DLP products forum - Recent Threads

DLPC964: Inquiry regarding issues with a custom board using the DLPC964

JH Shin — Fri, 16 Oct 2026 08:16:40 GMT

Part Number: DLPC964

Hello,

My customer has a few questions regarding the use of the DLPC964.

Q1. I understand that there are firmware versions 1.0, 1.1, and 2.0 for the DLPC964, and that the choice depends on the FPGA being used. Is this correct?

Q2. Should I use Version 1.0 when working with the XC7VX415T?

Q3. For Version 1.0, only the .MCS file is available. Could you please provide the .bit file as well?

Q4. I have designed a custom board using the same circuitry as the EVM and successfully completed the flash download (verified via the readback feature in Vivado). However, the 'DONE' and 'heartbeat' signals on the DLPC964 IC are not active. Even though I have loaded the firmware intended for the EVM, the board is not functioning correctly. Could you please provide some advice on how to proceed with debugging this issue?

Q5. If the flash memory recommended by TI is unavailable, are there alternative flash products that can be used instead? If so, please provide a complete list of suitable alternatives.

Thank you.

RE: DLPC964: Inquiry regarding issues with a custom board using the DLPC964

Tristan Bottone — Tue, 20 Oct 2026 20:03:31 GMT

Hello JH,

Let’s continue this conversation through our private message thread. I just sent you a response there!

Best Regards,

Tristan Bottone

RE: DLP3010: Connecting FPDLINK3 Deserializerto DLP3010

Kevin Rivera — Tue, 20 Oct 2026 18:11:57 GMT

Hi Neta,

Any front end (such as a FPD-Link Serializer-Deserializer) can be used as long as the DLPC can support that source. The length of this FPD-Link Serializer-Deserializer COAX cable does not impact the DLP performance as long as it receives proper data.

As the DLP3010 is going EOL, I would recommend looking at the DLP391TP DMD paired with a DLPC8445. The DLP391TP is listed as 4K resolution but this is through XPR, the native resolution is 1080p. The DLPC8445 can take in FPD-Link as a input source. The DLPA3082 or DLPA3085 PMIC can also be used to provide all necessary power rails to the controller and DMD.

The key aspect here is that there is a distance requirement between the DLPC and the DMD. This length can be found in the DLPC8445 datasheet, Table 8-5.

These lengths must be followed. If not, the DLP system will not operate properly.

Please let me know if you have any more questions!

Thank you,

Kevin Rivera

DLP3010: Connecting FPDLINK3 Deserializerto DLP3010

Neta Sherer — Tue, 20 Oct 2026 10:26:23 GMT

Part Number: DLP3010

Hi.

Some questions regerds the integration of DLP3010:

Is there any FPDLINK3 Deserializer which can support a single COAX input to a system (see high level block diagram as a refernce)
Assume a Power-Over-COAX architecture: What is the recommended voltage over the COAX in such a system?
What is the recommended PMIC to support a solid power supply in such a system?
Note: we know that there are more components on this system (LED, EEPROM), but we are now focusing on this specific issue.

Waiting for your reply

DLP3940S-Q1: DMD Remote Temperature Reads -64°C and Causes Shutdown — DLP3940S-Q1 + DLPC231S-Q1 + TMP411 (Remote Channel Issue)

jim liu — Tue, 20 Oct 2026 00:35:07 GMT

Part Number: DLP3940S-Q1

Hi TI Experts,

We are debugging a DMD temperature sensing issue on a DLPC23x-based system and would appreciate your help.

System setup:

Controller: DLP3940S-Q1 + DLPC231S-Q1 platform
Temperature sensor: TMP411AQDGKRQ1
Firmware version: 1.8.0.B1h
FPGA: bypassed in this build

DLPcomposer setup:FlashImage_p0.39_800_RDP.xml

Observed behavior:

Local temperature (TMP411 local channel) reads correctly and changes reasonably with ambient conditions.
Remote temperature (DMD channel) is abnormal:
1. Right after power-up, it reports a fixed +64°C that does not change.
2. After the system runs for a while, it switches to a fixed −64°C and stays there permanently.
3. The value does not track the actual DMD temperature at all.
The invalid remote reading eventually causes the DMD to be shut down (the system parks the DMD and disables illumination). We changed the thresholds so that the systerm could run futher.
There is no error or history error listed shows if there is an low speed bus disconnect.

