[help needed] UK Mazda MX5 miata NB 1.8'99 - Future Turbo project r0.5 #38

Your chance to introduce yourself and your vehicle
puff
contributor
contributor
Posts: 2961
Joined: Mon Nov 11, 2013 11:28 am
Location: Moskau

Re: UK Mazda MX5 miata NB 1.8'99 - Future Turbo project r0.5

Post by puff »

nice to have a scope at hand to troubleshoot that sort of stuff ;-)
User avatar
RadMx5
Posts: 77
Joined: Wed Sep 20, 2017 3:04 pm

Re: UK Mazda MX5 miata NB 1.8'99 - Future Turbo project r0.5

Post by RadMx5 »

nice to have a scope at hand to troubleshoot that sort of stuff ;-)
Yeah, I must say it was my best investment ;)

I've changed the crank pull-up resistor (R290) from 1k to 2.7k and now it revs higher :) I think a 4.7k should do for the final value. I also tried adding an extra ground wire directly from the battery, but it didn't really make any difference to the signal.

The frequency goes now above 400Hz.
[video]https://www.youtube.com/watch?v=EiZFR9T0t_Q[/video]
[video]https://www.youtube.com/watch?v=mkljPQDqG5Q[/video]

So I can say the problem is almost solved! :D :D

Now I need a wide band. I think there is no point doing anything more when I don't know whether it runs rich or lean. The stock narrow band doesn't seem to show real values in TS. The max value I have seen on the gauge was 12 and min was 10.

...and it would be nice to have a the MIL lamp working :) Does the bulb want GND or 12V control signal? I just want to test it without the ECU to make sure the bulb is actually in the dash and I'm looking at the right pin. And then which driver to use?
User avatar
andreika
donator
donator
Posts: 461
Joined: Mon Feb 13, 2017 2:35 pm
Location: Kiev

Re: UK Mazda MX5 miata NB 1.8'99 - Future Turbo project r0.5

Post by andreika »

RadMx5 wrote:The stock narrow band doesn't seem to show real values in TS.
You could try the latest rusEFI build with narrowband support using "Narrow to Wideband approximation".
User avatar
kb1gtt
contributor
contributor
Posts: 3758
Joined: Tue Sep 10, 2013 1:42 am
Location: ME of USA

Re: UK Mazda MX5 miata NB 1.8'99 - Future Turbo project r0.5

Post by kb1gtt »

In your separate ground , what gauge wire did you use, and how long is that wire?
Welcome to the friendlier side of internet crazy :)
User avatar
RadMx5
Posts: 77
Joined: Wed Sep 20, 2017 3:04 pm

Re: UK Mazda MX5 miata NB 1.8'99 - Future Turbo project r0.5

Post by RadMx5 »

Oh I didn't know about that andreika. Ill try it out :)


kb1gtt - from memory I think it was around 16gauge and about 2 meter. I have a spare reel of this wire so used it.
User avatar
andreika
donator
donator
Posts: 461
Joined: Mon Feb 13, 2017 2:35 pm
Location: Kiev

Re: UK Mazda MX5 miata NB 1.8'99 - Future Turbo project r0.5

Post by andreika »

RadMx5 wrote:Oh I didn't know about that andreika. Ill try it out :)
You might try something like this:
narrow-to-wide-example.png
narrow-to-wide-example.png (72.79 KiB) Viewed 19481 times
User avatar
kb1gtt
contributor
contributor
Posts: 3758
Joined: Tue Sep 10, 2013 1:42 am
Location: ME of USA

Re: UK Mazda MX5 miata NB 1.8'99 - Future Turbo project r0.5

Post by kb1gtt »

According to this on-line calculator, I would expect about a 0.2V drop from this 16AWG wire.
http://www.calculator.net/voltage-drop-calculator.html?material=copper&wiresize=13.17&voltage=12&phase=dc&noofconductor=1&distance=2&distanceunit=meters&amperes=4&x=79&y=10

Even 14 AWG is about 0.13V of drop. Don't forget connections and corrosion can also add voltage drops.
Welcome to the friendlier side of internet crazy :)
NBytevenom
Posts: 110
Joined: Mon Jul 24, 2017 8:23 pm

Re: UK Mazda MX5 miata NB 1.8'99 - Future Turbo project r0.5

Post by NBytevenom »

kb1gtt wrote:According to this on-line calculator, I would expect about a 0.2V drop from this 16AWG wire.
http://www.calculator.net/voltage-drop-calculator.html?material=copper&wiresize=13.17&voltage=12&phase=dc&noofconductor=1&distance=2&distanceunit=meters&amperes=4&x=79&y=10

