For any discussion not specifically related to your project
By jackscr
#50050
Hello,
I've been wrestling with the Vss input for some time now, and I managed to get it to be somewhat useable but I still get random spikes in the log sometimes even double the actual speed and last for a fraction of a second.
I am using a proximity switch through a filter and into the 0.4.3c board.
I attached a photo with a log and you can see the spikes in the red line .
I tried to set the smoothing factor in TS from 10 all the way up to 110 and it does not make a difference.
Oh and it seems to get worse when the speed changes (accelerating or decelerating )
Could this be due to some interference ? Has anyone encountered a similar issue?
Attachments
Vss spikes.png
Vss spikes.png (41.17 KiB) Viewed 5133 times
By jackscr
#50077
It's an Autonics Pr12-2DN it's a NPN N.O. , the signal was a bit messy directly from the switch but I ran it through a Hex Schmitt-Trigger Inverter .
By NickZ
#50079
and did you put some filtering caps on it? you should use the normal analog circuit the other sensors use. 470 ohm resistor and 0.1uf and 0.22uf caps.
User avatar
By PSIG
#50083
Spec's on that sensor show it has a response frequency of 1500 Hz. :shock: Also, it has a standard target dimension to read of 12mm x 12mm at a distance of under 1.4mm. I don't know your trigger setup, so is there any way you are outside of spec's on those or others?
User avatar
By Chris Wolfson
#50086
I don't know your setup, but if it is an engine, use something that anyone uses, because it works. Hall or inductive, anything else is exotic. In this case exotic = complicated and not sure to work well and durable. Even optical sensors have fallen out of favor for car applications. The Brit's liked them a lot.

Your sensor is made to limit travel with opening and closing doors or the movement of machine parts. I do not say that it can not be used as a trigger for an engine, but to me it seems just like the right part at the wrong place.
Maybe elaborate why you do not use a 8$ Hall or VR sensor for car engine applications, available at any corner of the world.
Seems you can not win anything with that kind of sensor, but loose a lot of time...
By jackscr
#50087
NickZ wrote: Wed May 05, 2021 10:13 pm and did you put some filtering caps on it? you should use the normal analog circuit the other sensors use. 470 ohm resistor and 0.1uf and 0.22uf caps.
Yes I used the TPS filter diagram for the input of the VSS
By jackscr
#50088
PSIG wrote: Thu May 06, 2021 12:40 am Spec's on that sensor show it has a response frequency of 1500 Hz. :shock: Also, it has a standard target dimension to read of 12mm x 12mm at a distance of under 1.4mm. I don't know your trigger setup, so is there any way you are outside of spec's on those or others?
Yes I know it's freq is pretty low but I am using it on the gearbox output with 8 "teeth" per rotation and that works out to a max off 500Hz also it seems to trigger ok and consistently I did check it with an oscilloscope, and even if it would lose sync at higher RPM it does not match the fact that I am getting spikes at any speed...
By jackscr
#50090
Chris Wolfson wrote: Thu May 06, 2021 3:08 am I don't know your setup, but if it is an engine, use something that anyone uses, because it works. Hall or inductive, anything else is exotic. In this case exotic = complicated and not sure to work well and durable. Even optical sensors have fallen out of favor for car applications. The Brit's liked them a lot.

Your sensor is made to limit travel with opening and closing doors or the movement of machine parts. I do not say that it can not be used as a trigger for an engine, but to me it seems just like the right part at the wrong place.
Maybe elaborate why you do not use a 8$ Hall or VR sensor for car engine applications, available at any corner of the world.
Seems you can not win anything with that kind of sensor, but loose a lot of time...
It is for the gearbox output so the rpm is 3 times less than actual. I had this sensor laying around and tried it out and it seemed to work fine but yeah I will look into a dedicated hall sensor if all of you agree this is a sensor issue not some weird interference or something else.

I've seen allot of people have issues with implementing VSS on speeduino for some reason, can you recommend a DIY sensor in the same format as the one I have that is known to work for this application? A normal VR or a Hall with magnetic trigger is off the table for me.
User avatar
By Chris Wolfson
#50092
The usual problem with sensors are where you power them from, where you ground them and what kind of noise you induce into all of these wires. If you go for a Hall sensor, use a shielded 3-wire cable. Ground and feed it at the ECU, connect the shield only to the ECU ground, leave it open at the sensor side. A VR sensor needs only 2 shielded wires, but you need a conditioner at the ECU that converts into a clean square. There are countless versions of both types, take what suits your situation best, but be so wise to use only automotive parts. Keep wires fixed, don't have them dangling around, this can give distorted signals. Think which wires you bundle with zip ties, some are better kept 2" apart.
Mount the sensor in a way that does not allows it to resonate. All kinds of bend sheet metal construction are a possible cause for problems. So try to fasten it solid, if possible. Just looking at some OEM sensor constructions should show anyone how a good sensor mount is done. Learn from the industry.
The Speeduino is a working construction, anything you do is experimental. Keep that in mind, before blaming the electronics.
By the way, a proximity sensor is not really made for rotation afaik.

It looks like you have a fuel supply issue. readin[…]

Will this have an updated version about this featu[…]

Vr Conditioner Noise when cranking

I did the fix, but it's not work. I put my setti[…]

Perhaps some different points and pictures. Instr[…]

Still can't find what you're looking for?