[rusEfi] frankenstein 0.11 on nissan sr20ve neo vvl #34

Your chance to introduce yourself and your vehicle
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: frankenstein 0.11 on nissan sr20ve neo vvl #34

Post by AndreyB »

It was rusefi bug https://github.com/rusefi/rusefi/issues/500

I've just fixed it. Have you migrated to latest version with new incompatible configutaions? You can migrate using currenttune TS files

As a work-around you can use TunerStudio but in TS the formula would be
rpm 4500 >
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
matt
donator
donator
Posts: 271
Joined: Fri Aug 26, 2016 11:32 am
Location: Malaysia

Re: frankenstein 0.11 on nissan sr20ve neo vvl #34

Post by matt »

so i need to figure it first in console with this command?

-set_fsio_output_frequency 3 0
-set_fsio_output_pin 3 PC14
-set_fsio_expression 3 "rpm > 4500"

and then in TS it should be:
TSrus.png
TSrus.png (257.08 KiB) Viewed 18461 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: frankenstein 0.11 on nissan sr20ve neo vvl #34

Post by AndreyB »

Are you using latest version? Because my answer would depend on that. Please help me help you.
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
matt
donator
donator
Posts: 271
Joined: Fri Aug 26, 2016 11:32 am
Location: Malaysia

Re: frankenstein 0.11 on nissan sr20ve neo vvl #34

Post by matt »

got flash stm32 with latest firmware. but in tuner studio state a wrong version. .ini file i use in the latest bundle.
Untitled.png
Untitled.png (178.87 KiB) Viewed 18455 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: frankenstein 0.11 on nissan sr20ve neo vvl #34

Post by AndreyB »

Just finished WRONG VERSION - give it time to build new bundle.

With current version your command should work, in your case TS is just to verify. Everything looks good to me.
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
matt
donator
donator
Posts: 271
Joined: Fri Aug 26, 2016 11:32 am
Location: Malaysia

Re: frankenstein 0.11 on nissan sr20ve neo vvl #34

Post by matt »

still no working. i dont know where i doing wrong.
consolefsio.png
consolefsio.png (116.07 KiB) Viewed 18443 times
MAIN_rfi_report_2017-11-27_18_44_16.csv
(8.68 MiB) Downloaded 719 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: frankenstein 0.11 on nissan sr20ve neo vvl #34

Post by AndreyB »

fsioinfo?
Why do you think that anything is wrong? The overflow message in the console does not mean anything bad.
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
matt
donator
donator
Posts: 271
Joined: Fri Aug 26, 2016 11:32 am
Location: Malaysia

Re: frankenstein 0.11 on nissan sr20ve neo vvl #34

Post by matt »

Ok, If massage flow going right I will check my mosfet. I don't get any switching above rpm 2000 at my test light.
matt
donator
donator
Posts: 271
Joined: Fri Aug 26, 2016 11:32 am
Location: Malaysia

Re: frankenstein 0.11 on nissan sr20ve neo vvl #34

Post by matt »

Do dyno tuning yesterday, not so happy with the result. Engine can't pull over 6200 rpm. Feel like ignition cut or something else, but i'am 100% sure it cause by trigger error.
24254989_10213488134069607_922797880118729301_o.jpg
24254989_10213488134069607_922797880118729301_o.jpg (166.19 KiB) Viewed 18406 times
trigger error.png
trigger error.png (97.83 KiB) Viewed 18406 times
2017-11-30_21.28.12.msl
(1.43 MiB) Downloaded 700 times
matt
donator
donator
Posts: 271
Joined: Fri Aug 26, 2016 11:32 am
Location: Malaysia

Re: frankenstein 0.11 on nissan sr20ve neo vvl #34

Post by matt »

here is another log.
2017-11-30_21.38.28.msl
(1.99 MiB) Downloaded 696 times
Abricos
contributor
contributor
Posts: 849
Joined: Mon Aug 18, 2014 12:32 am
Location: Carteret, NJ 07008

Re: frankenstein 0.11 on nissan sr20ve neo vvl #34

Post by Abricos »

How do you know your engine is mechanical good to spin more than 6000 rpm ???
matt
donator
donator
Posts: 271
Joined: Fri Aug 26, 2016 11:32 am
Location: Malaysia

Re: frankenstein 0.11 on nissan sr20ve neo vvl #34

Post by matt »

Factory claims this engine engage exhaust solenoid at 6500 rpm. So I assumed it save to rev up to 7000-7500 rpm.
User avatar
kb1gtt
contributor
contributor
Posts: 3758
Joined: Tue Sep 10, 2013 1:42 am
Location: ME of USA

Re: frankenstein 0.11 on nissan sr20ve neo vvl #34

Post by kb1gtt »

What technology is in side the cam / crank sensor? I suspect it's optical. Do you have any intuitions if this sync issue is hardware or software related? Do you have a scope? Can you check the electrical signal? Do you have a schematic of how your cam / crank signal wires are wired?
Welcome to the friendlier side of internet crazy :)
puff
contributor
contributor
Posts: 2961
Joined: Mon Nov 11, 2013 11:28 am
Location: Moskau

Re: frankenstein 0.11 on nissan sr20ve neo vvl #34

Post by puff »

In the Russian-speaking topic, there are reports on intermittent trigger sync errors, presumably, during cranking and shuting-down. But there were such errors in the error log from the console. Do you see such errors?

