SN74LVC2G74: Wrong activity

Shu-Cheng LIN

Part Number: SN74LVC2G74
Other Parts Discussed in Thread: SN74LVC1G17, SN74HCS74-Q1, SN74AUP1G97, SN74LVC1G97, SN74HCS08-Q1

Hi Sirs,

Sorry to bother you.

We have use 2G74, schematic as below

We meet the output Q had wrong active. We tried remove D3, R4, R384. But still have wrong activity

Could you help check it?

Waveform CH1 is pin5 Q, CH2 is pin1 CLK

1. Not ready the positive pulse then to change state.

2. Put switch button to make positive pulse, but Q didn't have any activity.

3. Not change state on positive pulse.

4. Not change state on positive pulse.

over 4 years ago

0 Clemens Ladisch over 4 years ago

Guru 244360 points

The PULSE_CON1 signal bounces, i.e., the edges are not clean. If you tell the oscilloscope to use a higher sample rate and zoom in, you will see that there are multiple rising and falling edges.

Mechanical buttons must be debounced. When done in hardware, usually with a low-pass filter; see, for example, IC that converts momentary push button switch to maintain switch.

0 Dylan Hubbard over 4 years ago in reply to Clemens Ladisch

TI__Mastermind 42670 points

Hi Shu-Cheng,

Please refer to this training video on how to properly use a DFF for button toggling:

https://training.ti.com/convert-momentary-switch-toggle-switch?cu=1134826

You will need a debounce circuit and Schmitt-triggers. You can add external schmitt-triggers like the SN74LVC1G17 or a device with internal Schmitt-triggers like the SN74HCS74-Q1. The CLR pin will also have issues since there is a large RC circuit tied to it. The slow rise time can cause damage to the device.

0 Shu-Cheng LIN over 4 years ago in reply to Dylan Hubbard

Guru 11785 points

Hi Sirs,

Thanks for your reply

Could you help check

Does TI have AND Gate solution which include schmitt-triggers function?
About The CLR pin will also have issues since there is a large RC circuit tied to it., have suggest value on RC?

0 Clemens Ladisch over 4 years ago in reply to Shu-Cheng LIN

Guru 244360 points

1. SN74AUP1G97 or SN74LVC1G97

2. If you drive the CLR pin with a slow signal, you must put a Schmitt-trigger buffer in front of it. (You can also configure a '97 device as buffer.)

0 Dylan Hubbard over 4 years ago in reply to Clemens Ladisch

TI__Mastermind 42670 points

Hi Shu-Cheng,

We are actually coming out with a new family of logic devices that will have integrated schmitt-triggers on all inputs. Along with the SN74HCS74-Q1, we have the SN74HCS08-Q1 which is an AND gate with schmitt-trigger inputs.

Since the device has a transition rate requirement of 10ns/V max (@3.3V), really any practical RC would violate this. Therefor, as Clemens said there must be a schmitt trigger there.

I'd recommend replacing the NC7S08 with the SN74HCS08-Q1. Since it contains four 2-input AND gates, you can use one of the remaining three as a Schmitt-trigger buffer for the DFF CLR pin. Then you have two more ANDs to use maybe in other spots where an AND gate or a schmitt trigger buffer is needed.

0 Shu-Cheng LIN over 4 years ago in reply to Clemens Ladisch

Guru 11785 points

Hi Sirs,

Thanks for your reply.

About your reply of CLR pin,

We currently have no control over the CLR pin.
At the beginning, RC only wanted to make the initial value of the output LOW.
So if i changed it to 1k, and 0.1u, have any risk?

0 Clemens Ladisch over 4 years ago in reply to Shu-Cheng LIN

Guru 244360 points

LVC requires at most 10 ns/V at 3.3 V, so you'd have to design your RC to be no longer than 33 ns, which would be pretty much pointless.

0 Shu-Cheng LIN over 4 years ago in reply to Clemens Ladisch

Guru 11785 points

Hi Sirs,

Thanks for your reply.

We have one more question that My q# direct connect to d, looks it's followed application schematic but why IC didn't work.

It's seems need to have add resistor between them then IC will be work, I tried a few times, also changed the board. looks have same phenomenon
So can you help to explain this phenomenon?

Thanks!!

0 Clemens Ladisch over 4 years ago in reply to Shu-Cheng LIN

Guru 244360 points

In theory, connecting Q directly to D should work.

Are U24_QN(_R) connected to anything else?

0 Shu-Cheng LIN over 4 years ago in reply to Clemens Ladisch

Guru 11785 points

Hi Sirs,

Thanks for your reply.

As shown in the figure, U24_QN I think it should refer to the pin3.
At present only string resistor connected to the pin2.

It has not connect to anywhere.

In other word, is there any risk on our application if we add 1k ohm?

Thanks!!

0 Clemens Ladisch over 4 years ago in reply to Shu-Cheng LIN

Guru 244360 points

I do not see why it is needed, but it has no risk.

Logic

Logic forum

SN74LVC2G74: Wrong activity