• State
  • Locked Locked
  • Replies 5 replies
  • Subscribers 64 subscribers
  • Views 1327 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.

LM5045 current limit with sense resistor?

azadeh vkl
Prodigy 170 points
Other Parts Discussed in Thread: LM5045

In all application notes and evaluation boards related to LM5045, the current limit is implemented by CT(current transformer).
Although in datasheet, it is written that is possible to use sense resistor, I couldn't find any LM5045 example circuit with sense resistor.
in my application, I used a 15mohm shunt resistor instead of CT to detect the current limit 16A. in overcurrent condition the voltage on resistor would be 0.250v.
which is multiplied on 3 using an opamp circuit. then is connected to the CS pin through a RC filter with fc=1.6MHz which is 8 times greater than f=200KHz (R=1K,C=100pF).
without output load the performance is good. But with the 40w light bubble as an output load, the CS pin goes above 0.75v and the LM5045 switches to hiccup mode.
In another test,I tied the RES pin to the GND and disabled the hiccup mode, but the CS pin surpassed the 0.75v level.
Is there any other parameter that I had to consider but didn't? what's wrong with my design?

the schematic and waveform of CS and RES pins are attached if it helps.

any help would be appreciated


  • 0
    TI__Mastermind 34465 points
    Most likely switching noise is entering your current sensing circuit. You could try replacing your non-inverting opamp with a differential opamp.
    You need to pay particular attention to board layout and ensure that the low level control signals are routed away from the high current power traces.
    Regards,
    John Griffin
  • 0 in reply to John Griffin1
    Prodigy 170 points

    Thanks for your suggestion about amplifier.

    I removed the opamp circuit and modified the shunt resistor value to 45mohm,  and connect the sense resistor pin to the CS pin through the RC filter.
    since the shunt voltage can reach to 0.750v in 16A, so its expected that the overcurrent can be detected.
    but the problem is remained and not be solved. with 40w output load the CS pin raised to 800mv, and LM5045 goes to hiccup mode again.
    Is there any sample circuit which uses sense resistor in LM5045 current limit, that can help me to select the true resistor ?

  • 0 in reply to azadeh vkl
    TI__Mastermind 34465 points
    The problem is that noise on the return path of the power MOSFETS (the main power ground) is entering the low level sense ground of the LM5045 controller. I haven't seen your board layout but it is most likely that the high level current transitions are using the low level ground of the LM5045 in their return path. This is one of the big problems with resistive sensing of primary current.
    The current limit is 750mV and you say that you are measuring 800mV.Can you explain how you are measuring this signal. ?
    Where are you grounding the scope and how long is the ground lead?
    You should measure without a ground trace and connect the barrel of the scope directly to AGND pin.
    Can you attach the scope waveform and the schematic with BOM.

    Regards,
    John Griffin
  • 0 in reply to John Griffin1
    Prodigy 170 points

    I have designed it in a two layer PCB. 

    About the Ground traces, all the components( SS,RES,CS filter, RT,RD1,RD2) are connected to the AGND.
    only the negative pin of capacitor of vcc Pin is attached to PGND. The AGND and PGND and thermal tab are connected together directly under the IC. The PGND is directly connected to the shunt resistor ground by a short trace. It was tried to avoid adjacency of high current traces with LM5045 components.

    could you give me an email ID to send you the PCB layout and schematic?

    About the probe and CS waveform, in case of using 45mohm sense resistor, I connect the probe directly on two leads of C301 which are CS pin and AGND.
    So the scope is gruonded to the AGND. but again in this capture It can be seen that the CS pin goes above 800mv.

    could you help me please what is wrong here?

    Regards,

    Azadeh

  • 0 in reply to azadeh vkl
    TI__Mastermind 34465 points
    You can contact me at
    johngriffin@ti.com

    Regards,
    John