Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronic 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.

Register Log in

2:1 Multiplexer with open drain Outputs IC

malli_1729

Full Member level 5
Joined
Feb 13, 2007
Messages
247
Helped
47
Reputation
94
Reaction score
46
Trophy points
1,308
Location
INDIA
Activity points
3,120
In our design,

we want to connect a GPIO (3.3V) to 2 devices (selected by another GPIO, one device is 3.3V and other device works with 1.8V.

I am seeing 2:1 mux with Open drain outputs so, that i can connect each output to any voltage(1.8V/3.3V) tolerant device.

any other alternate way to fulfil this in single IC is fine.
 

malli_1729

Full Member level 5
Joined
Feb 13, 2007
Messages
247
Helped
47
Reputation
94
Reaction score
46
Trophy points
1,308
Location
INDIA
Activity points
3,120
@bking: that's a straight forward 1:2 demux.
Out2 in my diagram is 1.8V (not 3.3V complaint)- these devices with 3.3V VCC will generate Min Voh as 2.4V which will damage 1.8V device....T
That's the reason i asked for open drain outputs....
 

bking

Member level 5
Joined
May 15, 2012
Messages
85
Helped
28
Reputation
56
Reaction score
28
Trophy points
1,298
Location
Maryland, USA
Activity points
1,762
Hmm... I didn't have the post with the picture until now. I see you also flipped it to a 1:2.

How about a 74LS156. That has open collector drivers.
 

KlausST

Super Moderator
Staff member
Joined
Apr 17, 2014
Messages
17,460
Helped
3,943
Reputation
7,884
Reaction score
3,813
Trophy points
113
Activity points
115,755
Hi,

An analog MUX usually is bidirectional. I think you don´t need this.
OpenCollector outputs need a pullup. This may be waste of power or slow rising edges.

Thus I think lowVoltage powerd single gates with 3.3V input tolerance could be a good solution without drawback in signals.

But there are informations missing:


HTML:
IN  | SEL || OUT1  | OUT2
-------------------------
 0  |  0  ||   0   |  ?  
 0  |  1  ||   ?   |  0  
 1  |  0  ||   1   |  ?  
 1  |  1  ||   ?   |  1

Please replace the "?" with your desired values.

Klaus
 

malli_1729

Full Member level 5
Joined
Feb 13, 2007
Messages
247
Helped
47
Reputation
94
Reaction score
46
Trophy points
1,308
Location
INDIA
Activity points
3,120
@bking: Those are old IC's with big packages. I can't fit in my board.

- - - Updated - - -

@KlausST: Can you specify which IC can generate this.
I see configurable gates generate 2:1 mux because gates will have multiple inputs and single output.I haven't see any configurable gates which can generate 1:2 DEMUX
 

KlausST

Super Moderator
Staff member
Joined
Apr 17, 2014
Messages
17,460
Helped
3,943
Reputation
7,884
Reaction score
3,813
Trophy points
113
Activity points
115,755
Hi,

@KlausST: Can you specify which IC can generate this.
I see configurable gates generate 2:1 mux because gates will have multiple inputs and single output.I haven't see any configurable gates which can generate 1:2 DEMUX
First you need to give complete informations.
--> replace the "?" with the logic levels you desire.

Klaus
 

malli_1729

Full Member level 5
Joined
Feb 13, 2007
Messages
247
Helped
47
Reputation
94
Reaction score
46
Trophy points
1,308
Location
INDIA
Activity points
3,120
@KlauST: Thanks,

all '?''s will be replaced by tri-state/disabled as like SN74LVC1G18( 1-of-2 Noninverting Demultiplexer With 3-State Deselected Output).

Basically based on Selection pin IN state should be transferred to OUT1/ OUT2. OUT1 and OUT2 should be Open drain outputs.
 

