Miata Turbo Forum - Boost cars, acquire cats.

Miata Turbo Forum - Boost cars, acquire cats. (https://www.miataturbo.net/)
-   ECUs and Tuning (https://www.miataturbo.net/ecus-tuning-54/)
-   -   VVT cam trigger settings in TunerStudio? (https://www.miataturbo.net/ecus-tuning-54/vvt-cam-trigger-settings-tunerstudio-104300/)

HugoW 12-06-2020 02:39 PM

VVT cam trigger settings in TunerStudio?
 
Hi,

I have my NB2 1.8 VVT engine running, but the VVT does not work. The wiring and all is OK, I think my settings are, too, but the log shows no VVT duty during the drive. I suspect my cam settings may mess up. In TunerStudio (I have a Speeduino 0.4.4 ECU) I can only chose 'Single tooth cam' or '4-1 cam'. Looking at the VVT cam, it's not single tooth and also not 4-1:
https://cimg0.ibsrv.net/gimg/www.mia...67256fa877.jpg


I think it doesn't hurt anyway to have the correct cam settings, and I hope it helps to get the VVT going. All suggestions are welcome,
Cheers,

Hugo
https://cimg7.ibsrv.net/gimg/www.mia...e3c2599242.png
https://cimg8.ibsrv.net/gimg/www.mia...90552f862b.png
https://cimg9.ibsrv.net/gimg/www.mia...3c04f82374.png

HowPrayGame 12-06-2020 03:07 PM

I don't think speeduino has any way to control VVT unless you have customized the board for it

HugoW 12-06-2020 03:14 PM

https://wiki.speeduino.com/en/boards/V04

Pin 38.

Hugo

deezums 12-08-2020 01:53 AM

Pick 99-05 miata as the trigger, the cam/secondary trigger type will be greyed out as it's fixed to the trigger pattern. Set closed loop mode in VVT settings, program cam angle instead of duty cycle. Tunerstudio can't live rename tuning pages, those are all hardcoded in the .ini file. Speeduino devs will probably find a way around that soon enough.

Note it's as experimental as it says it is and there are plenty of issues currently reported on it's function. Seeing as how it looks like it's based on MS3 VVT control with even less control inputs, I doubt it could ever work worth a damn in it's current form. Although I do find it amusing someone managed to pull VVT off on a mega, while all of B&G's brains couldn't make it work on a MS2...

HugoW 12-08-2020 06:45 AM

Oh, bummer. I changed my crank wheel to the OEM Mazda 323 one, 36-1, for better resolution. I'l stick to open loop and maybe I should grind two teeth off of the cam, making it a 1 tooth cam. Or I should go back to the standard Miata crank trigger wheel. Should any firmware update contain proper closed loop, I'll get to that. And I need to figure out a way to measure angle vs PWM value on the current system.

Hugo

HugoW 12-08-2020 09:22 AM

If I put the original trigger wheel back in place (quickest action) what angle BTDC is the standard setting for that? I know I need to check and adjust, but to get it going again.

Cheers,

Hugo

deezums 12-08-2020 11:56 AM

You can't use a 36-1 for VVT, not even on megasquirt, not with the miata cam pattern. It specifically says VVT works for MIATA patterns only right now, that would be 99-05 miata.

You can't run open loop VVT on a miata. It does not work that way. It's a hold duty arrangement, and the duty to hold the valve neutral changes all the time. So, open loop you get all the advance or no advance and nothing repeatable in between.

You need to go read the manual. I don't even have a speeduino and never have and yet I know how it works, their documentation is not terrible, so what's your excuse.

If it is the correct trigger on the correct engine it is safe to assume you can run no trigger offset to start.

Ted75zcar 12-08-2020 12:02 PM

Fill the gap between the 2 teeth with conductive material on the cam wheel and you can pretty much run any trigger pattern you want (MS3).

Edit: you might have to cut the singlet off too now that I think of it.

