• State
  • Locked Locked
  • Replies 5 replies
  • Subscribers 63 subscribers
  • Views 1903 views
  • Users 0 members are here
Support feedback
Options
Options
Related

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.

TPS61500 problems

Andreas Pabst
Prodigy 30 points
Other Parts Discussed in Thread: TPS61500, TPS40211

 

 

 

Dear support team / Dear Community,

 

We have a problem concering the TPS61500. Below you will find a description of our project and our problem.

 

 

We are working in a project for a diploma thesis. The goal of this project is to build a lamp which is able to emit light in most colors of the RGB- model.

To achieve this goal we are using three differnt types of high power leds as illuminants. Each Type emmits light in a specific color. In our case they emit light in red, green and blue. These colors will be added together and produce a new color. 

 

We have successfully implemented a circuit to dim three Osram Golden Dragon high power LEDs. (one each for red, blue and green)

The next step in the project is to implement a new circuit to dim 10 LEDs for each color to iluminate a bigger area. To accomplish that objective we have searched for a LED driver based on the following criterias:

 

* output voltage must be high enough to power 10 LEDS (1 LED: 3.43V forwarding voltage, 0.7A current) 

* brightness dimming, which should be defined by an external PWM signal (pure PWM dimming)

* overvoltage protection 

* input voltage between 5V and 20V

 

 

After a lot of research we found the device TPS61500 (See datasheet: http://focus.ti.com/general/docs/lit/getliterature.tsp?genericPartNumber=tps61500&fileType=pdf).

The following values are based on the datasheet:

 

* 2.9-V to 18-V Input Voltage Range

* 3-A, 40-V Internal Power Switch

* Analog and Pure PWM Brightness Dimming

* ...

 

In our opinion this device should perfectly meet our needs.

So we designed our circuit based on the "Typical Application" Figure 2. Pure PWM Dimming Method. Here is the schematic of our application:

 

 

Description of the Schematic:

* The LEDs are mounted on a seperate print and are connected serially.

* The PWM Signal is assigned to the PIN 4 (EN) of the TPS and to the FET Q1.

 

According to the datasheet the following values were calculated (a detailed description can be found in the datasheet on the pages: 10-14)

* C1: 4.7uF

* C2: 1uF

* C3: 47nF

* C4: 470nF

* C5: 100nF

 

* R1: XXX

* R2: YYY

* R3: 0.27

* R4: 51k

* R8, R9: 1k

 

* L1: VLF10040 - 6R8N4R5: 6.8 uH

* D1: Schottky-Diode 30BQ040

* Q1: 2N7002 

* Q2: Si2319DS

 

Our main problem: 

 

In our setup we have a Vin of about 12V and a Vout of 36V. While we tested the circuit we noticed that the output voltage is not high enough to power 10 LEDs. The output voltage was as high as the input voltage. For that reason the LEDs aren't iluminating any light. Therefore we assume that we have a mistake in the part of the circuit where the TPS61500 scales the voltage up to Vout.

 

 

Things that were tested and worked as they should:

* Selection of the switching frequency: R4 = 51k which sets the frequency to 2MHz

* Switching of the FETs Q1 & Q2 

 

At this point we are stuck and hope that you can help us to solve our problem and help us to finish our project in time.

We really appreciate any help and are looking forward to hearing from you soon.

 

 

If there are any questions according to the Text above feel free to contact us at any time. You can either write an answer in the forum or send me an email to 

"andreas.pabst@sz-ybbs.ac.at"

 

 

Thankfully,

Andreas Pabst

 

  • 0
    TI__Guru**** 166126 points

    From Vin=12V, the TPS61500 should work for you.  What values did you select for R1 and R2?  These values must be installed to set the over-voltage protection level.  For your application, that would be the max 38V.

    For Vin=5V, the TPS61500 will not work.  The TPS61500's 3-A current limit will prevent it from providing the power that you need.  You might consider the TPS40211.

  • 0 in reply to Jeff F
    Prodigy 30 points

     

    Dear Jeff,

     

    Thank you for your fast reply.

