Conceptual switch question

[Atmaweapon]

I think my last thread was too specific and poorly framed to be very useful, so here's hoping this is a more answerable question: Is there an easy way to have a switch lock out another switch?

For instance, say I have two push-button switches in parallel and only want one to be active at a time... is there a simple/cheap way to design that, or do I have to purchase a pre-made one?

 

[Mike_2545]

Do you remember the old push button car radios?

 

[Mike_2545]

How about a rotary switch,12 pole, wired in series with a push button?

 

[3v0]

I think if you visit a thrift store you can find just that sort of switch on some blenders. You did not say how many positions you needed.

3v0

 

[Dean Huster]

You can use momentary-contact pushbutton switches to operate small relays. N.C. contacts of relay A in series with the coil of relay B (and vice versa) provide the electrical lockout. It takes two poles for the control (one N.O. to seal the pushbutton and one N.C. for the lockout) and a third pole for the circuit you're trying to control.

The same could be done with digital logic.

And there's always someone here who'll suggest using a processor of some sort.

The old "push-push" switches (such as the type used for vertical control on the original Tektronix 465 and 475) are mechanical self-cancelling types and you used to be able to find similar ones by the droves in the surplus catalogs.


Dean

 

[Mike_2545]

Yes, Blenders! I knew that! Thanks 3vO, he might be too young to remember the old time radios, not to mention they might be hard to find. Geeze, blenders....

 

[3v0]

I like to use a uC for everything. My choice would be to use a uC with tactile switches and relays. But an old blender does not require programming

 

[Atmaweapon]

I do remember those old radios (and blenders) but I don't know if those mechanical throws will work here. I'm trying to allow for 14 different outputs (via 14 different NO SPST push buttons currently) that will supply current to different combinations of LEDs. I could program this I suppose but working with an IC seems like more trouble/cost than needed.

 

[Mike_2545]

The mechanical throws will work for leds. If there is some squirrel-y logic circuit maybe not, but pure leds, no problem.

 

[Sceadwian]

IC's seems expensive and troublesome only if you don't hunker down and learn how to use them. The wiring and expense of the switches alone is going to be more than the micro controller. The hard part is learning to use them, there is so much information (Nigels tutorials for one) on the net right now it's rediculous. It depends on how much and what kind of work you're willing to put into it. I'm sure you could do a purely physical switching setup, but it will be more complicated and less flexible than something made using micro controllers.

 

[Mike_2545]

Here is a 18 position rotary switch. Dial the led and have one push button on the common lead.

 

[3v0]

I do not think you are going to find a push button switch with 14 buttons.

Unless you can go with the rotary switch the uC is the way to go. It will also be the least expensive route.

The uC is under $5. Switches are maybe 20 cents each. Depending on how may leds per postion you may need 14 transistor or mosfets. If you can get by with a single LED the uC may be able to drive it directly.

The uC code is simple.

Time to tell us about the LEDs.

 

[Atmaweapon]

Thanks everyone. Mike, rotary switch is a great idea that I didn't appreciate fully at first glance - it'll work perfectly. I'm just going to get an 8-position rotary and have a SPST switch for left or right which will get me my 14 combos. Beautiful.

Thanks again.

 

[Sceadwian]

What you don't see, is the million combination's you'd get with a micro controller, being able to sequence each line independently at any given moment.

 

[Mike_2545]

I remember when I took the plunge and bought my first micro, I didn't know what to expect, AVR butterfly, I never could get anything to work. 100 bucks down the drain.
Then I found Basic stamps, love them, they do everything I want, and are easy to work with. I guess, if you are not comfortable potentially blowing good money on something you know nothing about, its safer sticking with what you know will work.

By the way, anyone want a cheap never been programmed AVR butterfly?

 

[Sceadwian]

Butterfly 100 bucks? Where did you get ripped off from?
They were sold at 20 bucks a piece. They WERE notoriously buggy, but they did work.
I didn't even spend that much on my STK500, and that's a full development platform for AVR's. Even the current Dragon development kit only runs 49 dollars (in stock) from Digikey, and they're USB programmable with full JTAG programming and debug support.

Basic stamps are a blight upon micro controller kind. Not because they're pic, but because they're a slow pic loading instructions from eeprom.

 

[Mike_2545]

I got the whole kit including; book, CD some other wires & bits, I guess the cost was around 65.00 now that I think of it. Still cant get past the first lesson in the book.

C Programming for Microcontrollers. Compiler errors or some such nonsense.

I know basic stamps are s-l-o-w, but when one thing won't work for you, find another way.

 

[Sceadwian]

I'm sorry Mike, but you definitely got ripped off, the documents for the Butterfly were fully available from the day it was released, including source code and schematics for it in PDF format. As far as a physical interface I think it had a build in RS232 converter.
If you can do it with a basic stamp, bless you. They're functional but not something I would recommend to anyone. For what they offer they're massively over priced, and even more massivly under functional.
I can do more with 50 cent AVR part than all but the fastest (read: expensive) of the basic stamp modules.

 

[Mike_2545]

I was for a short time looking around the net for some classes or some way to introduce total green cadets to the AVR or PIC. I found the price paid for the basic stamp includes a really good forum/tech support on any mundane tid-bit of misunderstanding. I also learned the basic language as a kid on my TRS-80, so that was not so foreign to me.

Believe me I would rather not pay 50 bucks for a micro, maybe someday I'll retry the AVR. Until then Basic is all I have...

 

[3v0]

Mircochip is selling the PICkit2 for $30 now that the 3 is out. I would either get one of these or a Junebug which has the tutor onboard.

Either way for about $30 to $50 you are up and running. You can use the student C18 compiler, BoostC demo or the swordfish basic demo version, all free. Any you can get help here. That counts big time.

I am not saying PICs are better then AVRs but if you want to get answers here PICs are the way to go. AVR questions are often answered with "you have a better chance of finding an answer on avrfreeks". Which is true.