“lacks the ability to remove human error”.

Which human is causing the error?

 

I am guessing that changing drives while the computer is operating is rather disruptive. It will probably lead to loss of data and software, even. That almost calls for a mechanically interlocked selector scheme that requires a shut down and reboot to make the change. The alternative could be to have all of the drives active and connected at all times, and a hardware controller system to select one of them.
BUT clearly we do not have enough of an understanding to provide a workable suggestion.
We need to understand the purpose of the multiple drives as well as the frequency of changing the selections. Presently we are "in the DARK"

 

Actually, though it took some teasing, I think the TS's requirement is clear to me - you do have read his posts carefully and put yourself in his shoes, so to speak.

For reasons of his own, he has 6 boot drives on his PC, probably 6 build environments he wants to be completely separate,no multibooting.

The scenario is simple: he comes to his PC, selects his environment (drive) and powers up the PC. After that, no pressing of buttons will change anything until he powers off the PC, so he can safely leave it running while attending to other things.

The solution I proposed meets that brief perfectly at minimum component count and cost. If I was building it for myself from scratch I'd ditch the relays in favour of proper high-side switches for reliability which would give current monitoring on the 12v rail and could offer some potential early warning on drive failure. I'd also consider swapping out the push buttons for simple touch pads and using a touch IC, or an MCU with touch inputs, as mechanically and financially easier.

The TS made the classic error, because he knows the problem domain so well, of focussing on the solution rather than describing the problem well.

While it can be frustrating here (on both sides), I'm glad of this human trait. I made a lot of money as a consultant helping clients articulate their requirements clearly in small, well formed, testable, user stories which could be drip-fed to the development and Engineering teams. It requires the ability to put yourself in the client's head and see it from their viewpoint.

 

Why make it more difficult than necessary.
Use six of these lockable ON/OFF toggle switches.



You have to make a conscious effort to screw up.

 

Why make it more difficult than necessary.
Use six of these lockable ON/OFF toggle switches.

View attachment 323122

You have to make a conscious effort to screw up.

Suppose you have small children in the house and there's a delivery at the door...

That could be a few hours work lost...

Think outside the box - excuse the pun - but the existing box does that already with a latching push button. .

 

Suppose you have small children in the house and there's a delivery at the door...

That could be a few hours work lost...

Think outside the box - excuse the pun - but the existing box does that already with a latching push button. .

If you have small children then your house is already children proof, or it should be.

 

There's no such thing! You've not met my kids , though kids no more... If there was a way, they'd find it!

 

For drive selection at "switch on", guards for toggle switches have been available for many years, and that would be one fairly secure option. In fact, a rotary selector switch with a cover that was mechanically held latched when the power switch was in the "ON" position would be a failure-proof scheme requiring no electrical circuit additions. And unlike relays or electronic circuits, it would be immune to those short transient interruptions that cause problems in my location. In addition, it would be rather resistant to little fingers playing where they should never be,

 

Good idea. Use a rotary switch but omit the knob. Almost child proof.

 

Use a rotary switch interlocked with the power switch so that it can only be rotated when the power switch is in the OFF position. That could be as simple as a skirted knob on the selector switch that has notches for the toggle of the power switch. No intermediate moving parts to add to the complexity.

Certainly this would be a huge change from what the original post implied the issue was. An excellent example of why the question offered in the first TS posting should contain more information.
But I do like my concept of using JK flipflops for a first response detection scheme. That is vastly less complicated than some of the versions I have designed in the distant past. Why didn't I think of this long ago??