Even if i had a scope, I can't imagine a simple way to track down this issue (is it a hardware issue, or a software bug). Any ideas? I'd say using a trigger signal sniffer parallel to frankenso could help, but I am not aware of such a device.A logic analyzer + some parsing script?

On the other hand, this could be solved analytically. That would require some knowledge on trigger wheel configuration and the conditioning circuit...
matt
donator
donator
Posts: 271
Joined: Fri Aug 26, 2016 11:32 am
Location: Malaysia

Re: frankenstein 0.11 on nissan sr20ve neo vvl #34

Post by matt »

Sorry, I have no skills on software/hardware. Iam more to engine build, don't have any scope. I book dyno machine for 1 hour, all job done in rush so I really forgot to log it in rusefi console. Yes, it's optical sensor pull up 5v.
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: frankenstein 0.11 on nissan sr20ve neo vvl #34

Post by AndreyB »

Can you reproduce this issue outside of dyno? What is your trigger wheel on this vehicle?
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: frankenstein 0.11 on nissan sr20ve neo vvl #34

Post by kb1gtt »

What is your internal pull up resistor? It is possible that is to large or to small. If it's to large, the current is lower and could be more susceptible to electrical noise issues, as well if it's to small, then the current will go up which could over drive the sensor. Do you know what current it should be? Also at high RPM, it may need some coils or caps to change the impedance. It is really hard to diagnose such issues with out a scope. If you could get one and capture the trigger pulses at the ECU harness connector, that would tell me many things. The signal at that point won't be a square wave, it will have rounded corners. I can determine allot from how those corner are rounded. If I knew how they are rounded, I could tell you if this is the source of the problem, and if it is, then I could make some suggestions about adding coils or caps to correct the rounding issues. However without a scope trace to work from, it's very difficult to determine if this is software or hardware.

Can the self simulation generate pulses like this? Can that decode RPM's that are this high or higher?
Welcome to the friendlier side of internet crazy :)
matt
donator
donator
Posts: 271
Joined: Fri Aug 26, 2016 11:32 am
Location: Malaysia

Re: frankenstein 0.11 on nissan sr20ve neo vvl #34

Post by matt »

ok found why it was. my stm32dics faulty. just now i try to load with fresh firmware and it fail. this is 3rd stm32disc going dead.
puff
contributor
contributor
Posts: 2961
Joined: Mon Nov 11, 2013 11:28 am
Location: Moskau

Re: frankenstein 0.11 on nissan sr20ve neo vvl #34

Post by puff »

there should be some reason why they become faulty.
matt
donator
donator
Posts: 271
Joined: Fri Aug 26, 2016 11:32 am
Location: Malaysia

Re: frankenstein 0.11 on nissan sr20ve neo vvl #34

Post by matt »

2 of them can't program. In stm utilities I see some of hex/numbers (I don't know what it is) not fully erase. If load new firmware, this number still there. 1pcs got power supply issue. Plug in mini USB ok, but when plug in Frankenstein no power on. Use older stm firmware also same result.
User avatar
kb1gtt
contributor
contributor
Posts: 3758
Joined: Tue Sep 10, 2013 1:42 am
Location: ME of USA

Re: frankenstein 0.11 on nissan sr20ve neo vvl #34

Post by kb1gtt »

Have you updated the STM's firmware? Are you using OEM STM's or are you using china knock off STM's? See how to update firmware found here.
https://rusefi.com/wiki/index.php?title=Manual:Software:User/en#Load_firmware

We have not had problems with STM's going bad for a long time. It would seem that something is very wrong with this setup.
Welcome to the friendlier side of internet crazy :)
matt
donator
donator
Posts: 271
Joined: Fri Aug 26, 2016 11:32 am
Location: Malaysia

Re: frankenstein 0.11 on nissan sr20ve neo vvl #34

Post by matt »

1,2 from left not able to load firmware. This two got over current issue, then I use stm utilities v3.9.0exe issue seen away. Number 3 from left bought last 2 month, use v3.9.0exe firmware. It function if plug in mini USB, but not 5v ecu regulator.
IMG_20171206_191648.jpg
IMG_20171206_191648.jpg (1.08 MiB) Viewed 18293 times
User avatar
kb1gtt
contributor
contributor
Posts: 3758
Joined: Tue Sep 10, 2013 1:42 am
Location: ME of USA

Re: frankenstein 0.11 on nissan sr20ve neo vvl #34

Post by kb1gtt »

Some years ago, we drafted some notes about broken STM32 which can be found here. Perhaps that might help.
https://rusefi.com/wiki/index.php?title=Hardware:stm32f4discovery_repair
Welcome to the friendlier side of internet crazy :)
matt
donator
donator
Posts: 271
Joined: Fri Aug 26, 2016 11:32 am
Location: Malaysia