KlausST

Super Moderator
Staff member
Joined
Apr 17, 2014
Messages
17,460
Helped
3,943
Reputation
7,884
Reaction score
3,813
Trophy points
113
Activity points
115,755
Hi,

all '?''s will be replaced by tri-state/disabled as like SN74LVC1G18( 1-of-2 Noninverting Demultiplexer With 3-State Deselected Output).
Why you want tristate? It is an undefined logic level. The receiver may see HIGH or LOW, or oscillation...

Klaus
 

malli_1729

Full Member level 5
Joined
Feb 13, 2007
Messages
247
Helped
47
Reputation
94
Reaction score
46
Trophy points
1,308
Location
INDIA
Activity points
3,120
"Why you want tristate? It is an undefined logic level. The receiver may see HIGH or LOW, or oscillation..."
--> when it is tri-stated other output will be connected to receiver.

Actual case is,
I have a JTAG chain, with 2 devices connected to a connector (where an ICE or equalent will access the chain to program each device one by one).
Device 1 - 3.3V device
Device 2 - 1.8V device(last in chain)

I need to have option to bypass device2 when required.
TCK,TMS,TST & TDO going to Device2 are going to voltage translating switch(SN74AVC4T245 - name U1) - this translates 3.3V logic to 1.8V logic.
which can disabled by driving OE#

Device1 TDO - can go to Device 2 TDI(in image above it is OUT2) through voltage translating buffer (Device1,Device2 both are in chain) or directly to Connector TDI(when Device2 bypassed- in image it is OUT1)
to implement this,i need a demux(U2)

Case1 : Device2 is bypassed in chain - demux will send IN to Out1
Out2 will be tri-stated,and U1 also disabled by driving OE# of U1


Case2: Device1,Device2 are in chain
demux will select Out2 (from IN),
Out1 will be tristate and U1 will be enabled(OE# LOW), because U1 is enabled TDO from device2 will go to TDO of the connector.
 

KlausST

Super Moderator
Staff member
Joined
Apr 17, 2014
Messages
17,460
Helped
3,943
Reputation
7,884
Reaction score
3,813
Trophy points
113
Activity points
115,755
Hi,

Daisy chaining JTAG devices should need no extra logic. It can simply be done with JTAG protocol.
You should be able program to program one device only or both as you like.

You are not the first one to have this problem, thus I recommend to look for existing solutions.
The device manufacturers should provide such informations.

Many JTAG inputs are 3.3V or even 5V tolerant. Please check datasheet.

In case you still want a special solution, then please post a draft/schematic about your idea.
The better your informations the better our chance to help.

Klaus
 

malli_1729

Full Member level 5
Joined
Feb 13, 2007
Messages
247
Helped
47
Reputation
94
Reaction score
46
Trophy points
1,308
Location
INDIA
Activity points
3,120
"Daisy chaining JTAG devices should need no extra logic. It can simply be done with JTAG protocol.
You should be able program to program one device only or both as you like.
"
--> Multi voltage jtag chain requires voltage translators.
If there is a need to bypass/Include certain device on need basis - yes we need to add glue logic to do it.

Below is the ruff implementation.

jtag_chain.jpg
 

std_match

Advanced Member level 4
Joined
Jul 9, 2010
Messages
1,083
Helped
409
Reputation
818
Reaction score
404
Trophy points
1,363
Location
Sweden
Activity points
8,276
The original request can be solved with a single 74HC4066 and 4 resistors:

level_conversion_demux.png

The unmarked resistors should not be critikal. Use 10K or whatever.
The 1k resistor will affect the maximum speed for the 1.8V TDO output. Reduce the value to increase the speed.
 

std_match

Advanced Member level 4
Joined
Jul 9, 2010
Messages
1,083
Helped
409
Reputation
818
Reaction score
404
Trophy points
1,363
Location
Sweden
Activity points
8,276
I think it would be better to use a MUX on the signal going to the TDO pin on the connector.
If you need to disconnect the ASIC TDI pin when the ASIC is bypassed, use another mux before the 3.3V to 1.8V level converter.
One 74HC157 contains 4 dual-input muxes controlled by the same select input, and should solve the problem.
 

Toggle Sidebar

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Top