For sure I could be the one on the wrong track. I'm not right, just trying to be of help.

Nope, you were on track.., that's exactly how I read it.

There MUST be some way to reset the switches.

On my MCU example for a game show indicator, as in post #5 it was a simple timeout after 3S. The TS' requirement is after power off which needs no further parts (assuming a master 'ON/OFF' switch), but an extra reset button could easily be added (reset MCU- no extra software needed).

 

It is clear to me that the TS has not specified the problem in enough detail for us to solve it.

For one thing, where are these switches located? The mechanical piano switch only works if they are all in the same location. Are they? Or are they meters apart? (For other reasons, I doubt that the piano switch is what the TS wants)

Do they all control the same load or different ones? Since no connection to a load is shown we do not know that.

Are they momentary switches or not? Let’s suppose they are toggle switches. If a switch is toggled on when it is “disabled”, then the other switch is toggled off, what happens? If three of them are toggled on when “disabled” and the original switch is toggled off what happens?

What, exactly does “disabled” mean? Can you still operate the switch when it is “disabled”, or does operating it simply have no effect? Is there an indication that it is disabled?

@Mr6ft6: Do yourself (and us) a favor by telling us the problem you are trying to solve. Then there might be a chance you will get a solution.

 

What would be the best way so when any one switch is pushed all of the others will not be able to supply voltage. Only one switch in the circuit could be live at any time. 12 vdc 3 Amp load maxView attachment 322757

non shorting rotary switch

 

non shorting rotary switch

If you believe “disabled” means off, and all the switches are in the same location, there is still the problem that switching from one circuit to another will briefly activate circuits in between on the switch.

 

Your original question sounded like a radio button circuit, so called because it functions like the pushbuttons on an old car radio; press any button, and all other buttons are released, but able to be pushed.

Now it sounds like what you want is a game show circuit. Pressing any button locks out all other buttons until some kind of reset functions happens; pressing a reset button, bouncing power, etc. There are many examples of game show circuits on the innergoogle, using neon bulbs, relays, transistors and diodes, IC's, etc. The unique thing about your question is the output power level, but any of them, driving a group of output power transistors, will do what you want.

Do you have a fixed maximum number of switches, of do you need a design that is n-expandable?

ak

 

The circuit shown in post #10 has a separate reset button. This input could be replaced by a 555 circuit (triggered by the quad-AND gate output) to automatically reset the output after a desired period of time, if that’s the desired operation.
Also, the number of switches can be easily increased to eight with the use of an 8-input AND gate. Or many other inputs by a combination of AND gates. It may also be possible to use the Q outputs of the flip flops with a diode OR and an inverter (NOT).
The circuit had been in use for many years. The linked article describes it and its build in more detail.

 

I already suggested the solution in post#19, but I was not able to provide any details.
A "J-K flipflop" (CD4027) will provide a latched change of state upon the rising edge of a CLOCK pulse. So that takes care of latching the first button press. To lock out all of the others, the first "Q" output is "ORed" with allot hhe other FFs to change the "J" inputs on all of the ICs to zero, and the "K" inputs of all of them to "ONE". That leaves all of the other FFs locked out from changing states until the one triggered FF is reset. The delay time in which a second trigger might set another FF is three CMOS propagation delays. With a 5 volt logic supply that could be as great as 1200 NANOSECONDS.
Running the logic on 12 volts will reduce that delay time quite a bit.
Of course the input will need a buffer to assure a clean clock pulse, and a bit of noise suppression in the form of a pull-down resistor. Also, there will need to be a connection to change the inputs from "J" to "K" as soon as any FF is triggered. To reset, a single pulse to the "Reset " input of all the FFs will be required. For the higher output current requested, a buffer/driver driven by the all inputs "OR" can be used. So there you have the whole design , which is simple, and could even use a diode-OR function if there are a lot of buttons to be watched.

The CD 4027 is a commodity IC, available from a number of different manufacturers. Also it is not very expensive.
Probably the latching drive to the J and K inputs can be a CD4013 FF driven to set by the OR gate.
That is about as simple as it can be.
To lock out any button simple put a logic HIGH on it's Reset input.

 

What would be the best way so when any one switch is pushed all of the others will not be able to supply voltage. Only one switch in the circuit could be live at any time. 12 vdc 3 Amp load max

So I'm sorry I'm not clear. I would like to change the circuit so that only one switch at a time could ever be powered up. Even if someone were to push the button for one or even all those switches only one first pushed would complete the circuit at a time until power is shut off

"when any one switch is pushed all of the others will not be able to supply voltage"
"only one first pushed would complete the circuit at a time until power is shut off"
Seems to me the only two statements indicate when one switch is pressed no other pressing of buttons can ever change the state of the circuit.
"Only one switch in the circuit could be live at any time."
This statement doesn't help to clarify the need; which we don't yet understand, nor do we understand where these switches are to be located in relationship to the other switches.

We've had no other input than just those two quoted statements (which are quoted in part, not in its entirety).

When one button is pressed all others are locked out. Some members are thinking only one switch can be active when pressed but are approaching the solution as if (for example) button 3 is pressed and output 3 becomes active. (Yes, I know nobody is talking about outputs. But this is a necessary assumption). Now, #3 has been pressed and all others are locked out. YOU CAN'T PRESS ANOTHER BUTTON AND CHANGE THE OUTPUT. All other switches are locked out. So somewhere there has to be a way of de-energizing #3 (my example). A way of de-energizing ALL switches simultaneously.

The purpose of this project eludes me.

Unwatched.