HugoW 12-08-2020 04:40 PM

@deezums; thanks for all the input. I must admit I only read the maual before the purchase, some 3 years ago. In this manual I read there are only plug-and-play versions for the non-VVT MX5s, hence I treaded this motor as a 4 cylinder 1.8, not necessarely from Mazda. Could have been a Ford for all I care (oh, it actually is, largely). And as more other engines are getting VVT to work with a PWM controlled valve letting oil pressure through (see some clever Scandinavian dudes in BMWs) I thought this would be the route for me. I understand now the MX5 cam trigger pattern messes up things, but that can be changed. As far as I am aware the teeth on the end of the cam are on a pressed on ring, I can make any ring I want. I could make a 36-1 on the cam, altough those would be darn small teeth. But making it a single tooth is relatively simple, and single tooth cam in combination with 36-1 on the crank is a combination Speeduino can handle.

However, you have thrown another spanner in the works which I'd like to know more about. Namely that the cam angle is not a (or close to a) constant relation to the PWM percentage. Could you point me towards a more in-depth explanation as to why? I would greatly appreciate that. Ican imagine oil temperature being a factor, but that should stabilise when the engine is warm (abeit it will take longer then coolant warm-up). But reading between your lines, I have a feeling there is more to it. I gladly learn.

BTW, I have no doubt you are right and I am wrong, but still I like to learn why. I'm not trying to prove you wrong.

Cheers,

Hugo

deezums 12-08-2020 07:03 PM

There are different versions of VANOS, some are like vtec on a RPM on/off switch, and some are like the miata and constantly variable. Some go a step further and have two solenoids, one for advance and one for retard.

As for different cam patterns, I don't think speedunino makes use of anything more than pulsed teeth. A poll level cam trigger would be nice for faster sync, but only if it's treated as a poll level and not a long tooth. In that case lopping off all but one is going to be about the same in the end.

As for why you can't run open loop PWM, it's because the neutral hold duty, the current that holds the valve middle of it's travel, is constantly changing. On top of that, the amount (or lack of) current to the valve controls how fast it advances and retards, plus the supply/return from the actuator depends on viscosity and pressure. Opening that valve destroys the common fixed relationship between RPMs and oil pressure. What RPM you are currently sitting at will change that, as well as engine load to a much lesser extent. You could run it open loop, but I would only run it 0 or 100% valve duty otherwise you have no good idea what it's actually doing.

As for control inputs, I believe you need to pause and kill all VVT controls and outputs below a certain RPM. When or where depends on the particular car. I made a thread a while back, something like "anyone running VVT on haltech" where a lot of issues with how MS3 (and speeduino) attempt to control the valve are brought up. My car cannot run on one set of PID values across the entire RPM range, it will not hold a steady target across half the engine operating range one way or another. Besides that, it sucks off idle or makes a ton of mechanical noise. Output resolution sucks on multiple levels. Most people end up making lots of compromises to get "acceptable" vvt control out of MS3.

HugoW 12-09-2020 01:24 AM

Great, thanks, I learned a lot from that. One more question; what would happen if I turn off VVT all together? I believe the system helps economy and mid range torque. As I will fit a small GT2554r trubo after I get the normally asperated set-up down, economy will be out the window (I know my right foot) and I hope on early boost due to the small trubo. With a boost control valve, I think I can get plenty of torque. I'm 'only'aiming for 200 rwhp.

Cheers,

Hugo

der_vierte 12-09-2020 02:42 AM

Hugo,

with a turbo as small as a 2554, there is enough low end torque all the time. VVT is a "nice to have", but not a "must have" in that case. Plenty of people are very happy with their low end even on 1.6's with small turbos

deezums 12-09-2020 03:03 AM

Yes, that's pretty much how it works. It makes no more power than a 99-00, and it makes about the same to a little bit more torque through the middle while getting better emissions. Those numbers assume it's a US spec miata, which has a worse intake manifold.

If you don't mind loss of performance below ~5500RPM the actuator should lock out with no oil flow. Otherwise it's a loss of around 10% torque for a large chunk of normal driving range, which makes it feel about like a 1.6.

