Shop OBEX P1 Docs P2 Docs Learn Events
easyest way to use a prop based logic analyzer on a 5v system? — Parallax Forums

easyest way to use a prop based logic analyzer on a 5v system?

rwgast_logicdesignrwgast_logicdesign Posts: 1,464
edited 2012-09-23 23:56 in Propeller 1
so a while back i built a board spefically to use as a standalone logic analyzer running propalyzer. recently ive been doing alot mre experements with the bs2 than the prop since its so quick to band code out

i want to use my LA board to read bs2 signals im wondering what the best option is.. i have some of the octal buffers / line drivers parallax sells 74sn244 i belive. savage circuitz shows multiple ways to connect a 5v system to the prop, i also know alot of people just use a 10k resistors between the bs2 and the prop which is what im hoping will work.

so my question is what the simplest way to use propalyzer with the bs2? im just wondering if all the regular methods will work or if the low output of the bs2 will register as a hi with the prop?

Comments

  • Mark_TMark_T Posts: 1,981
    edited 2012-09-20 15:50
    A low output from any CMOS chip will be close to 0V, similarly a high-output will be close to Vdd. The 10k resistors are the simplest solution, but will affect bandwidth somewhat. A resistor divider using 1k and 2k2 resistors will give good bandwidth, but will interfere with any signals that have pull-ups.

    You could pull up Prop inputs to 3V3 with 2k2's or 4k7's or so and feed the 5V inputs via 1N4148's so the circuit-under-test can only pull-down - this has the advantage of working with any voltage from 3V3 up (24V PLCs perhaps?) and working directly with open-drain or open-collector outputs. And 1N4148's are cheap and fast.
  • rwgast_logicdesignrwgast_logicdesign Posts: 1,464
    edited 2012-09-20 16:09
    hmm ill look around need to do this now i know i have 1n4001s is there voltage drop to hi or somethin? how about the line drivers the la has 100mhz bandwith will a buffer effect that?
  • Phil Pilgrim (PhiPi)Phil Pilgrim (PhiPi) Posts: 23,514
    edited 2012-09-20 16:13
    Use 74LVC-series logic buffers on your inputs.They are fast, run from 3.3V, and the inputs are 5V-tolerant.

    -Phil
  • kwinnkwinn Posts: 8,697
    edited 2012-09-20 19:29
    rwgast_logicdesign wrote: »
    hmm ill look around need to do this now i know i have 1n4001s is there voltage drop to hi or somethin? how about the line drivers the la has 100mhz bandwith will a buffer effect that?

    1N4001's are not very good for this application. They are power rectifiers. You need a high speed signal diode like the 1N4148 or 1N914.
  • rwgast_logicdesignrwgast_logicdesign Posts: 1,464
    edited 2012-09-20 21:29
    ya i got curious and looked up both data sheets, after reading them i can see why the 1n4148 is a better fit due to its fast times.

    i was reading through the 74clc244 sheet i dont see anything about time or bandwidth how do you figure out how fast you can drive these chips?

    also im curious when working with 74x series whats the difference in hc, vlc, sn etc...
  • Phil Pilgrim (PhiPi)Phil Pilgrim (PhiPi) Posts: 23,514
    edited 2012-09-20 21:43
    It's LVC, rw. Here's the datasheet for the 74LVC244:

    On page 1, you will note the following:
    Supports Mixed-Mode Signal Operation on
    All Ports (5-V Input/Output Voltage With
    3.3-V VCC)

    Most logic families do not have this feature.

    Also, on page 4, note the figures for tpd (propagation delay). With Vdd = 3.3V, it ranges from 1.5 to 5.9 ns. By comparison, the system clock period for the Prop (80 MHz) is 12.5 ns. So this chip should have no trouble keeping up with your Prop-based logic analyzer.

    -Phil
  • RDL2004RDL2004 Posts: 2,554
    edited 2012-09-20 21:44
    Logic Chip Families

    7400 Series TTL Logic Chips (check the second section "7400 series derivative families")
  • frank freedmanfrank freedman Posts: 1,983
    edited 2012-09-20 22:09
    RDL2004 wrote: »
    Logic Chip Families

    7400 Series TTL Logic Chips (check the second section "7400 series derivative families")

    Download this from Texas Instruments:

    http://www.ti.com/general/docs/lit/getliterature.tsp?baseLiteratureNumber=sdyu001


    Frank
  • RDL2004RDL2004 Posts: 2,554
    edited 2012-09-21 06:34
    Download this from Texas Instruments...

