Yahoo Groups archive
Thread
2008-11-18 by xolang1
Hi I have an ATmega128 utilizing the ADC to read voltage from a thin film vibration sensor that generates voltages from 0 - 70V. Im currently using a simple resitor voltage divider to scale down this 0-70V to 0-5V. To protect the ADC pin, i have a 5V Zener across the ADC input. Question: is this set-up sufficiently protected? any suggestions? Thanks in advance for the help. chris
2008-11-18 by Jim Wagner
On Nov 17, 2008, at 9:19 PM, xolang1 wrote: > Hi > > I have an ATmega128 utilizing the ADC to read voltage from a > thin film vibration sensor that generates voltages from 0 - 70V. > > Im currently using a simple resitor voltage divider to scale down > this 0-70V to 0-5V. To protect the ADC pin, i have a 5V Zener > across the ADC input. > > Question: is this set-up sufficiently protected? > any suggestions? > > Thanks in advance for the help. > > chris > > > A zener is not a particularly good choice. It is too "soft" at its breakdown. Further, if it is really rated at 5.0V, it will have some tolerance (probably 10% for an ordinary zener) and this could make the breakdown as low as 4.5V. I would use a small reverse-biased Schottky diode from the ADC input to Vcc. Let the source resistance of the divider limit the current. This will keep the input from exceeding the max of Vcc+0.3V that is a common spec. Jim Wagner Oregon Research Electronics [Non-text portions of this message have been removed]
2008-11-18 by Dao Viet Dung
All of the ATmega series themself have protected diode inside. So you should find the other way to protect it. --- On Tue, 11/18/08, Jim Wagner <wagnerj@proaxis.com> wrote:
From: Jim Wagner <wagnerj@proaxis.com>
Subject: Re: [AVR-Chat] Protecting the ADC
To: AVR-Chat@yahoogroups.com
Date: Tuesday, November 18, 2008, 12:30 PM
On Nov 17, 2008, at 9:19 PM, xolang1 wrote:
> Hi
>
> I have an ATmega128 utilizing the ADC to read voltage from a
> thin film vibration sensor that generates voltages from 0 - 70V.
>
> Im currently using a simple resitor voltage divider to scale down
> this 0-70V to 0-5V. To protect the ADC pin, i have a 5V Zener
> across the ADC input.
>
> Question: is this set-up sufficiently protected?
> any suggestions?
>
> Thanks in advance for the help.
>
> chris
>
>
>
A zener is not a particularly good choice. It is too "soft" at its
breakdown. Further, if it is really rated at 5.0V, it will have some
tolerance (probably 10% for an ordinary zener) and this could make the
breakdown as low as 4.5V. I would use a small reverse-biased Schottky
diode from the ADC input to Vcc. Let the source resistance of the
divider limit the current. This will keep the input from exceeding the
max of Vcc+0.3V that is a common spec.
Jim Wagner
Oregon Research Electronics
[Non-text portions of this message have been removed]
[Non-text portions of this message have been removed]2008-11-18 by xolang1
Appreciate the help guys. that was informative. the datasheet indeed says the ADC's are internally diode protected. so then it seems the current circuit would suffice. thanks! --- On Tue, 11/18/08, Dao Viet Dung <vietdung79@yahoo.com> wrote:
From: Dao Viet Dung <vietdung79@yahoo.com>
Subject: Re: [AVR-Chat] Protecting the ADC
To: AVR-Chat@yahoogroups.com
Date: Tuesday, November 18, 2008, 4:17 PM
All of the ATmega series themself have protected diode inside. So you should find the other way to protect it.
--- On Tue, 11/18/08, Jim Wagner <wagnerj@proaxis. com> wrote:
From: Jim Wagner <wagnerj@proaxis. com>
Subject: Re: [AVR-Chat] Protecting the ADC
To: AVR-Chat@yahoogroup s.com
Date: Tuesday, November 18, 2008, 12:30 PM
On Nov 17, 2008, at 9:19 PM, xolang1 wrote:
> Hi
>
> I have an ATmega128 utilizing the ADC to read voltage from a
> thin film vibration sensor that generates voltages from 0 - 70V.