Re: frankenstein 0.11 on nissan sr20ve neo vvl #34

Post by matt »

good info. thanks.

anyway i got fix one of them yesterday(power supply issue). i change U1 that i belive is 5v to 3v regulator.
User avatar
kb1gtt
contributor
contributor
Posts: 3758
Joined: Tue Sep 10, 2013 1:42 am
Location: ME of USA

Re: frankenstein 0.11 on nissan sr20ve neo vvl #34

Post by kb1gtt »

U1 on the STM32 can only handle 6V max on the input. If U1 was broken, I would look closely at 5V power supply on the Frankenso. Did you purchase or build your Frankenso? We know a common issue is for U1001 on the Frankenso to have trouble soldering pin 3. This pin is on a large copper plane and you have to heat up that copper plane to get a good solder connection. If you do not, then you get 12V on the line which should be 5V. I know that @ does a several tests on the Frankenso boards he sells, which verifies this chip is making 5V. Once verified The W23 jumper is then installed to connect 5V to the rest of the 5V sensitive chips. If the 5V looks to be solid, then my next guess of what went wrong could be as simple as ESD on the 5V pins. Also the Frankenso 5V regulator is a push style regulator, and does not pull down to 5V. This means that if additional current gets into the 5V and the devices on 5V do not consume this current, then the 5V will raise above 5V. Potential sources of stray current getting into the 5V include the rail clamping diodes on the op-amps, or on the clamping diodes on the MAX9926. The op-amp has 10k current liming resistors like R242, which should limit this stray current, as even a 105V voltage on the input would only put 10mA of current into the 5V, and your STM32 likely consumes more than 10mA. So this is not likely, but it could be a possible path. The MAX9926 is a more likely possible source, but then again R102 and R107, etc make a 10k current liming resistor.

Hmmmm, your cam / crank appears to be an optical or a hall. Those are usually fairly safe as they typically have low voltage. However if you had some electrical noise, some how coupling into those wires, I could see how that could cause problems. That's only a 1k pull up resistor. If you had 15V on the cam / crank that would dump 10mA into the 5V. Do you know if you returning signal is 12V or 5V? This circuit is typically a pull down device, so it should only see 5V. However I don't really know what your dealing with.

As well keep a close eye on the 5V that goes through W49. That is another potential place where external energy can pull up the 5V. Try to keep that away from sources of high frequency or high voltage.
Welcome to the friendlier side of internet crazy :)
matt
donator
donator
Posts: 271
Joined: Fri Aug 26, 2016 11:32 am
Location: Malaysia

Re: frankenstein 0.11 on nissan sr20ve neo vvl #34

Post by matt »

Good explanations. Thanks!!

Anyway, i'am using Frankenstein at this engine. 5v regulator I copy same as speeduino circuit. Reason I do this because I can easily sources those components locally.
matt
donator
donator
Posts: 271
Joined: Fri Aug 26, 2016 11:32 am
Location: Malaysia

Re: frankenstein 0.11 on nissan sr20ve neo vvl #34

Post by matt »

After 2 weeks frustration with Speeduino, i think this is right time to switch back to Rusefi. Ok, I'am planning to swap Nissan trigger disc to Mitsubishi 4/2 to get more resolution. In TS i set to single coil, but in console I see it only give 3 spark @1000rpm. One of the sprak give 33.++ dwell which is too high. Can i set to wasted spark, tie together spark A and B and feed it to single coil distributor?

Other thing is I still can't setup my shift ligh.(this reason why I planning to switch to Speeduino :lol: )

all this massage was write at console:-
set_fsio_output_frequency 3 0
set_fsio_output_pin 3 PE3
set_fsio_expression 3 "rpm > 5500"

and this:-
set_fsio_expression 3 "rpm > fsio_setting(3)"
set_fsio_setting 3 5500

but nothing effect at my PE3.

Log attched and trigger signal produce from Ardustim.
MAIN_rfi_report_2017-12-13_21_15_44.csv
(7.58 MiB) Downloaded 581 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: frankenstein 0.11 on nissan sr20ve neo vvl #34

Post by AndreyB »

Are you doing writeconfig to save the configuration?

Can you please post tunerstudio tune file one you've applied all changes?
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
matt
donator
donator
Posts: 271
Joined: Fri Aug 26, 2016 11:32 am
Location: Malaysia

Re: frankenstein 0.11 on nissan sr20ve neo vvl #34

Post by matt »

Yes. I do writeconfic. Confused here, I need write confic one times for all or three times?
Yesterday I log with console and seems rev got to 7k rpm. I can't upload log file, maximum allowed in forum is 1mb. I do minor rewiring, my cas ground seems not wired properly to board.
matt
donator
donator
Posts: 271
Joined: Fri Aug 26, 2016 11:32 am
Location: Malaysia

Re: frankenstein 0.11 on nissan sr20ve neo vvl #34

Post by matt »

I mean:
set_fsio_output_frequency 3 0--- writeconfig
set_fsio_output_pin 3 PE3----writwconfig
set_fsio_expression 3 "rpm > 5500"-----wirteconfig

Or, writeconfic once after all massages?
Post Reply