Clock Divider

[flagada]

First post!


I have built the Yusynth Clock Divider on stripboard. It actually works fine, but I do have a very strange observation:

The different divisions are not found at the IC pins described in the schematic. For instance, the 1/4 division should be at pin 10 of the 4017 (according to the schematic) but instead, it occurs at pin 1. In turn, the division of 1/5 that should be found at pin 1, is found at pin 5. And so on. The first 3 divisions are at the right pins, the others are in a different place.

[flagada]

This is the schematic:

[Grumble]

maybe you switched pins 3 and 2 on your stripboard, because the sequence is 2 4 7 10 1 5 6 9 since pin 3 is output 0 and is not connected to the switch.

[flagada]

Ooops! I made a mistake in the table I made (see first post). The masterclock (1/1) appears at pin 2 of the IC, not pin 3. In my clock divider that works as it should. (error has been corrected).

@ Grumble: sorry for the confusion. But no, I did not switch pin 2 and 3 on my stripboard.

[PHOBoS]

What exactly do you mean with the divisions appearing at those pins ?
Do you mean setting the step switch to those pins, because all the divisions should be measured at pin 3.
Also make sure pin 13 is connected to GND, it could cause some weird behaviour if it isn't.

[Grumble]

[PHOBoS]

Actually it is AND-ed with the CLK, but yes. Although for measuring it doesn't make much difference
except for the 1/1 division as pin 3 just stays high.

[Grumble]

Your'e right, AND-ed

[flagada]

Hi PHOBoS, I'll try to be clearer.

First, I put all three rotary switches completely CCW (all common terminals are connected to pin 2 of their 4017). All the LEDs blink at the same rate i.e. the rate of the clock signal. So far so good.

Then I put SW2 in the 1/2 position (common gets connected to pin 4). The second LED blinks at half the rate of the two other LEDs. That's good too.

Next: SW2 in the 1/3 position (common connected to pin 7). The second LED blinks at a third of the rate of the other two. Still working as it should.

But then: SW2 in the 1/4 position (common connected to pin 10): the second LED blinks only once and then it stops blinking. If I connect the common of SW2 to pin 1 of the 4017 it blinks like it should, at a rate of a fourth of the rate of the other two.

Likewise, if I want the LED to blink at a rate of, say, 1/7 of the clock signal, I have to connect the common to pin 9, while the schematic says it should be connected to pin 6.

This behaviour is the same for all three switches/4017's.

Yes, Pin 13 is connected to ground (all three of them).

[PHOBoS]

[flagada]

[PHOBoS]

hmm skipping, I've seen that happen before, see this thread, but that had to do with pin 13.
It could have something to do with the CLK signal but the discrete schmitt triggers should provide a clean transition.
It's rather odd that it skips just that output. Do you have any decoupling caps on the board ?

[flagada]

Yes, pin 16 of each 4017 has a 100nF to ground. Also the 10R resistors at the supply rails are there. Instead of the 10uF elco's, I put 47uF elco's.

[flagada]

I did some more testing and I got more confused...

The results above are obtained when I use my MFOS sequencer as a clock (gates of 0 to +12V, or 0 to +6V). When I use a LFO (-12V to +12V) the results are the same. The clock signal at the collectors of Q2, Q5 and Q8 is clearly visible on my scope.

But when I use another LFO (-4.5V to +4.5V) the clock signal is still visible on the collectors, but the clock divider behaves totally different:

With the rotary switches fully CCW, the LEDs blink according to the masterclock (OK). But with the switches in positions 1/3, 1/4, 1/6, 1/8 the LEDs don't blink at all. In position 1/5, the LEDs blink at a rate of 1/4, in position 1/7 the LEDs blink at a rate of 1/5. In position 1/2, the LEDs sometimes blink, and sometimes they don't.

And when I pull the wire from the common of the rotary switch, the LED starts blinking, but now at a rate of 1/7 of the master clock (instead of 1/10).

Somehow with this new LFO, three steps are being skipped??

[blue hell]

hmm .. interesting :-)

- What supply voltage do you use on the divider? (asking because the gate propagation delays depend on the supply voltage, getting higher for lower supply voltages, and there might be a race condition in the reset signal).

- What does the supply voltage look like on the scope?

- What does the RST input signal on pin 15 of the chips look like on the scope? (or even .. when you put the probe on the RST line do the symptoms change?)

[flagada]

Hee hallo!

I am using +15V and 0V. On the scope it looks very clean, no ripple or something. The power supply can deliver at least 1,5A and it is not at all working at it's top.

Actually I don't see anything on the RST pins (15). What should I see exactly? The symptoms do not change when I touch pin 15 with my probe (did not check all of those symptoms because there are a lot of them!)

I discovered that the RESET input of my module does not work. If I apply a gate (12 V) to the reset input (R39) there is no reset. When i connect the emitter from Q11 to +15V for a split second, then there is a reset.

[blue hell]

Hoi hoi :-)

Ok .. supply seems fine then and delay times are minimal, which is good.

You should see the same as on the common from the step switch (as long as no external reset is being applied) .. a narrow upgoing impulse, like : _|_

When that is not happening ... are D3, D8, D13 properly oriented? Also .. what is the signal level at the RST input?

Tying Q11's emitter to + causing a reset seems to indicate proper orientation for D2, D7, D12, and also it seems to indicate an unrelated error in the circuit around Q10 and Q11.

[flagada]

Well I don't see such an impulse at pin 15 (none of them). D3, D8 and D12 are oriented in the right direction. I did a diode test and they seem to be OK. The signal level at RST (pin 15) is ) 0 Volt.

While probing around with my scope I noticed something remarkable at pin 3. I'll do some more testing tonight and report back here.

[PHOBoS]

It depedns on the scope but I wouldn't expect the pulse at the reset input to be visible as it is very narrow.

[blue hell]

Right ... my work scope prolly spoiled me a bit

[flagada]

I guess my scope should be good enough to see them. It's a 60MHz Tektronix.

[flagada]

Ok, I did some more measurements on pin 3 with the rotary switch in different positions (MFOS sequencer as clock, gates of 12V). I noticed something that I think is strange when the switch is in position 1/4.

In all positions except position 1/4, the voltage at pin 3 switches between 0 and 15V at the same rate the LED is blinking. When the LED is on, pin 3 is high (15V), when the LED is off, pin 3 is low (0V). (In position 1/1, pin 3 is always high). I guess this is normal.

But when the switch is in position 1/4 (common connected to pin 10), the LED does not blink at all. The voltage at pin 3 remains low. However, I see tiny but regular spikes of 15V at a rate of 1/3 of the clock. I guess this is strange...

[wackelpeter]

I don't know if your scope has a led/lamp for triggering, mine has and i can see when an trigger Impulse reaches the scope even when i can't see it on the scope screen because it's too narrow and slow for my timing divisions.

At an earlier point i would have guessed to check the pulldown resistor on the RST Input, but when it only occurs at that certain step then it's not likely that this resistor is too small and thus limiting the Impulses too much...

When you disconnect the pin that doesn't work properly then all other divisions work correct?

Have you measured your rotary switch for all positions to the common? And especially if in this certan position if there is any other connection between the pin and the common?

[flagada]

[blue hell]

I'm out of ideas .. except for one ... grounding issues maybe