Hall sensor spikes?

Ask for and share advice on using the PicoScope kit to fix vehicles here.

Hall sensor spikes?

Postby baging » Wed Feb 22, 2012 8:51 am

Hi,

I'm trying to find signal errors on my car using an Megasquirt, and now when i got the Picoscope i finally was able to see the signals. But i'm not sure what's going on here with my hallsensor.... anyone that could explain?
Is this some kind of interference or is it ok? The "windows" looks ok to me??

Image

Frozen it looks like this....

Image

/Stefan
baging
User
User
 
Posts: 16
Joined: Wed Feb 22, 2012 8:45 am
Location: Sweden

Re: Hall sensor spikes?

Postby Alan » Wed Feb 22, 2012 9:37 am

Hi Stefan,

I suspect all is OK here. It shows a hall effect switching from battery voltage to 0V. Its more common for the ECU to feed a hall effect with a clean 5V supply but 12V is not unusual.

There is a lot of noise on the 12V part of the signal. Assuming the sensor has 3 wires (GND, signal, 12V) I would also scope the 12V on channel B. I suspect the interference is coming from injectors and if the hall effect is the crank sensor then my guess is that you have a 8 cylinder car with sequential injection. (On the other hand it could be a 4 cylinder wasted spark ignition signal causing the noise).

Anyway if you do capture the above waveform, please upload the .psdata file and also details of the vehicle / engine.
Alan
Pico Staff Member
Pico Staff Member
 
Posts: 239
Joined: Thu May 25, 2006 8:02 am

Re: Hall sensor spikes?

Postby baging » Wed Feb 22, 2012 10:10 am

Thank you for you quick answer!

I'll take some more measurments during the day and upload the data...
The setup is an Opel C20XE 4 cylinder engine with Megasquirt 2 3.0, VR cranksensor, Hall camsensor and sequential ignition. Coils is VAG 4-pin COP's.....

Thanks again :)
baging
User
User
 
Posts: 16
Joined: Wed Feb 22, 2012 8:45 am
Location: Sweden

Re: Hall sensor spikes?

Postby baging » Wed Feb 22, 2012 11:46 am

Ok, a new measurment with both hall and 12V feed to the hallsensor!

For some reason i couldn't attach it here, so i'll provide you with the url to my server instead....

http://www.mek1.se/misc/7/hall2.psdata

Another thing that hit me was that i might be on the totally wrong track? My thought is that the voltage "spikes" is as you say interference from either injectors or coils, and when they "level out" (as marked on the image below) it rather indicates a problem with feed to either coils or injectors? The car runs totally fine up to approx 5000rpm and then starts to miss badly. The Megasquirt log has suggested "tooth loss" but this makes me wonder....??

Image
Last edited by baging on Thu Feb 23, 2012 11:02 pm, edited 1 time in total.
baging
User
User
 
Posts: 16
Joined: Wed Feb 22, 2012 8:45 am
Location: Sweden

Re: Hall sensor spikes?

Postby Alan » Thu Feb 23, 2012 5:50 pm

Hi,

I think the upload / attachment issue is due to the large file size (13MB). I will check what our limit is. I saved one waveform from the buffer below and all seems OK.

As its a cam sensor then the 4 dips you are seeing in the 12V feed is probably from injectors. The 1V drop in supply voltage each time an injector seems a bit excessive to me. It may be normal for your system but I would check the condition of the battery (use the battery test option in PicoDiagnostics).

The spikes are probably interference from nearby ignition - such interference is a fact of life with scope diagnostics. The test leads are well screened, but the wiring loom in the vehicle is not so there is often genuine noise in signal you measure.

hall2.jpg

hall2a.psdata
(941.65 KiB) Downloaded 225 times
Alan
Pico Staff Member
Pico Staff Member
 
Posts: 239
Joined: Thu May 25, 2006 8:02 am

Re: Hall sensor spikes?

Postby baging » Thu Feb 23, 2012 11:01 pm

Thank you very much for your input! The more you explain and the more i use it, the more i see :)

That 1V drop in my injectors is something i'll have to adress!! As you say, it could be normal with the Keihin injectors i'm using, but will look at it!

Took some additional measurments today on the VR-signal. Before and after the conditioner, and with the correct amount of hysteresis, it seems to have done the trick! This scope is priceless!! :)

Image
baging
User
User
 
Posts: 16
Joined: Wed Feb 22, 2012 8:45 am
Location: Sweden

Re: Hall sensor spikes?