    That is an excellent reference.

    I got one of these last year for 10 cents (with free shipping from TI).
    Though it's not up-to-date it's still useful.

    attachment.php?attachmentid=85246&d=13165318321
  • rwgast_logicdesignrwgast_logicdesign Posts: 1,464
    edited 2012-09-21 16:25
    Thanks for the Docs guys!

    Is the book still available for fairly cheap? I really would rather have books most of the time
  • PublisonPublison Posts: 12,366
    edited 2012-09-21 17:02
    rwgast_logicdesign wrote: »
    Thanks for the Docs guys!

    Is the book still available for fairly cheap? I really would rather have books most of the time

    When it was available from TI..cheap. Now available on amazon and ebay...not so cheap.
  • rwgast_logicdesignrwgast_logicdesign Posts: 1,464
    edited 2012-09-21 23:46
    ok so i stuck one of these on a protoboard and got it translating one line looking through the data sheet i want to make sure im write about a few things before i make all the connections

    if i drive oe1 and oe2 high that will set the input side of port1 and port2 into hi z essentially not allowing the ports to recive any data? essentialy i plan to jumper the oe pins hi when the ports not in use if this is true.

    the chip says its inputs are 0 to 5v so o can also use this to analyze 1.8 logic?

    the output ports are set to vcc so i can translat from input voltage to whatever i set vcc at with in spec? im sorda confused becuase it says vcc is 2.7 to 3.6 in one place and 1.2 to 3.6 in another
  • Phil Pilgrim (PhiPi)Phil Pilgrim (PhiPi) Posts: 23,514
    edited 2012-09-22 00:20
    rwgast_logicdesign wrote:
    if i drive oe1 and oe2 high that will set the input side of port1 and port2 into hi z essentially not allowing the ports to recive any data?
    The inputs are always high-Z. Setting the /OE's high makes the corresponding outputs high-Z. (Always be sure to include the slash / when discussing inverted inputs or outputs.) If you do this, though, the Propeller's inputs will be floating, so you might want to add pull-ups or pull-downs to these pins.
    the chip says its inputs are 0 to 5v so o can also use this to analyze 1.8 logic?
    The answer to that question is on page 3 of the datasheet that I cited above.
    the output ports are set to vcc so i can translat from input voltage to whatever i set vcc at with in spec?
    Yes.

    Now, given this information and that from the Propeller datasheet, what could you set Vcc for the 74LVC244 to in order to accommodate 1.8V to 5V logic?

    -Phil
  • rwgast_logicdesignrwgast_logicdesign Posts: 1,464
    edited 2012-09-22 22:14
    i just want you to know i havent been avoiding you homework just got home, im gonna go through the prop manual now.... wouldnt vcc be set to 3.3vs though?

    the line driver is going to accept any input voltage between 0-5 volts then shift it to 3.3. athough im not sure of the prop thresh holds and how 1.8 will be read as low thats why ima break the manual out :)
  • rwgast_logicdesignrwgast_logicdesign Posts: 1,464
    edited 2012-09-22 23:21
    Ok so in reference to the 1.8v logic question i looked through the prop manual and found

    prop VCC= 3.3 obviously
    prop Hi Logic =1.65+ or the top 1/2 of VDD this is correct?

    So for a 1.8v system .9+ would be high....

    This means a propeller can only detect a 1.8v systems hi if there inbetween 1.65 and 1.8, the prop will detect everything between .9 and 1.65 as logic low. So I guess after this I get a bit confused...

    The 74lvc244 doesnt shift any logic levels it just provides a buffer with 0-5v inputs and an out put of vcc. So im probably totlly wrong but I would think the way to go about this is to set the vcc of the line driver to 1.8v and then add pull ups on the inputs so all the high levels would be 1.8 and anything under 1.65 the propeller would detect as 0? :/ this is probabhly totally off?
  • Phil Pilgrim (PhiPi)Phil Pilgrim (PhiPi) Posts: 23,514
    edited 2012-09-23 01:09
    The high output of the buffer will be whatever its Vcc is, and the low will be zero. So if you set buffer Vcc to 1.8V, a "high" on the input of the buffer can be anywhere from 0.65 * 1.8 = 1.17V on up to 5V. If you're willing to call 1.17V "high" under all circumstances, this would work. But you could raise Vcc to 2.0V, in which case a "high" on the input of the buffer would have to be at least 1.3V.

