing corrections on, ie any of these correction tables.
Screen Shot 2019-05-26 at 10.18.14 AM.png
Screen Shot 2019-05-26 at 10.18.14 AM.png (64.57 KiB) Viewed 22435 times

Make sure all those are zero in the idle cells when you're checking timing.

-Also, if you ever have the crank pulley off, or the timing covers, you need to verify that the timing marks on the crank pulley with the mark on the crank gear. My crank pulley had slipped by almost 10 deg when i first checked timing! so I had to make a new mark for tdc.

-I would bet that it is close or perfect since we have the same car :lol:

Notes on injectors:
- I have red top rx7 460cc injectors. They are about 520cc on NB fuel pressure. So the fuel map is definitely going to be off, but you can fix that with autotune when its up to temp.

-The tune you stole from me has injector dead times from the rx7 injectors and these are different from the stock injectors and will cause it to run poorly. I cant find the old dead times I was using, but I believe I used @ or @s settings for that.

Good Luck!

Re: 2004 Mazda MX5 1.8 VVT #50

Posted: Sun May 26, 2019 2:49 pm
by MgFoster
Question for you Gwendal, is your tachometer working?

Re: 2004 Mazda MX5 1.8 VVT #50

Posted: Sun May 26, 2019 3:12 pm
by Crazy Striker
Thanks for the infos, I can't find the correct dead times it's awful.
I checked the marks with the stock ECU at Idle and it was at 10 degrees fine.
I checked the timing when cranking with RusEFI (with fuel off) when it is fixed at 10 degrees.
For the roughness it might be from the IAC badly configured, I will try that too
I have no issue with the tacho, when the car started last month it was ok up to 7k RPM

Gwendal

Re: 2004 Mazda MX5 1.8 VVT #50

Posted: Sun May 26, 2019 4:06 pm
by MgFoster
Do you mind sharing how your board is configured for the tach? What channel you go in and what jumpers you have installed? I question if Im even using the right pin from the nb2 harness.

Thanks,

Re: 2004 Mazda MX5 1.8 VVT #50

Posted: Sun May 26, 2019 4:33 pm
by Crazy Striker
I just followed the wiring from the Miata 2003 alt. page and it worked on the first try. The only thing that do not work for now is the CEL...
I was wondering if for the cranking problem I couldn't use a row set at 450 RPM (with the cranking limit set at 400 RPM) set withe the cranking timing on the spark table, it would smooth out the transition between the states

Re: 2004 Mazda MX5 1.8 VVT #50

Posted: Sun May 26, 2019 4:41 pm
by MgFoster
Yes I thought about that too, definitely could work. Give it a shot!

Re: 2004 Mazda MX5 1.8 VVT #50

Posted: Sun May 26, 2019 6:32 pm
by kb1gtt
Very cool to see users helping other users. That's great to see tunes and such being shared.

Re: 2004 Mazda MX5 1.8 VVT #50

Posted: Sun May 26, 2019 8:10 pm
by Crazy Striker
MgFoster I may have found your tacho issue.
I do not use the tacho output but the dizzy.
It works fine for me, feel free to try it, I attached my working config for this.

Gwendal

Edit : lookng at my files I just found we don't have the same ignition outputs, it may explain why my engine was not running properly, as only 2 cylinders were ignited :lol: :lol:

2nd edit : What is even funnier is that all this time I was unable to start because my 2 coil packs were inverted in the configuration file :o :o :o I feel so stupid. In the second screenshot, PC9 should be first and PE14 second. That explains kickbacks and exhaust loud bangs. Let's just hope I didn't break anything (it seems that starter morot can severly suffer from kickback)

Re: 2004 Mazda MX5 1.8 VVT #50

Posted: Sun May 26, 2019 9:47 pm
by MgFoster
I have used multiple configs of tacho and dizzy out. Still can't seem to get it to work :(

Glad you found an issue! The engines are pretty durable, I doubt you broke anything, and my starter has taken LOADS of abuse with no signs of dying yet.

Does it run good now?

