Shop OBEX P1 Docs P2 Docs Learn Events
Propeller 1 voltages — Parallax Forums

Propeller 1 voltages

Buck RogersBuck Rogers Posts: 2,185
edited 2015-01-29 20:49 in Propeller 1
Hello!
Would someone with slightly more knowledge of the Prop please correct or confirm this assertion? The Prop 1 is a 3.3v device. To be used with 5v TTL designed logic it requires an appropriate set of translators, but to be used with CMOS logic such as the CD4000 series it does not.

I have here a regular Prop 1, which is the P8X32A Propeller Microcontroller - 40-pin DIP package Propeller microcontroller, and the majority of the materials contained within this kit, Propeller DIP Plus Kit which is 130-32305 and everything is located at http://www.parallax.com/product/130-32305 .

As for what I am planning, let's just say it can be called an interesting project.

Comments

  • Peter JakackiPeter Jakacki Posts: 10,193
    edited 2015-01-28 21:17
    Hello!
    Would someone with slightly more knowledge of the Prop please correct or confirm this assertion? The Prop 1 is a 3.3v device. To be used with 5v TTL designed logic it requires an appropriate set of translators, but to be used with CMOS logic such as the CD4000 series it does not.

    I have here a regular Prop 1, which is the P8X32A Propeller Microcontroller - 40-pin DIP package Propeller microcontroller, and the majority of the materials contained within this kit, Propeller DIP Plus Kit which is 130-32305 and everything is located at http://www.parallax.com/product/130-32305 .

    As for what I am planning, let's just say it can be called an interesting project.

    Welcome to the 21st century Buck! Only 4 more to go!

    CD4000 series was really meant to be operated at 12V so just saying CD4000 series does not make it compatible, besides why would you use them? To translate from the Prop's 3.3V to 5V I use 74HCT devices powered from 5V but they accept anything above 2V as a logic high so they work well as translators. In fact I tend to use single gate devices and so I can choose to invert or not I just use 74HCT86 exclusive-0R and tie the other input high or low. So these are not 74HC but 74HCT to indicate the TTL 2V high compatibility.

    As for 5V to the 3.3V input of the Prop all that is needed is current limiting in the form of a series resistor, I normally use 10K and although some may use much lower values there is no need to and we need to keep the current limit to an acceptable level, especially if you have quite a few inputs all sitting at 5V.
  • evanhevanh Posts: 15,921
    edited 2015-01-28 22:20
    Peter, you're not the first person to make that mistake. I guess it was spread by word of mouth. The 4k series was, unlike every other logic type out there, spec'd for a wide range of 3-15 volts. However, at the 3 volt end of the range it was something like 5 times slower than the 15 volt end. Presumably a linear relationship. But I guess that's true for all CMOSes. They're just not spec'd.
  • Duane DegnDuane Degn Posts: 10,588
    edited 2015-01-28 22:59
    As Peter said, a series resistor is all that is needed when receiving input from a 5V device. When sending data to a 5V device the 3.3V logic is usually enough to register as logic high.

    The series resistor may be as low as 3K but I generally use 10K like Peter. One reason to use a lower resistor value would be if the signal is bidirectional. While I haven't found many 5V devices which won't properly interpret 3.3V logic, I have had trouble when the 5V device is linked through a resistor.

    So when do we get to find out what this "interesting project" is?
  • Peter JakackiPeter Jakacki Posts: 10,193
    edited 2015-01-28 23:11
    evanh wrote: »
    Peter, you're not the first person to make that mistake. I guess it was spread by word of mouth. The 4k series was, unlike every other logic type out there, spec'd for a wide range of 3-15 volts. However, at the 3 volt end of the range it was something like 5 times slower than the 15 volt end. Presumably a linear relationship. But I guess that's true for all CMOSes. They're just not spec'd.

    Oh, I do know that it will work down to 3V and it is very slow even at 12V which is the voltage that was used for a lot of early industrial control for reasons of noise immunity. So, no mistake, just not worrying about getting into all that CD4000 trivia and nonsense, it is so 70's!

