The circuit I'm attempting to construct:

enter image description here

I'm constructing the circuit shown in the above image, but I'm experiencing strange behavior when I solder on the switches and test the circuit. The switches are normally closed and are meant to break the connections between the NOR gate inputs and ground. The NOR gates (NTE74S02 TTL gates) are arranged as a SR NOR latch that is meant to turn on the lamp when the switch attached to the SET NOR gate's input is opened and turn off the lamp when the switch attached to the RESET NOR gate's input is opened. The voltage source is 5V DC.

After soldering on the switches and connecting the voltage source, the lamp is meant to remain off until the SET switch is pressed. It is meant to turn off when the RESET switch is pressed. However, the lamp remains persistently on when the voltage source is connected, even after I press the RESET switch a few times. Here are pictures of the solder joints connecting the SET switch to the SET NOR gate:

enter image description here enter image description here

The circuit works as intended if pressure is applied to the completed solder joints on the contacts of the SET switch. This behavior suggests that the joints are cold joints, but use of a multimeter indicates that the joints are in fact connecting the switch contacts and the wires. I also made sure that I used flux on the contacts and the wires.

Has anyone else encountered a similar problem with solder joints?

  • \$\begingroup\$ what pictures?? Thanks for adding them \$\endgroup\$ – I. Wolfe Mar 15 '15 at 3:59
  • \$\begingroup\$ Sorry. Fixed the links. \$\endgroup\$ – 2PI Mar 15 '15 at 4:01
  • \$\begingroup\$ The link for the switches I'm using: amazon.com/Momentary-Panel-Mount-Button-Switch/dp/B001TL79JE/…. I know additional info should be added onto the post, but I don't have enough reputation to post more than two links. \$\endgroup\$ – 2PI Mar 15 '15 at 4:02
  • \$\begingroup\$ Those joints look pretty bad. I would resolder and after soldering clean off all that flux and stuff. Hard to tell with all the junk on the joints but looks like its just not a very secure solder job. \$\endgroup\$ – I. Wolfe Mar 15 '15 at 4:04
  • \$\begingroup\$ I'm pretty sure I cleaned the contacts and the wires I used. Do you think the added gunk was caused by the improper application of flux? \$\endgroup\$ – 2PI Mar 15 '15 at 4:05

The problem here is there is no input to the NOR gates when the switch is open.

Basically, when you have an undriven logic input, the input is undefined. This means it could be high, it could be low, it could toggle randomly as a function of local electrical fields. This is colloquially called a floating input, because the voltage on the input "floats" around as a function of local electromagnetic influences.

Since it seems like you want the inputs to be high except when the switch is closed, you need to add a resistor between the inputs and the positive rail, to pull the inputs into a known state when the switch is open. A few KΩ should be fine.

Ok, more troubleshooting. First, I'd stick pull-ups in there anyways. Relying on the internal behaviour of the input buffer makes me uncomfortable.

Next, rather then just wondering if your solder joints are bad, just measure them! You should see ~Vcc at the input from the switch when the switch is open, and ~Gnd when it's closed. This should be quite easy to test, even with a cheap multimeter.

  • \$\begingroup\$ I'm using TTL gates, whose inputs automatically assume a high state when left floating. Source: allaboutcircuits.com/vol_4/chpt_3/10.html, near the bottom of the page. \$\endgroup\$ – 2PI Mar 15 '15 at 4:07
  • \$\begingroup\$ Although I do agree in general, the asker did say that they were using a Schottky TTL chip which should treat open as high. \$\endgroup\$ – Ignacio Vazquez-Abrams Mar 15 '15 at 4:08
  • \$\begingroup\$ @IgnacioVazquez-Abrams - I looked at the datasheet, and it was a empty NTE datasheet. \$\endgroup\$ – Connor Wolf Mar 15 '15 at 4:09
  • \$\begingroup\$ Here's the link to the datasheet: alldatasheet.com/datasheet-pdf/pdf/140292/NTE/NTE74S02.html. \$\endgroup\$ – 2PI Mar 15 '15 at 4:09
  • 2
    \$\begingroup\$ Anyways, the point remains. You have inputs that have no external pull-up/down, to which you've connected what is basically a great long antenna. \$\endgroup\$ – Connor Wolf Mar 15 '15 at 4:13

1) You understand that the set switch needs to be open to operate the latch. Are you quite certain that your switches are NC (normally closed)? Those things are pretty rare. And the switch symbols you are using would ordinarily be used for NO (normally open) switches. I suspect you are using NO switches, and your latch is normally in the forbidden state - that is, with both outputs low. Activating the reset does nothing to the output your looking at, since the high input keeps that output low. I suspect that what's happening is that when you put pressure on the set switch contacts, your fingers are providing a low enough resistance to constitute a logic low input, and the set output goes high.

2) You have the operation of the latch backwards. If the set switch is opened, the input (particularly if you take Connor Wolf's advice and add 1k pullup resistors) will go high, and the output you are monitoring will go low.

3) TTL (Schottky or otherwise) is not happy driving grounded loads. You will do much better to tie your indicators to +5 rather than ground. This will mean that they turn on for a logic 0, but you can get around this by adding an unused gate configured as an inverter to drive the "lamp" - and I hope that you mean LED, not an incandescent bulb.

  • \$\begingroup\$ I checked the switches with a multimeter. I'm pretty sure that the switches are NC. As for your second point, I thought that the output of a SR NOR latch goes high when the set input goes high? (Source: ece.msstate.edu/~reese/EE3714/fflatch/img003.GIF).By the lamp, I meant LED. The software I was using (Oregano) to draw the circuit didn't have an LED symbol. \$\endgroup\$ – 2PI Mar 15 '15 at 8:12
  • \$\begingroup\$ Link doesn't work. Actually, your question about which output goes high is one I've dealt with before: electronics.stackexchange.com/questions/151717/… Basically, if you're using active high logic, you've misunderstood your latch, and the two outputs are swapped. \$\endgroup\$ – WhatRoughBeast Mar 15 '15 at 12:54
  • \$\begingroup\$ Fixed the link: ece.msstate.edu/~reese/EE3714/fflatch/img003.GIF \$\endgroup\$ – 2PI Mar 15 '15 at 17:30

First, if you apply finger pressure to a solder joint to make it work and the circuit gets flaky when you let go, there's either something wrong with the switch or the joint. I'd suspect the joint and resolder it, but first I'd make sure that the switches weren't flopping around when when I "fixed" the joint since that'll just make more trouble. I don't know how familiar you are with soldering, but it's absolutely imperative that the connection not move while the solder is cooling.

Second, - someone already mentioned this, but it's important enough to bear repeating - you should be pulling the manually switched NOR inputs up to Vcc, and sinking current, not sourcing it, through the LED; like this:

enter image description here

Take a look at the excerpt from TI's data sheet, below, and you'll see that the recommended maximum output current is 1mA, and at that current the output voltage can fall to 2.7V. As the output current is increased, the output voltage will keep falling and eventually won't be able to drive the latch input it's connected to properly and the circuit will fail.

enter image description here


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.