Schematic:

I have attached the relevant schematic section showing the TMP411 connections and the DMD temperature diode pins. Please review if there is anything wrong.

What we have considered so far:

Based on the behavior, we suspect one of three root causes:

Hardware connection issue — the remote diode path (D+/D−) may be open, shorted, or improperly connected, causing the TMP411 to report a fixed invalid value.
TMP411 configuration issue — incorrect register settings (e.g., N-factor, series resistance correction, conversion rate, or ALERT masking) may cause the remote channel to be misread.
Firmware issue — since the temperature calculation (converting the TMP411 raw reading into a temperature value and handling faults) is performed in the DLPC23x firmware, a firmware bug or incompatibility could produce the fixed +64°C / −64°C readings and the subsequent DMD shutdown.

Questions:

TMP411 configuration: Are there specific TMP411 register settings (e.g., N-factor, series resistance correction, conversion rate, or ALERT masking) that are required or recommended when using the TMP411 with the DMD sense diode on the DLPC23x platform? Could an incorrect configuration cause the remote channel to be falsely detected as open?
Hardware checklist: From a hardware perspective, are there any known pitfalls in the DMD temperature diode routing (e.g., series resistance, filtering capacitors, parasitic leakage, or grounding) that could produce a −64°C even though the diode itself is intact?
Firmware behavior: Does firmware 1.8.0.B1h have any known issue related to TMP411 remote temperature handling? Is the temperature monitoring path independent of the FPGA, or does bypassing the FPGA change how the temperature data is read or interpreted?

We would be grateful for any guidance on how to isolate whether this is a TMP411 configuration issue, a board-level connection issue, or a firmware/error-handling behavior.

Thank you for your support.

Best regards,

Jim Liu

RE: DLP3940S-Q1: DMD Remote Temperature Reads -64°C and Causes Shutdown — DLP3940S-Q1 + DLPC231S-Q1 + TMP411 (Remote Channel Issue)

Jason Thompson — Tue, 20 Oct 2026 16:21:33 GMT

Hi Jim,

Normally, when -64C reading occurs there is either a poor connection between the TMP411 and DMD or the DMD is not making good connection to the PCB. In a rare case, the diode itself on the DMD might be damaged. It is unlikely related to a FW issue, specifically because you can measure the local temp on the TMP411 without issue. This would rule out these FW. The other register settings for the TMP411 and N-factor are fixed and same with other DMD chips, so it is not likely related.

Did you try to swap the DMD? Or, even re-seat the DMD onto the PCB?

Jason

RE: DLP5531LEQ1EVM: Lens breakage

Jason Thompson — Tue, 20 Oct 2026 14:47:29 GMT

HI David,

Unfortunately, there is not anyway to swap the lens as it is a custom designed module. The EVM is out of warranty and is really only meant for evaluation purposes in a lab. What type of use condition was the EVM under that caused it to break?

Jason

DLP5531LEQ1EVM: Lens breakage

David lim — Tue, 20 Oct 2026 12:59:23 GMT

Part Number: DLP5531LEQ1EVM

We purchased the DLP5531LEQ1EVM from Mouser last year, but the lens broke during use.

So I would like to swap this lens, and I’m wondering if I can purchase the lens separately.

Or can you send the entire product to me for repair?

RE: DLP3010: Connecting FPDLINK3 Deserializerto DLP3010

Neta Sherer — Tue, 20 Oct 2026 14:06:50 GMT

Hi Kevin,

Thanks for the quick reply.

We are building a module for defense application, so we can move quickly for a different DLP module.

The main purpose is to simplify the connectivity of the DLP to larger distances but in a wired way.

For that matter we prefer to use a COAX cable between FPDLINK Serializer-Deserializer with Power Over Coax.

Anyway, it looks like the DLPC3433 has DSI interface, can we connect this interface to one of the FPDLINK DSI Desriailizers?