Even 14 AWG is about 0.13V of drop. Don't forget connections and corrosion can also add voltage drops.
Connections as in adding more grounds around the car, or more things on the circuit? Just curious because I've added a couple ground wire from my engine to my chassis.
User avatar
kb1gtt
contributor
contributor
Posts: 3758
Joined: Tue Sep 10, 2013 1:42 am
Location: ME of USA

Re: UK Mazda MX5 miata NB 1.8'99 - Future Turbo project r0.5

Post by kb1gtt »

Connector like pin and socket, ring terminals or flat Spade. Those also as voltage drop.
Welcome to the friendlier side of internet crazy :)
User avatar
RadMx5
Posts: 77
Joined: Wed Sep 20, 2017 3:04 pm

Re: UK Mazda MX5 miata NB 1.8'99 - Future Turbo project r0.5

Post by RadMx5 »

andreika - I tried to use the narrow to wideband approximation but it didn't really make any difference and I was unable to type exactly the same values as you.
rusEFI narrow to Wideband.png
rusEFI narrow to Wideband.png (39.12 KiB) Viewed 19428 times
I have also tried replacing the R290 with a 5K resistor and it seems to be just a bit better, but not perfect. I didn't scope the signal with the 5K resistor yet. But it revs better. Occasionally I get a trigger error message in the Tuner Studio around 7K RPM.

About the ground volt drop - I'll try adding the thickest wire I can get :) direct from the battery to the rusEFI, and will show you the results if it makes any difference.
User avatar
andreika
donator
donator
Posts: 461
Joined: Mon Feb 13, 2017 2:35 pm
Location: Kiev

Re: UK Mazda MX5 miata NB 1.8'99 - Future Turbo project r0.5

Post by andreika »

RadMx5 wrote:andreika - I tried to use the narrow to wideband approximation but it didn't really make any difference
This ain't gonna work, you also need to set the EGO type:
ego_type.JPG
ego_type.JPG (17.93 KiB) Viewed 19427 times
User avatar
RadMx5
Posts: 77
Joined: Wed Sep 20, 2017 3:04 pm

Re: UK Mazda MX5 miata NB 1.8'99 - Future Turbo project r0.5

Post by RadMx5 »

Ok,thanks I'll try it out. I had to download the latest rusefi_bundle to see these new options.

It would be very useful if we could have a big download button somewhere on the rusefi.com main page with the latest rusefi_budle. So any person who visits the website easily sees a big DOWNLOAD button with the latest changes. What you think? would it be possible?
User avatar
kb1gtt
contributor
contributor
Posts: 3758
Joined: Tue Sep 10, 2013 1:42 am
Location: ME of USA

Re: UK Mazda MX5 miata NB 1.8'99 - Future Turbo project r0.5

Post by kb1gtt »

I think the theory was that some day there will be stable releases, which include tested features. Then the bundles would be for experimental changes.
Welcome to the friendlier side of internet crazy :)
User avatar
AndreyB
Site Admin
Posts: 14292
Joined: Wed Aug 28, 2013 1:28 am
Location: Jersey City
Github Username: rusefillc
Slack: Andrey B

Re: UK Mazda MX5 miata NB 1.8'99 - Future Turbo project r0.5

Post by AndreyB »

RadMx5 wrote:It would be very useful if we could have a big download button somewhere on the rusefi.com main page with the latest rusefi_budle. So any person who visits the website easily sees a big DOWNLOAD button with the latest changes. What you think? would it be possible?
How about now? https://rusefi.com
Very limited telepathic abilities - please post logs & tunes where appropriate - http://rusefi.com/s/questions

Always looking for C/C++/Java/PHP developers! Please help us see https://rusefi.com/s/howtocontribute
Abricos
contributor
contributor
Posts: 849
Joined: Mon Aug 18, 2014 12:32 am
Location: Carteret, NJ 07008

Re: UK Mazda MX5 miata NB 1.8'99 - Future Turbo project r0.5

Post by Abricos »

russian wrote:
RadMx5 wrote:It would be very useful if we could have a big download button somewhere on the rusefi.com main page with the latest rusefi_budle. So any person who visits the website easily sees a big DOWNLOAD button with the latest changes. What you think? would it be possible?
How about now? https://rusefi.com
Not sexy ...
Maybe put on side ...
Attachments
Screenshot_20171107-181000_1_1.jpg
Screenshot_20171107-181000_1_1.jpg (78.4 KiB) Viewed 19405 times
User avatar
RadMx5
Posts: 77
Joined: Wed Sep 20, 2017 3:04 pm

