Document created by mayp Employee on Jul 3, 2015
Version 1Show Document
  • View in full screen mode

Q1. What are the key benefits of these fault protected switches?

A1.  There are a number of key features on these parts

  • Overvoltage fault protection to +/-55V levels in powered and unpowered state
  • Secondary supply rails to allow for user defined fault thresholds and clamp levels
  • Fault diagnostics via digital fault flags to indicate fault is present (FF pin), and  to indicate specific source channel in fault (SF pin)
  • Very low drain leakage in a fault condition, meaning other channels can continue to operate as normal.
  • Industry leading performance for charge injection and switch capacitance for fault protected switches and multiplexers
  • High ESD rating (up to 4kV HBM ESD) and latchup immune for additional robustness


Q2. What are the system benefits of overvoltage fault protection with secondary supplies?

A2.  Powering the switch with a higher supply voltage means it can operate in the lower, flatter RON region of the device. This optimizes distortion and noise performance. The secondary supply voltages can be set to protect downstream components such as Amplifiers, ADCs, or DACs. The overvoltage fault protected switches will clamp a fault to the corresponding secondary supply. The switch can also withstand up to +/-55V at its input while still protecting downstream components to the secondary supply levels.  The minimum supply voltage for the secondary supply is 4.5V.


Q3. How do the fault diagnostics work?

A3.  The voltages on the input channels of the ADG5248F are continuously monitored and the state of the switch is indicated by an active low digital output pin, FF. A HI signal (>2V) indicates “no fault” condition and a LO signal (<0.8V) indicates an overvoltage has occurred on any of the input channels.

       Once a fault has been detected, the SF pin can then be used to determine which specific channel is in fault. The SF pin is also an active low digital output pin. The F0, F1, and F2 digital input pins are used as decoder pins. The user cycles through all 8 combinations of (F2, F1, F0) and monitors the SF pin to diagnose which of the 8 input channels is in fault.


Q4. What are the system benefits of fault diagnostics?

A4.  For systems that are sensitive during a start-up sequence, the active low operation of the fault flags allow the system to ensure that the ADG5248F is powered on and that all input voltages are within normal operating range before initiating operation.  The FF pin introduces a means to diagnose systems in fault programmatically and can be used to stop long or expensive tests if a fault occurs.

       The fault flags also offer more intelligence to the system, preventing differential signals that over-voltage on both channels (e.g. ADG5249F dual 4:1 Mux)) presenting as “normal” operation.

       The specific fault pin allows the user to debug which input channel is in fault and can help with system debug (e.g. deciding which input sensor isn’t working and needs to be replaced). The system can use this information to skip over faulty channels and continue measurements on other channels which aren’t in fault, rather than collecting faulty measurement data.


Q5. Can other channels really continue to operate as normal when another channel is in fault?

A5.   Yes, one of the key benefits of the ADG5248F is that the drain leakage is still very low (<90nA over temperature), even when there is a fault on one of the input channels. Most competitor solutions have very high drain leakage when one of the channels is in fault, meaning the leakage can swamp sensitive measurements such as signals from Thermocouples. For example, the MAX4508 has 1uA leakage at 85C and 100uA leakage at 125C which would make other channels completely unusable for sensors sensitive to leakage.

        The other solution which customers use for overvoltage protection is using series resistors to protect multiplexed inputs. The major drawback of that approach is that an over-voltage event on OFF channels injects noise to good ON channel due to current flow in substrate leading to corrupt data. With the ADG52xxF parts, there is no current flow in the substrate ensuring no corruption of data.


Q6. The ADG5248F offers ±55V over-voltage protection.  What are the keys things I need to understand about this Over-voltage protected switch compared to using a standard switch in my application?

A6.  The ADG5248F is ideally suited as a protective element in signal chains that are sensitive to both charge injection and overvoltage signals. During normal operation the ADG5248F behaves like a standard multiplexer and can pass signals within the secondary supply rails, thereby protecting sensitive downstream components.  The ADG5248F offers 0.8pC charge injection with excellent leakage and switch capacitance specs ensuring optimum performance in multiplexed systems (e.g. instrumentation and process control applications).  

       The ADG5248F offers ±55V protection on the inputs with respect to GND.  When the voltages on the Sx inputs exceed either rail by ~0.7V the switch will turn off and present high impedance to the input.  The inputs can withstand up to ±55V; exceeding this limit may damage the ESD protection on the part. 

