Hardware problems... RESn gets triggered by touching it with a bit of wire

Wossname · 2012-01-21 12:23

I built a propeller dev board months ago and it seems to work just fine, however I've just noticed an unfortunate problem where the chip seems to go into it's reset state if I touch the RESn pin with a piece of wire or a pair of pliers or any metallic object. The wire doesn't even need to be connected to anything, just a 3" bit of breadboard wire will do the trick. I have a 10K pullup resistor on it and a reset switch to ground in this configuration...

         &#61463; +3v3
         &#9474;
         &#9474;
         &#61628; 10K
         &#9474;
         &#9474;
RESn  &#9472;&#9472;&#9472;&#9515;
         &#9474;
         &#9474;
         / reset switch                            
         &#9474;
         &#9474;
         &#61464; GND

I've got this exact same problem on a second Propeller design of mine (which is running entirely different code).

Edit:

Even if I connect RESn directly to +3v3, the Prop will still reset if I also touch the RESn pin with a metallic object!
What the hell is going on here?

How do I fix this problem?

Phil Pilgrim (PhiPi) · 2012-01-21 13:51

What kind of filter and bypass caps do you have in your designs?

-Phil

Cluso99 · 2012-01-21 16:33

This is expected. Your body will reduce the 3v3 at the reset pin sufficiently to cause a reset. You can touch just about any reset pin on any chip and get the same result. However, a long wire is different as itacts like an aerial and depends on other factors like RFI. As Phil said, do you have correct bypass capacitors and how stable is your power supply. All these things matter.

kwinn · 2012-01-21 19:11

Not only RFI is involved. Our bodies pick up quite a bit of 60Hz from the surrounding wiring. Try putting your finger on a scope probe.

Duane C. Johnson · 2012-01-21 19:58

Just put a capacitor between the RES pin and the 3.3V pin next to it.
I use 22nF but the value is not critical.
If to large the PropPlug may not reset correctly when required.

Duane

Heater. · 2012-01-22 01:56

As well as the cap I guess you could reduce the pullup resistor, to say 1k. Makes the circuit stiffer to such outside influence at the cost of more current when the reset switch is closed.

Wossname · 2012-01-23 01:40

I normally use 100nF bypass caps, normally positioned underneath the Prop (within the DIP socket footprint) near the power pins. I always use a decent fixed value regulator (7803 or LD33V or something like that) with the smoothing caps recommended in their respective datasheets.

I'll try the 22nF cap in parallel with the pullup as suggested.

Thanks for the advice guys.

Peter Jakacki · 2012-01-23 03:46

Wossname wrote: »

I built a propeller dev board months ago and it seems to work just fine, however I've just noticed an unfortunate problem where the chip seems to go into it's reset state if I touch the RESn pin with a piece of wire or a pair of pliers or any metallic object. The wire doesn't even need to be connected to anything, just a 3" bit of breadboard wire will do the trick. I have a 10K pullup resistor on it and a reset switch to ground in this configuration...

<snip>

Edit:

Even if I connect RESn directly to +3v3, the Prop will still reset if I also touch the RESn pin with a metallic object!
What the hell is going on here?

How do I fix this problem?

How do you fix this problem??? Why that's easy, just don't touch the reset pin with a metallic object, that's how!
No really, I'm amused that you find this an unfortunate problem. If I touch the input to an audio amplifier with a piece of wire or my finger then I hear a hum, how do I fix it? Remove the finger!

Okay, now that I've had my fun I am guessing that with the reset pin connected to Vdd and touching it still results in a reset seems to smack of a mains to ground leakage. I am assuming that in every case that you are holding the wire and if your board is powered from a switch-mode mains power supply then you may have the dreaded mains leakage problem caused by the EMI suppression capacitors that form a voltage divider across the mains to the output common. Anyhow, I could be wrong but we have so little to go on, a photo of the complete setup is better than hit-n-miss guessing. Have you tried powering the board from a battery just to see if it still plays up?

Wossname · 2012-01-23 04:07

I run variously from batteries or wall warts depending on my mood/laziness. The issue is common to all three of my dev boards.

I'm concerned about this purely from a reliability point of view. I don't want my board resetting itself because of some random environmental factors.

Just tried adding a 10nF cap in parallel with the pull-up and the problem seems to have improved a bit - now I have to try a bit harder to make it happen, but it's not great. I'll live with it for now.

Peter Jakacki · 2012-01-23 05:43

The strange thing is why it resets even after you tie reset to Vdd. I am also curious as to whether this "dev board" is one of your designs or someone else's. Once again, a photo may save dozens of posts to and fro.

Ariba · 2012-01-23 09:49

Have you connected BOEn to Ground?

Andy

Wossname · 2012-01-26 05:50

Peter Jakacki wrote: »

The strange thing is why it resets even after you tie reset to Vdd. I am also curious as to whether this "dev board" is one of your designs or someone else's. Once again, a photo may save dozens of posts to and fro.

It's my own layout but it's based on the schematic in the Propeller manual. Two other people I know have the exact same problem with their own personal designs also based on the manual diagram.

Ariba wrote: »

Have you connected BOEn to Ground?

Yes, because the schematic in the manual suggests to do that.

Andy

Hardware problems... RESn gets triggered by touching it with a bit of wire

Comments