DLPA2000: DLAP2000 buck-boost regulator

Hello User,

Welcome to the E2E forums and thank you for your interest in DLP ® technology. 

I just want to double check that this is a custom system you have designed and not the DLPDLCR2010EVM? If it is a custom design were you basing it off of this reference design? (https://www.ti.com/tool/TIDA-00325)

The voltage measurements you have taken seem to be outside of the normal operating conditions of the DLPA2000 for VLED. The below specifications are from the DLPA2000 datasheet. Additionally what LEDs are you using with your design? 

I see from the DLPA2000 datasheet here that there are these fault conditions descriptions that perhaps may be what is occuring.

What do the waveforms look like at a lower brightness for VLED?

Regards,

Alex Chan

Hello User,

How do you change the display brightness to maximum (from what value to what value)? Do you change the RGB LED current to achieve the display brightness?

Can you still have communicate with the DLPC3430 when the display freezes? 

Regards,

Lori 

Hello, thank you for your response.

Yes, it is a custom design.

I’ve checked the datasheet, and based on the measurements I sent for VLED, I agree that the voltage was too high. I managed to stabilize it by increasing the capacitance on VINL of the DLPA2000. Previously, it was 150µF, and now it’s 440µF (2x220µF). On the VLED pin, we currently have a 47µF capacitor (which we haven’t changed).

With this adjustment, the voltage level has stabilized so far, as shown in the attached pictures. However, the problem persists when I try to exceed 90% of the display brightness.

Do you think increasing the 47µF capacitor on VLED would help in this case?

VLed at minimun brightnees.

VLed at 50% of the brightnees.

VLed at 90% of the brightnees.

Hello User,

Please give our team some time to look into this issue we will need to check with the hardware team.

Was your custom design previously reviewed by anyone from TI?

Regards,

Alex Chan

Hello, 

No it was not reviewed. Also can you confirm the wave form I should have in Vled pin. 

Hello user,

Thanks. Our team will need to look into this and get back to you by the mid of next week.

Regards,

Lori 

Hello Alvaro,

I don't immediately see anything wrong with your Vled waveform. Could you please check the status of each of these power lines before and after the fault occurs?

Additionally can you please confirm the firmware you are using?

Best,

Max

Hello Maximus,

I have reviewed the power lines from Figure 5 and observed that the V6V line drops when the Buck-Boost regulator stops working. As shown in the following pictures, the rest of the signals do not appear to be affected.

Channel 1: VLED

Channel 2: V6V

Channel 3: LS_OUT

Same Pic (Zoom in)

Hello,

Thank you for this information, could you please send me your PMIC schematic? I am discussing this issue internally with our hardware team to see if they have seen this issue before.

Best,

Max

Hello,

This is the schematics or the PMIC, also this the firmware version I got:

System software version:
Register: 0xD2
Value: 0x01

Flash Build Version:
Register: 0xD9
Value: 0x0B

C106 and C107 are not populated.

Hi Alvaro,

Thanks for the quick response, is your Vin 5V here or something else?

Best,

Max

Hello Alvaro,

More specifically could you check the system power input for glitches or sags as well as check if RESETZ or INTZ go low?

Best,

Max

Hello,

Between 3.7V and 4.2V.

Hello,

Ch1: VLED

Ch2: Vin

Ch3 RESETZ

Ch4: INTZ 

Hi Alvaro,

Thank you for this, let me get back to you once I have evaluated it with our hardware team.

Best,

Max

Hi Alvaro,

Could you please confirm the following:

• How many systems are you having the issue on, is it just one or all?
• Do the LEDs still function after you experience this behavior? 
• Why is your input voltage varying so much? I believe you mentioned this was 3.7-4.2V 
  • How are you powering your board?
• Where did you get your firmware from? Was it the TI DLP Pico Firmware Selector tool or else where and what was the .img file named?
• Can you share your LEDs or LED spec and how they are connected to your board? 
• Is your proj_on signal stable while you experience this behavior?
• Are you using 100 milli ohms as your Rlim sense resistor? Our reference designs use 39 milli ohms and thus our default firmware is configured for 39 millohms as well.

Best,

Max

Hello Maximun,

• How many systems are you having the issue on, is it just one or all?

            In all of them.

• Do the LEDs still function after you experience this behavior?

            Yes, they still work but the mage gets frozen.

• Why is your input voltage varying so much? I believe you mentioned this was 3.7-4.2V 
  • How are you powering your board?

           Because, the boar is powered by a battery. We experiment this proble even when the battery is fully charged. 

• Where did you get your firmware from? Was it the TI DLP Pico Firmware Selector tool or else where and what was the .img file named?

       The firmware we got from the light enginee vendor.     

• Can you share your LEDs or LED spec and how they are connected to your board? 
• Is your proj_on signal stable while you experience this behavior?

        Yes, proj_on is stable.

• Are you using 100 milli ohms as your Rlim sense resistor? Our reference designs use 39 milli ohms and thus our default firmware is configured for 39 millohms as well.

       We are using 100 milli ohm.

Hello Max,

Thank you for your time and support. The issue was resolved by adjusting RLIM for our setup. We found that a range of 0.15 < RLIM < 0.3 works effectively.

If you have any further suggestions regarding this change, I would appreciate your input.

Thank you in advance for your assistance.

Best,

Alvaro