PicoScope 7 Automotive
Available for Windows, Mac, and Linux, the next evolution of our diagnostic scope software is now available.
Noise in ECU signals..
-
Walterronny
- Newbie
- Posts: 5
- Joined: Sun Jul 21, 2024 2:42 pm
Noise in ECU signals..
Hi! new here and new user with picoscope. As i mentioned in the Introduction group im looking for try to decode a kline of a Civic +02 . for try first time i got a logic analyzer, very cheap and works good but i want to go more advance due it was still difficult to try to decode using logic analyzer, here an example:
a bit of zoom:
as you can see, all the signal are perfectly with no noise. there is a rare thing i noticed with this analyzer but i will comment later.
Ok, decided to get picoscope,im sure it will works better for my purpose/goal (sorry for my english!)
ok then decided to use it, i was learning how software good and there is something wrong i dont know what's wrong im doing or software or hardware can't do that.
here is a first example, i saved all test but just for a first check.
you can notice it's not "square" signals, to 1ms/div get that..also it's still show per waveform and i dont know if it's possible to get all waveform signals togheter.
decide to "play a bit" with 1x and 10x, sorry i dont know how it's say in english right now. so i get this
stil not enough lenght for signal.
if i change to 50ms/div,i get good lengh signal
but if i zoom it this is what happens
so 5ms could be ..OK
but you can see 200v scale..not good
what im looking for is to try to extract the bytes that control some sensors for a project im working on.
https://www.hondata.com/help/kmanager/i ... output.htm
here is a helpfull page that shown good info but still it's impossible to get that kinda signal so far also i dont know how it's possible go get that kinda decoder.
so far we got that 0x82 is for a sensor im looking for but only get it using the cheap logic analyzer. im measuring in the E24 pin, according that website say it's 12v and battery voltage when engine is runinng but i dont know how the cheap logic analyzer works good since it will works only up to 5v max input.
Also in the workshop manual says this
so that lost me , anyway there is not problem if works to 12 or 5v, main problem is i dont know why dont get clean signal and id like a bit help for getting the right way to try to decoder this signals.
i add another test i did today.
I readed like 10 times the great article of Ben about Kline but looks like it's used a OBD2 reader , this time will be different i think due this is other Kline protocol that is used for get CLT sensor in cluster working,etc. Any help will be appreciate it a lot!! thanks in advance!
W.R
a bit of zoom:
as you can see, all the signal are perfectly with no noise. there is a rare thing i noticed with this analyzer but i will comment later.
Ok, decided to get picoscope,im sure it will works better for my purpose/goal (sorry for my english!)
ok then decided to use it, i was learning how software good and there is something wrong i dont know what's wrong im doing or software or hardware can't do that.
here is a first example, i saved all test but just for a first check.
you can notice it's not "square" signals, to 1ms/div get that..also it's still show per waveform and i dont know if it's possible to get all waveform signals togheter.
decide to "play a bit" with 1x and 10x, sorry i dont know how it's say in english right now. so i get this
stil not enough lenght for signal.
if i change to 50ms/div,i get good lengh signal
but if i zoom it this is what happens
so 5ms could be ..OK
but you can see 200v scale..not good
what im looking for is to try to extract the bytes that control some sensors for a project im working on.
https://www.hondata.com/help/kmanager/i ... output.htm
here is a helpfull page that shown good info but still it's impossible to get that kinda signal so far also i dont know how it's possible go get that kinda decoder.
so far we got that 0x82 is for a sensor im looking for but only get it using the cheap logic analyzer. im measuring in the E24 pin, according that website say it's 12v and battery voltage when engine is runinng but i dont know how the cheap logic analyzer works good since it will works only up to 5v max input.
Also in the workshop manual says this
so that lost me , anyway there is not problem if works to 12 or 5v, main problem is i dont know why dont get clean signal and id like a bit help for getting the right way to try to decoder this signals.
i add another test i did today.
I readed like 10 times the great article of Ben about Kline but looks like it's used a OBD2 reader , this time will be different i think due this is other Kline protocol that is used for get CLT sensor in cluster working,etc. Any help will be appreciate it a lot!! thanks in advance!
W.R
- Attachments
-
- key on engine off then engine turn ON.psdata
- (157.01 KiB) Downloaded 212 times
Last edited by Walterronny on Fri Aug 02, 2024 4:50 pm, edited 1 time in total.
-
Walterronny
- Newbie
- Posts: 5
- Joined: Sun Jul 21, 2024 2:42 pm
Re: Noise in ECU signals..
Edited: didnt know why i was getting error for uploading BBCode URL but i think it's fixed, dont know how to delete post..sorry!
-
Steve Smith
- Pico Staff Member
- Posts: 1671
- Joined: Sun Aug 25, 2013 7:22 am
Re: Noise in ECU signals..
Hello and thank you for the post
Looking at your logic analyser images, they look like a graphical representation of the signal you are measuring (almost text book) as opposed to the scope that is capturing the analogue voltage in real-time.
In your 3rd image (signal not square) this can be attributed to several factors of which sample rate, vertical resolution (8-bit) and probe impedance will contribute. We should also consider if some of what you have captured (signal not square) is physically present.
Using the Pico x10 probes is a good idea as these have an increased impedance and will prevent your scope influencing the serial data under test
Given you are capturing a 12 V signal use a x10 probe with +- 20 V input range to improve vertical resolution
To obtain more data (i.e. a stream of data) and so more “packets” set your scope to 200 ms/div
The hope the image below will help
Once you have the relevant number of buffers (Waveforms) apply the decoder as per Ben’s article
I hope this helps, take care.......Steve
Looking at your logic analyser images, they look like a graphical representation of the signal you are measuring (almost text book) as opposed to the scope that is capturing the analogue voltage in real-time.
In your 3rd image (signal not square) this can be attributed to several factors of which sample rate, vertical resolution (8-bit) and probe impedance will contribute. We should also consider if some of what you have captured (signal not square) is physically present.
Using the Pico x10 probes is a good idea as these have an increased impedance and will prevent your scope influencing the serial data under test
Given you are capturing a 12 V signal use a x10 probe with +- 20 V input range to improve vertical resolution
To obtain more data (i.e. a stream of data) and so more “packets” set your scope to 200 ms/div
The hope the image below will help
- 1
I hope this helps, take care.......Steve