How to connect my Roland Synth?

[Enzo]

Now the details: So, I still would need PU resistors and at what value?
Also, what value for PD resistors?
Thus, when foot SW closes +5V enter into AND gate allowing A# (or B) to be added to C and when the switch is open the PD resistors earth out the gate and only C gets through.

Not quite as you describe. The footswitches are normally closed types. Thus they keep the second input of each gate LOW. This prevents Acc C from sending the High (+5) on to A#. Then when you press the footswitch, that UNgrounds the gate input, and PU brings it up to +5 - high. Now, the high gets to A#. The footswitch enables the gates, in other words.

R values? PD I suppose 10k would be a good one to try. ANd PU maybe 1k? The PUs are on most of the time, but 1K only draws 5ma. I guess the PD could be 1k as well, after all, only one of them will have current at a given time.

Remember that the ROland has the matrixed keys, and we have to connect two things together. You could emulate the Roland circuit in logic, but it would not be a couple simple gates. Wwe would gave to find the pattern and polarity of the Roland strobe, then we could gate that strobe through to the appropriate return when the rmote key is pressed. That woul dbe a little similar to the footswitch method

[jjj]

You are right! I don't know what I saw... I must have been too exited.

Important: The +5V for the And gates PS have to be connected after the button SW so that the footswitch only works when an accordion major bass button is pressed.
Note: When accordion button C is activated it pulls one AND gate input high... I hope it won't cause problems when then the foot switch stays closed (on earth)? It shouldn't, because the AND gate needs two (not one) high inputs to output the +5V..
So... I describe it once more in my words:
When accordion button C is activated only C goes to 4066 (and then to Synth SW), because foot SW is (normally) closed and the +5V from the button are earthed out via PU resistor and PD resistor so, no current flows to AND gate!
Whereas, when accordion button C is activated and one of the foot switches opened +5V flows to (either A# or B) AND gate and then to 4066 etc. so, that I will hear two tunes. Yea, that's what I need!
Amazing, for it's almost what I wanted to achieve with my two resistor foot switch circuit, only that we now added AND gates! I guess the advantage is that even lower currents are needed to change the state from +5V to ground?

[Enzo]

If we are using CMOS logic ICs, the only current involved will be that through the resistors.

[jjj]

I just finished drawing up the 4066B connections for 12 bass notes and 12 chord notes with the new 1 switch per Synth key layout:

The pin # are in reverse, because I glue the IC's upside down... and then wire them up.
In regard to the second (to switch) project, which can be successfully accomplished with 6 chips of 4081, I still like to try out my rather simple foot switch circuit idea... just with one note and just for the sake of learning something from it...I fear might even work?! When I finished the above 3 chips 4066 for bass I just connect two 2K Pots and then slowly lower the resistance until I get the result (of course I won't go lower than 0.5K). If it works I then measure the current when PU is earthed. I think there is as much current involved as with AND gates? Actually, the foots switch circuit is only activated when I press an accordion button and it's not always the same accordion button. I still cannot understand that it won't work.

[jjj]

I forgot to mention that each 4066 gate has a 22K PD resistor!

[jjj]

Hi Enzo,

Today I finished wiring up the 3 bass IC's (3 x 4066B) and anxiously did my test to find out if my simple foot switch circuits delivers and it did! It works that good that I really can forget AND gates!
Result:
At first I build a test circuit with two 2K2 Pots and it worked immediately!
But then I thought to increase resistance and swapped the 2K2 Pots for 50K Posts. The result was even better:
The PU resistor worked at 5K6 (and stopped working at 8K6)
The PD resistor worked at 15K (and stopped working at 21K3)
Hence 5K6 and 15K are pretty high values, meaning the 4066 gates draw very little current! I mean 1K would also work, but draws more current.
So, I ask myself: what's the advantage with using AND gates?

My simple circuit is so much easier to wire up that using 6 x 4081...

[Enzo]

Use whatever works. I perfer to do logic chores with logic chips rather than resistors. They are simple and reliable, and can be expanded upon readily, and any sort of combinational things can be done in a snap. CMOS chips use almost no current, and even that only when changing states. Most of the current your CMOS circuit uses will be in the pullups.