    @Duane: BTW, 3.3V may work back into a 5V device but the signal is right on the threshold unless the 5V device is TTL compatible. I know the Stamp has TTL inputs available. The minimum high voltage for standard 5V CMOS is 3.6V and one easy way to achieve a logic high without a translator is either to bump up the Prop's supply to 3.6V or else if possible insert a diode in the 5V supply so that the 5V stuff ended up working at 4.4V and so the high threshold was dropped to less than 3.3V. A pullup to 5V will also work if you drive the pin as "open drain" so that the output ends up in current limit mode with the pin sitting at around 3.8V. (Note: you can never achieve true open-drain but at least you are not trying to drive it at 3.3V, just letting it float up as far as it will go)
  • evanhevanh Posts: 15,921
    edited 2015-01-29 01:35
    It's no more 70's than dropping in a bunch of diodes or running components out of spec. If one isn't needing top speed and can do without the ground bounce then the 4k parts are just fine.
  • Buck RogersBuck Rogers Posts: 2,185
    edited 2015-01-29 16:47
    Duane Degn wrote: »
    As Peter said, a series resistor is all that is needed when receiving input from a 5V device. When sending data to a 5V device the 3.3V logic is usually enough to register as logic high.

    The series resistor may be as low as 3K but I generally use 10K like Peter. One reason to use a lower resistor value would be if the signal is bidirectional. While I haven't found many 5V devices which won't properly interpret 3.3V logic, I have had trouble when the 5V device is linked through a resistor.

    So when do we get to find out what this "interesting project" is?

    Hello!
    Okay Duane yes that makes much better sense.

    As for
    So when do we get to find out what this "interesting project" is?
    I am still working on the details behind it.

    My other problem is that the kit supplied (via the website) materials for having a Prop 1 talk to a Labview example. I've never been good at figuring out the "whys and wherefores" of how Labview manages itself, and even communicates with the outside world. It happens I successfully taught myself a GUI based test and measurement application called VEE that was originally invented by HP and is now managed by their inheritors. So that's the basis of part of my interesting project.
  • tonyp12tonyp12 Posts: 1,951
    edited 2015-01-29 19:23
    The Prop is a cmos 3.3V device, its input are Schmitt triggered (with no hysteresis) at around 1.65V.

    How to interface to a 5V devices, if the device is the input get the HCT version it needs to be powered 4.5-5V and a 2.5V on it input is considered high so props 3.3v is OK.

    The ACH version, can be powered from 3v to 6V but its input as high needs to be at least 70% off its Vcc, so if at 5V a Props 3.3v is not enough, use a 10K pull-up on its input pin and Prop will only sink when you want a low state.

    Get a 74LVC1G17W5 buffer, even if powered at 5V a 3V on its input is considered high, great to use if the Vcc is not predetermined and could be 3.3V or 5V (as in breadboard applications)
    http://www.mouser.com/ProductDetail/Diodes-Incorporated/74LVC1G17W5-7/?qs=%2fha2pyFadui5%2fGGEC%2fVyhsh8Cj7I8hsRkyIiiIDXfKltPkDseSRDxw%3d%3d
  • Peter JakackiPeter Jakacki Posts: 10,193
    edited 2015-01-29 19:42
    tonyp12 wrote: »
    The Prop is a cmos 3.3V device, its input are Schmitt triggered (with no hysteresis) at around 1.65V.

    How to interface to a 5V devices, if the device is the input get the HCT version it needs to be powered 4.5-5V and a 2.5V on it input is considered high so props 3.3v is OK.

    The ACH version, can be powered from 3v to 6V but its input as high needs to be at least 70% off its Vcc, so if at 5V a Props 3.3v is not enough, use a 10K pull-up on its input pin and Prop will only sink when you want a low state.

    Get a 74LVC1G17W5 buffer, even if powered at 5V a 2.5V on its input is considered high, great to use if the Vcc is not predetermined and could be 3.3V or 5V (as in breadboard applications)
    http://www.mouser.com/ProductDetail/Diodes-Incorporated/74LVC1G17W5-7/?qs=%2fha2pyFadui5%2fGGEC%2fVyhsh8Cj7I8hsRkyIiiIDXfKltPkDseSRDxw%3d%3d

    @Tony, just a technical note as Schmitt triggers always have hysteresis in that once they trip a threshold then that threshold changes which prevents jitter around the otherwise fixed threshold. So the Prop does not have Schmitt trigger inputs as the threshold is fixed.
  • tonyp12tonyp12 Posts: 1,951
    edited 2015-01-29 20:13
    Maybe Schmitt is not the right word, but Comparator-Trigged inputs as a pretty hard trigger at 1.65V is not the standard 70% of Vcc that cmos have
  • evanhevanh Posts: 15,921
    edited 2015-01-29 20:49
    I haven't actually tested this but I think that 70% is just the minimum voltage at whence they guaranteed a high input is treated as high, ie: The comparator reference will never go higher than that. There will be an equivalent, 30% presumably, for the maximum voltage of a low as well. In reality, generic CMOS is right at the 50% mark.

    The Prop1 makes good use of this fine switching 50% level for it's 1-bit ADC circuit. The Prop2 is going to have a programmable switching level - for each input individually I think.
Sign In or Register to comment.