Shop OBEX P1 Docs P2 Docs Learn Events
74HC595 problems with Prop... Not SX? — Parallax Forums

74HC595 problems with Prop... Not SX?

Chicago MikeChicago Mike Posts: 88
edited 2010-01-28 23:38 in Propeller 1
So I have a Nixie clock that I"ve been working on for years on and off, and I had everything working with an SX. Of course in the name of adventure, I decided a few months ago to make it work with the prop instead and relayed the board anyway. Everything worked with the SX. So anyway, I'm still using a chain of 3 74HC595 chips to get the data. I was using shiftout with SX, but thought I would use the '75HC595_Regular' obex device for the prop. Below is my code. Basicly... Though no wiring has changed I'm getting nothing but zeros· out of this. Can anyone see if I'm missing anything.
<code>
CON
o_nixrck······· =· 9·· 'OUTPUT· Nixie Data Driver Line. rck Pin (Latch Pin)
o_nixsck······· =· 10· 'OUTPUT· Nixie Data Driver Line. sck Pin (Clock Pin)
o_nixser······· =· 11· 'OUTPUT· Nixie Data Driver Line. ser Pin (Data Pin)

OBJ
··nixie2 : "74HC595_Regular"

PUB Setup
· dira := %00000000_00001101_10011111_11010000·························· ' Setup inputs and outputs of active pins as stated in CON
· outa~······························································ ' Set all used output pins low
· waitcnt(1_000_000+cnt)
· temp:=1
·· nixie2.Start(10,9,11, 3)
· repeat
··· temp:=temp+1
··· nixie2.out(temp,1)
··· nixie2.out(temp,2)
··· nixie2.out(temp,3)
··· waitcnt(1_000_000 + cnt)

</code>

So·shouldn't that increment each tube set by one? I just get zeros. INteresting thing. So the 74HC595 also seemed to need 5V on its power pin even·though it states it can run down to 2V. I'm of course giving 3.3 out of the rck, sck, and ser pins·with nothing in the way. When I tried giving the 74hc595 3.3 volts I got nothing... NOt even zeros. by the way, I"m using the ogilumen nixie driver set.

Any help would be great!

Comments

  • Timothy D. SwieterTimothy D. Swieter Posts: 1,613
    edited 2010-01-24 01:47
    Chicago Mike -

    A couple items to check out:

    1. Maybe setup a test bench to verify you can power the 74HC595 and still receive a 3.3V input. I looked at the data sheet on NXP's web site, it looks you can because of the input voltage levels, but maybe you have a different variety of the IC. I agree that it should also be able to be powered from 3.3V based on what I see in the data sheet.

    2. I haven't reviewed the specific 74HC595 object, however since it looks like you are passing the pin numbers to it I would venture a guess that the 74HC595 objects setups up its own pins in its own cog. You are also setting up the pins in the calling object as well. That means two cogs are controlling the same output pin and you will get garbage on the output. trying removing the DIRA/outa from the calling cog.

    Post some pictures of your project when you get a chance.

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    Timothy D. Swieter, E.I.
    www.brilldea.com - Prop Blade, LED Painter, RGB LEDs, 3.0" LCD Composite video display, eProto for SunSPOT
    www.tdswieter.com
  • VIRANDVIRAND Posts: 656
    edited 2010-01-24 01:52
    If the 74hcXXX is not responding to 3.3V with 5V power, it is some minor bug or has a simple solution.
    If the 74hcXXX is powered by 3.3V then it probably is not outputting enough to turn on the nixie driver circuit.

    Whatever the driver is, it has to switch 150V or so, and probably only has enough gain for 5V input.

    IF you say you are only getting zeros in the nixie tubes, that could be because the zeros are closest to
    the common electrode and barely light. Another possibility is something pulling the clock or reset or enable pins
    too far from the threshold of useable 0/1 voltage, if there are such pins on the 74hcXXX.

    Post Edited (VIRAND) : 1/24/2010 2:01:09 AM GMT
  • mynet43mynet43 Posts: 644
    edited 2010-01-24 03:30
    I have a circuit with a number of 74HC595's on it, to drive 7-segment LED's. I found that the 3.3V output from the Prop will not reliably drive the inputs of the HC devices. The spec calls for a minimum of 3.15V to trigger a high with a VCC of 5V.

    I needed to use the 5V VCC, so I ended up using 74HCT365 buffer chips to raise the driving voltage for the HC inputs. If you use a 3.3V VCC, then it should work fine.
  • VIRANDVIRAND Posts: 656
    edited 2010-01-24 05:10
    In my experience, CMOS would respond symmetrically to an input, with the midpoint being the boundary
    between zero and one, except in some cases where there was linear response. Both responses are
    expected because two "complementary" transistors of opposite but equal polarity response are
    used in CMOS logic. It makes no sense to me that newer HCT would raise the threshold in a trend
    of lower voltage logic power supplies. But I can't say it isn't so. If HCT is 200 times faster than OLD
    CMOS, then maybe that speed cannot be achieved by true complementary circuits.

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    VIRAND, If you spent as much time SPINNING as you do Trolling the Forums,
    you'd have tons of awesome code to post!     (Note to self)
  • mynet43mynet43 Posts: 644
    edited 2010-01-24 14:57
    The HCT series is designed to match the input and output voltages of TTL logic. That's why it's able to work with the loaded 3.3V input from the Prop and also drive the inputs of the HC devices. It's all in the spec sheets.
  • Chicago MikeChicago Mike Posts: 88
    edited 2010-01-28 21:03
    I still haven't much luck with getting this to work..(granted I didn't have much time this week). Even though we are pretty sure 3.3 is acceptable, I opted to try a bus transceiver level shifter. (The SN74LVC4245ADW) A really nice and easy 3.3 to 5v 8 channel shifter. its really cool! However it didn't help the issue at all.

    Timothy, i removed the dira and outa referneces and that didn't do anything either. (I will post pictures sooner or later.. two years of 10 minutes here and there adds up [noparse]:)[/noparse] BTW, nice DMX object, I've used it several times). I think tomorrow I'm going to use a scope at work and have a look at what's coming out.
    .
  • Timothy D. SwieterTimothy D. Swieter Posts: 1,613
    edited 2010-01-28 23:38
    Post your code Chicago Mike and we may find something else.

    You could use Hanno's ViewPort right from home to get a feel for how the signals are (or are not) changing on the Propeller pins without connecting an oscilloscope. ViewPort is handy for this - check out the lab for a tutorial on it. I think ViewPort is free for a trial.

    Sometime I hope to revisit my DMX512-A object and update it further. Since I wrote the object several years ago I have too have used it many times, but I have realized there are some improvements that can be made.

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    Timothy D. Swieter, E.I.
    www.brilldea.com - Prop Blade, LED Painter, RGB LEDs, 3.0" LCD Composite video display, eProto for SunSPOT
    www.tdswieter.com
Sign In or Register to comment.