Some background:
I have search a variety of forums and threads for something(simple) to accomplish this but I am thinking my problem is relatively unique. I am a jack of all trades with some 60s-70s Navy training in electronics. I am a hobby woodworker, pen-maker, knife-maker, electronics repairman, etc plus most anything else that is different. I just took up making knives and need to elector-chemically etch my logo into the blade which meant building a switchable power supply, DC to etch and AC to mark. Attache is a pic of the front of the unit and some screen prints of the inside circuitry. The circuitry pics are sort of combination wiring and traditional circuit diagrams(the work for me) but I think they can be followed. Basically I have a modded PC pwr supply with a buck circuit added to the +12v rail to give me 24v for "Etching". There is a 24vac CT bell xformer that gives me the AC for "Marking"(darkening the etch), it is controlled by an SCR to give me variable AC for other purposes. The power supply works great and I have already been using it.

The Problem:
The etching/marking output form my "utility" power supply is available at the banana terminals at the lower left. I want to add LED indicators at all the small black circles you see on the front panel. I have many of them working already and others will not be a problem however, indicating which supply, AC or DC is currently at the etching output with LEDs has me stumped. I have exerted the lower left section of the circuit diagram out to a more readable pic and have added one thought I had for a circuit but of course it doesn't work. It does however give folks the idea of what I am trying to do.

Originally I was going to light the LEDs with the left terminals of the ETCH/MARK switch and had it all together when I realized I needed to use that side of the switch to switch commons between the AC and the DC. So now I have the problem of how do I light the LEDs without a major circuit change(read that as major$$$)

I really enjoy tinkering with lots of different things so if some one thinks they might have an idea that "might" work. Let me know what it is and I will play with it and see what happens. I have no time restraints(retired) only $$$ restraints. ;-)

Thanks for looking/reading my post.

 

I don't see an editing feature for posts but I wanted to add that I am new to this forum, I am a member of several woodworking forums so I may or may not be using this forum correctly, if not someone please indicate what I am doing wrong and it will be taken care of in future posts.

As I mentioned earlier I have some nave training, I am a Vietnam era vet and have probably forgotten more about electronics then I have learned over the past 50 years. Just hit the big 70 this past july so don't be too hard on me. ;-)

Edit: Now I found the editing tool, good thing it wasn't something deadly.

 

Are you using Bulbs or LED's?

 

If your DC and AC supplies are completely isolated, just wiring one LED back to the DC GND and one back to the AC GND would suffice to do what you want. But if there's ever any voltage potential between them, you'll blow out LEDs.

You could do it pretty simply with a three-pole switch, and use one pole to switch a low-voltage supply to one of two LEDs, separate from the rest of the wand circuitry.

You could rectify and filter the want output and threshold it with a 12V zener to drive a 12V relay when the voltage got above 12-15V. Then you could use the relay to switch between the two LEDs as long as they were powered off the same rectified, filtered wand output so when the wand was off, both LEDs would be off.

Wand output --> bridge rectifier --> capacitor --> 12/25V tie point --> 12V zener diode ("backwards") --> 12V relay --> switch ground between two LEDs fed back to 12/25V tie point.

 

You can actually use a 2 lead bipolar LED like these" http://www.allelectronics.com/make-a-store/item/led-202/5mm-yellow-green-bi-polar-led-2-lead/1.html There are mounting clips that can be used for the panel.
The brightness might be different.

In any event, you can use a single RED/GREEN bipolar LED and ACwould be yellow and DC could be green or red. You can sise the resistor as (24-Vf)/50 mA assuming a 20 mA-50 mA nominal LED. RED+GREEN = Yellow. You can buy plugs to plug the other hole. Vf is usually between 1.8 and 3 V or so.

Otherwise, you end up with something messy Say a zero crossing comparitor folloed by a missing pulse detector for AC.
For DC you might have to use the logic of NOT AC AND V>(some voltage).

In your case you can also full wave rectify and filter and then compare the voltage. In one case you'll get 12-1.2 or so for DC and for ac you wil get a DC voltage of ~24*1.4-(2*0.6). Somewhat tricky, but nnot imposible. Then add some comparitors and glue logic.

Anyway, that's 3 approaches and 1 almost desiged.

EDIT:

You dont have room for a 3PDT switch. but you may have room for a switch and a 3PDT relay.
One pole does the LED thing and the other two poles do your switching. This a far easier option, but it doesn't tell you that there is AC or DC present. It just tells you the mode or there is "supposed to be AC" and "Suppoed to be DC" at the termials. Your relay would have to be AC/DC rated andd probably be a relay aux contact. Even without a 3PDT relay, you can use two: One for the AC/DC high current contact and one one SPDT contact for the LEDs. This relay could be tiny.

Some of the stuff above might require another supply.

