Page 1 of 3

Electrical noise

Posted: Mon Jul 22, 2019 7:07 pm
by melvinvdb
Hi guys,

I've been struggling with trigger issues for a few months now.
I'm using a 60-2 trigger setup with a single tooth cam signal.
I can't get the single tooth cam to work because of noise.
Here are some oscilloscope measurements that will tell everything:
CAM signal
CAM signal
NewFile21.png (46.1 KiB) Viewed 13071 times
This is the CAM signal
CAM signal
CAM signal
NewFile22.png (78.39 KiB) Viewed 13071 times
Also CAM signal

Now here it gets interesting. This is the 5v signal inside the Speeduino:
5v
5v
NewFile24.png (39.4 KiB) Viewed 13071 times
As can seen here this signal is also very noisy.
And here is the 12v signal coming into the ECU:
BATT
BATT
NewFile23.png (36.5 KiB) Viewed 13071 times
My question is:
Is this not normal for a battery power supply on a running engine?
If not, this may also solve a mystery because I have broken two AEM 30-0300 widebands in my car (under warranty) that all failed differently. Maybe those widebands broke because of the voltage spikes/drops.

When I disconnect power of two ignition coils it looks like the noise gets less. When I disconnect power from all 4 ignition coils with the engine running at around 3k the noise immediately disappears.
I'm running these coils which are LS6 D585 knockoffs (herko B045HE):
https://www.ebay.nl/itm/Set-of-4-Herko- ... 3090305476

So my questions:
- Is this normal on a 12v power line?
- If not, should I blame the coils or am I missing something like capacitors at the coils?

Thanks :)

Re: Electrical noise

Posted: Mon Jul 22, 2019 7:47 pm
by LPG2CV
Hi

Not my usual area to help in. :)

Just for clarity, do you have VR, or, HE sensors?

And, if using HE, are you using a VR conditioner?

Re: Electrical noise

Posted: Mon Jul 22, 2019 8:10 pm
by melvinvdb
I'm using VR for the 60-2 wheel. Using ford zetec vr sensor with conditioning circuit.
For CAM signal I'm using a hall sensor.

When using VR only (wasted spark/semi sequential fuel) I'm having just a few (very minor) sync issues.
It was pretty bad but after changing the cable from a very small shielded cable to a larger 0.75mm2 shielded cable it got less sync issues. Also moved the air gap from 1.5mm to 0.6mm which also made it a lot better. Still having some minor sync issues.

The problem starts to get a whole lot worse when enabling the CAM signal. The voltage spikes can be seen in the tooth logger and I think the ECU is reacting on it by thinking it got a CAM tooth. As the CAM signal is only a single tooth per CAM revolution this messes up everything...

Re: Electrical noise

Posted: Tue Jul 23, 2019 1:04 am
by PSIG
LPG2CV wrote: Mon Jul 22, 2019 7:47 pm And, if using HE, are you using a VR conditioner?
melvinvdb wrote: Mon Jul 22, 2019 8:10 pm For CAM signal I'm using a hall sensor.
Still need this answer. ;)
melvinvdb wrote: Mon Jul 22, 2019 8:10 pmWhen using VR only (wasted spark/semi sequential fuel) I'm having just a few (very minor) sync issues. … The problem starts to get a whole lot worse when enabling the CAM signal. The voltage spikes can be seen in the tooth logger and I think the ECU is reacting on it by thinking it got a CAM tooth. As the CAM signal is only a single tooth per CAM revolution this messes up everything...
While you have a lot of low-level noise; the 'scope of your cam signal does not indicate it is enough to cause a false trigger. Not that trace or others you posted, anyway. ;)

Noise is typically better to attack than defend against. Play with the signal filters, run the cam signal through the conditioner, separate wires, twist wire pairs, spiderweb your grounds, and other standard stuff. But if that is not enough, go hunting for the noise source(s). Your 'scope can make a good detector.

David

High-Low_trigger_Vs.png
High-Low_trigger_Vs.png (48.03 KiB) Viewed 13057 times

Re: Electrical noise

Posted: Tue Jul 23, 2019 7:27 am
by melvinvdb
Thank you for your answer.
For the hall I'm not using the VR conditioner. If I would like to use it should I connect the hall signal to VR2+ and leave the VR2- floating or to gnd?

The captured signal was just a screenshot from a random moment. There are sometimes larger surges and spikes that will trigger the ECU. That trigger can be seen in the tooth logger. Also having a 5v that spikes to 8v and drops to -1v seems unhealthy for the atmega2560?

I've found a wiring diagram that recommends capacitors at the coil power supply and trigger signal:
Image
Maybe that will solve my issue? Seems a bit ugly to just add a few through hole capacitors to my custom wiring loom.

Re: Electrical noise

Posted: Tue Jul 23, 2019 8:32 am
by PSIG
melvinvdb wrote: Tue Jul 23, 2019 7:27 amFor the hall I'm not using the VR conditioner. If I would like to use it should I connect the hall signal to VR2+ and leave the VR2- floating or to gnd?
That would depend on the specific conditioner type. I would just try it both ways.

melvinvdb wrote: Tue Jul 23, 2019 7:27 amThe captured signal was just a screenshot from a random moment. There are sometimes larger surges and spikes that will trigger the ECU. That trigger can be seen in the tooth logger. Also having a 5v that spikes to 8v and drops to -1v seems unhealthy for the atmega2560?
Agreed.
melvinvdb wrote: Tue Jul 23, 2019 7:27 amI've found a wiring diagram that recommends capacitors at the coil power supply and trigger signal:
Maybe that will solve my issue? Seems a bit ugly to just add a few through hole capacitors to my custom wiring loom.
Yes, the supply cap is always a good idea to limit noise through the wiring harness, as it effectively 'shorts' the coil discharge, instead of running it through the positive wiring back to the coil. So, in an indirect way, it is more of an EMR noise capacitor than decoupler. I would connect at the coil connector, and hide it behind there. ;) Some using the LS coils prefer to ground the signal return with the coil power ground near the coil, instead of or in-addition to using the little cap for signal noise reduction.

I know you are up and running, but for others: note that diagram is for LS1, not the LS2/7 D585 "truck" coils, which is identical, but with the latch on the other side. Which side the locking latch tab is on is important. This makes the wiring order exactly backwards for some reason only GM knows. :roll:

David

LS2 Ignition Coil Wiring.jpg
LS2 Ignition Coil Wiring.jpg (44.59 KiB) Viewed 13036 times

Re: Electrical noise

Posted: Tue Jul 23, 2019 9:44 am
by melvinvdb
Hi David,

I'm using the max9926 with conditioner board v3.

I have 1uF 50v mlcc ceramic capacitors and 1uF 100v electrolytic capactors laying around. Which one would be best?
Ceramic non polarized or electrolytic polarized caps?

Re: Electrical noise

Posted: Tue Jul 23, 2019 4:23 pm
by runis6
Hello, also i am struggling with this problem and also i have 60-2 wheel without cam sensor. I am try to use a lot of ignition coils and it is not the problem. I have that the speeduino just like freeze and need to restart, i can drive about from 1 - 100km when it just freeze randomly (sometimes it like misfire) and need to restart.

Reinis!

Re: Electrical noise

Posted: Tue Jul 23, 2019 9:43 pm
by PSIG
melvinvdb wrote: Tue Jul 23, 2019 9:44 am Ceramic non polarized or electrolytic polarized caps?
Non-polarized and usually film types. I would suggest 100V minimum, as most I see are rated 100-300V for use on the coil (+) terminal to ground.

David

Re: Electrical noise

Posted: Wed Jul 24, 2019 10:50 am
by NickZ
runis6 wrote: Tue Jul 23, 2019 4:23 pm Hello, also i am struggling with this problem and also i have 60-2 wheel without cam sensor. I am try to use a lot of ignition coils and it is not the problem. I have that the speeduino just like freeze and need to restart, i can drive about from 1 - 100km when it just freeze randomly (sometimes it like misfire) and need to restart.

Reinis!
Are you connected to anything at the time and what are you connect with Eg: usb, bluetooth etc?