• Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!

Led Mod

Overclockers is supported by our readers. When you click a link to make a purchase, we may earn a commission. Learn More.

bluedude2288

Member
Joined
Sep 17, 2002
Hey, i have a dsl modem from ameritech, i want to do a mod of extending the led's cords so i can moubt them somewhere, but i can't open the modem, i was wobdering if there was somw circuit like with a phototransitor that i could place over the lights, and it would see them blinking and blink an led on the end of it
 
Sure there is !
I can scrounge up a drawing if you wish to have. It will be a very flexible unit, being that flexible it also require alot of soldering, so I hope you are ready for it. It will demand 5 or 12volts, maybe 5 resistors per channel(per LED you wish to observe) a op-amp gate per channel and of coz the photo transistors and output LEDs.
Intrested?

Another "way" is to set up cheap 2mm plastic opto fibre infront of the modem and route them to where ever you wish to see it. Only warning will be that the light will be a bit faint.
Send me a PM here on the forum if you wish me to work on it.
 
Paxmax said:
maybe 5 resistors per channel(per LED you wish to observe) a op-amp gate per channel and of coz the photo transistors and output LEDs.
Intrested?
:eek:

Wouldn't this be an easier way? (Maybe you can even leave out the BC547...didn't look for the TIL81 specs):
ModemLed.jpg


**NOTE: It doesn't have to be 5V, and you need to calculate the resistor, as it depends on the LED you wanna use.

Hope this helped.
 
KILLorBE said:
Wouldn't this be an easier way? (Maybe you can even leave out the BC547...didn't look for the TIL81 specs):

Hope this helped.

Nah, I don't think it's going to work that good. It's going to "strangle" itself. When the BC 547 opens, the voltagedrop over the TIL81 is going to be very very low, and subsequently it will not conduct any more uA to the base of BC 547. Try to improve it KILLorBEE, you are on the right way to solve it.

Though I think this solution is rather crude, it might work with refinement. I have used a similar solution for detecting red LED pulses coming out from energy meters(my daytime occupation).

Remember to include a resitor to define a split voltage to feed the base of the BC 547, split between TIL81 and a *new* resistor. This will also help the transistor to close faster.

One problem might be detecting blue, green and yellow leds with a photo transistor. Their light spectrum is usually far out of peak sensitiviety of a photo transistor, they are usually more centered around IR. Red LEDs can usually be detected. LOL! Be sure that the TIL81 hasn't got a daylight filter! If it has, it won't detect a s**t from red, blue, green leds.
 
bluedude2288 said:
thanks for all the info
Well, if thats enough and you got it to work, good ! I'll refrain from posting the more complicated solution. :)
 
no, if you have info and want to post it, go ahead, i haven't built it yet, i'm just waying out all the possibilities.
 
Umm... how do I upload pics?? I absolutely suck at forum techniques.
Anyone knows? /Paxmax
 
Paxmax said:
Umm... how do I upload pics?? I absolutely suck at forum techniques.
Anyone knows? /Paxmax

You could upload them at The Forum Is Down, copy the link and paste it between the image tags (I should look like this your link here[/img* Note: * should be ]), or you could just paste the link, or you could attach it when you reply (It's under the window in which you type your reply: Attach file....Browse).

[quote="Paxmax"][B]Nah, I don't think it's going to work that good. It's going to "strangle" itself. When the BC 547 opens, the voltagedrop over the TIL81 is going to be very very low........................

.....................they are usually more centered around IR. Red LEDs can usually be detected. LOL! Be sure that the TIL81 hasn't got a daylight filter! If it has, it won't detect a s**t from red, blue, green leds.[/B][/QUOTE]
I've build the circuit, but I couldn't get a TIL81 so I used a BPW40...it works fine...BUT it won't detect **** from ANY LED :( (should've mentioned that in my previous post:eh?: , You'll need a phototransistor that'll detect light with a wavelenght as low as ~550nm or lower (565nm for Green, 595nm for yellow and 625nm for red leds).

At least I can test IR remote controls with the circuit:p

BTW: I would love to see what you came up with Paxmax.
 
Last edited:
Ok, I'll try to see if I can get a JPG output of my CADint program. (Excellent electric cad program BTW, free *limited* copy available)
The circuit we use, is made for detecting LEDs and opto communication at 9.6k baud. I'll be back.... The recent version will detect a very faint red LED, I must test it for green and orange LED aswell before posting. (I just don't wanna post anything that doesn't work. Ok?) If the photo transistor can't detect green/orange I must redesign it for LDR(light dependent resistor) instead.
Also, where do you guys live? I live in Sweden, I'm guessing that the parts available might be different. What store would you like to purchase parts from? (might be good to know)
 
Auurgh... There is a hefty difference between a faint red LED and a faint green LED. When I measure the OHMs over the phototransistor I got(Infineon SFH300-3) I get like 3Kohms with RED light. Take your average green light and Whooa! we didn't get more than 120 K ohm! The yellow fell inbetween with 50 something K ohms. Just for your info, you'll need to optimize every LED sense circuit with different splitter resistors. I'll do some work on this circuit during my lunchbreak.
 
'key !
I got a design of some sorts ready now.
Ledsense1.jpg

Well, I still haven't checked out your electronics store, so I don't know which components to recommend. as I said earlier, I used Infineon SFH300-3 phototransistor and it detected even green, but with a heavy penalty. Also this and the other design posted will work differently during different ambient light conditions. Some ambient light makes it easier to open the phottransistors since you have a "base saturation" effect, but if the sun shines on your modem and you haven't sheiled it off at all your LEDs will shine up at once especially the green and yellow circuit thoose are super super sensitive and will go off at nothing. The phototrasistor I used has a "-3" marking, that denotes the sesitivity factor. "-3" is very sensitve and it doesn't get much better than that. I belive some photodiodes might be more sensitve than phototransistors, might be worth looking into.
As I mentioned before you can use LDR, those have a wider sensitivity range a bit beyond phototransistors. Thoose are however really slow in shifting and at theese low levels of energy it might take a sec or two to shift, I'm not so sure it'll work good for flashing LEDs.
 
Might wanna test the crude design aswell....
This is a refined version of KILLorBE 's design.

crudeLEDsense.jpg

It will be pretty sensitive as it is. Might wanna change the R1 to something even higher for green LED. Maybe 220K.
 
Last edited:
Back