LSF0108-Q1: LSF0108 connection b/w Vref_B and EN

Ryuuichi machida

Guru 12685 points

Part Number: LSF0108-Q1
Other Parts Discussed in Thread: LSF0108, LSF0102

Hi,

Let me confirm about below.

According to datasheet, there is following description in section 9.2.1.2.1 Bidirectional Translation

"the EN input must be connected to Vref_B"

It seems that EN pin should be connected to Vref_B in this device. But, if user want to control EN function and Vref_B separately, is it available ?

If you have the reason which EN pin should be connected to Vref_B, could you please tell me ?

(For instance, To avoid device is broken.)

Best Regards,

Machida

over 8 years ago

0 Clemens Ladisch over 8 years ago

Guru 317000 points

I've looked into Voltage-Level Translation With the LSF Family (SLVA675), but that description isn't any better.

Anyway, it shows the device's internals, and EN just controls the gates of the pass transistors. So you certainly can disable the translator by driving EN low.

I guess what "the EN input must be connected to Vref_B" actually means is that

EN must not be connected to Vref_A, and
EN must be high for the translator to work.

0 Emrys Maier over 8 years ago

TI__Guru* 97046 points

In order to translate with this device, the EN pin must be tied directly to the Vref_B pin. Additionally, this voltage node must be allowed to float during normal operation -- ie the voltage is set by the bias current through the 200kohm resistor, not by a device driving the pin directly. Here is a common method for adding a disable to this device (this example is with the LSF0102, but the LSF0108 is functionally identical, just with more channels):

You can see above that the voltage node at the EN pin is forced low when the nFET is turned on (control = high), but then is allowed to 'float' to the appropriate bias voltage when that nFET is turned off (control = low).

0 Chuchen Wang over 7 years ago in reply to Emrys Maier

TI__Expert 3470 points

Hi Emrys,

I have similar question from my customer. Per datasheet, a 200kohm pull-up resistor is required to allow Vref_B to regulate the EN pin. My customer has released a PCB with Vref_B directly connected to power. Can you elaborate on if this would cause any issues?

Thanks,
Chuchen

0 Emrys Maier over 7 years ago in reply to Chuchen Wang

TI__Guru* 97046 points

Hi Chuchen,
Removing the 200 kohm pull-up resistor will result in unregulated current flow from VrefA to VrefB, definitely burning a huge amount of power and probably causing a thermal failure (fire).

0 Chuchen Wang over 7 years ago in reply to Emrys Maier

TI__Expert 3470 points

That wouldn't be good. Thanks for the quick response Emrys.

0 Emrys Maier over 7 years ago in reply to Chuchen Wang

TI__Guru* 97046 points

Always happy to help. Keep an eye out for our LSF family series of training videos. I have completed 5 of them so far and will be releasing the whole set within the next few weeks.

0 Chuchen Wang over 7 years ago in reply to Emrys Maier

TI__Expert 3470 points

One more question from the customer:

It looks like for the 1.8V side for I2C/SPI/GPIO no pull up is needed. However for the 5.0V side, if the signal is bidirectional or if the 1.8V IC is outputting a digital signal to A1, B1 will need a pull up to 5.0V.

Please confirm the below diagram.

Thanks,

Chuchen

0 Chuchen Wang over 7 years ago in reply to Chuchen Wang

TI__Expert 3470 points

Hello,

Any update on above?

Thanks,

Chuchen

0 Emrys Maier over 7 years ago in reply to Chuchen Wang

TI__Guru* 97046 points

I2C is an open-drain standard, and as such needs pull-up resistors on both sides of the LSF device. I would recommend large values on the 3.3V side (4.7k to 10k are typical values).

The other 3.3V lines should be fine without pull-up resistors.

0 Chuchen Wang over 7 years ago in reply to Emrys Maier

TI__Expert 3470 points

Thanks Emrys. Did you mean the other 1.8V lines should be fine without pull-up resistors?

0 Emrys Maier over 7 years ago in reply to Chuchen Wang

TI__Guru* 97046 points

My apologies, I didn't realize this was a 1.8V down-translation.... Vref_A needs to be set to 1.8V instead of 3.3V, then it will work as I mentioned.

0 Chuchen Wang over 7 years ago in reply to Chuchen Wang

TI__Expert 3470 points

Hi Emrys,

Couple more questions from the customer:
1. But you can confirm that for the SPI interface only the MISO requires a pull up on the 5.0V side?
2. Also for GPIO signal output from the 1.8V side requires a resistor pull up to 5.0V on the 5.0V side of the LSF device?

Thanks,
Chuchen

0 Emrys Maier over 7 years ago in reply to Chuchen Wang

TI__Guru* 97046 points

Yes, both of those question-statements are correct.

There are 3 cases in which a pull-up would be required on the 1.8V side of the LSF (in a 1.8V to 5V translation setup)
1. The 1.8V driver is open-drain.
2. The LSF is setup for a lower threshold, such as 1.2V.
3. The 1.8V input has a large leakage value (>1uA typical)

For your application on the SPI and GPIO lines, I believe that none of these cases are true, so a pull-up to 1.8V is not required.

0 Emrys Maier over 7 years ago in reply to Emrys Maier

TI__Guru* 97046 points

I'm happy to announce that we have officially released a series of 8 videos that explain how to use the LSF family of translators in detail. These videos are part of a new training series called "The Logic Minute" and are available here:

The Logic Minute

Please let me know if I can be of further assistance.

0 Emrys Maier over 7 years ago in reply to Emrys Maier

TI__Guru* 97046 points

I updated the link above. It turns out that the link I originally posted was for internal use only - sorry about that!

Logic

Logic forum

LSF0108-Q1: LSF0108 connection b/w Vref_B and EN

The Logic Minute