Axon 1 Glitch

[moltrex84]

Hello,


I have a small issue with the Axon 1. I’m interfacing 4 SRF-05 and sometimes in the middle of the code execution the axon glitches out and suddenly the flashing leds in the SRF-05 stop blinking and the axon is no longer functional but the power led stays on, and the Sabertooth stay with the last command …

This glitch is random … I don’t know the source of the problem. Maybe the axon1 need more current and it turn off and back on again.



Connection to the axon is a voltage regulator. Axon is power by 8VDC.  The SRF-05s are powered by a 5VDC regulator.

Any Ideas

[knossos]

Sounds to me like the the axon is reseting.  Check that your battery pack is charged and that it can provide enough current for all the devices you have connected.

[Admin]

In addition to what knossos said, if the problem still occurs when the battery is fully charged, add a capacitor between power and ground on the regulated 5V line.

The capacitor should be between 5uF and 20uF, any type is fine.

[moltrex84]

**I have the Axon connected to a two 12 VDC in series car batteries "24 VDC" using a regulator to regulate to a 8VDC ... So it have juice to drink for a while ... Axon 1 sees 8VDC always using the same common as all the system, is not floating.

What is the purpose of the Capacitor between the 5VDC regulated pin and the ground The regulator comes from the regulator of the ADC ports ...

[Admin]

The purpose of the capacitor is to stabilize short-term voltage spikes/drops that can be caused by whatever hardware you attach to the Axon. It has some already, but under rare circumstances a reset can still occur on the original Axon.

The best thing to do is flip the Axon over and solder the cap to the pins on the bottom of the Axon.

[moltrex84]

I will try that tonight ... I let you know ...  Thanks

[knossos]

Correct me if I'm wrong, you have 2 12v batteries in series for a total of 24v.  The 24v is running into an 8v regulator.  The 8v regulator is feeding the Axon.  The Axon's on board 5v regulator is feeding the SRF-05's.

What kind of voltage regulators are you using?  The SRF-04 which is very similar to the SRF-05 uses 30 - 50 mA.  4 of these would be about 120 - 200 mA.  The onboard regulator on the Axon is limited to 1.5A max.  If you are feeding it from another voltage regulator, then you are further limited by that voltage regulators max.

More importantly though, is the 8v voltage regulator getting hot?  Typically the higher the voltage applied to the regulator, the more the regulator has to dissipate as heat.  Voltage regulators usually have thermal protection and shut off if they are getting too hot.  I would guess this is the problem you are having with your Axon.  Consider getting a switching regulator which are much more efficient and can handle greater differences between input and output voltages.

[moltrex84]

Correct me if I'm wrong, you have 2 12v batteries in series for a total of 24v.  The 24v is running into an 8v regulator.  The 8v regulator is feeding the Axon.  The Axon's on board 5v regulator is feeding the SRF-05's.

True.

What kind of voltage regulators are you using?

L7808CV

More importantly though, is the 8v voltage regulator getting hot?

Thats correct, but I connect it directly to a adjustable external high quality Power supply and the problem continues ... But I'm aware that is not reliable to have the regulator be that hot, and it have a heat-sink.



Also I have a digital output that keep a 5VDC relay in normaly-open "I'm energizing the coil with signal and ground"... But I also discard this by connecting it directly to the Power Supply 5VDC and the glitch continues ...

I also tried different capacitors from 14uF to 34uF in the regulated ports "ADC" to suppress the spikes, but the glitch continues.

[Admin]

I'm with knossos on the voltage regulator not being sufficient. Get a second one and put it in parallel.

[knossos]

Thats correct, but I connect it directly to a adjustable external high quality Power supply and the problem continues ... But I'm aware that is not reliable to have the regulator be that hot, and it have a heat-sink.

Just to clarify, when you say you connect it directly to an adjustable external high quality power supply, do you mean you connect the 8v regulator, or the Axon itself to the power supply?  Also what voltage is the external power supply set for?

Assuming you haven't already, I would try adjusting the external power supply to about 6v and bypass the 8v regulator.  If the problem continues, I would remove everything but power to the Axon and add in components one at a time until the fault occurs.  I would also check my actual voltages from the sources and regulators with a multimeter.  If you have a method of checking current, I would check currents as well.  Also, just to make sure, all your grounds are connected together, right?

[moltrex84]

Just to clarify, when you say you connect it directly to an adjustable external high quality power supply, do you mean you connect the 8v regulator, or the Axon itself to the power supply?

The Axon to the Power supply ...

Also what voltage is the external power supply set for?

Is variable, I set the voltage to 6VDC.

grounds are connected together, right?

They are all connected together.


-------------------------------------------------------------------------
I connect the sensor to a external power supply ... and the bug continues.

I would try adjusting the external power supply to about 6v and bypass the 8v regulator.  If the problem continues, I would remove everything but power to the Axon and add in components one at a time until the fault occurs.

I was thinking about that, because is not very obvious what is making the fault.


-------------------------------------------------------------------------------

(The project and what is connect it so far)
I'm making an autonomous electric wheelchair.

I have connected a Sabertooth, compass, SRF-05, and i'm using a digital out put to trigger a 5VDC relay ...

The wheelchair have been command with the laptop for a long time without any problems. The bug begins as soon as 4 SRF-05 were connected.

[moltrex84]

I was testing today and I found that the microcontroller is not restarting... Because I have in the code to turn on for 5 seconds a led and then turn off the led... This is executed only once at the beginning.

When the axon goes in the glitch it doesn't restart.

Instead the sabertooth keep sending the last value to the motors and the SRF-05 stop blinking, the microcontroller leds still turn on ...


--------Notice I have a delay betweens SRF-05 read

int sensingDelay = 7;
      

        distanceRead(sonar0);
      dist_sonar0=sonar0.distance.cm;//store sonar value
      delay_ms(sensingDelay);

        distanceRead(sonar1);
      dist_sonar1=sonar1.distance.cm;//store sonar value

         delay_ms(sensingDelay);
        distanceRead(sonar2);
      dist_sonar2=sonar2.distance.cm;//store sonar value
   delay_ms(sensingDelay);
      
        distanceRead(sonar3);
      dist_sonar3=sonar3.distance.cm;//store sonar value

   delay_ms(sensingDelay);

[knossos]

At this point then, I would probably take a close look at my code to make sure it wasn't something I was inducing.  I would use a lot of debugging statements printed via UART to see when the problem is happening.  If the problem isn't your code, this could also help narrow down which item is causing the glitch and why.