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.

OPA1664: OPA1664 over heat

Part Number: OPA1664
Other Parts Discussed in Thread: DAC8718

Hi Team,

Please see below circuit of DAC output, this circuit is working fine,
Please see the net name of VOUT_1...3, If unfortunately, I apply the +24V input on these output pins means circuit/ OPA1664 op-amp get damaged.
For that, we provided the 80mA restartable fuse at the end of output voltage and another end of the fuse terminal connected to +24V input voltage. At that time, we observed the OPA1664 part is getting more heat.
Please suggest to me the circuit to protect, if the +24 VDC input on that DAC output( VOUT_1...3)
And Whether we applied the fuse on output is the correct methodology. Please suggest on this also.

Best Regards,

Dinesh KS.

  • Hi Dinesh,

    Your image did not go through. Please use the icon circled in red to upload an image file, use the icon circled in green to upload a standard file. 

    Best,

    Raymond

  • Hi Raymond,

    For the same DAC and Op-AMP circuitry in the above mail loop,
    We are providing 4.096V (from REF5050IDR) reference voltage to Vref pins of DAC8718, then the amplitude will be 6x gain of +/-12.2V?,

    But +/-15 V Operating voltage given in to OPA1664. whether this correct? or need to provide the +/-12V to operate OPA1664? for the function of amplitude signals.

    Regards,
    Dinesh KS

  • Hi Dinesh,

    If I understood your issues correctly, here is what I did. 

    1. Add input and output protection for OPA1664. 5kOhm will limit the input and output current for the op amp and DAC.

    2. Add TVS at Vcc and Vee at 16V (momentary short). If the output short is prolonged short, then 16V zeners should be used. 

    3. Lower the BW to approx. 13MHz. Previous there is a Q in the circuit. 

    4. Increase R2 from 600Ohm to 800Ohm, that will limit the output short current. I did not place bidirectional TVS in the simulation, leave it as is as you designed. Also, keep your diode protections as is as you designed (I did not know what you are using, so I just pick one from sim. library). So ignore IN5817 schottky and use yours. 

    How many OPA1664 output may be shorted at the same time potentially? I think you are likely using  quad package, each part is designed to short one output at any given time (up to 50mA). If you have more than one shorted outputs at any given time, we may have to limit additional shorted current to keep the OPA1664 cool (less heat dissipation).  

    /cfs-file/__key/communityserver-discussions-components-files/6/OPA1662-Input-and-Output-Protection-10282020.TSC

    Please test the above configuration in your circuit and see if OPA1664 and/or DAC are getting hot. Please let us know your findings.   

    Best,

    Raymond

  • Hi Raymond,

    Thanks for the reply,

    Do you mean changing the R2 (ferrite bead) value from 600 ohms to 800 ohms will not allow 24V external short?

    For this cause, we don't want to add the PTC resettable fuse at the end of the Vout signal, correct? so will not allow the external voltage short with op-amp output voltage.

    In the DAC8718SRGZT DAC, V-out depends upon REF-A and REF-B voltage (V-out = 6 x REF) or AVDD/AVSS voltage?
    For REF voltage means, we are using the REF5040IDR part to get the 4.096V REF voltage.

    So Amplitude Voltage will be (4.096Vx 6 gain) +/- 12.3V, is that correct?

    So we need to provide a +/-12V operating voltage to the Op-AMP or it will Exist +/-15V voltage?

    If we provide +/-12V for AVDD/AVSS pins in the DAC8718SRGZT DAC instead of +/-15V in the present circuitry, what will happen?

    If you are suggesting +/-12V for OPA1664, then need to change the AVDD/AVSS supply to +/-12V instead of +/-15V?

    If possible, shall we do the Web-ex meeting?


    Best Regards,

    Suresh S
     

  • Hi Suresh,

    Do you mean changing the R2 (ferrite bead) value from 600 ohms to 800 ohms will not allow 24V external short?

    Can you provide me the part number in the ferrite bead, so that I can simulate the short circuit properly? I am interested to see the datasheet of the ferrite beads. Using 800 ohm resistor, I was trying to see if there is an excessive short current that go through the Op Amp. If you provide me with actual components at OPA1664, I can simulate it again tomorrow. 

    Please provide me the nominal load condition at output OPA1664. How many output ports may be shorted at the same time? Are you using Quad Op Amp package?

    For this cause, we don't want to add the PTC resettable fuse at the end of the Vout signal, correct? so will not allow the external voltage short with op-amp output voltage.

    Yes, If you configure the output properly, you will not need resettable fuse. 

    In the DAC8718SRGZT DAC, V-out depends upon REF-A and REF-B voltage (V-out = 6 x REF) or AVDD/AVSS voltage?
    For REF voltage means, we are using the REF5040IDR part to get the 4.096V REF voltage.

    So Amplitude Voltage will be (4.096Vx 6 gain) +/- 12.3V, is that correct?

    yes, the OPA1664's input will be 24.576Vp-p, or +/- 12.29Vp at its input. 