>
> Im currently using a simple resitor voltage divider to scale down
> this 0-70V to 0-5V. To protect the ADC pin, i have a 5V Zener
> across the ADC input.
>
> Question: is this set-up sufficiently protected?
> any suggestions?
>
> Thanks in advance for the help.
>
> chris
>
>
>
A zener is not a particularly good choice. It is too "soft" at its
breakdown. Further, if it is really rated at 5.0V, it will have some
tolerance (probably 10% for an ordinary zener) and this could make the
breakdown as low as 4.5V. I would use a small reverse-biased Schottky
diode from the ADC input to Vcc. Let the source resistance of the
divider limit the current. This will keep the input from exceeding the
max of Vcc+0.3V that is a common spec.
Jim Wagner
Oregon Research Electronics
[Non-text portions of this message have been removed]
[Non-text portions of this message have been removed]
[Non-text portions of this message have been removed]2008-11-18 by Kathy Quinlan
NO, the internal diodes are not designed for this level of protection, depending on your external resistor values, you could exceed the maximum sink current of the device. An external reverse-biased Schottky will provide the extra protection needed. I believe in belts and braces, I would rather have a device in the field that does not lock up than a device that I saved a cent or two on in manufacturing. Regards, Kat. xolang1 wrote:
> > Appreciate the help guys. that was informative. > > the datasheet indeed says the ADC's are internally diode protected. > so then it seems the current circuit would suffice. > > thanks! > > --- On Tue, 11/18/08, Dao Viet Dung <vietdung79@yahoo.com > <mailto:vietdung79%40yahoo.com>> wrote: > > From: Dao Viet Dung <vietdung79@yahoo.com <mailto:vietdung79%40yahoo.com>> > Subject: Re: [AVR-Chat] Protecting the ADC > To: AVR-Chat@yahoogroups.com <mailto:AVR-Chat%40yahoogroups.com> > Date: Tuesday, November 18, 2008, 4:17 PM > > All of the ATmega series themself have protected diode inside. So you > should find the other way to protect it. > > --- On Tue, 11/18/08, Jim Wagner <wagnerj@proaxis. com> wrote: > From: Jim Wagner <wagnerj@proaxis. com> > Subject: Re: [AVR-Chat] Protecting the ADC > To: AVR-Chat@yahoogroup s.com > Date: Tuesday, November 18, 2008, 12:30 PM > > On Nov 17, 2008, at 9:19 PM, xolang1 wrote: > > > Hi > > > > > > I have an ATmega128 utilizing the ADC to read voltage from a > > > thin film vibration sensor that generates voltages from 0 - 70V. > > > > > > Im currently using a simple resitor voltage divider to scale down > > > this 0-70V to 0-5V. To protect the ADC pin, i have a 5V Zener > > > across the ADC input. > > > > > > Question: is this set-up sufficiently protected? > > > any suggestions? > > > > > > Thanks in advance for the help. > > > > > > chris > > > > > > > > > > > A zener is not a particularly good choice. It is too "soft" at its > > breakdown. Further, if it is really rated at 5.0V, it will have some > > tolerance (probably 10% for an ordinary zener) and this could make the > > breakdown as low as 4.5V. I would use a small reverse-biased Schottky > > diode from the ADC input to Vcc. Let the source resistance of the > > divider limit the current. This will keep the input from exceeding the > > max of Vcc+0.3V that is a common spec. > > Jim Wagner > > Oregon Research Electronics > > [Non-text portions of this message have been removed] > > [Non-text portions of this message have been removed] > > [Non-text portions of this message have been removed] > > > ------------------------------------------------------------------------ > > > No virus found in this incoming message. > Checked by AVG - http://www.avg.com > Version: 8.0.175 / Virus Database: 270.9.4/1795 - Release Date: 17/11/2008 5:24 PM > >
2008-11-18 by xolang1
the source resistance should be around 50-75k which should be sufficient for current limiting. for over voltage protection, reverse Ge diodes (0.3V) will be placed from adc to vcc and gnd to adc. any thing missed or inappropriate? --- On Tue, 11/18/08, Kathy Quinlan <kaqdialup@iinet.net.au> wrote:
