low voltage tripping bounce?

[Magnethead494]

On the first video, that looks like the scope was set to 5 volts/div, maybe?

Have you tried any of my suggestions yet?

No idea what it was set to- the menu is stuck in japanese/chinese right now and I really don't know how to use a scope effectively, even if it were in english. I know how to move positions and that's about it.

I've not gotten to your suggestions yet due to time and money, but I'm going to add the 430's to the controller end and start there since I have them and it can't hurt. I'm still not sure how to ground the shielding without losing the ability to remove the cable. I only have access to the troublesome cars for 4 hours every wednesday so I'm trying to pack in what I can afford to do. Right now I'm working for the racetrack 2 nights a week at 7.25/hr just to pay for gas to school and back. Right now I'm balancing assembler and java homework, and trying to make these things work.

 

[Magnethead494]

Here is today's back-and-forth.

Hi Steve,

In the processor, how are you sampling the button inputs to determine a valid switch closure? By this I mean, are you looking for a low for a certain period of time, or just any detected low condition?
Regards,

Randy

Randy,

It checks for a logic low lasting longer than 40ms. It will only listen for one trigger, so holding the switch down doesn't make it scroll, must be tapped multiple times.

Thanks,
Steve

Hi Steve,

I would think that 40mS would be more than enough.

When it does bad things, what exactly does it do?

Randy

Randy,
A random number will "roll" up or down, like the switch has been triggered. The direction and which number, are both random. Sometimes it's quickly, sometimes it is slowly.

Does the power supply end look right? It's just 2 basic regulators, but I don't know of the 1000uF cap and 400V fast acting flyback diode are enough to filter the noise generated by unprotected solenoids, motors, ec

Steve

Steve,

When the numbers change, does the dash display change to match?

The regulators should be ok, and the diode is really not doing you any good. I would add an input protection diode in series with the input power, before the filter cap. This will block negative spikes, and, believe it or not, racers will connect the power backwards and blow up your unit if the diode isn't there.

Have you disconnected the comm cable at the controller with the engine running to see what it does? That would tell you if it was noise in the comm, or noise getting into the controller from a different direction.

Something keeps getting me back to the 40mS. The ignition noise we have had to deal with has always been in the uS range, so 40 mS should be more than enough. This makes me want to look into the code and see if it is really doing the right thing. I would like to connect a signal generator to a switch input line, generate various pulse widths of negative going pulses and see how the display responds.

Regards,

Randy

Randy,

I set the diode there to act as a flyback against solenoids releasing or motors shutting off. But I understand your point about an inline catch diode.

I disconnected the controller and left the cable laying in the seat and it still misbehaved, but I never did unplug the cable from the main box. It was on my to-do list for wednesday but I was in a time crunch.

My O-scope has a 1KHz frequency generator on it that I could probably use...

Thanks,
Steve