One concern I have is logic levels. if it is working, it is working and we ought not worry further, but while the logic will toggle states at more or less 2.5v with a 5v rail, it will not be reliable until highs and lows are past their thresholds of 30% and 70% of rail. What voltage is appearing under both on and off states at points A#, B, and C?

[jjj]

ON state: C= 4V29 // a#= 3V02

OFF state: C= 4V28 // a#= 2V41

Last evening I twisted & soldered 24 little sausages, looking like this:

x-->I--5K6--x--15K--->I-x
(i.e. From C-gate of 4066 via diode and 5K6 to a#-gate of 4066 and via diode to foot switch and ground)
Remember, on gates of 4066 is a 22K PD resistor.

So now I'll have only 24 of these sausages and six 4066 for bass and chord notes to complete my project. The three 4066 for bass are done and fully working. The 12 diode blocks with 13 diodes ea. are ready connected, too. So, now I'll have to wire up the three 4066 for chords.

[Enzo]

Yep, there is the problem. If it works for you, then don't worry about it. But with the resistors, you are asking the circuitry to differentiate hi and low from 3.2 and 2.4 volts. True logic levels want hi to be over 3.5v and lo under 1.5v. This might work for you, but in some other circuit, it might not.

[jjj]

I could have even increased the voltage difference, but it was not necessary. Those resistor values I chose were actually a bit lower in resistance than the crucial point at which the switch properly worked. So I went a little more down in resistance to ensure safe operation. It must be that the 4066 gates allow this narrow voltage margin, what else?
Tonight I glued these 'sausages' (with UHU, German Universal glue) and wired it up, but didn't have the time to see if it all works OK. - Jeeze, the glue was so good that I was able to fully solder the connections in position.
Tomorrow I'll be in Santiago (120km from here) and so, I'll continue my project the day after. Provided I wired it up properly, I can't see why it shouldn't work. It's not critical.
I always create a little wiring plan. That makes it also easier to find faulty connections or problems, later on.
But your AND gates were certainly the professional solution. It's good to know for future projects. I learn best from practical examples. Most of the time I looked up other circuits to find a solution.
Music is my love and so, it will make me happy... for the rest of my life. I also enjoy writing books. See my books at: http://www.lulu.com/ (enter Drinda into the search window) and read the Preview...
What do you really enjoy a lot?

[Enzo]

Again, if it is working reliably for you, then that is sufficient.

The whole point of logic gates is that you do NOT have to worry about critical voltage levels. You are very close on your threshold. The chip turns off and on because it MUST change state SOMEWHERE in the center of voltage swing. But you might find that next time you build this or worse, next time you need to replace a 4066 and you have to use a different brand, the new chip won't switch at the same voltage.

Logic chip makers specify that their chips will reliably change when a signal crosses a threshold. They guarantee that. And that is why the gate maker says logic hi and logic low rather than a specific voltage at the input. They will not guarantee you ANY form of performance in those center voltages. In the case of 5 volt logic (CMOS can operate at any suply voltage from 3 to 18) that band between 1.5 and 3.5 volts is called the forbidden zone.

Imagine a toggle switch. when you push the toggle lever, you must push it past the center point until you reach where it snaps to the other position. As long as you push it so far, it will reliably snap to the other position. Now imagine you grip the lever tight and move it just to the center point. Now move it bit by bit until it snaps one way. Them, still holding, move it back the other way until it just snaps. Now arrange barriers so the toggle lever can move no farther than those two points. It just barely works by a slim margin. That is analogy to what you have done with the resistors.

What do I enjoy? I like making people smile, and even better, making them laugh.

[jjj]

Back from S'go...
Yes, I understand you perfectly, for I experienced it while adjusting the pots. A little more it was on a little less it was off. I then went a little more in direction of safe operating range, but kept the resistance as high as possible, because the higher its resistance value the less load on the PS.
Luckily I have got heaps of same type 4066B. I won't use them in my lifetime... Knowing that it might cause problems with other types of 4066, I better stick a few of them into the button box. End of risk!
So, tomorrow or so, I'll try to finish the whole project.
I let you know how things are working out. Since you are well informed on my project, you'll be best positioned to shoot any mistake.
I'm glad to be able to do these things without electronics education...
I did far more complex projects and also fixed my electronic organ Elka E49. I have all circuits schematics and special parts for it.