Author |
Message |
xiphocoleon
Joined: Oct 04, 2013 Posts: 40 Location: Brooklyn
|
Posted: Wed Aug 20, 2014 11:56 am Post subject:
2 cascaded 4017s problems Subject description: Counters are not counting correctly |
|
|
Hi everyone -- I have created an 8-step sequencer from 2 cascaded 4017s. The bread-boarded version of this circuit works. However, after soldering this together (through-hole, DIP style) on a very small board, the circuit is producing glitchy behavior. Now, after four straight days of checking all the connections, replacing the chips, and resoldering some nodes, I'm really stuck as to what could be going wrong. If you have worked with 4017s, 4049s, or op-amp inputs to circuits, could you offer me troubleshooting advice?
I have checked these chips in the working prototyped version, so let's assume the chips work fine. The 4017s are 4017BE.
At this point, the fishiest area is the clock signal. The clock leaves the 324 with a 0 to +10 V pulse, which on the prototype is enough voltage to make the 4017 count. However, when connected to the 4017, the 0 to +10 V pulse sometimes is "crushed" -- it looks like something is pulling current away from the Clock pin on the 4017. I have checked all nodes on my soldered board with a connectivity checker, and I have not found any nodes accidentally soldered together.
If you could please check the attached schematic and give me troubleshooting tips from your own experience that would be helpful. I am also wondering if the close proximity of the chips is causing the chips to interfere with each other. Also, I have heard that placing a 0.1 uF cap across V and GND for each 4017 is a good idea to reduce counting glitches. Thank you.
If possible I'll also post a picture of my crowded soldered board.
Description: |
|
Filesize: |
49.95 KB |
Viewed: |
328 Time(s) |
This image has been reduced to fit the page. Click on it to enlarge. |
|
|
|
Back to top
|
|
|
xiphocoleon
Joined: Oct 04, 2013 Posts: 40 Location: Brooklyn
|
|
Back to top
|
|
|
PHOBoS
Joined: Jan 14, 2010 Posts: 5591 Location: Moon Base
Audio files: 705
|
Posted: Wed Aug 20, 2014 12:32 pm Post subject:
|
|
|
I see 2 things that could be a problem.
The transistor connected to the clock should have a resistor connected between the base and ouput of the opamp.
And the emitter of the transistor connected directly to gnd with the LED connected between the collecter and series resistor (R9).
The other thing is that you have some outputs connected together:
IC3E & IC4C
IC4B & Q8 of 4017A
Decoupling capacitors for every chip is always a good idea and can make the difference between a circuit functioning correctly or not.
Depending on what you're using it for you might want to buffer the CV outputs with an opamp. _________________ "My perf, it's full of holes!"
http://phobos.000space.com/
SoundCloud BandCamp MixCloud Stickney Synthyards Captain Collider Twitch YouTube |
|
Back to top
|
|
|
xiphocoleon
Joined: Oct 04, 2013 Posts: 40 Location: Brooklyn
|
Posted: Wed Aug 20, 2014 5:37 pm Post subject:
|
|
|
Thank you so much PHOBoS for checking this out. I've been going a little craaaazy picking this thing apart... Also, for background, I started this design with this schematic: http://electro-music.com/forum/phpbb-files/thumbs/t_16_step_4017_247.png
PHOBoS wrote: |
The transistor connected to the clock should have a resistor connected between the base and ouput of the opamp.
And the emitter of the transistor connected directly to gnd with the LED connected between the collecter and series resistor (R9). |
Oops, I forgot to include it here, but I used a 1 Mohm resistor from pin 14 to the transistor base. However, for the second comment here, are you suggesting a resistor b/t base and emitter? Or?
PHOBoS wrote: |
The other thing is that you have some outputs connected together:
IC3E & IC4C
|
Right now if I leave these together, a voltage can trigger a 4017A reset. However, is this not a good design for that? For instance, since pin 15 of 4017A needs to be connected to the output of IC3E, is it a bad idea to allow an op amp output to connect here also?
