Double flash strobe circuit problem?

Ask and answer questions, share tips and resources for installing lighting and other electronics in your models.

Moderators: Sparky, Moderators

Post Reply
User avatar
Often wrong Soong
Posts: 291
Joined: Thu Oct 23, 2008 11:50 am
Location: Northamtonshire, UK

Double flash strobe circuit problem?

Post by Often wrong Soong »

I need help from you experts!! I've built up a circuit on a breadboard to get the double strobe flash for the JJprise using a schematic I downloaded a few years ago, but it's not doing what I expected. Here's a link to the schematic:

http://i546.photobucket.com/albums/hh42 ... 4f4935.jpg

I'm using all of the proper components except for C1 - it says to use 1uf 10v, but I only had 1uf 50v. I doubt that is the problem but I'm not sure.

The problem I'm having is this: the LED is lit all of the time and has 2 peaks of brightness in the cycle instead of the usual double strobe (unlit, then 2 quick flashes)

I don't know enough about reading circuit diagrams to know if the circuit is doing what it should, or if I've got something connected incorrectly.

Help!! :oops:

Wayne
User avatar
tetsujin
Posts: 2350
Joined: Mon Feb 28, 2005 6:08 pm
Contact:

Post by tetsujin »

A higher voltage rating on a cap is fine: the voltage rating is just how many volts you can put across the terminals before it explodes.

It's hard to diagnose just what's going on with the circuit, but it helps to understand how the circuit is supposed to work:

The 555 is being used to generate a clock signal - at certain regular intervals it generates pulses which trigger the other ICs to change their state.

The 74590 is a binary counter. Each time it gets a clock pulse in, it increments the count on the Qa-Qh lines. (It's a binary counter, so it goes 00000001, 00000010, 00000011, etc.) Four of its outputs are being used: Qe - Qh.

Qh Is the highest-order bit of the counter. Its state changes only once every 128 clock pulses. It's being used to turn the 744051 off for half the time. The three lower-order bits are used to control S0-S2 of the 744051.

744051 takes a three-bit binary number in (at S0, S1, S2) and uses that to decide which of eight lines (A0-A7) to connect to another pin, called A in this diagram. In this circuit, A is wired to positive voltage, and A0 and A2 are wired to the transistor which turns on the LED. So the LED should turn on when (S0, S1, S2) = (0, 0, 0) or (0, 1, 0). If the 555 were running at 256 cycles per second, then the LED would be on for 1/8 second, off for 1/8 second, on for 1/8 second, off for 5/8 second and then repeat.

I don't really understand from your description what's going wrong with the circuit's behavior. I think one thing to check would be that transistor at the far right: if you got gate and drain reversed, for instance, then current would always flow through the transistor, and the output of the 744051 would probably change the LED's brightness a bit but not turn it on and off. The pin layout of transistors varies a lot, so make sure when you're checking it that you've got the right pinout for the transistor you're using.

(EDIT): Incidentally: I don't know why someone would design a circuit like this (clock + binary counter + binary decoder) when you could do this with two ICs instead of three by using a decade counter like the 4017. But for that matter, you could do this with a microcontroller and then you'd only have one IC. :)
Last edited by tetsujin on Mon Oct 07, 2013 12:27 pm, edited 2 times in total.
---GEC (三面図流の初段)
There are no rats.
The skulls eat them.
User avatar
Often wrong Soong
Posts: 291
Joined: Thu Oct 23, 2008 11:50 am
Location: Northamtonshire, UK

Post by Often wrong Soong »

tetsujin wrote:A higher voltage rating on a cap is fine: the voltage rating is just how many volts you can put across the terminals before it explodes.

It's hard to diagnose just what's going on with the circuit, but it helps to understand how the circuit is supposed to work:

The 555 is being used to generate a clock signal - at certain regular intervals it generates pulses which trigger the other ICs to change their state.

The 74590 is a binary counter. Each time it gets a clock pulse in, it increments the count on the Qa-Qh lines. (It's a binary counter, so it goes 00000001, 00000010, 00000011, etc.) Four of its outputs are being used: Qe - Qh.

Qh Is the highest-order bit of the counter. Its state changes only once every 128 clock pulses. It's being used to turn the 744051 off for half the time. The three lower-order bits are used to control S0-S2 of the 744051.

744051 takes a three-bit binary number in (at S0, S1, S2) and uses that to decide which of eight lines (A0-A7) to connect to another pin, called A in this diagram. In this circuit, A is wired to positive voltage, and A0 and A2 are wired to the transistor which turns on the LED. So the LED should turn on when (S0, S1, S2) = (1, 0, 0) or (0, 1, 0). If the 555 were running at 256 cycles per second, then the LED would be on for 1/8 second, off for 1/8 second, on for 1/8 second, off for 5/8 second and then repeat.

I don't really understand from your description what's going wrong with the circuit's behavior. I think one thing to check would be that transistor at the far right: if you got gate and drain reversed, for instance, then current would always flow through the transistor, and the output of the 744051 would probably change the LED's brightness a bit but not turn it on and off. The pin layout of transistors varies a lot, so make sure when you're checking it that you've got the right pinout for the transistor you're using.
That's a really great explanation of the circuit, thanks Tetsujin :D

I thought it might be something to do with the transistor so I'll check that out. It's doing exactly as you said, the led is pulsing rather than flashing.

Thanks buddy :8)

W
User avatar
opal_1970
Posts: 57
Joined: Mon Oct 31, 2011 4:17 pm

Post by opal_1970 »

tetsujin wrote:But for that matter, you could do this with a microcontroller and then you'd only have one IC. :)
I totally agree with you there, I use the Arduino board (30$) to create the code and then export it to a cheap AT-tiny microcontroller (1$ -2$).
Being that you code what you want, the sky is really the limit regarding the type or amount of blinking patterns.
-------------------------------------------
"I honor those who have gone where I have yet to go, who have seen what I have yet to see; for they are the true pioneers. May your journey be an adventure." -- Nomad farewell to the dead


Current build: TOS Galactica (Take 2)
User avatar
Often wrong Soong
Posts: 291
Joined: Thu Oct 23, 2008 11:50 am
Location: Northamtonshire, UK

Post by Often wrong Soong »

Yeah microcontrollers are the way to go, no question there. I need to find the time to go through the learning curve and get away from complex circuits.

The transistor was the problem in the end, I swapped it out for another and it worked fine...must have had a bad one in the bunch or something. The circuit works fine now but takes up a bit of room when assembled, good job the model is a decent size.

Thanks Tetsujin :8) I'm off to dive into microcontroller theory!!

Wayne
Post Reply