 

The switch/relay combination won't really work because his switches are center-off.

The bipolar LED would require him to re-think his LED arrangement, and the panel is already cut and labeled.

My suggestion was to full-wave and rectify the voltage but simplify the voltage comparison by thresholding the voltage to a relay with a zener diode. Single-component voltage detection as opposed to comparators and glue logic.

 

Before addressing everyone's ideas please note that I made an oops in transferring info from the overall circuit diagram to the smaller version. I show +12vDC as an input to the top of the switch it is actually from the buck circuit and will vary from 1.5 to ~30vDC.

Are you using Bulbs or LED's?

Hi ISB123, thanks for posting. My intention is to use LEDs but I think I would have the same problem with bulbs. Most of the LEDs are already installed and working, the front panel pic is early on.

If your DC and AC supplies are completely isolated, just wiring one LED back to the DC GND and one back to the AC GND would suffice to do what you want. But if there's ever any voltage potential between them, you'll blow out LEDs.
You could do it pretty simply with a three-pole switch, and use one pole to switch a low-voltage supply to one of two LEDs, separate from the rest of the wand circuitry.
You could rectify and filter the want output and threshold it with a 12V zener to drive a 12V relay when the voltage got above 12-15V. Then you could use the relay to switch between the two LEDs as long as they were powered off the same rectified, filtered wand output so when the wand was off, both LEDs would be off.
Wand output --> bridge rectifier --> capacitor --> 12/25V tie point --> 12V zener diode ("backwards") --> 12V relay --> switch ground between two LEDs fed back to 12/25V tie point.

Thanks for posting Inspector. The first thing I will try is your "straight to commons" LED idea, that might just work, the PSs are isolated without having a common ground. As KISS in a post farther down noted, there is not enough room for a 3pole switch. BTW if you look at the original schem closely you will see that the 12.3 or 25.6vAC switch does just as you are suggesting. Someone also mentions using a relay, I may have room for that and I happen to have a relay, have to check its coil voltage though. As for your last idea I will have to line that out on some paper. My brain is too old to get a really good picture of what you are saying. ;-)

You can actually use a 2 lead bipolar LED like these" http://www.allelectronics.com/make-a-store/item/led-202/5mm-yellow-green-bi-polar-led-2-lead/1.html There are mounting clips that can be used for the panel.
The brightness might be different.
In any event, you can use a single RED/GREEN bipolar LED and ACwould be yellow and DC could be green or red. You can sise the resistor as (24-Vf)/50 mA assuming a 20 mA-50 mA nominal LED. RED+GREEN = Yellow. You can buy plugs to plug the other hole. Vf is usually between 1.8 and 3 V or so.
Otherwise, you end up with something messy Say a zero crossing comparitor folloed by a missing pulse detector for AC.
For DC you might have to use the logic of NOT AC AND V>(some voltage).
In your case you can also full wave rectify and filter and then compare the voltage. In one case you'll get 12-1.2 or so for DC and for ac you wil get a DC voltage of ~24*1.4-(2*0.6). Somewhat tricky, but nnot imposible. Then add some comparitors and glue logic.
Anyway, that's 3 approaches and 1 almost desiged.

EDIT:
You dont have room for a 3PDT switch. but you may have room for a switch and a 3PDT relay.
One pole does the LED thing and the other two poles do your switching. This a far easier option, but it doesn't tell you that there is AC or DC present. It just tells you the mode or there is "supposed to be AC" and "Suppoed to be DC" at the termials. Your relay would have to be AC/DC rated andd probably be a relay aux contact. Even without a 3PDT relay, you can use two: One for the AC/DC high current contact and one one SPDT contact for the LEDs. This relay could be tiny.
Some of the stuff above might require another supply.

Thanks for posting KISS, I will have to study on your first ideas for a while but I like the relay idea. As noted above I have a relay but not sure of the coil voltage. I will be looking at that first Thanks again for posting.

 

The switch/relay combination won't really work because his switches are center-off.
The bipolar LED would require him to re-think his LED arrangement, and the panel is already cut and labeled.
My suggestion was to full-wave and rectify the voltage but simplify the voltage comparison by thresholding the voltage to a relay with a zener diode. Single-component voltage detection as opposed to comparators and glue logic.

Hi again Inspector, I don't have a problem with both lights being off if the switch is in the center position(they were real cheap BTW couldn't find any on-none-on for similar pricing lol). If I use the existing switch, one set of contacts to operate the LEDs from my 12vDC supply like the AC switch, I can use the second set of contacts to activate the relay and just ignore the center off position. If I was to happen to leave it int the center position the relay would simply default to one of the lights... unless maybe I can spring for a latching relay... hmmm. Is there an emoj for thinking out loud?

Edit: I think I will run over to eBay or Amazon and check into latching relays.