PHOBoS wrote: |
IC4B & Q8 of 4017A
|
Here I think it is correct to connect these together. This way CE is low for 4017A, allowing it to count. When Q8 goes high, this will stop 4017A counting and through IC4B it will make 4017B's CE low, allowing it to count. What do you think?
Thanks again!
I also included another shot of the crowded board.
Description: |
here is a shot of the top of the crowded circuit |
|
Filesize: |
3.45 MB |
Viewed: |
288 Time(s) |
This image has been reduced to fit the page. Click on it to enlarge. |
|
|
|
Back to top
|
|
|
xiphocoleon
Joined: Oct 04, 2013 Posts: 40 Location: Brooklyn
|
|
Back to top
|
|
|
AlanP
Joined: Mar 11, 2014 Posts: 746 Location: New Zealand
Audio files: 41
|
Posted: Wed Aug 20, 2014 11:06 pm Post subject:
|
|
|
No idea on the issues, but is there any reason why you are using two 4017's, rather than one using the Q0 to Q7 outputs, with Q8 going to Reset? |
|
Back to top
|
|
|
PHOBoS
Joined: Jan 14, 2010 Posts: 5591 Location: Moon Base
Audio files: 705
|
|
Back to top
|
|
|
xiphocoleon
Joined: Oct 04, 2013 Posts: 40 Location: Brooklyn
|
Posted: Thu Aug 21, 2014 11:37 am Post subject:
|
|
|
AlanP wrote: | No idea on the issues, but is there any reason why you are using two 4017's, rather than one using the Q0 to Q7 outputs, with Q8 going to Reset? |
My goal was to get 8 separate gates from a 4017 circuit. As far as I've tried, I can only get one long, continuous gate (with 8 discrete voltage changes) from one 4017. By using two 4017s and only using every other count, I could get 8 separate gates. Basically, as a pitch control for a modular synth VCO, I could create separate notes, instead of one long sliding glissando sound.
What do you think? Can you get 8 separate gates from a 4017?
Thanks.
Thomas |
|
Back to top
|
|
|
xiphocoleon
Joined: Oct 04, 2013 Posts: 40 Location: Brooklyn
|
Posted: Thu Aug 21, 2014 11:43 am Post subject:
|
|
|
PHOBoS wrote: | With CMOS you usually don't want any outputs connected to other outputs, including opamps.
If one output goes low and the other high you're basically shorting the power through the chips.
You can avoid this by using an OR gate which can be made with some diodes and a resistor.
|
Thank you so much for this tip.
PHOBoS wrote: |
I'm not sure how you want the envelope part to work,. but I expect it's a reason why you're using 2 4017's and skipping every other output.
However I think the same thing could be achieved with just 1 4017 and 2 flipflops.
|
Yes, with the separate gates, when the envelope turns on it adds attack and release to the gates, basically makes the pitches sound all warped out. I would like to look into the one 4017 and 2 flip flops design.
PHOBoS wrote: |
And I would use a summing amp to mix CV out from the sequencer with the master pitch, modulation and AR voltages.
But maybe this already works exactly as you want it.
|
This would probably be a good idea to protect the circuitry behind this output, even if it's already well-defended enough behind diodes. I do have an extra op-amp left on my 324.
And the Canal Street Plastics in NYC printed my panel in less than 12 hours. Definitely recommend them. |
|
Back to top
|
|
|
xiphocoleon
Joined: Oct 04, 2013 Posts: 40 Location: Brooklyn
|
Posted: Thu Aug 21, 2014 11:57 am Post subject:
|
|
|
Also, do you have recommendations for diodes, for instance the ones used connected to the 4017? I have weird experiences with diodes. For this type of signal here I'd probably use 1N914s because I have many on hand, but sometimes only "rectifier" (not sure if that's the right categorization) diodes (4003, 4004) work. For instance, I've found in the envelope circuit here that I've made, 4003s work best. |
|
Back to top
|
|
|
xiphocoleon
Joined: Oct 04, 2013 Posts: 40 Location: Brooklyn
|
Posted: Thu Aug 21, 2014 12:00 pm Post subject:
|
|
|
And, from Ud4 (in your schematic) do you think there should be a diode from the op-amp output to the 4017 clock? |
|
Back to top
|
|
|
PHOBoS
Joined: Jan 14, 2010 Posts: 5591 Location: Moon Base
Audio files: 705
|
Posted: Thu Aug 21, 2014 12:30 pm Post subject:
|
|
|
xiphocoleon wrote: | Yes, with the separate gates, when the envelope turns on it adds attack and release to the gates, basically makes the pitches sound all warped out. |
ah yes that's what it looks like it would do
Quote: | What do you think? Can you get 8 separate gates from a 4017? |
Quote: | I would like to look into the one 4017 and 2 flip flops design. |
You can get seperate gates by using AND gates after the outputs and add the CLOCK signal to them.