 

The purpose of this project eludes me.

Yes, it eludes all of us, since he has given a clue.

 

OK, clearly we need additional information. I guessed that the goal was a "first response detector" like used on some quiz shows in the past. Or to detect whose "Pinewood Derby" car crossed the line first.
If the goal is to allow only one power consuming device to be operated at any one time, that is very different. Really, though, the exact same scheme could be used, with the difference being that after one device were done being used a reset would need to be requested. The reset would be one input to reset whichever had been set, so while it would require more buttons it would not require more logic beyond the original.

 

Hard to select a solution to very vague problem... And the TS is apparently lost in the internet somewhere else... And yes, there is a plethora of possible solutions from simple relay, transistor switches, or uProcessor based. But not knowing what the TS is capable of would have an impact on any solution.

 

Ok thanks for everyone's replies,
So this is what I'm trying to modify. It work great but lacks the ability to remove human error. It is a ssd drive sata power controller. It is designed so the you can power up to 6 drives at a time or in any combination. The problem is that if the computer is powered on with one of the drives and someone were to push one of the physical buttons it causes problems in windows and the bios. In my scenario Iwill never want more than one drive powered at time due to multiple operating systems and configurations that are on the drives. I don't want to do multiboot. The numbered 4 pin connectors are the output power to each drive with (Ground, 12vdc, Ground, 5vdc). Would like a solution that I could remove connections on rear of unit and install relay board something? Even if another button is pushed it would not power up any other drives until computer was shut off and different switch was selected. No more than 1 of the rear connetions powered at any time. Thank you everyone for any help you may give me!

 

Ok thanks for everyone's replies,
So this is what I'm trying to modify. It work great but lacks the ability to remove human error. It is a ssd drive sata power controller. It is designed so the you can power up to 6 drives at a time or in any combination. The problem is that if the computer is powered on with one of the drives and someone were to push one of the physical buttons it causes problems in windows and the bios. In my scenario Iwill never want more than one drive powered at time due to multiple operating systems and configurations that are on the drives. I don't want to do multiboot. The numbered 4 pin connectors are the output power to each drive with (Ground, 12vdc, Ground, 5vdc). Would like a solution that I could remove connections on rear of unit and install relay board something? Even if another button is pushed it would not power up any other drives until computer was shut off and different switch was selected. No more than 1 of the rear connetions powered at any time. Thank you everyone for any help you may give me!

Then the safest, simplest and, probably, cheapest solution is the one I proposed in post #5, using a low-cost micro-controller and 6 DPDT relays to switch 5v and 12v rails. Here I've extended it to show all 6 relays including an LED per channel (might be built-in to push button), to indicate which channel is selected. I've also added a 'master reset' ie 'all off' button, not essential as default is all off on power on. but switching off and on the main power supply unnecessarily is a well known cause of failure. Total cost of 5 pcb, inc 1 assembled PCB excluding push buttons and output connectors (type unknown) inc 8-15day shipping to US from JLCPCB around $17

Noting your comment about installing this after the existing board, then this will do it. Just replace push buttons 1 - 6 shown here with 10k resistors to the 5v output of connectors 1 - 6. Which ever goes to 5v first will activate respective relay until power off (or reset pressed).

 

Here's my version that uses a diode matrix, push button switches and magnetic latch relays.

The schematic shows only three switches. The same may be extended to six.

S1, S2 and S3 are momentary push button switches.

'NO' contacts of the relays K1, K2 & K3 are to be used for switching the voltage.

Freewheeling diodes for the relay coils and LED indicators are not shown.

-Nandu.

 

Here's my version that uses a diode matrix, push button switches and magnetic latch relays.

The schematic shows only three switches. The same may be extended to six.


S1, S2 and S3 are momentary push button switches.

'NO' contacts of the relays K1, K2 & K3 are to be used for switching the voltage.

Freewheeling diodes for the relay coils and LED indicators are not shown.

-Nandu.

Clever, the electromagnetic version of the flip-flop solution ealier, but similarly doesn't meet the later requirement that once selected all other presses are locked out until overall reset, does it?

 

Thank you for clarifying what you are actually trying to accomplish.

So, it sounds to me like what you really want is to allow switching the drive only when the PC is powered off, right?

I don't see any practical way to accomplish that by electronically modifying that device.

Here is a silly solution:

How about building a box with a door to access the front panel buttons. And the door is automatically closed whenever the PC is powered on.

 

Clever, the electromagnetic version of the flip-flop solution earlier, but similarly doesn't meet the later requirement that once selected all other presses are locked out until overall reset, does it?

Yes, Irving, it does not meet the later requirement that I have overlooked.

Many thanks for pointing it out.

Nandu.

 

I don't see any practical way to accomplish that by electronically modifying that device.

The MCU version has a couple of spare inputs and could facilitate that.

 

The MCU version has a couple of spare inputs and could facilitate that.

Yes, I didn’t mean the circuit was difficult, I meant that integrating it with the box was difficult. Those appear to be latching, multipole switches. I can’t see that any add on circuit involves less work than replacing the box completely.

 

Yes, I didn’t mean the circuit was difficult, I meant that integrating it with the box was difficult. Those appear to be latching, multipole switches. I can’t see that any add on circuit involves less work than replacing the box completely.

Quite, and sourcing matching connectors from old box to new board, and then for the new board to drive, adds a lot of expense. To be honest, I'd junk the innards of the box and replace it with the suggested circuitry, repurposing, if possible, those output connectors. Might be possible to remove the latching mechanism and repurpose the buttons too.