Page 9 of 10

Re: VSS help please

Posted: Tue Jan 27, 2015 7:52 am
by Bernard Fife
Addicted,

Because I can't reproduce your issue here at all, I suspect that there may be some issue with your hardware. This might explain why it came up suddenly. If this was mine, I would have a very, very good look at the area around the processor area of the PCB to make sure that are no stray bits of snipped-off leads, solder, or residual flux on the traces, vias, solder pads, jumpers, or other connections to the microprocessor.

Re: VSS help please

Posted: Wed Jan 28, 2015 1:10 am
by Addicted
Hi Lance,

When you bench test what do you use as a signal generator? The 555 on a transtim? or something a little more advanced? If you cannot replicate my issue I'd like to see if I can replicate your working setup...

One thing I can guarantee is the board being clean.. it has been cleaned at work, where we build US military radios... :D

Re: VSS help please

Posted: Wed Jan 28, 2015 6:43 am
by Bernard Fife
Addicted,

In my testing I use a variety of signal sources, including actual VR sensor and a wheel mounted on a bench, the trans stim, and my signal generator (with a number of wave forms).

You might still want to look at the board as there may even be a manufacturing defect.

What was the code you were running when it worked for you? Have you tried running that code again? Did it help? The answers would be helpfully diagnostic.

Also, the scope shot is too slow to see if there is any spikes in the signal - do you have any that show one or two pulses at a time? You might need to look at a bunch of these before deciding is there is any noise.

Re: VSS help please

Posted: Thu Jan 29, 2015 12:41 am
by Addicted
Morning Lance,

I've booked a day off work tomorrow to have another look at this. The biggest indicator for noise is that it registers speed even when the shaft isn't rotating, depending on the position of the shaft the processor pin will either see 5Vref or ground. If I increase the masking in 5% steps until that speed is zero it doesn't then pick up and signal at all when moving, which would lead us to think the noise amplitude was as big as the signal, but I don't see it on the scope.

I've a few things to try tomorrow, including implementing a schmitt trigger as I'm pretty convinced the problem is hardware not software...

Normal VR circuit or Schmitt Trigger?

Posted: Mon Feb 02, 2015 12:37 am
by Addicted
I'm taking a different route with my VSS signal and adopting a 12 tooth VR wheel and sensor. I have all the components needed to build a schmitt trigger so am wondering if I should just go staight ahead and build that or start with the 'as manual' VR circuit first?

Any opinions?

Re: Normal VR circuit or Schmitt Trigger?

Posted: Mon Feb 02, 2015 10:54 am
by Bernard Fife
Addicted,

Because we don't know what the particular issue might be, it is very hard to say what the best course of action is.

Re: VSS help please

Posted: Tue Feb 03, 2015 12:26 am
by tobmag
Addicted.

I have had the ame issue in my car that seems to generate a lot of electronic noise by it self....I also had the speedo reading up to 50kph while standing still.

I did try an Schmitt trigger but it was not reliable enough so ended up with another solution, you will find it on page 4 in the post below.
There you also fins another solution from Ashford on page 5

I did also try some externa VR conditioner using the MAX9926 circuit but they went in th egarbage bin as well.
That curcuit is a very good one but extremly sensetive for any noise in th electronic.
On the bench it was the far best I tried, in the car it was worse than the standard VR build....

viewtopic.php?f=4&t=678


BR//Tobmag

Re: VSS help please

Posted: Tue Feb 03, 2015 12:47 am
by Addicted
tobmag wrote:Addicted.

I have had the ame issue in my car that seems to generate a lot of electronic noise by it self....I also had the speedo reading up to 50kph while standing still.

I did try an Schmitt trigger but it was not reliable enough so ended up with another solution, you will find it on page 4 in the post below.
There you also fins another solution from Ashford on page 5

I did also try some externa VR conditioner using the MAX9926 circuit but they went in th egarbage bin as well.
That curcuit is a very good one but extremly sensetive for any noise in th electronic.
On the bench it was the far best I tried, in the car it was worse than the standard VR build....

viewtopic.php?f=4&t=678


BR//Tobmag
Thanks Tobmag, actually I'm very aware of that thread as it was reading your posts on the subject that made me ask the question as to whether I should implement the normal GPIO VR circuit or just go straight to the schmitt trigger. I've assembled all the components to allow me to test a few of the suggested solutions on seperate PCB's..

I've spent 6 months under the mistaken impression that my problems were caused by Lance's software interupts, so I can only apologise to Lance for that. The root of all my issues appears to be a noisy ground plane in the car and with it being an OEM loom my only solution would be to build a new one from scratch..

Re: VSS help please

Posted: Tue Feb 03, 2015 1:35 pm
by tobmag
Addicted,

No need to build a new GPIO there is plenty of space inside the box for an additional pcb.
Ill try to remember to post a pic of the inside of my box.

Or are you talking about the loom?


BR//Tobbe

Problem with VR1 circuit

Posted: Sun Feb 15, 2015 6:46 am
by Addicted
I've built the VR1 circuit as per the manual and am feeding a signal from a vr sensor and a 32 tooth wheel.

I get a lovely clean signal into the circuit, but it doesn't make it past R37.. In these two photos trace one is the signal at the input (bottom of R59), trace two is either side of R37. Can anyone offer any advice as to what may be happening?

Image[/URL

[URL=http://s7.photobucket.com/user/mickeyma ... h.jpg.html]Image


The only difference is that I have used BC547 transistors instead of 2N3904, but the manual suggests it should be fine. I initially fitted the BC547s backwards as their pinouts are opposite to 2N3904 and assumed that was the reason for the signal loss... However they were replaced correctly with new ones and it made no difference..