If you do this the outputs will only be high as long as the CLOCK is high.
By adding the CLOCk to a flip flop first you will have a 50/50 duty cycle so on/off times will be equal,
but it will run at half the speed (which is the same effect you have now).
You would still need the AND gates but since you're mixing all the outputs together I think it might be
possible to do this with just 1 gate, maybe a flip flop can be used for this. (since you'll get at least 2 in one chip).
I'm not sure about it yet, but I'll see if I can come up with a way to do it. _________________ "My perf, it's full of holes!"
http://phobos.000space.com/
SoundCloud BandCamp MixCloud Stickney Synthyards Captain Collider Twitch YouTube |
|
Back to top
|
|
|
PHOBoS
Joined: Jan 14, 2010 Posts: 5591 Location: Moon Base
Audio files: 705
|
Posted: Thu Aug 21, 2014 12:37 pm Post subject:
|
|
|
xiphocoleon wrote: | Also, do you have recommendations for diodes, for instance the ones used connected to the 4017? |
Pretty much all standard diodes except for zenerdiodes should work. I usually use 1N4148 but 1N914 or 1N400x should be fine too.
xiphocoleon wrote: | And, from Ud4 (in your schematic) do you think there should be a diode from the op-amp output to the 4017 clock? |
There is no other (output) signal connected to that ouput, so that's not needed. _________________ "My perf, it's full of holes!"
http://phobos.000space.com/
SoundCloud BandCamp MixCloud Stickney Synthyards Captain Collider Twitch YouTube |
|
Back to top
|
|
|
elmegil
Joined: Mar 20, 2012 Posts: 2177 Location: Chicago
Audio files: 16
|
Posted: Thu Aug 21, 2014 2:08 pm Post subject:
|
|
|
xiphocoleon wrote: | AlanP wrote: | No idea on the issues, but is there any reason why you are using two 4017's, rather than one using the Q0 to Q7 outputs, with Q8 going to Reset? |
My goal was to get 8 separate gates from a 4017 circuit. As far as I've tried, I can only get one long, continuous gate (with 8 discrete voltage changes) from one 4017. By using two 4017s and only using every other count, I could get 8 separate gates. Basically, as a pitch control for a modular synth VCO, I could create separate notes, instead of one long sliding glissando sound.
|
That's a cool idea.
The way I skinned this cat was to do clock AND gate to generate discrete gates, but that is limited to the length of your clock signal. So a second 4017, an extra AND gate chip.... come se come sa. |
|
Back to top
|
|
|
xiphocoleon
Joined: Oct 04, 2013 Posts: 40 Location: Brooklyn
|
Posted: Sun Aug 24, 2014 7:01 am Post subject:
|
|
|
I reworked my circuit and redid it with the diodes in the newest schematic. It all works so far!
...except the first step will not trigger! .........
I think this is a common problem. It has something to do with the timing of the clock.
I have found that if I put a low resistor (10k or less) b/t 4017A's 15 (reset) and ground on my perfectly working breadboard prototype, I can also get the first step not to trigger. However, when I remove the 10k resistor b/t 15 and ground on the soldered circuit it does not fix the problem.
I have scoped the clock signals. They enter both my prototype and soldered circuit at the same speed. However, the soldered circuit sends out pulses that are slightly longer than the protoype version. And this is causing a timing overlap so the first step will not trigger. Any ideas why this is happening? I have put decoupling caps on both 4017s and the 4049. Could this affect it? There is also a polarized cap on the board near 4017A pin 15. Could this affect it?