    You could also switch-select the buffer Vcc, depending upon which logic family you're testing (up to 3.3V).

    -Phil
  • rwgast_logicdesignrwgast_logicdesign Posts: 1,464
    edited 2012-09-23 02:14
    im not understanding how you are getting the VLH on the input buffer, why are you multiplying vcc by .65 i must have missed something. everything else your saying makes perfect seanse.

    also i got these chips for 45cents there a small fairly easy to use solution so why would someone use something like this

    https://www.sparkfun.com/products/8745

    this thing is more money, bigger and requires more than one a hi and low reference voltage..
  • Phil Pilgrim (PhiPi)Phil Pilgrim (PhiPi) Posts: 23,514
    edited 2012-09-23 10:16
    why are you multiplying vcc by .65
    It's in the 74LVC244 datasheet on the page I pointed out above.

    -Phil
  • rwgast_logicdesignrwgast_logicdesign Posts: 1,464
    edited 2012-09-23 12:15
    ok so this acually raises a question ive wondered for along time

    i kept looking at page3 of my data sheet only to see an internal layout of the chip logic.. i searched for .65 and turned up nothing and i realized that u linked to the ti data sheet while i was using this one from nxp

    http://www.parallax.com/portals/0/downloads/docs/prod/compshop/74LVC244ADatasheet.pdf

    im using a ti chip but if i were using an nxp would all the info be interchangeable, for common chips like this? why would nxp make no mention of the .65 multiplyer
  • Phil Pilgrim (PhiPi)Phil Pilgrim (PhiPi) Posts: 23,514
    edited 2012-09-23 12:21
    I just looked at the NXP datasheet. I'm astonished at how poorly characterized the DC operating conditions are. I'd go with the TI part, just to be on the safe side, although the two are probably virtually identical.

    -Phil
  • rwgast_logicdesignrwgast_logicdesign Posts: 1,464
    edited 2012-09-23 13:22
    I acually am using the Ti part in this case i have no idea why i was looking at the NXP sheet.

    im glad to hear the ti and nxp parts are probably the same. i have some counter chips i wanted to use and they are NXP, just went with the lowest price on digi key. these are also 74xx logic chips. The reason theve been sitting is becuase NXPs data sheets werent very helpful i couldnt really figure out some of the details maybe ill give the TI sheets a look see, as id really like to use them as a hardware counter for the bs2
  • rwgast_logicdesignrwgast_logicdesign Posts: 1,464
    edited 2012-09-23 22:20
    Hmm so i just gooked this board up to my logic analyzer and if i connect input 3 the two inputs before it show the same data! So I figured out the sfix for this is to pull down the inputs, but i dont have any room to put pull downs on the inputs..... I do on the output that go to the propeller but pulling the outputs down doesnt seen to fix this problem, would pulling the inputs up work maybe? or something with the /OE pins. This really sucks I didnt realize these chips would have so much cross talk on there pins!!
  • Phil Pilgrim (PhiPi)Phil Pilgrim (PhiPi) Posts: 23,514
    edited 2012-09-23 23:06
    You never want to leave a logic input floating if you can avoid it. In your case the "crosstalk" is simple capacitive coupling. Pull-ups will work just as well as pull-downs. Doing something with /OE won't solve anything but will just move the problem to the Propeller pins.

    -Phil
  • rwgast_logicdesignrwgast_logicdesign Posts: 1,464
    edited 2012-09-23 23:56
    hmm ok well, well im just going to have to redo the board i guess, is this a genral rule for micros like the prop without internal pull up/downs too?

    the diagram on savage circuits twords the bottom
    http://www.savagecircuits.com/forums/content.php?309-Mixed-Voltage-Systems-Interfacing-5V-and-3-3V-Devices

    shows the inputs being pulled to ground, i could make that work.. i dont understand how the current being inputed wouldnt just flow to ground without acuall resistors.
Sign In or Register to comment.