Continue to Site

Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

TXS0108PWR connected back to back

Status
Not open for further replies.

zainka

Full Member level 2
Full Member level 2
Joined
Nov 18, 2004
Messages
137
Helped
5
Reputation
10
Reaction score
3
Trophy points
1,298
Location
No(r)way
Activity points
1,659
Hi
I have a design requirement where two systems which have different signal levels (and which may vary from about 2.2V to 5V) are to be connected.
Thought I should solve this by using two TXS0108 from Texas, connected back to back (or more correct A-side to A-side).

Looking at schematic below. TXS0108 have 8 bidirectional channels, and A-Side must have lower voltage than the B-Side, according to datasheet, and both VccA is therefor tied to 2V in our design. I will never know which of the B-Side (B1 or B2 in the image below) that will have highest voltage, but each VccB has its own power source given by the system connected to.
What I call "the system" is a range of different boards (already existing) with I/Os but with different signal levels. This is thought to be a generic solution for all these systems to tie them together,

Image shows two TXS0108 channels connected together on the A-sides and with Rs in between to avoid possible short-circuits if B1 and B2 tries to drive simultaneous
In a normal situation, however, B1 and B2 of a given channel will be one input and one output.

I need a second opinion if this will work or if the TXS0108 are incapable to be used this way, where two A channels are connected together in the same power domain..

1653995235989.png
 

Hi,

I do not see a major issue with this approach, as long you are able to keep VccA on each board almost identical (and connect GND).

The question would be, what kind of signals are translated by the TXS0108PWR ICs. Is it a communication like SPI or more "static" signals? If a high frequency communication is maintained, you might consider probagation delay introduced by both ICs. If a SPI communication is used, one migth observe the CS line, to avoid shorting of the IO pins. You could also use one pin (if not all are used until now) in open drain mode with a dominant LOW signal level, to indicate the bus is already used by one board. You could also use an 8 to 3 decoder (e.g. [1]) to observe the current pin states.

But of course, this considerations depend on your planed operation, available (free) TXS0108 pins and if there is a MCU, which can be used by means of free GPIO pins as well as if the SW can be modified.

[1] https://www.ti.com/product/SN74LS148?keyMatch=SN74LS148&tisearch=search-everything&usecase=GPN

BR
 

Hi, one of the board uses JTAG, and we will reduce clock to avoid problems with propagation delay.
Both VccA's are tied together to same power. Yes we do connect GND.

74LCX245 requires a direction bit setting all channels to same direction. TXS0108 have individual direction pr. channel and autosens. We could have two 245's, one for each direction, but then again, for some configurations we have the situation that a bit may change direction.
 

Status
Not open for further replies.

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top