Author Topic: Datasheet question  (Read 377 times)

0 Members and 1 Guest are viewing this topic.

Offline SlimEddieTopic starter

  • Newbie
  • Posts: 6
  • Country: lt
Datasheet question
« on: August 13, 2024, 02:50:13 pm »
Hi guys,

I think either I am missing something or there is a mistake in a datasheet. I am looking for some one to give your thoughts on the meter.

I was in the process of replacing TI's GaN fets on a halfbridge and replacing isolated supply with a bootstrap. I ended up picking NV6128 GaN fets to replace LMG341xR070, as they have some excess features that we don't need and the price can be reduced.

The issue I have run into, I am driving the half-bridge from F28049C TI's MCU, and the PWM level is 0..3.3V, while the GaN fet driver accepts 0..5V (with 1V and 4V thresholds).
So by following the datasheet for NV6128 https://eu.mouser.com/datasheet/2/1451/NV6128_Datasheet-3076917.pdf, on page 15 Fig. 26 the manufacturer suggests the use of SN74LVC1G34 buffer. What got me confused is how can this buffer with a supply voltage of 5.5V max be used in the configuration that is shown in the attached image? and in terms of using it on high side, wouldn't connecting it to SK fry it because SK is connected to the Source which can be at quite a high voltage?

 

Offline selcuk

  • Frequent Contributor
  • **
  • Posts: 251
  • Country: tr
Re: Datasheet question
« Reply #1 on: August 13, 2024, 08:56:12 pm »
It is on Figure 22 in the datasheet and SK is not connected to PGND, Source or CP. For example in Figure 23 they are all connected and it doesn't have the buffer. You may read "8.4. Source Kelvin (SK) Ground Pins" part for further information. The reference ground for the PWM section becomes SK and not PGND for that application.
 

Offline SlimEddieTopic starter

  • Newbie
  • Posts: 6
  • Country: lt
Re: Datasheet question
« Reply #2 on: August 14, 2024, 01:28:55 pm »
Yes I do understand what you mean, as it is not connected externally, but as far as I understand there is still a direct connection to the source of the fet, albeit that the main source current doesn't flow out of the kelvin connection, wouldn't this pin be at the potential of a switching node (point between high side fet source and low side fet drain). Wouldn't this potentially create a high potential difference between the supply pin and ground pin of a buffer?
 

Offline selcuk

  • Frequent Contributor
  • **
  • Posts: 251
  • Country: tr
Re: Datasheet question
« Reply #3 on: August 14, 2024, 02:16:03 pm »
On the "simplified schematic" in the datasheet, it looks like there is an internal connection from SK to S. That is definitely a short. But check Figure 1 in the below document. S pins are shorted together but SK is not. I believe there is a similar internal connection here. Nevertheless, it is hard to decide that when SK and S are shorted on internal drawings in the datasheet.

https://www.onsemi.com/pub/Collateral/AND9932-D.PDF
 

Online dietert1

  • Super Contributor
  • ***
  • Posts: 2415
  • Country: br
    • CADT Homepage
Re: Datasheet question
« Reply #4 on: August 14, 2024, 03:16:13 pm »
S = Source
SK = Source kelvin connection, load current free
Certainly SN74LVC1G34 can't work in that circuit as its max supply is 5.5 V. Even VDZ = 6.2 V is a bit high. Its supply must be somewhere else.

Regards, Dieter
 
The following users thanked this post: SlimEddie


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf