Hello From iran

[hossein]

Hello guys, I am hossein from iran, i am a sysadmin and have some experience with programming and a bit with micros .
I want to run my father's old bike cbr250 hurricane with your EFI magical, so i bought a frankenstein board and assembled it , it works.


My challenge is :
The crankshaft tooth number is 7 and has one gap so i have 8 tooth(s).
So i have nothing pattern like this in my mind now, but i guess some ways - not sure witch one is useful -

What I did :
I wrote some c code to find how the trigger is working....
And trying to find how can i use this tool for 18200 RPM?

What i have :
just one coil is triggered
4 injector running

My questions :
why one coil ?
Is some way to deal with 18200 rpm red-line ?
what is the
operation mode

And at least :
Thanks for making open-source tools and programs & sorry for poor English .

[AndreyB]

Hello!

Pictures?

Have you found "enable self_stimulation" in console? Can you post consile Engine Sniffer picture showing a problem with one coil? What is your setting about coils? Can you post your CurrentTune.msq file with your whole tune?

There is a red line setting somewhere . With this relatively easy trigger shape I hope 18K would work but I am really looking trying you testing it!

[hossein]

Thanks for reply

Have you found "enable self_stimulation" in console?

1 ) No i have not found "enable self_stimulation" in console"

Can you post consile Engine Sniffer picture showing a problem with one coil? What is your setting about coils?

2 ) 2017-04-10_20_49_33rpm_0_maf_NaN.png ( attached in efi-img.tar.gz)

Can you post your CurrentTune.msq file with your whole tune?

3 ) snapshot472.png
I searched for CurrentTune.msq and nothing was found in "snapshot_20170402_2324_rusefi" but :

Code: Select all

$ cp  ../../latest/code/tags/tag20160101_9139/misc/logs/1993_escort_gt/2015.12.20/CurrentTune.msq . 

Code: Select all

$ grep -i hardlimit CurrentTune.msq 
<constant digits="2" name="rpmHardLimit" units="rpm">18500.0</constant>
<constant name="cutFuelOnHardLimit">"true"</constant>
<constant name="cutSparkOnHardLimit">"true"</constant>
<constant digits="0" name="noAccelAfterHardLimitPeriodSecs" units="sec">3.0</constant>

but nothing is changed in console yet.
snapshot473.png

[AndreyB]

Trigger signal is the top line, that does not look like 8/1 skipped, where is the skip?

'enable self_stimulation' is something you can type in the command or you can see it in 'messages' on the right side.

You need to install http://www.tunerstudio.com/index.php/downloads it's easier to configure there.

[hossein]

Thanks for reply.
That is a simple trigger, made by a simple code on rpi2 - attached.
I will try.

[AndreyB]

I do not follow. If you are trying to simulate a 8 tooth with one tooth missing you do not need anything external, rusEfi can do it itself.

What kind of trigger does not vehicle have? What exactly are you trying to do now?

I believe a bunch of pictures and more details on what EXACTLY you are trying to archive RIGHT NOW would help me help you.

[kb1gtt]

Hello and welcome along.

Please provide information about your engines crank wheel. We understand you need to decode 8 teeth with a gap. Is this 10 position with 2 missing teeth? Are the teeth wide, or narrow? Are these teeth evenly spaced? Do you have a picture of your wheel? A picture would answer many questions about what crank wheel pattern you are trying to decode.

Once we know the crank wheel you are tying to use, we can help guild you about if self simulation has already been done. It is very likely that you do not need an external piece of hardware for simulation. The rusEFI firmware can generate simulated crank wheel pulses as well as decode them. However we need to know your crank wheel pattern.

[hossein]

my crankshaft

[AndreyB]

I see "total 8", "skipped 1" skipped wheel signal - that's how you can configure this with rusEfi, this would allow simulated signal on pin PD1 and with self_stimulation you will have virtual callbacks into the logic.

On the engine sniffer snapshots you've shared I do not see the skipped tooth. At the moment I believe that your external simulator is not right, I could be wrong.

[hossein]

OK, i did what you said, it works, but please tell me that "Hard Limit RPM" is changeable.
Thanks.

[AndreyB]

Did you get tunerstudio working? Will post a screenshot of where that setting is located in a bit

[AndreyB]

Here you are

[hossein]

It's working now ...
Thanks @ @
And now just last question:
I changed trigger type to custom toothed wheel and
total tooth -> 8
skipped toooth -> 1
and then i have no green led - crank input -, so i decided to use honda cbr 600.
Anything needs to change for custom trigger ?

