Continue to Site

Welcome to our site!

Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

  • Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

Help with Hall sensor signal amplification or filtering.

Status
Not open for further replies.

treefiddy

New Member
I have a signal strengths or noise problem with a Hall effect sensor and a programmable engine management computer. Apparently the ems has some noise issues and is quite sensitive to signal noise or the output signal has diminished. The sensor is placed over my camshaft and it senses the cam angle.

Is there any way that I can filter out the signal or amplify it a little bit with out hurting anything? I Lost my variable valve control system because the reference was strange and I had to shut down the system to avoid engine damage.


If anyone can help or give me ideas, it would be grate!
 
Do you have an oscilloscope? Without knowing anything about signal levels, signal to noise ratio, the shape, frequency, amplitude of the signals, we are shooting in the dark...
 
Last edited:
Use a single rail op amp with the hall sensor in voltage follower mode...this drops the impedance of the signal and can be tweaked to amplify a bit if necessary. Note that the op amp and hall sensor need to be very close coupled to prevent noise from being amplified as well. I'd suggest the higher temp. spec opamps...perhaps an LM258.
Ideally u need to 'scope the signal at the ecu input to be certain of it's integrity and determine if your signal wire needs to be rerouted away from noisy wiring/sources, like a starter motor, injector wire or tach signal.

Also, be sure that the hall sensor ground IS the same as the ECU ground. Best to tie it in directly.
 
Last edited:
Are you getting full output from the sensor?
 
Last edited:
Where could I find an oscilloscope that wont brake my pockets? It seams that all of them are 400+

As far as the output of the sensor, it gets choppy sometimes at idle and sometimes during acceleration. Sometimes left will read while right wont. Weird thing is that the stock ecu has no problem with the synchronization routine. But the aem is just to sensitive to this I think.

The ground son the sensor are the stock ones, the harness is unmodified in that area. I wonder if I'm getting noise from bad coil insulators.

If there where a place where I could rent an o.scope then that would be nice, since I would use the scope to fix this and for an other simple project That I will be posting soon.

Thanks for the replays so far!
 
Find a Ham or UberGeek Electroniker (like me) in your neighborhood.
 
Most people over here don't know what that is lol.


I was watching some vids in youtube about some portable china o.scopes that works up to 1mhz. Seams to me that for my application they will work. It would also be nice to have it around just in case of any new projects.

So far the one that I have seen is the DSO nano. Looking for more info on it or other similar devices.
 
Is it possible for me to use a high pass filter or a capacitor to clean the signal out? I'm really lost in the woods here guys! Please help me out, been dealing with this issue for the past 6 months with little to no progress what so ever.
 
i get my scopes on ebay for 100$ they are just little pocket ones though,

if you want to go cheap i'v seen others on ebay, like usb scopes which arnt much


a parallel capacitor would smooth your signal out , but we dont know if that's your problem,


you could be detecting extra background magnetic noise, your sensor may not be detecting enough, 2 very sepreate ways to deal with these,, also how do you know if you management computer is aok?

my first guess is that the problematic sensor(right) is loose, and falling out of detection range when the engine vibrates
 
treefiddy; Can we get some more background on your problem?

As far as the output of the sensor, it gets choppy sometimes at idle and sometimes during acceleration. Sometimes left will read while right wont. Weird thing is that the stock ecu has no problem with the synchronization routine. But the aem is just to sensitive to this I think.
Seems like you are trying to run two computers off one set of sensors. A left one and a right one. Am I right? Sounds like you added one of the computers a programmable engine management computer. Is that when you started having problums? My first thought is that it may not be a good idea to wire two computers to the same hall effect sensor. Andy
 
I'm running a single Engine management system. Its a Aem EMS Series 1. I talked to a company that deals with other Engine management solutions including this one, he told me that the signal strength on my car "2004 Subaru Impreza Wrx Sti" is low and that that particular engine management computer has problems picking up the signal and that their choice for EM solutions " a Hydra Ems" has an option to pull up the signal so to eliminate this problem. I have been analyzing this for quite some time now and all the evidence points that the problem is the AEM ems not picking up the weak signal. What I mean with the stock ecu is that she has no problem picking up the signal but she has a more advance noise filtering and signal boosting electronics.

I can't use the stock ecu due to some other limitations on my engine setup that requires the EMS but now I have this problem with the ems and I paid 1300 for this thing. The company won't admit that the design is flawed so I'm stuck attempting to fix their error. So instead of filtering it out, is there a way to step the signal up like .5 volts or so across the entire range with a boosting circuit using a pot?
 
treefiddy;
I can't use the stock ecu due to some other limitations on my engine setup that requires the EMS but now I have this problem with the ems and I paid 1300 for this thing. The company won't admit that the design is flawed so I'm stuck attempting to fix their error. So instead of filtering it out, is there a way to step the signal up like .5 volts or so across the entire range with a boosting circuit using a pot?
Sorry to hear your delema. The only true way to fix thare box is to 1st get a look at signal from the sensors with a scope, otherwize you are taking shots in the dark. Or find someone that has alraedy fixed the problum. Let us know when you get your hands on a scope. Andy
 
Status
Not open for further replies.

Latest threads

New Articles From Microcontroller Tips

Back
Top