Re: UK Mazda MX5 miata NB 1.8'99 - Future Turbo project r0.5

Post by RadMx5 »

Yes, it's a lot better and easier now for me, but I think it could be more visible for new people. I have quickly drawn something :)

A date of publication would be useful, so people can check just by going on the rusefi.com whether they have the latest bundle or not.
And russian if you don't mind adding some information about what was changed/added?

Therefore everything would be in one place nice, clear and easy to access for people. Anyone would know what was changed, and could test the new features without searching the forum for new software changes.
What you think? Does it make sense?
Attachments
download latest bundle icon_.rar
(97.05 KiB) Downloaded 348 times
User avatar
AndreyB
Site Admin
Posts: 14292
Joined: Wed Aug 28, 2013 1:28 am
Location: Jersey City
Github Username: rusefillc
Slack: Andrey B

Re: UK Mazda MX5 miata NB 1.8'99 - Future Turbo project r0.5

Post by AndreyB »

RadMx5 wrote:A date of publication would be useful, so people can check just by going on the rusefi.com whether they have the latest bundle or not.
And russian if you don't mind adding some information about what was changed/added?
Let's continue this discussion at https://rusefi.com//forum/viewtopic.php?f=13&t=1293
Very limited telepathic abilities - please post logs & tunes where appropriate - http://rusefi.com/s/questions

Always looking for C/C++/Java/PHP developers! Please help us see https://rusefi.com/s/howtocontribute
User avatar
RadMx5
Posts: 77
Joined: Wed Sep 20, 2017 3:04 pm

Re: UK Mazda MX5 miata NB 1.8'99 - Future Turbo project r0.5

Post by RadMx5 »

So I had a little chance to play with the narrowband approximation and as you can probably tell it is not easy to tune with it... it seems to work. I think it goes from reach to lean as it should do, but it is not easy to read the stoichiometric mixture from it. Maybe I'll record a video showing how it behaves.

But to be honest I'm still waiting to get the wideband. I've got some questions:

1. VE table - when I increase the values does it practically mean I increase the amount of fuel injected into the engine?
2. The best way to find injector dead time?
3. My VE table is filled with a value of 80. When I decrease the value to for example 45 it is still idling, but less than 45 engine stops. Somewhere online I saw that usually, the VE tables start at 25 and then go up to around 120, it's a bit confusing...
4. I think I have purple injectors (hard to see without removing the intake manifold) regarding this: http://miataturbo.wikidot.com/fuel-injectors
they are 265cc, when I change the injector flow in tuner studio from original 200cc to 265cc then the VE value must be even bigger than 80 to keep the engine idling. Should it not be the opposite way, when injector flow is bigger the VE value should be smaller? I'm a bit confused with this as well...
5. Knock sensor circuit I'm not using it yet but is there actually any point of using it? is there a way of retarding the spark timing when knock detected or is it working at the moment just as a knock indicator?
6. What wideband sensor do you recommend for the 1.8 miata and rusEFI? Should I bother buying the controlling gauge as well?

I think this is all I want to ask for now :) Thanks!
puff
contributor
contributor
Posts: 2961
Joined: Mon Nov 11, 2013 11:28 am
Location: Moskau

Re: UK Mazda MX5 miata NB 1.8'99 - Future Turbo project r0.5

Post by puff »

I thought that VE is a function of your camshaft, your inlet and your exhaust system. but probably yes, the higher the VE value, the more air your engine sucks in and the more fuel you want there (imho)
User avatar
AndreyB
Site Admin
Posts: 14292
Joined: Wed Aug 28, 2013 1:28 am
Location: Jersey City
Github Username: rusefillc
Slack: Andrey B

Re: UK Mazda MX5 miata NB 1.8'99 - Future Turbo project r0.5

Post by AndreyB »

RadMx5 wrote:5. Knock sensor circuit I'm not using it yet but is there actually any point of using it? is there a way of retarding the spark timing when knock detected or is it working at the moment just as a knock indicator?
6. What wideband sensor do you recommend for the 1.8 miata and rusEFI? Should I bother buying the controlling gauge as well?
Knock sensing in rusEfi is not working and not finished.

