1
\$\begingroup\$

Consider a simple latch (two cross coupled inverters). The positive feedback makes the circuit operate in either of the two stable states if the overall gain is greater than unity. If there is a trigger that pushes the operating point beyond the switching point or pulls it down below the point, the output changes its state. Is the only difference between this and a schmitt trigger is that there are two trigger points in a schmitt trigger?

How do these circuits differ in operation in terms of positive feedback?

\$\endgroup\$
6
\$\begingroup\$

They differ in the level of the signal that is required to change their state.

In a Schmidt trigger, a relatively small, well-controlled change of signal level is needed. The input is designed to accomodate this signal, and the input current will be relatively low. In the case of an HC14 supplied with 5v, the hysteresis, the difference in switching levels, is about 1 volt, and the current peaks up to about 400uA when switching.

When two inverters connected together, they can be forced from one state to the other and back again by driving one of the inputs high or low. However, to do this, the input voltage must be taken to the logic thresholds of the gate, and the input current must 'fight' the short-circuit output current of the other gate.

You can make two inverters into a 'tamer' schmidt, by using a resistor to feedback the logic state from the second gate back to the input. This reduces the input current as it only has to supply the current through the resistor.

\$\endgroup\$
3
  • \$\begingroup\$ Thank you! Is it a good idea to place a capacitor in feedback instead of resistor to prevent loading? \$\endgroup\$
    – JGalt
    Jul 25 '16 at 15:19
  • \$\begingroup\$ It's a good idea to parallel a small capacitor with the feedback to improve the sharpness of the switching. It's necessary to keep the resistor in place as well, for it to work at DC. \$\endgroup\$
    – Neil_UK
    Jul 25 '16 at 18:20
  • \$\begingroup\$ Especially at DC, wouldn't just a capacitor work fine? It would prevent current flow from output of second inverter to the trigger source. \$\endgroup\$
    – JGalt
    Jul 27 '16 at 12:52
-2
\$\begingroup\$

Schmidt triggers and latches are two different electronic components or topologies both of which utilize positive feedback. An automotive analogy might be a pick-up truck and a station wagon. Both are designed to carry cargo and be driven by a human being. Yet, both play somewhat different roles and serve different situations in the automotive world.

The biggest difference between the Schmidt trigger and a latch is that the ST has one input, while a latch has two inputs (Set & Reset). Also, the ST is basically an analog device. It has an analog input which is responsive to non-logic voltage levels (the high-going and low-going threshold voltages), and a logic/digital output (it has only two stable states: saturated to the low supply rail or saturated to the high supply rail).

On the other hand the latch is a purely digital device. Both inputs are strictly logic level signals and the output produce a logic level.

So, yes, they are similar in that they both use positive feedback. However, they are used for quite different purposes by the designers of electronic circuits and systems.

\$\endgroup\$
10
  • \$\begingroup\$ Your idea that a latch has two inputs is not true. This only counts for the simplest SR latch. There are also D ant T type latches which have only one input, JK has two but enables you to use a state where both inputs can be high. \$\endgroup\$ Jul 24 '16 at 18:44
  • \$\begingroup\$ You cannot really make a statement like 'latches have set and reset'. Using this as your analogy is just going to confuse someone. You can create a latch with just 2 inputs and 1 output (1 input = data, 1 input = clock, 1 output = out). Extra inputs are just that, extra. \$\endgroup\$
    – jbord39
    Jul 24 '16 at 19:03
  • \$\begingroup\$ Arturas: Please give me a part number (e.g. 74HCT ) for a latch which has only one input. I guess you would call it a "latch once latch" - kind of like a write-only memory chip. \$\endgroup\$
    – FiddyOhm
    Jul 24 '16 at 19:39
  • 1
    \$\begingroup\$ @FiddyOhm: it's fine that you don't understand the internals of these devices, but I don't see why you come answer incorrectly to just confuse another person \$\endgroup\$
    – jbord39
    Jul 24 '16 at 19:59
  • 1
    \$\begingroup\$ Link to 8 bit D-latch. Note it is not a D-flip flop. nxp.com/documents/data_sheet/74HC_HCT373.pdf \$\endgroup\$
    – jbord39
    Jul 24 '16 at 22:16

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.