 

Forget the latching relay. They're a pain in the butt.

I like that idea of running the LEDs directly off one pole of the switch and running a DPDT relay off the other pole of the switch.

You just need a relay whose contacts can handle the currents you'll be using, and it'll default to one or the other positions (ETCH vs MARK) even when the toggle switch is in the center. So the LEDs would follow the switch, and the relay would just leave the terminals connected to either the ETCH or the MARK circuits when the switch was in the off position.

You could always use two separate relays to achieve the center-off isolation (assuming you had room for the relays). Just use the N.O. terminals on both for the different Etch/Mark power sources, and the common terminals on both go to your Etch/Mark output terminals. The coils of the two relays would be connected to opposite terminals on your Etch/Mark toggle switch.

I think that your schematic shows the LEDs on the 12/25VAC switch being driven by 5V. I think you'd be better off driving the aforementioned relay(s) with 12V. Makes for more sure contact closure.

 

I'll post a couple of circuits late tonight after I get back home and have time to fire up my OrCAD.

 

Forget the latching relay. They're a pain in the butt.
I like that idea of running the LEDs directly off one pole of the switch and running a DPDT relay off the other pole of the switch.
You just need a relay whose contacts can handle the currents you'll be using, and it'll default to one or the other positions (ETCH vs MARK) even when the toggle switch is in the center. So the LEDs would follow the switch, and the relay would just leave the terminals connected to either the ETCH or the MARK circuits when the switch was in the off position.
You could always use two separate relays to achieve the center-off isolation (assuming you had room for the relays). Just use the N.O. terminals on both for the different Etch/Mark power sources, and the common terminals on both go to your Etch/Mark output terminals. The coils of the two relays would be connected to opposite terminals on your Etch/Mark toggle switch.
I think that your schematic shows the LEDs on the 12/25VAC switch being driven by 5V. I think you'd be better off driving the aforementioned relay(s) with 12V. Makes for more sure contact closure.

"Pain in the butt", Inspector, I will take your word for it, I don't have much experience with latchers and what I do have is 50 years of so old. Besides I saw that latch relays run about $10(dang whats with this $sign before the number convention of ours, I always seem to type the number first then have to go back and put the $sign in, what's wrong with 10$, but, I digress). I can probably pick-up 2 std 2PST relays for the price of one latch, back to more research.

I'll post a couple of circuits late tonight after I get back home and have time to fire up my OrCAD.

Thanks I will watch for them.

 

I like that idea of running the LEDs directly off one pole of the switch and running a DPDT relay off the other pole of the switch.

I like that idea too. I tend to thow out ideas no matter how bizzare and then start simplifying. It's actualy a good way providing you can convince your boss it is. The method was mentioned in a professional publication that I get.

And then... we find out the DC voltage is variabe. Should have second guessed that.

 

If your DC and AC supplies are completely isolated, just wiring one LED back to the DC GND and one back to the AC GND would suffice to do what you want. But if there's ever any voltage potential between them, you'll blow out LEDs....

Thanks Inspector, this idea worked great and I purposely connected the two commons together for an instant and the only thing that happened was both lights fire in both positions.

Thanks again, I will update my schem and post it here.

 

I am eager to see the revised schematic. How did you deal with the 1.5 to ~30Vdc issue mentioned in post #7?

 

I am eager to see the revised schematic. How did you deal with the 1.5 to ~30Vdc issue mentioned in post #7?

Well, he has fixed voltages available. What issue was it?

 

I am eager to see the revised schematic. How did you deal with the 1.5 to ~30Vdc issue mentioned in post #7?

Well, he has fixed voltages available. What issue was it?

Marcf/Inspector, I really appreciate the continued interest in my post. I just sat down to update my schem and saw your posts. A little more background here my help understand the situation and use of my PS. Yes the DC etching voltage can be varied but is typically between 12 and 24vDC. The only time I will be using anything lower then 12vDC(not etching, playing with some external gadget or something) it will come from the other side of the panel and the etch side of the panel will be off. This is a universal PS for my other hobbies as well, not just for etching.

I have attached a couple more schems showing the update and some other corrections(oopses) in the main schem. For example I am powering some of my LEDs from the +12vDC rail but the schem showed it coming from the +5vDC. The schem showed the etching wand on the positive side of the DC where it is actually on the negative side. The positive side goes to the work piece(a pic is attached of an etched knife showing the etch and purpose of the etch side).

Just to reiterate, I really appreciate everyone's ideas. I may not have used them for this but they provide me with some refreshment of my old electronics knowledge and my be useful on some later project. I also hope what is posted here might be helpful to someone just getting into electronic projects.
Many many thanks.

BTW some of the LEDs not currently connected may turn out to be troublesome, if so I will be back.