HugoW 12-09-2020 04:43 AM

OK. I'll try putting back the stock 4-tooth crank gear this weekend, choose the 99-05 trigger pattern in TS , set it to closed loop and see what I get. If I don't get proper performance, I'll simple set it to no VVT in TS and run without. I read somewhere I should keep the oil feed on, as it lubricates the first bearing. This is my concern on cancelling the VVT without replacing the head, will that bearing get enough oil.

Hugo

curly 12-09-2020 10:09 AM

By disconnecting the 2 pin plug on the solenoid, the cam gear will remain locked, and all oil paths will remain stock. This is the best way to "delete" VVT, and is easily reversible.

HugoW 12-10-2020 03:51 PM

1 Attachment(s)
Hmmmmm... Thanks, BTW, Curly, for the info.

Tonight (I froze my butt off in the workshop) I changed the 36-1 triggerwheel back to the stock MX5 one, with the four unevenly spaced teeth. I went into TS and:
Under 'Settings', "Trigger settings":
et trigger pattern to Miata 99-05
Set trigger angle to zero (as per the manual, I read it again!)
Trigger filter to weak
I left trigger edges at 'Falling' as I never changed that in the first place from the base tune, so why start changing now.
Under 'Startup / Idle', 'Crank settings':
Fix cranking timing with trigger is on
The manual tells me to set 'Use new ignition mode' on, but I don't have that feature there.
I also set the VVT settings on, closed loop, and left everything else stock per the base tune. The VVT table hasn't changed, it is still called VVT Duty cycle.

The car started with the 36-1 wheel fairly decently, and I drove the car from the nearby carpark to the workshop (half a mile). Now with the changes made, the car doesn't start. I think it fires to early, backfires. I had a look at the trigger sensor and teeth.
At 10 degrees BTDC there is a tooth very close to but just before the sensor:
https://lh3.googleusercontent.com/IW...-no?authuser=0

At TDC that tooth is certainly passed the sensor:
https://lh3.googleusercontent.com/cl...-no?authuser=0

I've attached my current tune. I think there is something wrong with the timing, but the sensor hasn't moved from when it ran on the stock ECU and with the 36-1 wheel it triggered at 80 BTDC, exactly per the wheel's spec. Any advice is again appreciated. It's a bit disgruntling driving in, fixing an issue and not being able to drive out.

Cheers,

Hugo

HugoW 12-11-2020 02:02 AM

2 Attachment(s)
As per a previous problem, I reflashed the ECU using SpeedyLoader and loaded the current tune. Now the engine does not give any sign of life, except for cranking. I made a composite log, which shows all pulses nicely (attached) and I noticed an error message at the bottom of the screen:
https://cimg0.ibsrv.net/gimg/www.mia...198a0fc746.png
"Unable to read Logger data. Timeout of 2000 ms. exeeded". Googling gives doesn't give me anything I understand / can use. except for the complete phrase.

I'm in need of a little success here, please chime in if you know what's going on.

Cheers,

Hugo

shuiend 12-11-2020 07:16 AM

Did you put the 4toothe wheel on the correct side/orientation? I am almost positive you can put in on backwards and it will cause the car not to start.

HugoW 12-11-2020 08:09 AM

I think I did. I put the 36-1 wheel on the way the 4 tooth wheel was originally, and now I put the 4 back the way the 36-1 is on. I googled a pic:
https://shop.mazmart.com/content/ima...96-97_550.jpeg
I put it together with the raised middle of the wheel like on the pic. Is that correct?

[EDIT] I realise putting on a 36-1 backwards doesn't matter. Putting on the 4 tooth one does. Worth a try. [/EDIT]

Hugo

curly 12-11-2020 10:15 AM

The raised middle part sits against the back of the harmonic damper


All times are GMT -4. The time now is 10:43 AM.


© 2024 MH Sub I, LLC dba Internet Brands