Circuitbenders Forum
Circuitbenders Forum => Circuitbending discussion => BENDING TIPS => Topic started by: Dj Task Manager on October 15, 2009, 11:01:35 AM
-
Hi all,
Im building a running light sequencer circuit to use to sequence devices (like what you might have seen on m-z3r0's dd-drum machines). Im working form a this schematic for cascaded 4017 chips...;
http://saint.419.removed.us/runlit.html (http://saint.419.removed.us/runlit.html)
its sort of working a bit but a kind of strangely; it counts through the outputs from the first chip, then for each cycle of the this it advances the second chip one step. WIERD!
Can anyone see any glaring errors with it? or have i just made some mistake?
I was thinking perhaps the diode might be the wrong way round in the diagram...
Cheers!
-
I think you may have connected the second 4017 to the divider pin 12.
That would seem the most likely problem... but it is one way of cascading.
-
sadly no, pin 12 is left all alone on both chips. I'll try turning that diode round perhaps. And have a good think before doing anything rash.
-
hmmm, turning the diode round hasnt worked! arrrgh!
Its really annoying cos the case ive made for it looks amazing. Damn damn dman. lol.
-
The diode and resistor are there so that the clock doesn't drive the second 4017 unless pin 11 of the first one is high.
What happens is this: The 555 generates clock pulses which makes the first 4017 clock until pin 11 comes on. Pin 11 is wired to pin 13 - Inhibit - which stops the first counter from counting any further. While pin 11 was low it pulled the far end of the 33k resistor low, stopping the second chip from receiving clock pulses. Now the second 4017 will clock - pin 3 is ignored because it will be on when the chip is reset - until its pin 11 goes high, triggering the first 4017's reset pin. When the first 4017 resets, its pin 3 (output 1) resets the second 4017.
Check that you've got that all-important connection between pin 11 and pin 13 of the first chip.
-
Thanks gordon, but yes they are connected.
Maybe i'll try different values for the resistor...
-
Have you tried the 18-step LED sequencer from Bill Bowden's site?
http://www.bowdenshobbycircuits.info/page5.htm#4017-7.gif (http://www.bowdenshobbycircuits.info/page5.htm#4017-7.gif)
It's a little different from the one you linked to.
and if you can get that one to work, your case needn't be wasted.
-
After a little research ive discovered its a good idea to decouple this circuit cos of the 555. Im going to try this with a 10uf electrolytic cap and 0.1uf ceramic cap in parallel between + and gnd this evening. fingers crossed!...
-
no luck :'(
-
Get the first stage working *first*. Then make up a simple kind of logic probe (LED and some sort of buffer, perhaps a NAND gate with its inputs tied together) and watch the state of the clock and inhibit pins for the second chip.
-
I put this queery on another forum and the current final suggestion is a different schematic using a 4093 logic thingy with the 4017 decade counters. Im going to try this one out now.
To anyone who is interested heres the thread on the other forum:
http://www.electro-tech-online.com/electronic-projects-design-ideas-reviews/98869-cascading-4017-troubles.html (http://www.electro-tech-online.com/electronic-projects-design-ideas-reviews/98869-cascading-4017-troubles.html)
If anyone spots any errors with it or whatever pipe-up, i need to get these lights flashing!!!
-
that newest schematic may be the answer. maybe the original circuit you had was meant to do exactly what its doing. another way to go might be to use a mc40192 with a mc14515. never tried them myself. basically you'd get a 16 step sequencer that is bidirectional?... i'm going to try it out soon, i'll share my findings.
-
Sorry, I hadn't read the forum post. The original circuit will work, but it *will not* drive LEDs directly (and that'sa bad idea in any case). Build some buffers to drive those LEDs.
You can't connect and LED between the output of a 40-series chip and ground and expect it to work well because the chip can't source enough current. You'd need to connect it (with a suitable series resistor) between the pin and +5V and then arrange to turn the pin *off* to turn the LED *on*. The reason for this is that the output of the chip consists of two transisors, one to +5V and one to 0V. Due to the way the chip is made the one to 0V can turn on harder than the one to +5V - a bit like the difference in strength between your bicep and tricep muscles.
Now, go build a simple logic probe and get that first circuit working, because it *does* work. The point about 555s being utter shit still stands.
-
sorry gordon, i am a terrible unforgiveable noob.
I kind of follow, but any chance you could do a quick diagram to show the buffer and how it connects to the circuit.
Thank you kindly
Mike
-
Can anyone do a quick sketch to show how to make and connect the buffers and resistors type situation?
Honestly this is going to be a sweet machine when its done, and will truely help me kick out the jams.
I really want to get it finished! And its my birthday today!
Cheers
Mike x
-
happy birthday! you might not have to buffer each output, you could run each output through a signal diode(cuts the signal even more) and then connect all of the cathode sides to a signal booster of some kind. maybe a simple amplifier circuit? i dont know if this would actually work but its what i would try if i had more time. good luck.
(http://i35.photobucket.com/albums/d194/bendies/buff-1.jpg)
-
Thanks for that, but im affraid its over my head. Maybe I'll have to read a book or something. Damn it.