• State TI Thinks Resolved
  • Locked Locked
  • Replies 5 replies
  • Answers 1 answer
  • Subscribers 47 subscribers
  • Views 248 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

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.

THS3491RGTEVM: Damage during evaluation activity.

Ryan Yong
Prodigy 10 points
Part Number: THS3491RGTEVM
Other Parts Discussed in Thread: THS3491

HI TI forum,

I recently used this evaluation board to measure the frequency response of THS3491RGT when in A = 10V/V, using a range of Rf and Rg (feedback / gain resistors). 

I used an R&S FPC1500 spectrum analyser to measure the frequency response. 

I also used a Gw Instek GPS3303 dual power supply power the evaluation board. The outputs of the power supply was set to +15V 0.2A and -15V 0.2A. 

I connected the inputs / outputs as follows:

Vin - : No connection 

Vin+ : FPC1500 general output (0.2Vpp, variable frequency) 

Vs-  : -15V 0.2A (from power supply)

Vs+ : 15V 0.2A (from power supply)

Vout : FPC1500 RF input port 

GND : to power supply ground. 

I have done the following to this board:

1. Replace and install Rf and Rg several times with different values to record the output vs. different feedback resistors 

2. Remove and install R7 to record how the termination resistor affects the output 

3. Connected 

After installing a different set of Rf and Rg, I noticed that on the power supply, Vs- pulls 0.15A and Vs+ pulls 0.1A. 

When both Vin- and Vin+ are not connected, pins 10 and 11 measured -10V and pin 1 measured -1V.  

I understand that I may have damaged the evaluation board, but I'm hoping if someone here could help suggest some more debugging steps to identify how I damaged it so that I avoid damaging another evaluation board. 

  • 0
    TI__Genius 16846 points

    Hello Ryan,

    I will assist with looking into the behavior you are observing on your THS3491RGTEVM.  May I ask if you ordered the EVM from TI directly, had it sent to you by the HSAMPs product line or TI field team, or if you bought the EVM from a distributor, such as Arrow?

    This may be difficult or non-trivial depending on tools & experience, but have you attempted to do an A-B-A swap and try a new THS3491 on the EVM with the issue?  

    Are there shorts or other issues present with the THS3491RGT or the EVM?  I know you described pins 10 & 11 being pulled down towards negative supply (-10V) and pin 1 measures -1V.  The VOUT and VFB pins should have an internal connection on the metal layer, acting as a short.  The VFB pin allows for easier layout and connections during PCB design.  It would be a problem if these pins were no longer connected in a "short" but had impedance between them internal to the device.

    I will think over your comments and review the EVM documentation.

    Best,

    Alec

  • 0
    Prodigy 10 points

    Hi Alec, 

    Thanks for the reply. To answer your questions - 

    1. May I ask if you ordered the EVM from TI directly, had it sent to you by the HSAMPs product line or TI field team, or if you bought the EVM from a distributor, such as Arrow?

    The EVM was bought off Mouser. 

    2.  Have you attempted to do an A-B-A swap and try a new THS3491 on the EVM with the issue?  

    No,  I have not. 

    3. The VOUT and VFB pins should have an internal connection on the metal layer, acting as a short.  The VFB pin allows for easier layout and connections during PCB design.  It would be a problem if these pins were no longer connected in a "short" but had impedance between them internal to the device.

    V_out and V_FB are still shorted. Resistance between pin 1 (V_FB) and pin 10 / 11 (V_out) measures ~1.3 Ohms. 

    4. Are there shorts or other issues present with the THS3491RGT or the EVM?

    No, not that I can find. 

    A few other things that I think may have contributed to the damage might be that:

    a. I removed R7 (49.9 Ohm termination resistor) and connected J2 (Vin +) to a 50Ohm signal source. 

    b. I accidently connected the negative voltage supply (from power supply) to J5 (GND connector) and the GND cable (from power supply) to J6 (VS- connector) 

    Thanks again for walking me through this. 

    Ryan 

  • 0 in reply to Ryan Yong
    TI__Genius 16846 points

    Hello Ryan,

    Thank you for the quick responses.  I am thinking through potential behaviors and outcomes so far.  I will provide an update once I am able to share my thoughts. 

    May I also ask if you are in need of an additional EVM to continue prototyping or product evaluation?

    Once we determine possible cause and good operating procedure, you may end up wanting a fresh board, depending on your evaluation needs.

    Best,

    Alec

  • 0 in reply to Alec Saebeler
    Prodigy 10 points

    Hey Alec, 

    Yes, I'd need another EVM to continue testing the THS3491. I'm planning to chain this EVM with an in-house prototype to measure the overall system performance. 

    Cheers,

    Ryan 

  • 0 in reply to Ryan Yong
    TI__Genius 16846 points

    Hello Ryan,

    Thank you for confirming you would be needing an additional EVM.  May I as how long you operated the EVM with the VS- and GND connections swapped?  Was a signal being driven into the device during this time?  It is possible that the switching GND with VS- would provide an opportunity for electrical overstress to the board and/or device.  While there was a current-limit set, if the THS3491 was amplifying a signal while the thermal pad was connected to VS-, the part could overheat and cease to work as expected.

    Could you share the Rf and Rg resistor values you used for testing?  Since THS3491 is a current-feedback amplifier, the resistor selection for Rf does have an impact on the performance of the part.

    The termination resistor R7, present at the non-inverting input, assists with matching the signal generator, SMA cable, and signal to the amplifier's input circuitry.  Especially at higher speeds (frequencies), 50-Ohm equipment matching has a high impact on the performance of the circuit.  Reflections and signal loss are low when proper impedance matching is used.  I am not sure what investigation you were pursuing with the removal of R7 (BOM cost, curiosity, etc), but I would suggest using proper impedance matching and termination for high-speed designs.  This is especially true when using lab equipment to validate and test high-speed parts.  

    If you need the replacement EVM quicker than Mouser can provide, the EVM is available from TI.com.

    Best,

    Alec