    So we need to provide a +/-12V operating voltage to the Op-AMP or it will Exist +/-15V voltage?

    If we provide +/-12V for AVDD/AVSS pins in the DAC8718SRGZT DAC instead of +/-15V in the present circuitry, what will happen?

    If you are suggesting +/-12V for OPA1664, then need to change the AVDD/AVSS supply to +/-12V instead of +/-15V?


     

    If your max. input  to OPA1664 is at +/-12.3Vpp, then you will need operate +/-15V supply rails at OPA1664, since you want to output +/-12.3Vpp at the OPA1664 ouptput buffer with 2kOhm load. OPA1664's power rails have to be configured at the minimum 12.3+0.6=12.9Vdc or +/-12.9Vdc.   

    If possible, shall we do the Web-ex meeting?

    Who is regional AFE? Normally, the meeting is arranged via AFE. The audio application engineer is currently on vacation. I am just helping out to answer E2E questions. The answer to your question is yes, but you should contact your AFE or distributor in your region or area, who will arrange the best supports in your products. They have many resources than we do, if you need closed and timely support. 

    Best,

    Raymond

  • Hi Raymond,

    Thanks for the reply,

    We have manufactured and assembled the PCB as per Below Schematic. Now OPA1664 got damaged while applied the external 24V on VOUT Side.

    Ferrite bead Part number is FBMH3225HM601NT.

    We can prevent the circuit If we replace the 600-ohm ferrite bead with an 800-ohm resistor? If not please suggest how to prevent the OPA1664 and Diode with existed circuitry?

    We planned to add the diode between Vout (Anode pin) and external input (Cathode pin) to prevent the circuit, but it is worked for positive output supply and it will not work for negative supply output.

    FYI: There are 24 numbers of VOUT for that used 3 numbers of DAC.

    Please confirm this.

    Best Regards,

    Suresh S

  • Hi Suresh,

    What kind of loads do OPA1664s have to drive? I need to know what is the maximum load source/sink current at OPA1664's output.

    Ferrite bead Part number is FBMH3225HM601NT.

    I checked the part, the nominal impedance is 600 Ohm +/-30% at 100MHz, but DC resistance is only 0.042Ohm. So that is why you are damaging OPA1664 parts, if the output is shorted to 24Vdc. 

    I have another question. I believe that this is audio application. If you AC couple at the output, then OPA1664 will be safe from the output short to 24Vdc. The capacitor will block the DC signal, and you do not need to put 800 ohm resistor at the output. You still need to put protection limit the input current to OPA1664.

    Please provide me with the output requirements and I will check the circuit again. 

    Best,

    Raymond

  • Hi Raymond,

    Thanks for the reply!

    We are looking for a solution for the industrial-scale application, and there are 24 numbers of channels output.

    The maximum current is from 45-60mA as per the OPA1664 datasheet.

    Ok, We will replace the 600-ohm ferrite bead with an 800-ohm resistor in our circuit. is that correct? or we need to implement your suggested circuitry with diodes and 5K resistors in input? to prevent during the output short with 24VDC.

    Also, need to add a 2Kohm resistor pull-down resistor in the op-amp output?

    Best Regards,
    Suresh S

  • Hi Suresh,

    Ok, We will replace the 600-ohm ferrite bead with an 800-ohm resistor in our circuit. is that correct? or we need to implement your suggested circuitry with diodes and 5K resistors in input? to prevent during the output short with 24VDC.


    800 Ohm will limit the max. output current, since it is between 45-60mA. My suggestion is capacitive couple (like  1uf) at the OPA1664's output, and remove 800Ohm resistor. 

    Yes, you will need 5kohm resistors to protect the input pins of OPA1664s. You may even increase 22pf capacitor to lower the BW of the circuit, if this is for audio application. 

    I will try to simulate something and send it you at the end of the day.  

    Best,

    Raymond 

  • Hi Raymond,

    We don't have an issue with the present circuitry, and we are getting the proper output.

    We need the protective circuit for the below condition in the present setup/circuit without more re-work in the physical assembled board.
    If VOUT voltage get short with 24V DC / 0.5A input.

    The OP-AMP OPA1664 output is a VOUT,  and Op1664 is heating when 24VDC shorts.

    Best Regards,
    Dinesh KS

  • Hi Dinesh,

    This is the minimum changes that you have to do, since this is already a working product. By adding 1uf capacitor, it will block DC short at the output, and it will protect OPA1664 from DC short current. I assumed that the bidirectional TVS is adequate to protect the part from AC transient voltage at the output.  

    Here is the simulation.

    /cfs-file/__key/communityserver-discussions-components-files/6/OPA1662-Output-Protection--Min-Change-0302020.TSC

    The circuit has unity Gain BW at approx. 13MHz with phase margin of 62 degree. The  op amp circuit is stable. 

    If you have other questions, please let us know. 

    Best,

    Raymond

  • Hi Raymond,

    Thanks for the Reply!

    Best Regards,
    Suresh S