[AndreyB]

Unfortunately I have very limited telepathic abilities I believe you are providing very little information.

Please post a couple of Engine Sniffer screenshots where it does not run right, and a .csv log file from rusEfi console/out folder.

There is no gap on Honda trigger, why would you use it? It's same as using 8 total, skipped 0.



I still believe that the trigger signal you are trying with is not 8/1 skipped wheel.

[hossein]

The bts2140 is replaced by a circuit that based on irfz48-n and i have true spark.
Then i tried to run the engine... but !! after reading this site http://www.useasydocs.com/details/autotrig.htm and setting up trigger offset angle to 135< i have nothing i mean no spark on time no sound just nothing!.
I have tried to use a large scale of offsets from 80 (deg) to 180, and the results are similar, i mention : i can run it with carb and the old CDI transistor circuit.
In other word : how to calculate the right angle for trigger offset if the sensor is faced to tooth #3 at the TDC time.
he engine has two sensors, one on top face to face with tooth #3 on TDC #1, the other is 20-30 degree before tooth #7.

Thanks and i will apologize if i can not write correctly all important thing because of my poor English and mechanical knowledge.
Thank you all of you especially Russian .

[kb1gtt]

I'm still concerned about your electrical. That MOSFET is 55V, I believe you need 350V to 450V for spark under compression. Your 55V will probably spark when you put the plug on the side of the block, but I would expect failed spark when compression increases the amount of air between the spark plug gap. See some stuff about ignition modules are this thread.

http://rusefi.com/forum/viewtopic.php?f=4&t=286

[hossein]

I'm using a coil that running with 14V 7A : 5WY2843A => usage in PEUGEOT 504 ef7 engine for spark.
and this is my circuit -with some changes for biasing .

[AndreyB]

Does tachometer work when you are cranking your engine?

Does spark fire in bench mode without cranking your engine?

[hossein]

No it does not work.
But is it necessary to work ? i mean i mount the engine on a frame and i just run it step by step, and at the step one i need to spark, i fed the engine by its own carburetor.
Yes sometimes when the engine has tdc on any cylinder i can see the crank when i push the spark on 1-4 in tuner-studio program.

[puff]

I'd suggest looking at the engine sniffer in dev console while cranking.
Looking at your pictures, I can see two disks (so to say) - the bigger one with many teeth (does it have a missing tooth?) and the smaller one with 8-1 teeth. could it be that you are using the bigger one instead of the small one? how does the sensor look like and how do you connect it to the board?

[kb1gtt]

I still expect the irfz48-n may spark when plug is on the side of the block, but will fail when there is compression. I also expect that when it fails, it will damage up stream components, probably the port on the STM. This irfz48-n is bad as an ignition driver chip.

[hossein]

I still expect the irfz48-n may spark when plug is on the side of the block, but will fail when there is compression. I also expect that when it fails, it will damage up stream components, probably the port on the STM. This irfz48-n is bad as an ignition driver chip.

even using BC557 as a driver ?

[kb1gtt]

Correct, the BC557 does not matter. The problem is that the 55V driver will see far more than 55V which will either prevent spark, or damage the irfz48-n.

[hossein]

OK. i will change it, And the correct angle of offset is 135 in my case ?

[kb1gtt]

For the driver chip, I suggest you get one that specifies ignition in the list of applications. For example, this 410V chip from ST.

http://datasheet.octopart.com/STGB10NB37LZT4-STMicroelectronics-datasheet-12523196.pdf

I do not know your angle offset. Perhaps @ can guess. However he's very busy preparing for a race right now. I would guess he might tell you to use a timing light and adjust until the timing light indicates you are correct.

[kb1gtt]

Some other suggested chips include the two noted here.

http://rusefi.com/forum/viewtopic.php?f=4&t=286&start=26

[AndreyB]

the correct angle of offset is 135 in my case ?

timing light is the best solution, do you have access to one?

[matt]

i think it related with issue #297.

[hossein]

Thanks for replies.

And a question :
Is this chip usable or better to use for better performance and design ?
http://www.st.com/resource/en/datasheet/vb525sp-e.pdf

[kb1gtt]

Looks costly to me, but yes it should work. I see "Driving circuitry supply current ±200 mA" which means you need a chip like Frankenso hi/lo driver chip in addition to this chip. The STM32 can not drive this chip directly. As well, beware the diagnostic feedback. I would suggest you avoid using the diagnostic features, those will cause you problems. There are many design details not included in that datasheet which are required to make good use of those diagnostic features.