Voltagelevels that are SURELY detected as High / Low

StefanL38

hello,

Ok - as the answers to my last question are not fully satisfying me a new question:

What are the voltage-levels that are 100% SURE detected as logic HIGH or logic LOW
I guess with a powersupply of 3,3V
3,3V / 2 = 1,85V will be detected unsecure sometimes as low sometimes as high

So what is the range of a 100% surely logic LOW 0V - 1V ?
what is the range of a 100% surely logic HIGH 2,3V - 3,3V ?

I would appreciate if somebody could answer with NUMBERS like in the two lines above.

best regards

Stefan

Leon

It's impossible to say, it depends on how much noise is present.

Leon

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Amateur radio callsign: G1HSM
Suzuki SV1000S motorcycle

Ken Peterson

Stefan: The transition point would depend on the exact value of Vdd. It's almost exactly half of Vdd. Like Leon said, the range depends on noise.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔

StefanL38

yes i'm aware of that it depends on Vdd as i mentioned myself "powersupply with 3,3V"
saying Vdd is 3,3V and not 3,0V or 2,9V.

Anayway, so i'm gonna ask another question:

did anybody ever have a prop-application in a noisy environment with classical switches ?

If yes how did you connect the switches in YOUR application ?

This is NOT longer a question about "tell me a 150% supersaveway how to connect switches to an IO-Pin that works in every superspecial case"

It's simply a question "how have YOU - REALIZED the connection in ONE of your applications ?


best regards

Stefan

Leon

What sort of noisy environment? Do you mean electro-magnetic effects or noisy supply lines? The Propeller isn't any different from any other MCU in those respects. Careful attention to circuit design, PCB layout, filtering, transient suppression and screening are the usual solutions; you can't just consider one design aspect in isolation.

Leon

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Amateur radio callsign: G1HSM
Suzuki SV1000S motorcycle

StefanL38

hello leon,

that's NOT an answer to my question !

I'm NOT planning an application in a supernoisy environment.
I was asking: could somebody describe the switch-connection to the prop in ONE realized application
COMPLETELY regardless of an eventually specialized situation.

Just describe it.

And if you can only answer with a detailed specification:
a noiselevel averaged about 1 Million houses for a living-purpose like you find it on a livingroom-table
with a modern TV, somebody talking on a mobilephone 1m away

Or if you only want to make suggestion only for very specific numbers
Prop a LC-Display connected and some LEDs and that's all
specify a noiselevel about electromagnetic noise, noisy supplylines or whatever you like
and post a circuitry

best regards

Stefan

Paul Baker

The Propeller's PMOS and NMOS transitors are not completely matched, the NMOS is a little stronger than the PMOS (some model parameters weren't included in the spice transistor model which lead to the discrepancy of the simulated MOSFET strength vs the actual MOSFET strength). Because of this the input threshold is a little lower (around 1.6V). The input signal should not be allowed to "hang around" this voltage unless this is specifically your intent (such as the sigma-delta ADC), the logic state seen by the Propeller will randomly fluctuate between the two values. Also because of process variation a margin around this value should be observed, a ±150mV margin should be good in fairly low-noise environments. For higher noise environments this margin should be increased by the maxium instantaneous voltage due to noise. So if a 500mV peak to peak noise is present on the line, the margin should be ±400mV or greater.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Paul Baker
Propeller Applications Engineer

Parallax, Inc.

Post Edited (Paul Baker (Parallax)) : 7/3/2008 5:48:53 PM GMT

Leon

StefanL38 said...
hello leon,

that's NOT an answer to my question !

I'm NOT planning an application in a supernoisy environment.
I was asking: could somebody describe the switch-connection to the prop in ONE realized application
COMPLETELY regardless of an eventually specialized situation.

Just describe it.

And if you can only answer with a detailed specification:
a noiselevel averaged about 1 Million houses for a living-purpose like you find it on a livingroom-table
with a modern TV, somebody talking on a mobilephone 1m away

Or if you only want to make suggestion only for very specific numbers
Prop a LC-Display connected and some LEDs and that's all
specify a noiselevel about electromagnetic noise, noisy supplylines or whatever you like
and post a circuitry

best regards

Stefan

The CE requirements specify the EMC requirements that all equipment sold in the EU is required to meet. Here is a list of the relevant standards:

http://www.rnelectronics.com/resources/latest-standards.html

A simple pull-up resistor of 10k with short PCB tracks will work OK. I've never had any problems with it.

Leon

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Amateur radio callsign: G1HSM
Suzuki SV1000S motorcycle

Post Edited (Leon) : 7/3/2008 6:15:57 PM GMT

stevenmess2004

From the data sheet
Vlow max=0.3Vdd=1V
Vhigh min=0.6Vdd=2V

There was a case a while ago where someone was feeding in a fairly low frequency sine wave and having problems measuring the frequency of it using the counters because it was too long in the in-between region. So you do need to be a little careful.

You can actually make a nice switch debouncer out of a flip-flop but thats probably more than what you need.