Thanks. |
|
Back to top
|
|
|
PHOBoS
Joined: Jan 14, 2010 Posts: 5591 Location: Moon Base
Audio files: 705
|
Posted: Sun Aug 24, 2014 7:05 am Post subject:
|
|
|
Ok, so I don't think the 2 flipflop thing is gonna work. The problem is the voltage coming from the pots, for which you would need an
analogue switch or mux (maybe a single FET could do it). And then you'd still need to divide the clock so that's 2 chips already, so you might aswell use the
2 4017's and 4049. _________________ "My perf, it's full of holes!"
http://phobos.000space.com/
SoundCloud BandCamp MixCloud Stickney Synthyards Captain Collider Twitch YouTube |
|
Back to top
|
|
|
xiphocoleon
Joined: Oct 04, 2013 Posts: 40 Location: Brooklyn
|
Posted: Sun Aug 24, 2014 7:13 am Post subject:
|
|
|
The circuit I'm working with is the 2 4017s and 4049. No flip flops. It is basically the original circuit I started with, but I added the diodes and op amp that you drew in the post a couple back.
And this works on the breadboard.
But, on the soldered circuit it skips step one on 4017A (Q0/pin 3).
Anyone have ideas on how to fix this? |
|
Back to top
|
|
|
xiphocoleon
Joined: Oct 04, 2013 Posts: 40 Location: Brooklyn
|
Posted: Sun Aug 24, 2014 7:40 am Post subject:
|
|
|
Ah! I solved it with a 100k resistor b/t pin 15 of 4017B and pin 3 of 4017A. I will post my completed schematic here (after I clean up my mess) for anyone else to try. |
|
Back to top
|
|
|
PHOBoS
Joined: Jan 14, 2010 Posts: 5591 Location: Moon Base
Audio files: 705
|
Posted: Sun Aug 24, 2014 8:28 am Post subject:
|
|
|
I tried breadboarding it to see what happens, and got the same problem (skipping the first output).
So I started disconnecting everything from the first 4017 but it kept skipping the first step.
Untill I tied pin 13 directly to GND instead of through a 100K resistor, then it worked.
So I reduced the value and 10K seems to do it. _________________ "My perf, it's full of holes!"
http://phobos.000space.com/
SoundCloud BandCamp MixCloud Stickney Synthyards Captain Collider Twitch YouTube |
|
Back to top
|
|
|
xiphocoleon
Joined: Oct 04, 2013 Posts: 40 Location: Brooklyn
|
Posted: Mon Aug 25, 2014 10:25 am Post subject:
|
|
|
I am going to start a new thread and post my finished schematic and pic of my rig |
|
Back to top
|
|
|
prgdeltablues
Joined: Sep 25, 2006 Posts: 222 Location: UK
Audio files: 12
|
|
Back to top
|
|
|
xiphocoleon
Joined: Oct 04, 2013 Posts: 40 Location: Brooklyn
|
Posted: Tue Aug 26, 2014 6:09 am Post subject:
|
|
|
prgdeltablues wrote: | There's a circuit over on muffwigglers to sort out the problem of skipping the first step on a 4017, especially on reset. Might be what you need:
|
Thanks for posting this. Yes, they seem to be discussing my problem. The buffer created by two inverters was possibly causing my circuit to malfunction. |
|
Back to top
|
|
|
fonik
Joined: Jun 07, 2006 Posts: 3950 Location: Germany
Audio files: 23
|
Posted: Tue Aug 26, 2014 10:09 pm Post subject:
|
|
|
PHOBoS wrote: | xiphocoleon wrote: | Quote: | What do you think? Can you get 8 separate gates from a 4017? | You can get seperate gates by using AND gates after the outputs and add the CLOCK signal to them.
If you do this the outputs will only be high as long as the CLOCK is high. |
|
this is not necessarily a drawback. it is rather a feature than a bug IMHO:
use a PWM clock and drive an AR or LPG with your sequencer - the sequence will come to life. _________________
cheers,
matthias
____________
Big Boss at fonitronik
Tech Buddy at Random*Source |
|
Back to top
|
|
|
|