Postby Alan » Fri Feb 24, 2012 5:47 am

hall_error.jpg
hall_error.jpg (49.75 KiB) Viewed 3338 times


OK he above is interesting, I suspect the injectors are not firing where the arrows are showing and this does seem to be at higher RPM. I would be tempted to capture with 10 seconds across the screen, that would allow enough time for a snap throttle test. You can then zoom in after the event to see what happens. Make sure you have say 5M samples as you max samples.

Assuming you have a full PicoScope kit why not find the 12V feed to all the injectors (fuse box or relay normally, but not sure with the megasquirt system installed) and put the low amps probe around that. This will show whether the injectors are cutting out. If they are then the question is why. Monitoring injectors current, cam & crank (assuming you have 4 channels) would probably show it up.

I have been guessing a bit here - if you see the flat areas again, but the injectors are not cutting out it could be the voltage drop is being caused by ignition coils rather than injector coils. I would start at the battery to see if you have these 1V drops here. If yes, then the battery is a suspect. If no then look at where the 12V feed to the CAM sensor comes from - if its via an ECU then perhaps the ECU has a poor power feed or ground which might possibly be the cause of the issues at high RPM. On the other hand these voltage drops might be normal with the megaquirt system if the inectors take a lot of current - they just look a bit suspicious to me.

PS When you mention the conditioner above is that a module that takes the original crank signal and then turns it into a digital signal for the megasquirt unit? Not sure from your comments if adjusting this has fixed the original problem?
Alan
Pico Staff Member
Pico Staff Member
 
Posts: 239
Joined: Thu May 25, 2006 8:02 am

Re: Hall sensor spikes?

Postby baging » Fri Feb 24, 2012 8:59 am

Well, perhaps it's as simple as a Megasquirt sync problem.... when it fails to sync between cranksensor and camsensor i simply cuts out. And before the super-sensitive hysteresis adjustment it did lose sync occasionaly at higher rpm's.
That could be the problem with the injectors cutting out, but i will follow the route you've been describing!!
And after the adjustment, the problem seems to have gone away... for now anyway

Yes, the conditioner converts the crank signal to digital.

Injectors is getting the +12V feed from the fusebox, so the ECU is not involved in the feed.
Trying to find info about my battery. It's fairly small in size and unmarked, so don't know any of it's spec's.
The car is an Super 7, so there's no room for a bigger battery. But i will start there and see if the voltage drop starts there as soon as i get the low amps probe!

Same thing with the camsensor, it also get it's feed from the fusebox.

I don't have the full kit, so for now i'm missing the low amps probe... it's a 4-channel though and the probe is now ordered :)

Really appreciate you input here! /Stefan
baging
User
User
 
Posts: 16
Joined: Wed Feb 22, 2012 8:45 am
Location: Sweden

Re: Hall sensor spikes?

Postby Dan » Fri Feb 24, 2012 11:32 am

Hey.
Another old MS user here.

What injector driver are you using?
What injector impedance?
What ignition driver are you using? (if the CoPs are not internally driven)

Decent spec MS there - mine is a V2.2 MS1 with MS-Extra Circa '96 :lol: One day I'll have a car to use it on again.

Most importantly regards the MS - what firmware are you using? (You have to expect issues on Beta firmware - it's not for road testing - many people have chased shadows using such firmware!).
Knows pre-2000 Fords and R33's better than the back of my own hand.
Anything newer... I'm probably driving it :-(
Dan
User
User
 
Posts: 13
Joined: Mon Jan 30, 2012 11:28 am

Re: Hall sensor spikes?

Postby baging » Fri Feb 24, 2012 6:13 pm

Hi,

It's an MS2 3.0 with MS2/Extra firmware 3.2.1. and as far as i know it's tha latest stable version....

Injector drivers are internal MS 3.0
Injector impedance 13 ohms, Keihin KN-4 from an Honda Blackbird.
Ignition driver is internal in the COP's.... Bosch 06B 905 115R

Started this project witn an MS1 2.2 but had to upgrade it.... so now it sits on a shelf in my garage :)
Ordered the MS to my spec's so i haven't built it myself.... and since the day that MS found it's way in to my garage, i've had to go beyond the ordinary wiring diagrams i was used to.
Never in my wildest dream would i think i'd end up with an oscilloscope!! ;) :D
baging
User
User
 
Posts: 16
Joined: Wed Feb 22, 2012 8:45 am
Location: Sweden

Next

Return to Diagnostic discussion

Who is online

Users browsing this forum: No registered users and 0 guests