• State Resolved
  • Locked Locked
  • Replies 11 replies
  • Answers 2 answers
  • Subscribers 47 subscribers
  • Views 241 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.

TLV4111: High current linear drive

David Pfaltzgraff
Expert 2445 points
Part Number: TLV4111
Other Parts Discussed in Thread: ALM2402-Q1, ALM2403-Q1

Tool/software:

I have a requirement to drive a load (a high voltage DC-DC converter). The maximum voltage is 12V and the maximum current required is 300 mA. The output of the DC-DC converter is proportional to the input drive voltage.

My plan was to generate the control voltage with a DAC (0 to 3.3V). That is then amplified as needed, but I have not been able to find any buffer amps that would meet the drive capability. I had started by looking at the TLV4111, but later realized that the requirement goes up to 12V!

Any suggestions? (The simpler, the better.)

Thanks

  • 0
    TI__Genius 13736 points

    Hey David, 

    Yes, something like this is very possible. The 300mA is in somewhat of a strange range as it is between a high current drive OPA, and a power amp. 

    Since you need HV and 300mA current drive, I recommend looking at the ALM2402-Q1 or the ALM2403-Q1. These are dual channel parts, but you can always configure the other channel as an unused amp if you do not have a need for it : How to Properly Configure Unused Operational Amplifiers

    We often call these parts resolver drivers, but they are effectively just HV op amps with high current drive and good thermal packaging. Since these are op-amps, you can configure the device to gain up the DAC voltage and provide high output current (all in one stage).

    Please let me know if you want me to run any simulations for you.

    Thanks,

    Jacob

  • 0 in reply to Jacob Nogaj
    Expert 2445 points

    VERY interesting part. I never would have looked here. Will look into it further.

    Thanks

  • 0 in reply to David Pfaltzgraff
    TI__Genius 13736 points

    Hi David,

    Yeah, they are pretty interesting little devices. Feel free to reach out if you have any questions. We have simulation models and EVM's for all of the resolver driver amplifiers. 

    Thanks,

    Jacob

  • 0 in reply to Jacob Nogaj
    Expert 2445 points

    Hi Jacob,

    You offered to run a simulation, but I need a little more if you don't mind.

    While thinking about this, I realized that with two OpAmps, I could drive the power supply differentially since it does not have any real ground reference. To me the advantage is that when the outputs are the same (furthest from the supply rails), the current requirement is minimal. At the other end, when the full 12V is required, the current required is at its maximum. This would result in lower power dissipation in the driver section. (I think.)

    So, I've hacked up a quick schematic that I think will work. If you don't mind taking the time and commenting, I'd appreciate it. At the same time, I'm open to other suggestions.

    Oops! How can I attach a PDF file for you to view?

    Later - I hope!

  • 0 in reply to David Pfaltzgraff
    TI__Genius 13736 points

    Hey David, 

    No problem, I can absolutely review a design. 

    You can include files like this:

    Yes, the supply connections can be altered to optimize for power efficiency. 

    Since the op amp is a linear, the more voltage drop we have across the output stage transistor, the more power loss we see. 

    Feel free to include the schematic, and I will happily look over it.

    Thanks,

    Jacob

  • 0 in reply to Jacob Nogaj
    Expert 2445 points

    DiffDrive.pdf

    Hi Jacob,

    I hope this works. When I tried yesterday, it didn't take. This time it seemed to. The file is the result of an LTSpice run.

    In order to get it going, I just used generic OpAmps. In the final application, the two on the right would be replaced with ALM2402-Q1 and the 400 Ohm load should be 40 Ohms to match the power supply.

    At first I tried a single ended to differential circuit that I found on the 'net. While working with LTSpice (my skills are VERY rusty!) it didn't work. Then overnight I realized that if I just multiplied the DAC to make it 0-6V and then added/subtracted it from a 6V reference. I would have what I needed. The final outputs are both at 6V when the DAC is at 0 and then spread apart up to the max of 12V when the DAC goes to it's maximum.

    As I said before, I'm open to suggestions. The circuit is not very cost critical as 'production' will be less than 10. It's probably more space limited. The smaller the better.

    Thanks,

    Dave

  • 0 in reply to David Pfaltzgraff
    TI__Genius 13736 points

    Hey Dave, could you remind me again what the desired output range is? I assume 0V to 12V with linear scaling relative to the DAC. 

    Is the low side of the load terminating to GND, or a non ground referenced terminal?

    We can certainly make this circuit relatively small if needed. 

    Thanks,

    Jacob

  • 0 in reply to Jacob Nogaj
    Expert 2445 points

    The power supply being driven takes a voltage 0 - 12V and is not ground necessarily referenced as it's output is isolated. That's why I represented as just a resistor.

    My original thought was simply to drive it as ground referenced, but when I realized that I had two amps, I figured this configuration may (emphasis may) dissipate less heat. Most of the time, the power supply will have a 'moderate' voltage say around 8V and only occasionally be driven higher. It is not necessary to go all the way to 12V.

    Thanks, Dave

  • +1 in reply to David Pfaltzgraff
    TI__Genius 13736 points

    Hi Dave,

    Agreed, if you are using two amps, you might as well use differential drive. 

    I reworked the circuit a little to save a few amplifiers from your BOM.

    This new design only requires one ALM2402-Q1, and a 6V reference. 

    Note, I made my Vs 18V in simulation, but you could reduce this in the real design. 

    If you use 12V supply, the max Vout at 300mA would be about ~11.8V.

    Similarly, U2 will clip around 300mV if you try to output VEE at high current levels

    If possible, I would use a voltage higher than 12V is you want an accurate 12V out.

    If you can use something like 15V or 18V, we could also shift the output to not run into the VEE rail when using low output for U2

    Thanks,

    Jacob

  • 0 in reply to Jacob Nogaj
    Expert 2445 points

    Hi Jacob,

    Looks good! Just a quick note, I don't believe I'll be operating that close to the rails, but it's good to know that I can go higher if needed.

    Thanks, Dave

  • 0 in reply to David Pfaltzgraff
    TI__Genius 13736 points

    Hey Dave, great to hear that this design will be suitable. 

    Please let me know how it goes. 

    Thanks,

    Jacob