AMB Laboratories DIY Audio

[mcfrede]

Don't connect the LED directly across the power supply and ground! You need the current limiting resistor.

Oh, sorry, I did not mention the resistor, I would never connect it directly! I guess I am being a bit narrow-sighted on the main problem.

[linux-works]

what I'm saying is that if you have a set of contacts that go 'opposite' of how you want the led to come on, reversing the + and - on the led, reversing where its 'terminal' lead goes and which is its 'turn on' lead; that will invert its operation.

I also posted a pic a week or so ago about a simple transistor inverter using an NPN. if you pull the 'output' from the collector, you are inverting. if you pull the output from the emitter, it will not invert.

it would be worthwhile building that on a bread board, just for learning.

[amb]

If you look at the schematic of the ε24, it already has such a thing -- a relay that switches the polarity to reverse-parallel connected LEDs.

[mcfrede]

what I'm saying is that if you have a set of contacts that go 'opposite' of how you want the led to come on, reversing the + and - on the led, reversing where its 'terminal' lead goes and which is its 'turn on' lead; that will invert its operation.

I also posted a pic a week or so ago about a simple transistor inverter using an NPN. if you pull the 'output' from the collector, you are inverting. if you pull the output from the emitter, it will not invert.

it would be worthwhile building that on a bread board, just for learning.

I found it here and I have even actually bothered to try and draw the schematic as well as read up on transistors and actually trying to understand your simple circuit.

I will soon move on to try and locate where on the Epsilon24 I can connect the Base of the transistor as I reckon that is the key to succes (I do hope I will be able to find the 12VDC and GND). Alas, understanding the Epsilon24 schematic (though undoubtedly educating) is a somewhat more exhaustive exercise for me. Oh, and it is bedtime here in Denmark now...

As luck will have it I seem to remember that I have a some excess transistors and I know I have a variety of resistors lying about.

I hope to get back to you tomorrow with a nice picture of how I intend to wire it all up and then you can tell me if I have passed the examn!

[mcfrede]

So, I have had a busy day so I haven't made much progress, but here is my take on the schematic based on the picture by Linuxworks of his simple inverter.



I hope I have gotten it right thus far. I am aware I will only need one of the LED's, which one I will find out when I figure out the schematic for the Epsilon24.
I think I have figured out where I can tap into 12V and GND.

I shall get back!

[linux-works]

I think I powered mine from 5v, not 12. you'd want to increase the collector resistor to 10k or something like that.

you don't need the back to back LEDs, and the 500ohm was experimental based on brightness and also assuming 5v on the B+ side.

since you don't need the 2 LEDs, and more importantly, you don't need any middle off state, I would tie the single LED in series with its current limit R (a few hundred up to 1k) from collector to B+. the transistor is a switch and its C to E path is going to be open or closed. when you send in enough voltage to the base, the transistor 'switches on' and that CE junction goes to low resistance and you get a 'circuit' there. that's what we want to use, use the fact that, at the C point, you now have a low R path to ground. connect up a led from 5v on one end and then a limit resistor (don't want to burn out the led) and then go to the collector.

the 1k on the base is even very flexible on its value. you can use higher values as long as the transistor still switches on. don't go any lower than 1k, though, as a generic starter rule.