Optimum value for a reset pullup resistor is...?

Go To Last Post
10 posts / 0 new
Author
Message
#1
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Hello, Hope I can get some help on this easy one. Searched the data sheet quickly, no luck.

What size resistor is the best pull up resistor for a Atmega168 running on 3.3Volts to sit on the reset pin?

I have two chips, one works with a 78k pull up, the other works with a 1.5k pull up. Seems to work either way, power usage is of no concern.

What value resistor is the "best" so I can standardize?

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I use 10K ... but maybe my definition of "best" differs from yours.

Cheers,

Ross

Ross McKenzie ValuSoft Melbourne Australia

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Ok thank you =)

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I use 10k, too.

Yesterday, I read some Atmel document stating that 10k is the minimum. I'll see if I can find it again.

EDIT: It's in the datasheet chapter 25.3 (debugWIRE) : "Pull-up resistors on the dW/(RESET) line must not be smaller than 10k"

The DC characteristics (29.3) lists the internal reset pullup at 30k-60k, so there probably isn't much point in adding a weaker external pullup.

Sid

Life... is a state of mind

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I have always used 4K7.

John Samperi

Ampertronics Pty. Ltd.

www.ampertronics.com.au

* Electronic Design * Custom Products * Contract Assembly

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

js wrote:
I have always used 4K7.

Which, according to the datasheet, may at least prevent debugWIRE from working properly.

It's not clear whether the minimum of 10k applies even if you never want to use debugWiRE. It's only mentioned in the debugWIRE chapter.

But why not just use 10k anyway ?

Sid

Life... is a state of mind

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Quote:
may at least prevent debugWIRE from working properly.
I have used DW since when it came out without problems, in fact there was some research done by somebody here and 4K7 was better with some of the original Dragons.
Quote:
why not just use 10k anyway
Because I have used 4K7 in the reset line for about 20 years with other processors too not just AVR. The lower the reset line resistor the better the noise immunity as long as it doesn't prevent ISP or DW.
And I am a stubborn old guy. :lol:

John Samperi

Ampertronics Pty. Ltd.

www.ampertronics.com.au

* Electronic Design * Custom Products * Contract Assembly

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

John,

I'm quite new to designing electronics, so I'll take your word for it.

How about the decoupling cap ? The data sheet states (still in the debugWIRE section) that "Capacitors connected to the RESET pin must be disconnected when using debugWire." Is that true, in your experience ? It seems like a bit of a pain to have to unsolder something just to run the debugger...

Also, for reasons I can't remember I'm using a 100nF decoupling cap on the reset pin and it seems to work. But the AVR040 note says to use 4.7nF. Does it matter ? What do you use ?

Sid

Life... is a state of mind

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 1

Yes the cap on the reset line MUST be removed for DW to work as the reset line becomes the data line.

I have always used 100nF for bypass and on the reset line. It is really not necessary for reset purposes if one enables the BOD but very usefull if the board is working in electrically noisy environments.

Somebody worked out a long time ago that a 100nF cap was required across the supply pins of TTL chips, it seems to still work well and ol' guys don't like change. :wink:

John Samperi

Ampertronics Pty. Ltd.

www.ampertronics.com.au

* Electronic Design * Custom Products * Contract Assembly

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 1

There is no real "optimum".

The value can range from infinite (none used) to the minimum that will work with a programmer. This minimum appears to be in the range of 4.7K to 10K, depending on the programmer (and, maybe the power supply voltage).

So, if there is no optimum, why would you even use one?

The primary reason is probably when there is a long reset line that is capable of picking up external signals (radio transmitters, industrial equipment, heavy duty motors, switched high currents of all kinds, etc). The resistor helps to hold the reset pin more firmly in its operating logic level. This is particularly important since many AVRs use the Reset to enter the High Voltage Programming mode.

So, if your programming connector is less than a few cm from the MCU, the reset trace is short and the ability to pick up external stray signals is quite small. If it is longer, a resistor will help but, actually, a capacitor will help a lot more.

Unfortunately, a capacitor kills DebugWire and other single-wire debugging that depends on reset pin for serial debugging data.

Thus, anything you add is a compromise. And, there is no real optimum.

Jim

Jim Wagner Oregon Research Electronics, Consulting Div. Tangent, OR, USA http://www.orelectronics.net