I would recommend some innovate wideband with a gauge - the gauge and the controller are often sharing the same body these days.
Very limited telepathic abilities - please post logs & tunes where appropriate - http://rusefi.com/s/questions

Always looking for C/C++/Java/PHP developers! Please help us see https://rusefi.com/s/howtocontribute
mck1117
running engine in first post
running engine in first post
Posts: 1493
Joined: Mon Jan 30, 2017 2:05 am
Location: Seattle-ish

Re: UK Mazda MX5 miata NB 1.8'99 - Future Turbo project r0.5

Post by mck1117 »

RadMx5 wrote: 1. VE table - when I increase the values does it practically mean I increase the amount of fuel injected into the engine?
Yes. The way that speed-density fuel injection works is that the computer estimates the density of air in the intake, then the VE (volumetric efficiency) gives what percentage of that density makes it in to the cylinder through the open intake valve. An ideal engine would have 100% VE everywhere, but due to acoustics and the valve being open for a limited period of time, it varies with engine load and speed.
RadMx5 wrote: 2. The best way to find injector dead time?
Unless you have the data for your injectors, you have two options.
First, you can guess. This does work, but it isn't great. A guess is plenty to get the engine running, and is enough to have a perfectly drivable car.
The second option is to rig up a test bench to test amount of injected fuel vs. injection pulse width.
RadMx5 wrote: 3. My VE table is filled with a value of 80. When I decrease the value to for example 45 it is still idling, but less than 45 engine stops. Somewhere online I saw that usually, the VE tables start at 25 and then go up to around 120, it's a bit confusing...
4. I think I have purple injectors (hard to see without removing the intake manifold) regarding this: http://miataturbo.wikidot.com/fuel-injectors
they are 265cc, when I change the injector flow in tuner studio from original 200cc to 265cc then the VE value must be even bigger than 80 to keep the engine idling. Should it not be the opposite way, when injector flow is bigger the VE value should be smaller? I'm a bit confused with this as well...
Around 40-60% sounds about right for VE at idle. Here's what the VE table for my Volvo looks like. Keep in mind that it's a completely different engine (and has a turbo), but the general shape is common to most engines. As a rule of thumb the peak VE should be at or near peak torque on the engine, so for me, that's around 3500-4000 RPM.
Image

As for the injector flow doing something that doesn't make sense, that's strange. It sounds like something in the config isn't quite right. Are you willing to share your .msq tune file, and a datalog of the engine running?
User avatar
AndreyB
Site Admin
Posts: 14292
Joined: Wed Aug 28, 2013 1:28 am
Location: Jersey City
Github Username: rusefillc
Slack: Andrey B

Re: UK Mazda MX5 miata NB 1.8'99 - Future Turbo project r0.5

Post by AndreyB »

mck1117 wrote: Unless you have the data for your injectors, you have two options.
Would https://rusefi.com/forum/viewtopic.php?f=5&t=1085 be a 3rd option?
Very limited telepathic abilities - please post logs & tunes where appropriate - http://rusefi.com/s/questions

Always looking for C/C++/Java/PHP developers! Please help us see https://rusefi.com/s/howtocontribute
ZHoob2004
contributor
contributor
Posts: 153
Joined: Sun Apr 03, 2016 7:11 pm
Location: Tucson, AZ

Re: UK Mazda MX5 miata NB 1.8'99 - Future Turbo project r0.5

Post by ZHoob2004 »

RadMx5 wrote:6. What wideband sensor do you recommend for the 1.8 miata and rusEFI? Should I bother buying the controlling gauge as well?
I don't have any experience with them, but I've seen a number of megasquirt users recommending 14Point7 Wideband controllers and gauges

Their base offering is a controller and sensor with no display, perfect for integrating into an EFI system.

https://www.14point7.com/products/spartan-lambda-controller-2

http://forums.hybridz.org/topic/128106-14point7-wideband-o2-controllers-megaspartan-and-idash/

NO AFFILIATION - I CAME ACROSS THEM ON HYBRIDZ (link above) AND THOUGHT THEY LOOK NEAT
User avatar
andreika
donator
donator
Posts: 461
Joined: Mon Feb 13, 2017 2:35 pm
Location: Kiev

Re: UK Mazda MX5 miata NB 1.8'99 - Future Turbo project r0.5

Post by andreika »