> From: Kathy Quinlan <kaqdialup@iinet.net.au> > Subject: Re: [AVR-Chat] Protecting the ADC > To: AVR-Chat@yahoogroups.com > Date: Tuesday, November 18, 2008, 8:11 PM > NO, the internal diodes are not designed for this level of > protection, > depending on your external resistor values, you could > exceed the maximum > sink current of the device. > > An external reverse-biased Schottky will provide the extra > protection > needed. > > I believe in belts and braces, I would rather have a device > in the field > that does not lock up than a device that I saved a cent or > two on in > manufacturing. > > Regards, > > Kat. > > xolang1 wrote: > > > > Appreciate the help guys. that was informative. > > > > the datasheet indeed says the ADC's are internally > diode protected. > > so then it seems the current circuit would suffice. > > > > thanks! > > > > --- On Tue, 11/18/08, Dao Viet Dung > <vietdung79@yahoo.com > > <mailto:vietdung79%40yahoo.com>> wrote: > > > > From: Dao Viet Dung <vietdung79@yahoo.com > <mailto:vietdung79%40yahoo.com>> > > Subject: Re: [AVR-Chat] Protecting the ADC > > To: AVR-Chat@yahoogroups.com > <mailto:AVR-Chat%40yahoogroups.com> > > Date: Tuesday, November 18, 2008, 4:17 PM > > > > All of the ATmega series themself have protected diode > inside. So you > > should find the other way to protect it. > > > > --- On Tue, 11/18/08, Jim Wagner <wagnerj@proaxis. > com> wrote: > > From: Jim Wagner <wagnerj@proaxis. com> > > Subject: Re: [AVR-Chat] Protecting the ADC > > To: AVR-Chat@yahoogroup s.com > > Date: Tuesday, November 18, 2008, 12:30 PM > > > > On Nov 17, 2008, at 9:19 PM, xolang1 wrote: > > > > > Hi > > > > > > > > > > I have an ATmega128 utilizing the ADC to read > voltage from a > > > > > thin film vibration sensor that generates > voltages from 0 - 70V. > > > > > > > > > > Im currently using a simple resitor voltage > divider to scale down > > > > > this 0-70V to 0-5V. To protect the ADC pin, i > have a 5V Zener > > > > > across the ADC input. > > > > > > > > > > Question: is this set-up sufficiently protected? > > > > > any suggestions? > > > > > > > > > > Thanks in advance for the help. > > > > > > > > > > chris > > > > > > > > > > > > > > > > > > > A zener is not a particularly good choice. It is too > "soft" at its > > > > breakdown. Further, if it is really rated at 5.0V, it > will have some > > > > tolerance (probably 10% for an ordinary zener) and > this could make the > > > > breakdown as low as 4.5V. I would use a small > reverse-biased Schottky > > > > diode from the ADC input to Vcc. Let the source > resistance of the > > > > divider limit the current. This will keep the input > from exceeding the > > > > max of Vcc+0.3V that is a common spec. > > > > Jim Wagner > > > > Oregon Research Electronics > > > > [Non-text portions of this message have been removed] > > > > [Non-text portions of this message have been removed] > > > > [Non-text portions of this message have been removed] > > > > > > > ------------------------------------------------------------------------ > > > > > > No virus found in this incoming message. > > Checked by AVG - http://www.avg.com > > Version: 8.0.175 / Virus Database: 270.9.4/1795 - > Release Date: 17/11/2008 5:24 PM > > > >
2008-11-18 by Kathy Quinlan
By doing it that way you can also protect from reverse spikes, which is a great idea. xolang1 wrote:
> > > > the source resistance should be around 50-75k > which should be sufficient for current limiting. > for over voltage protection, reverse Ge diodes (0.3V) will be > placed from adc to vcc and gnd to adc. > any thing missed or inappropriate? > > --- On Tue, 11/18/08, Kathy Quinlan <kaqdialup@iinet.net.au > <mailto:kaqdialup%40iinet.net.au>> wrote: > > > From: Kathy Quinlan <kaqdialup@iinet.net.au > <mailto:kaqdialup%40iinet.net.au>> > > Subject: Re: [AVR-Chat] Protecting the ADC > > To: AVR-Chat@yahoogroups.com <mailto:AVR-Chat%40yahoogroups.com> > > Date: Tuesday, November 18, 2008, 8:11 PM > > NO, the internal diodes are not designed for this level of > > protection, > > depending on your external resistor values, you could > > exceed the maximum > > sink current of the device. > > > > An external reverse-biased Schottky will provide the extra > > protection > > needed. > > > > I believe in belts and braces, I would rather have a device > > in the field > > that does not lock up than a device that I saved a cent or > > two on in > > manufacturing. > > > > Regards, > > > > Kat. > > > > xolang1 wrote: > > > > > > Appreciate the help guys. that was informative. > > > > > > the datasheet indeed says the ADC's are internally > > diode protected. > > > so then it seems the current circuit would suffice. > > > > > > thanks! > > > > > > --- On Tue, 11/18/08, Dao Viet Dung > > <vietdung79@yahoo.com <mailto:vietdung79%40yahoo.com> > > > <mailto:vietdung79%40yahoo.com>> wrote: > > > > > > From: Dao Viet Dung <vietdung79@yahoo.com > <mailto:vietdung79%40yahoo.com> > > <mailto:vietdung79%40yahoo.com>> > > > Subject: Re: [AVR-Chat] Protecting the ADC > > > To: AVR-Chat@yahoogroups.com <mailto:AVR-Chat%40yahoogroups.com> > > <mailto:AVR-Chat%40yahoogroups.com> > > > Date: Tuesday, November 18, 2008, 4:17 PM > > > > > > All of the ATmega series themself have protected diode > > inside. So you > > > should find the other way to protect it. > > > > > > --- On Tue, 11/18/08, Jim Wagner <wagnerj@proaxis. > > com> wrote: > > > From: Jim Wagner <wagnerj@proaxis. com> > > > Subject: Re: [AVR-Chat] Protecting the ADC > > > To: AVR-Chat@yahoogroup s.com > > > Date: Tuesday, November 18, 2008, 12:30 PM > > > > > > On Nov 17, 2008, at 9:19 PM, xolang1 wrote: > > > > > > > Hi > > > > > > > > > > > > > > I have an ATmega128 utilizing the ADC to read > > voltage from a > > > > > > > thin film vibration sensor that generates > > voltages from 0 - 70V. > > > > > > > > > > > > > > Im currently using a simple resitor voltage > > divider to scale down > > > > > > > this 0-70V to 0-5V. To protect the ADC pin, i > > have a 5V Zener > > > > > > > across the ADC input. > > > > > > > > > > > > > > Question: is this set-up sufficiently protected? > > > > > > > any suggestions? > > > > > > > > > > > > > > Thanks in advance for the help. > > > > > > > > > > > > > > chris > > > > > > > > > > > > > > > > > > > > > > > > > > > A zener is not a particularly good choice. It is too > > "soft" at its > > > > > > breakdown. Further, if it is really rated at 5.0V, it > > will have some > > > > > > tolerance (probably 10% for an ordinary zener) and > > this could make the > > > > > > breakdown as low as 4.5V. I would use a small > > reverse-biased Schottky > > > > > > diode from the ADC input to Vcc. Let the source > > resistance of the > > > > > > divider limit the current. This will keep the input > > from exceeding the > > > > > > max of Vcc+0.3V that is a common spec. > > > > > > Jim Wagner > > > > > > Oregon Research Electronics > > > > > > [Non-text portions of this message have been removed] > > > > > > [Non-text portions of this message have been removed] > > > > > > [Non-text portions of this message have been removed] > > > > > > > > > > > ---------------------------------------------------------- > > > > > > > > > No virus found in this incoming message. > > > Checked by AVG - http://www.avg.com <http://www.avg.com> > > > Version: 8.0.175 / Virus Database: 270.9.4/1795 - > > Release Date: 17/11/2008 5:24 PM > > > > > > > > > ------------------------------------------------------------------------ > > > No virus found in this incoming message. > Checked by AVG - http://www.avg.com > Version: 8.0.175 / Virus Database: 270.9.4/1795 - Release Date: 17/11/2008 5:24 PM > >
2008-11-18 by Zack Widup
