SN74AVC4T774: Starup Behaviour of AVX and AVC devices.

Ashvin Srinivas15

Prodigy 135 points

Part Number: SN74AVC4T774
Other Parts Discussed in Thread: SN74AXC4T774, TXU0304, TXU0104, TXU0204, TXB0304

This is a question for both the AVX and the AVC parts.

We are having the VCCA coming up first and the VCCB coming up after a few seconds. In the meantime, the direction pins are pulled low and high to set the direction.
The OE is pulled low from power on VCC_A.
We are seeing a high current being taken on the VCCA during this time. VCCB is a floating voltage around 0.6Volts in these initial few seconds.
If we ensure that the OE is pulled high for the time till VCCB is not present and then comes down low, will we see this high current draw again due to a floating input on the B side,
Or will the OE enable high take care of isolating the input on the B side till the OE is pulled low?

over 1 year ago

0 Michael Ikwuyum over 1 year ago

TI__Mastermind 40265 points

Hi Ashvin,

1. For AVC, ensure output supply do not ramp up before input.

2. This activates the outputs. Note, if outputs are enabled prior to inputs ramping up, glitches are common for AVC.

3. If the above two points are applicable, it could help explain the current / glitches.

4. AVC is not recommended to have floating inputs at any time.

5. Disabling the OE places all outputs in high-impedance. However, floating the inputs means unknown and not recommended for the device.

I would further recommend the AXC family with a more robust power-up sequence and with internal pull-downs on the inputs. Hence, no concerns floating. Please see SN74AXC4T774, thanks.

Best Regards,

Michael.

0 Ashvin Srinivas15 over 1 year ago

Prodigy 135 points

Hello Michael,

The SN74AXC4T774 has internal pull downs of 71K and a recommended pull up of 7K max. Is this 71K pulldown active under all operating conditions? Is there any situation in which the pull downs are disabled?

0 Ashvin Srinivas15 over 1 year ago in reply to Ashvin Srinivas15

Prodigy 135 points

If yes, then there would be significant current drawn from the external driver into these pulldowns. Would these pull downs be disabled, if the port is configured as an output, for example?

0 Clemens Ladisch over 1 year ago in reply to Ashvin Srinivas15

Guru 302700 points

The input buffers are always active. The current through these pull-downs is included in the input leakage current (I_I) specification of ±8 µA. (But this value is not consistent with a resistor of 71 kΩ.)

0 Ashvin Srinivas15 over 1 year ago in reply to Clemens Ladisch

Prodigy 135 points

As our design is not able to afford 8uA of current at each input, is there any other level translator that you can suggest, which will not require these pull downs?

0 Clemens Ladisch over 1 year ago in reply to Ashvin Srinivas15

Guru 302700 points

How much current can you afford? (Please note that these is the theoretical maximum at the most extreme temperature.)

TXU0104/TXU0204/TXU0304 have lower leakage currents.

0 Ashvin Srinivas15 over 1 year ago in reply to Clemens Ladisch

Prodigy 135 points

Could you please suggest in the auto direction and direction controlled versions?

0 Clemens Ladisch over 1 year ago in reply to Ashvin Srinivas15

Guru 302700 points

What are the two voltages, the number of signals, and their directions? How many signals go in the same direction?

0 Ashvin Srinivas15 over 1 year ago in reply to Clemens Ladisch

Prodigy 135 points

VCCA=2.45 to 3.3V

VCCB=1.8V o 3.V

Three from A to B and one back from B to A.

0 Clemens Ladisch over 1 year ago in reply to Ashvin Srinivas15

Guru 302700 points

For this application, the TXU0304 is ideal. Why do you want to use another device?

0 Ashvin Srinivas15 over 1 year ago in reply to Clemens Ladisch

Prodigy 135 points

The added advantage in the SN74AXC4T774 was the flexibility to change the direction if we wanted to. Thanks for the help.

This resolved my issue.

0 Ashvin Srinivas15 over 1 year ago in reply to Ashvin Srinivas15

Prodigy 135 points

Will the TXB0304 also be suitable of my application as it is auto direction sensing? It does seem to need external pull downs though of greater than 20 K.

0 Clemens Ladisch over 1 year ago in reply to Ashvin Srinivas15

Guru 302700 points

The TXB works best without any external pull-ups/-downs. It requires V_CCA ≤ V_CCB during normal operation, and allows a current to flow between them otherwise. It also has restrictions on the drive strength and the load of the other devices.

0 Ashvin Srinivas15 over 1 year ago in reply to Clemens Ladisch

Prodigy 135 points

We are testing now with the SN74AXC4T774 . Our power supply sequencing is as mentioned in the first post. We are seeing a large (140uA) current inrush during the time the voltage comes up on the VCC_B. Is this to be expected? and is there something we can do to not have this inrush? Our VCCB is generated from a TPS62745DDR.

0 Michael Ikwuyum over 1 year ago in reply to Ashvin Srinivas15

TI__Mastermind 40265 points

Hi Ashvin,

140 uA do not exceed the abs max rating. Hence, no concerns.

If concerned, please ensure the supply is at a known voltage and not floating around 0.6 V to help further mitigate the 140 uA current concerns, thanks.

Best Regards,

Michael.

0 Ashvin Srinivas15 over 1 year ago in reply to Michael Ikwuyum

Prodigy 135 points

We tried powering up both the VCCA and VCCB together but see that there is an initial inrush of current taken at that time as well. This is on both VCCA as well as VCCB. Is this to be expected then? We have a 100nF capacitors on each of the supplies, What abs. max rating are you referring to?

0 Michael Ikwuyum over 1 year ago in reply to Ashvin Srinivas15

TI__Mastermind 40265 points

Hi Ashvin,

Below is the absolute max rating:

Could you please help double confirm if current is 50 mA? Also, please help double confirm how the current is measured and if at power supplies or at I/Os, thanks.

Best Regards,

Michael.

0 Ashvin Srinivas15 over 1 year ago in reply to Michael Ikwuyum

Prodigy 135 points

The current is measured at the power supplies itself and not at any I/O pin.

I was asking about a specification which mentions how much (inrush) current can be taken from the power supplies at VCCA and VCCB at turn on.

We do not have any large loads or capacitors on the IO pins on either side A or B.

0 Ashvin Srinivas15 over 1 year ago in reply to Ashvin Srinivas15

Prodigy 135 points

Could it be because of the 71K pull downs that is present of the AXC? We did not see this problem with the AVC part.

0 Michael Ikwuyum over 1 year ago in reply to Ashvin Srinivas15

TI__Mastermind 40265 points

Hi Ashvin,

The pull-downs are for the I/Os, preventing them from floating.

For the specification when VCC supplies are turned on, please see 5.5 of the data sheet for ICCA + ICCB max for up to 40 uA max.

I presume, you are powering up the device and measuring 140 uA at both supplies with some output current, I.e. Io not equal 0 and not similar to the data sheet tested condition. However, 140 uA on the supplies would not be a concern as this do not exceed the data sheet’s and max rating, thanks.

Best Regards,

Michael.

Logic

Logic forum

SN74AVC4T774: Starup Behaviour of AVX and AVC devices.