       One other point to note, 80V is the maximum voltage that can be present across the OFF switch, (Sx to Dx = 80V) and 80V is the maximum voltage that can be present from the Sx inputs to the supplies, (Sx to VDD or VSS = 80V).  Therefore, if you are using the part with 40V Single Supply the maximum voltage allowed on Sx is -40V and not -55V.

If you are switching signals with a bandwidth of 1MHz or greater, please reference the datasheet to understand the relationship between signal amplitude and bandwidth to ensure signal integrity.


Q7. What happens to the output during an over-voltage event?

A7.  The drain output will be clamped during an overvoltage fault condition. The ADG5248F, ADG5249F, and ADG5243F have secondary supplies (POSFV, NEGFV) which will be the clamp voltages on those parts. The switch will turn-off when the voltage on the input exceeds POSFV/NEGFV by 0.7V and the drain will pull to the rail (POSFV or NEGFV) that was exceeded.

       The ADG5208F and ADG5209F don’t have secondary supply rails, so the drain output will clamp to the primary supply rails (VDD, VSS) when the input exceeds VDD/VSS by 0.7V.


Q8. Can you explain Power-Off protection and its benefits?

A8.  The ADG5248F will provide protection to downstream circuitry against over-voltage conditions when it’s unpowered. This is very important for modules that may be unpowered but have signals present on the inputs.  Power-off protection to down-stream circuitry cannot be guaranteed using discrete protection components. 

       The ADG5248F Power Off protection guarantees the channel will remain in the OFF state and will Standoff up to ±55V i.e. no signal will get passed to the output protecting downstream components. On the ADG5248F, the supplies can be grounded or floating but GND must be present on the GND pin for the Power-Off function to work.  If the supplies are grounded during         Power-Off you will get ~500nA of leakage on the output and if the supplies are floating the leakage peaks to about 50uA.


Q9. My downstream component states that the inputs cannot exceed Vdd+0.3V, but the ADG5248F only turns off when the voltages on input exceeds Vdd by ~0.7V. Will this cause an issue in my system?

A9.  No. The output of our ADG52xxF parts clamp the OVP to Vdd+0.7V for ~90ns (Overvoltage Response Time, tRESPONSE), after which time the switch has fully turned OFF and output pulls to the rail that was exceeded.  So for ~90ns period, there will be some current flow but it would be more benign than a 1kV HBM ESD pulse. This should not cause any issues in the system once the downstream circuit has an ESD rating in excess of 500V HBM. Clamping the signals too close to the rails could trigger false Over-voltages if there is ripple on the power supplies so the 0.7V Vt is good balance for the system.


Q10. What will be the recovery time when switching from a channel in fault to a channel not in fault?

A10.  The timing will be the same as normal channel to channel switching; i.e. there is no additional timing to change from a fault channel to a non-fault channel: typically 210ns for Ton on the ADG5248F


Q11. Are these parts pin-pin compatible with existing parts?

A11.  Yes, the ADG5208F/ADG5209F are pin-for-pin compatible with the ADG5208/ADG5209 and ADG1208/ADG1209. They are drop in replacements if overvoltage protection is required. However, they are not spec-for-spec replacements so there may be some performance impacts, e.g. higher Ron, higher charge injection. There are no pin-for-pin alternatives for the ADG5243F, ADG5248F, or ADG5249F.

         The ADG54xxF family are the low RON alternative. They have been built using the same high voltage trench isolated process and have been optimized for low on resistance.


Q12. What is the recommended supply sequence?

A12.  The Golden sequence for power up is; Ground, VDD, VSS, Digital Inputs. The trench isolation and lack of logic supply pin (VL) means that it is not possible to cause latch-up in these parts by an improper supply sequence. However, it is still good practice to follow the Golden sequence whenever possible.


Q13. Can I leave the exposed pad floating?

A13.  The exposed pad is connected to the substrate which in this case is the most negative voltage, Vss. The exposed pad should be connected to Vss to enable more efficient heat transfer and increase reliability. There is no impact on switch performance if exposed pad is left unconnected but heat dissipation will not be optimized. Connecting the exposed pad to anything other than Vss potential may cause high current flow, affect the switch performance, and the long-term reliability of the part.