A couple questions: Are the resistors in the divider and the zener surface-mount devices or leaded devices? And what current is the vibration sensor capable of producing at 70 volts? Chip components have a tendency to develop minute cracks when under a lot of vibration. I've had that happen before where the cracked part was easily seen but I had it happen recently where I couldn't tell which part was bad and just about went around the bend trying to find it. Leaded parts are less likely to crack under vibration. Also, from my experience with zeners going bad, 90% of the time they become short-circuits when they go bad. If there's enough current behind the source to blow out the shorted zener, you're in trouble. Other than those two areas, I'd say your idea itself would provide adequate protection. Zack
On Mon, 17 Nov 2008, xolang1 wrote: > Hi > > I have an ATmega128 utilizing the ADC to read voltage from a > thin film vibration sensor that generates voltages from 0 - 70V. > > Im currently using a simple resitor voltage divider to scale down > this 0-70V to 0-5V. To protect the ADC pin, i have a 5V Zener > across the ADC input. > > Question: is this set-up sufficiently protected? > any suggestions? > > Thanks in advance for the help. > > chris > > > >
2008-11-18 by David VanHorn
On Tue, Nov 18, 2008 at 7:11 AM, Kathy Quinlan <kaqdialup@iinet.net.au> wrote: > NO, the internal diodes are not designed for this level of protection, > depending on your external resistor values, you could exceed the maximum > sink current of the device. Definitely correct. Although the internal protection diodes exist, you are never supposed to use them. A series resistor to limit current, schottky diodes to VCC and ground, and another resistor into the ADC input pin is about the minimum I'd use. For real protection, a buffer op-amp would be better. A lot depends on how much current this device can source, but I suspect it's pretty low. I did a battery management system recently that had to work with 6V - 48V input of any polarity, AND be "taser-proof" on the input leads. That was fun.
2008-11-18 by David Kelly
On Tue, Nov 18, 2008 at 09:38:50AM -0500, David VanHorn wrote: > On Tue, Nov 18, 2008 at 7:11 AM, Kathy Quinlan > <kaqdialup@iinet.net.au> wrote: > > NO, the internal diodes are not designed for this level of > > protection, depending on your external resistor values, you could > > exceed the maximum sink current of the device. > > Definitely correct. Although the internal protection diodes exist, > you are never supposed to use them. People these days tend to forget The Bad Old Days when most every IC was super sensitive to static discharge during handling. They are still sensitive but not anything nearly as bad as they once were when one could almost zap gates with ESP (Extra-Sensory Perception, just by thinking). There is/was an Atmel appnote on building an electric power meter where 240 VAC was connected directly to an input pin through a 1 or 10 megohm resistor, stating the protection diode was safe up to 1 mA. And then the appnote went to lengths disavowing the safety ... -- David Kelly N4HHE, dkelly@HiWAAY.net ======================================================================== Whom computers would destroy, they must first drive mad.
2008-11-18 by David VanHorn
> People these days tend to forget The Bad Old Days when most every IC was > super sensitive to static discharge during handling. They are still > sensitive but not anything nearly as bad as they once were when one > could almost zap gates with ESP (Extra-Sensory Perception, just by > thinking). Yes.. Many of the little details are getting lost. The interesting failure mode of most three terminal regulators, how to properly select a crystal and it's caps for a micro, proper PCB layout.. > There is/was an Atmel appnote on building an electric power meter where > 240 VAC was connected directly to an input pin through a 1 or 10 megohm > resistor, stating the protection diode was safe up to 1 mA. And then the > appnote went to lengths disavowing the safety ... YIKES! I suppose if the circuit were entirely within a non-conductive enclosure, and the user had NO way to touch it at all... Engineering IS the art of compromise, but sometimes you do wonder what the hell they were thinking.