     

    I am a little bit confused now. 

    I recalculated the values for R1, R2, Rfb to get 38 Vovp (Over voltage protection).

     

    Based on the formula from the Datasheet (Equation 1) I calculated 29.7k for R1 and 1k for R2.

     

     

     

    Furthermore i calculated the value for Rfb based on the Equation 2 from the Datasheet:

    Vfb should be 200mV and Iled should be 0.7A so i get 0.270 Ohm for Rfb (R3 in the formula from the Datasheet).

     

     

    Then I went on in the datasheet and wanted to recalculate Ip (Inductor peak to peak ripple). Based on the formula 4 from the datasheet.

     

    The Inductor Value L is 6.8uH, The forwarding voltage of the diode is 0.43V the Switching frequency is 2Mhz (I am using a 51k Resistor to define the frequency). Vout is 10*3.43V + 200mV which is a total of 34.5V and Vin is 12V.

     

    So i calculated 0.57922A as Ip.

     

    This is the point where i get quite confused... I don't understand why the Ip needs to be that high. I thought that the Ip should be as low as possible because when i calculate the Iled_max in the next formula, i get a very low 

    value for Iled_max.

     

    Ilim (over current limit ) = 0.85A

    n(efficiency) = about 92%

    Vout = 35.4V

    Vin = 12V

     

    Based on the formula from above i get a Iled_max (max current) of only 0.179A which is ways too low to power the 10 LEDs.. 

     

     

    I am really confused right now and I appreciate any help.

     

    Thanks in advance,

    Andreas Pabst

     

  • 0 in reply to Andreas Pabst
    TI__Guru**** 166126 points

    The ILIM in your equations above should be 3A not 0.85A.

  • 0 in reply to Jeff F
    Prodigy 30 points

     

    Hello Jeff,

    Thanks again for your help with our TPS61500 problems. We now have sucessfully implemented Analog Dimming. Our circuit is supplied with 12V as Vin and 

    is in position of dimming 8 Argus Red Osram Golden Dragon Power LEDs, which are supplied with 0.8A.

     

    The brightness of the LEDs is defined by the "ON-OFF-Ratio" of an external PWM Signal (The frequency of the signal is about 166Hz). 

    It can be set in 256 Steps(between 0: low and 255: high).

     

    Here we have a minor flaw. When we change the PWM Signal's ON-OFF Ratio from low: 0 to high 255 and vice versa, it takes the TPS61500 about 10 seconds to 

    get from about 0A to our desired 0.8A which represents the value 255 in the PWM Dimming.

     

    We really want to shorten that time because we are planing to change the color of the system gradiually and we therefore need the TPS61500 to change the 

    brightness faster. 

     

    Underneath you can find our circuit diagramm.

     

    R11 is 51k to get a switching frequency of about 2MHz

    C4, C5, and C6 are 1uF each to get a total of 3uF there. We used three Ceramic capacitor due to their low ESR value.

    Vovp is defined as 23.7V (180k + 3.3k for R1 and 10k for R2), because we have 8 LEDs with at forwarding voltage of about 2.6V each.

    There are 2 Feedback resistors R2 and R_FB with a total resistance of 0.249 Ohm , so that Iled is about 800mA

     

    We really appreaciate your help and are therefore grateful for any help you can provide to reduce the time it takes the circruit to turn the LEDs from low value to high value.

     

    Thanks in advance Andreas Pabst.

     

     

     

     

  • 0 in reply to Andreas Pabst
    TI__Guru**** 166126 points

    When going from 0 to 255, is EN low for more than 10ms?  If so, the IC is shutting completely down and must re-start up.  Lowering the value of the SS cap might help.  However, this will result in a larger inrush current at startup that might bog down the power supply to your system.

    Another problem might be the capacitor on DIMC.  A 220uF capacitor is a bit large and so there will be a long response time before the filter formed by the internal resistor and external C.  Try lowering it to 47uF and see if the startup is faster.  With such a low dimming frequency,  you need a larger capacitor.   You can keep lowering the capacitor until the ripple on FB gets too large (probably 50mVpp or so).