RadMx5 wrote:andreika - I tried to use the narrow to wideband approximation but it didn't really make any difference and I was unable to type exactly the same values as you.
There's one more important point: voltages must be arranged ascending, otherwise it won't work. Please correct your table (turn it upside down).
User avatar
AndreyB
Site Admin
Posts: 14292
Joined: Wed Aug 28, 2013 1:28 am
Location: Jersey City
Github Username: rusefillc
Slack: Andrey B

Re: UK Mazda MX5 miata NB 1.8'99 - Future Turbo project r0.5

Post by AndreyB »

Reminds me of https://github.com/rusefi/rusefi/issues/440 - anyone willing to implement?
Very limited telepathic abilities - please post logs & tunes where appropriate - http://rusefi.com/s/questions

Always looking for C/C++/Java/PHP developers! Please help us see https://rusefi.com/s/howtocontribute
User avatar
RadMx5
Posts: 77
Joined: Wed Sep 20, 2017 3:04 pm

Re: UK Mazda MX5 miata NB 1.8'99 - Future Turbo project r0.5

Post by RadMx5 »

Thanks for all your replies!
I haven't done anything on this project since my last post. I was waiting for some bits to arrive :)
Finally I got the Innovate controller with a Bosch wideband and a boost gauge (getting ready for the BOOST). Just need to mount them somewhere in the car and start some experiments.
IMG_1700.JPG
IMG_1700.JPG (1.46 MiB) Viewed 19183 times
User avatar
RadMx5
Posts: 77
Joined: Wed Sep 20, 2017 3:04 pm

Re: UK Mazda MX5 miata NB 1.8'99 - Future Turbo project r0.5

Post by RadMx5 »

I have added a thick ground wire from the battery to the ECU but it doesn't improve the CRANK trigger signal... So I'm still getting the trigger problem above 6500 RPM. For now, I just limit the RPM in TS to 6500 :roll:

Small update - gauges fitted :)
image1.jpeg
image1.jpeg (1.62 MiB) Viewed 19097 times
User avatar
AndreyB
Site Admin
Posts: 14292
Joined: Wed Aug 28, 2013 1:28 am
Location: Jersey City
Github Username: rusefillc
Slack: Andrey B

Re: UK Mazda MX5 miata NB 1.8'99 - Future Turbo project r0.5

Post by AndreyB »

Beautiful gauges placement!

I hope that with our simple Miata trigger 6500 is not the firmware performance limit. The way I am testing performance is with two discovery boards - I am using one to generate signal on PD1 & PD2 pins with disabled ignition and injection, and the other one as signal consumer. Sometimes with a cheap logic analyzer in between.
Very limited telepathic abilities - please post logs & tunes where appropriate - http://rusefi.com/s/questions

Always looking for C/C++/Java/PHP developers! Please help us see https://rusefi.com/s/howtocontribute
User avatar
kb1gtt
contributor
contributor
Posts: 3758
Joined: Tue Sep 10, 2013 1:42 am
Location: ME of USA

Re: UK Mazda MX5 miata NB 1.8'99 - Future Turbo project r0.5

Post by kb1gtt »

Do you have a log of the 6500 RPM issue? I wonder if high RPM jitter could be causing problems. I expect that this jitter would not exist on @'s simulation.

I wonder if the simulation could add some jitter and then see if it generates the same issues with some controllable amount of jitter. Could the simulation STM be programmed such that once every crank pulse, it generates a random number, then multiples that number with a constant, then adds or removes that number from the time for the next cam / crank event? Then with TS or the console, you can increase or decrease the constant to increase or decrease the amount of jitter. Is that a possible method to check the effects of jitter on the crank decoder?
Welcome to the friendlier side of internet crazy :)
User avatar
AndreyB
Site Admin
Posts: 14292
Joined: Wed Aug 28, 2013 1:28 am
Location: Jersey City
Github Username: rusefillc
Slack: Andrey B

Re: UK Mazda MX5 miata NB 1.8'99 - Future Turbo project r0.5

Post by AndreyB »

I bet we were close to 7000rpm during the races pretty often - see https://svn.code.sf.net/p/rusefi/code/misc/logs/1991_miata_hunchback/

Do not remember if there were trigger errors there but these logs could be a reference point.
Very limited telepathic abilities - please post logs & tunes where appropriate - http://rusefi.com/s/questions

Always looking for C/C++/Java/PHP developers! Please help us see https://rusefi.com/s/howtocontribute
Post Reply