• State TI Thinks Resolved
  • Locked Locked
  • Replies 9 replies
  • Answers 2 answers
  • Subscribers 16 subscribers
  • Views 1838 views
  • Users 0 members are here
Support feedback
Options
Options
Related

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

Analog switch TS5V330C: mismatch for the on resistance and delays

Steve Wood
Prodigy 140 points
Other Parts Discussed in Thread: TS5V330C, TS5A23159

Analog switch TS5V330C: 

1) What is the expected maximum mismatch for the on resistance (both switch positions) for signals near rail and ground?  Ron is typically 3 ohms with a max of 7 ohms at 1V input. 

2) Is there any data on expected delay mismatch between switches?

3) I assume that the four switches are monolithic?

4) S to D delays are given.  May I assume that EN to D delays are the same?

  • 0
    Prodigy 140 points
    I should restate question #4:
    4) The t_on and t_off delays are given, but are these delays the same for the IN control pin as the EN control pin?
  • 0 in reply to Steve Wood
    TI__Mastermind 45575 points

    Steve,

    I apologize that this older device datasheet is not characterized up to our current standards and is missing several specs that would be important to you.  What are your looking to do with this device that these parameters are critical to your circuit? Maybe I can find a new device that has these critical parameters for you more defined. 

    1) What is the expected maximum mismatch for the on resistance (both switch positions) for signals near rail and ground?  Ron is typically 3 ohms with a max of 7 ohms at 1V input.

    I see that this value is not stated in the datasheet.  Let me see if I can dig up this information in our archives. 

    2) Is there any data on expected delay mismatch between switches?

    I see that this value is not stated in the datasheet.  Let me see if I can dig up this information in our archives. 

    3) I assume that the four switches are monolithic?

    Can you describe what you mean by the switches are monolithic? 

    4) The t_on and t_off delays are given, but are these delays the same for the IN control pin as the EN control pin?

    No you cannot necessary assume that the delay for the control inputs are the same as the EN input. Let me see if I can dig up this information in our archives.

    Thank you,

    Adam

  • 0 in reply to Adam Torma
    Prodigy 140 points
    Hi Adam,

    Thanks for your prompt response. I'm looking for a very fast (<6ns for control signal to switch from one "throw" of the switch to the other),quad or dual, single-pole double-throw analog switch. It preferably can be supplied with up to 5.5V single supply, and accepts 3.3V CMOS control inputs. 'ON'-resistance must be low (Ron<7 ohms), and ideally it should match between switches. Matching of delays between switches is also desirable. This is part of a discrete 3-level high-speed DAC, that has a target SNR of 130dB, so these specs are critical. Dual-supply operation would be preferred, but I think this is too much to ask at this speed (and at anywhere near this price point).

    Response to your question:
    3) Can you describe what you mean by the switches are monolithic?
    I just mean that all 4 switches are on the same die, so some matching of R-on and propagation delays can be expected "for free".

    Regards,

    Steve
  • 0 in reply to Steve Wood
    TI__Mastermind 45575 points

    Steve,

    Thank you for the detailed requirements.  I think I have these requirements ranked in order of importance to your system needs.

    ton/tswitch: <6ns

    Configuration: 1:2 SPDT

    Supply voltage: 5.5V

    Number of channels: 2 or 4

    Ron/On-state resistance: <7ohms

    Vih digital control high voltage: <2.7 V

    delta Ron/Ron state resistance between channels : <1 ohm

    tskew/ delay between channels: <1ns

    Based on your requirements what do you think about the TS5A23159.

    ton/tswitch: 8ns

    Configuration: 1:2 SPDT

    Supply voltage: 5.5V

    Number of channels: 2

    Ron/On-state resistance: 0.7 ohms

    Vih digital control high voltage: 2.4 V

    delta Ron/Ron state resistance between channels: 0.05 ohms

    tskew/ delay between channels: not specified in datasheet

    The TS5V330C is monolithic as I'm 99% sure that all of our signal switches are.  We have a lot of switches there maybe an outlier in the hundreds of devices.  

    It looks like characterization was never done on the TS5V330C for the delta Ron or tskew. 

    Adam

  • 0 in reply to Adam Torma
    Prodigy 140 points

    Adam,

    Originally I needed a quad (4 matched switches), so I didn't look at any duals.  I'll take a look at the switch you suggest.


    Thanks,

    Steve

  • 0 in reply to Steve Wood
    Prodigy 140 points
    Hi Adam,

    I think I can make this switch work. Is it safe to parallel both switches in the TS5A23159? In other words, can I be assured that the break-before-make times of each SPDT switch in the package will overlap?

    Thanks,

    Steve
  • 0 in reply to Steve Wood
    TI__Mastermind 45575 points

    Steve,

    Could you draw a picture of the connection you are asking?  Sorry I'm not clear on your question.

    The break before make time specification for this device is from the time that NO FET will begin to turn off (90% of its final voltage) until the NC FET will turn on (90% of its final voltage.)

  • 0 in reply to Adam Torma
    Prodigy 140 points

    Connect COM1 to COM2

    Connect NO1 to NO2 to a low impedance source

    Connect NC1 to NC2 to a low impedance source

    Connect IN1 to IN2

    By the way, there is an anomaly in the parametric search for analog switches.  The TS5A23159 is shown with a max on-delay of 55ns, while the data sheet shows 16.5ns max for 4.5-5.5V.  The 55ns figure is for low voltage operation, so it makes this switch difficult to find if you're looking for a fast 5V switch.  Ron also shows up as 15 ohms.  I tried to attached a screen shot, but the browser is crashing when I do (Firefox on Ubuntu).

    Another question:

    The TS5A23159 data sheet for 4.5-5.5V operation shows the min "tOFF" as 1ns and min "tBBM" as 1ns.  The min "tON" is also shown as 1ns, but it seems to me that the following inequality must be observed for a SPDT switch:  tON > tOFF+tBBM  Could you clarify this please?

    Thanks,

    Steve

  • 0 in reply to Steve Wood
    TI__Mastermind 45575 points

    Steve,

    Thank you for the feedback on the parametric search.  We are constantly trying to improve this table for our customers but unfortunately there are hundreds of issues like you described above where a specific parametric can vary based on test conditions. 

    Here is some insight on how the table kinda works that might help your understanding.  In general if there is a typical value in the table it will state the value under the best test conditions usually 25deg C.  If there is a max value it will be the value under the worst test conditions usually 85 deg C.  If you are looking for low Ron I would filter on Ron typ which will give you values under good test conditions instead of Ron max which will give you wost test conditions.  Having the typical (being good values) and max (being bad values) we tried to accommodate customers designing for typical and wost case corners.  We are always open to suggestions on how to make this better.    

    To answer your timing question you need to look at how these timing values are measured.  The Ton and Toff specs are measured with a voltage (V+) placed on the COM pin and then measured how long it takes the NC and NO pins to reach 90% of (V+) after the IN pins transition.  The break before make is measured with a voltage (V+) placed on the NC and NO pins and shows how long of a time the COM pin voltage stays below 90% V+ after the IN pins transition.  

    Thank you,

    Adam