Re: 2004 Mazda MX5 1.8 VVT #50

Posted: Sun May 26, 2019 10:09 pm
by Crazy Striker
It's 1AM now in France so I can't test it but I will try tomorrow morning.
On stock ECU it still runs fine except for this issue :
I have a weird noise on cranking, after the key is released from starter, I have some heavy grinding during 2 or 3 seconds, and sometimes some clanking when idling which disappear when pressing on the clutch. It's not crank related (like a spun rod/crank bearing or anything) but comes from the gearbox or near it, so I assume I can have snapped something on the starter like the nose cone or teeth... I can buy a replacement starter assembly for 35€ in UK so if I find out something broke it's not so serious. I think it's easy to break the starter loose to inspect teeth aspect on it and on the flywheel so I may do that in a near future.
For the dizzy I can 100% confirm it will worked for me, using the pinout described on the 2003 alt page.

I keep you updated tomorrow on how it runs.

Again thank you for everything

Gwendal

Re: 2004 Mazda MX5 1.8 VVT #50

Posted: Mon May 27, 2019 8:56 am
by Crazy Striker
Very very very good news guys !

I can now crank the engine, cold or hot and on half E85 :D :D
I'm so thankful to you, I'm now closer than ever
I'm so happy but I still need to solve the fact that my engine runs pig rich, I may not have the right injectors dead times dialed in...

And because you seem to love it, here is a video of a hot cranking (I heated the engine running RusEFI).
https://youtu.be/2hUpqQAHfGs

Gwendal

Re: 2004 Mazda MX5 1.8 VVT #50

Posted: Mon May 27, 2019 9:10 am
by Crazy Striker
For the rich part I now know where the issue come from.
It's just that my MAP input seem to be improperly grounded which leads to a read value of 10 at all times. In fact my AFR were pretty correct and I need to solve that issue

Re: 2004 Mazda MX5 1.8 VVT #50

Posted: Mon May 27, 2019 9:47 am
by Crazy Striker
Hopefully the last mistake :D

Wrong input for AFR and wrong calibration. My engine was running slightly lean (16-17) and heated up quickly, now I've upped the entire VE table by 1.5 and it runs rich but on purpose, to be on the safe side.

I've done a short test run around my school and the throttle response is very poor, I have to tune the Accel enrichment

Re: 2004 Mazda MX5 1.8 VVT #50

Posted: Mon May 27, 2019 2:56 pm
by MgFoster
I have not touched accel enrichment and my throttle response is perfectly fine. Having a good VE table is critical.

Can you post screenshots of your cranking settings, cranking coolant temp multiplier, ignition advance and VE tables?
It should start just fine in 2-3 seconds..

-You want to get it up to operating temp,
-set the coolant temp multiplier to 1 at that temp,
-play with the base fuel width and IAC position until it starts reliably,
-and then only change the coolant temp fuel multiplier when you try to do a cold start or a start at different temperatures.

I take it you dont have the premium version of Tuner Studio yet? VE Analyze is a god send,

Re: 2004 Mazda MX5 1.8 VVT #50

Posted: Mon May 27, 2019 3:03 pm
by MgFoster
I believe these are the dead times you want to run with stock 265cc injectors.
Courtesy of stefanst's tune, which is what I based my initial tune from.
Screen Shot 2019-05-27 at 11.01.47 AM.png
Screen Shot 2019-05-27 at 11.01.47 AM.png (72.13 KiB) Viewed 22388 times
Attached is his tune if you want to try starting from there.

Re: 2004 Mazda MX5 1.8 VVT #50

Posted: Mon May 27, 2019 3:10 pm
by Crazy Striker
Hi !

I don't have any issue starting when hot, it catches after approx 3s of starter.
I disconected the VVT actuator first to get my VVT offset and it now works fine. for the trigger offset I use 0 with no issue at all. In fact with your offset of 12 degrees the car was unable to fire and made the loudest bang I've ever made with it.
For my settings I've attavhed the pictures and my tune to this post.
I have the registered version of TS with the Autotune so I will be able to fine tune it now that I have my AFR wired in properly.
I will change my injectors for yellow RX8 ones soon to switch to E85 with the stock ECU. I understand they should flow near 550cc/min with stock NB fuel pressure (4 bars). I have an AFPR and plan to convert to a return style circuit.


Please do not mind the loud bang maker row in the bottom of my ignition table
Gwendal