Bucket Brigade circuit Modification help

I have a circuit that I started with that I guess you could call a "rain drop" effect and circuit works "as advertised" when I breadboard it. However I am trying to modify it, without success. The fact that I don't know anything about shift registers doesn't help.

The circuit function is as follows:
8 LED's light and stay lit one at a time, then turn off one at a time until all are off. Then it repeats.

Modification: Add a mode switch to have the circuit perform 1-8 turn on, all off, repeat 1-8 turn on, all off. Rinse and repeat.
NOTE: I have run across a video of this but the it is set up with cascaded registers and the schematic is beyond my limited knowledge on this.

At the moment, when pin 13 (Q7) goes high (after 8 clocks, all leds on), it turns the transitor on which takes pin 1 (DSA) to ground and so the chip start counting low pulses and the next 8 pulses turn them off 1 by 1.

To turn them all off, you need to move the collector of the transitor from pin 1 to pin 9 (MR), and connect pin1 to +V.
Then, when pin 13(Q7) goes high, it will take pin 9(MR) low turning off all the LEDs. When Q7 goes low, MR goes back high and the chip start counting +ve pulses again.

I created a model of the 74HC184, and this circuit demonstrates the original and that the change above works.

Update: Added a delay so that the 8th led stays lite for a while.

btw, this is more of a ripple counter... a "bucket brigade" is an analog delay device that stores analog samples and acts like an analog version of a shift register.

https://www.e-repair.org/SIA Repair/SAD4096.pdf

I set the circuit up as you diagrammed it and it works great. I needed too bump the 10uF to a 100uF to get the last LED to sync.

The other half of the problem is how to make it selectable between the 2 modes with a slide switch or similar. This was my original dilemma , or at least part of it.

Any idea? Thanks a ton in advance for the all the help.

EdStraker said:

The other half of the problem is how to make it selectable between the 2 modes with a slide switch or similar.

Click to expand...

Could be laid out better, and you'll probably need to tweak the cap values:

Thanks a lot Buk. I breadboarded the circuit and it works perfectly. Exactly what I was looking for. I did have to make an alteration thought. Had to remove the 10pF Cap and 3.3k Resistor on the Reset pin of the last LED wouldn't light. Left all the other Resistors at 10k and it seems to function perfectly. Go Figure.

EdStraker said:

Had to remove the 10pF Cap and 3.3k Resistor on the Reset pin of the last LED wouldn't light. ... Go Figure.

Click to expand...

The 74HC164 model I created models the logical behaviour pretty well, but the physical/electronic properties not so much. This must be a case where the inherent propogation delays work in your favour.