RE: DLP3010: Connecting FPDLINK3 Deserializerto DLP3010

Kevin Rivera — Tue, 20 Oct 2026 13:45:19 GMT

Hello Neta,

Welcome to E2E forums and thank you for your interest in DLP Technology!

The DLP3010 is a part of a chipset with the controllers DLPC3433 and DLPC3438. This DMD can only be driven with either controller so it can not be connected directly to a FPD-Link Deserializer. This chipset includes a PMIC that would be recommended to use in the design.

The DLP3010 is also in a last time time buy state so I would not recommend creating new designs with this part. Could you share more about what your use case would be as well as desired resolution/frame-rate? If you share these, I can point you to a different part.

Thank you,

Kevin Rivera

RE: DLP3010LC: DLPC3478 + DLPA3000 + DLP3010LCFQK — correct firmware and DMD connection failure

Mark Hanninen — Tue, 20 Oct 2026 12:39:23 GMT

Hi Aaron,

1. Thank you, we've pulled our firmware from here.
2. No worries, thank you. For our PCB we're using DLPA3000.
3. 9.2, Red LED only.
4. We're successfully booting with the debug firmware and testing right now.
5. We do not have an external thermistor on our board. When you say it can be enabled, is that through the publicly available software or TI proprietary?

Thanks very much for your support!

Best,
Mark

DLP3010LC: DLPC3478 + DLPA3000 + DLP3010LCFQK — correct firmware and DMD connection failure

Mark Hanninen — Wed, 07 Oct 2026 20:54:41 GMT

Part Number: DLP3010LC

We are connecting a custom optical assembly to our PCB, which uses DLPC3478CZEZ and DLPA3000DPFD. Our intended DMD is DLP3010LCFQK.

With our existing projector-controller firmware, DLP3010FQK establishes a stable DMD connection, but DLP3010LCFQK does not.

Please identify the currently supported firmware image, download link, and release notes for DLPC3478 + DLPA3000 + DLP3010LCFQK.
Please confirm the firmware-selector settings required for our PCB, including PMIC selection, DMD pin mapping, and I²C address. We can supply the schematic and DMD flex/interposer connections.
Could firmware intended for legacy DLP3010FQK explain this failure with DLP3010LCFQK? What changes are required when migrating between these DMDs?
Which DMD identification/status results or boot diagnostics should we collect to distinguish a firmware compatibility failure from wiring, power sequencing, or signal-integrity problems?
Our PCB uses the DLPA3000’s CH1 LED output and provides an external thermistor input. Does the recommended image require configuration for our LED current limit and thermistor/divider characteristics?

RE: DLPC964: Inquiry regarding issues with a custom board using the DLPC964

Lee YeonJe — Mon, 19 Oct 2026 02:11:43 GMT

Hello, thank you for your prompt response.

I believe there may have been some misunderstandings, so I would like to clarify a few points and ask some follow-up questions.

We are not using a separate XC7VX415T FPGA. We are using the DLPC964 controller purchased directly from TI.
The link you referenced (DLPLCRC964EVM-SW) appears to provide the GUI and software tools for the DLPC964, which is not what we are asking about. We are looking for the MCS or BIN file that needs to be programmed into the Flash memory to boot the DLPC964. We obtained the file from the following link: DLPR964-FW v1.1. Could you please confirm whether this is the correct firmware package for this purpose?
Our issue is that the DONE_ signal is not being asserted. To clarify the required startup conditions, is it correct that DONE_ should assert when SYS_CLK100 and DMD_GTREFCLK are supplied, even while the system remains in the reset state? Or are there other prerequisites for DONE_ to assert?
Regarding Flash compatibility, my understanding is that TI only guarantees operation with the S25FL256SAGMFI011, and that compatibility with other Flash devices cannot be guaranteed. Is this correct?
The INIT_B signal goes High and remains High for approximately 2 ms before going Low again, even though PROGRAM_B stays High throughout the entire process.

The Flash read and write operations appear to be functioning normally. I also read back the Flash contents and verified that the readback data matches the original MCS file exactly. However, the DONE signal is still not asserted.

Thank you for your help and clarification.

RE: DLPC964: Inquiry regarding issues with a custom board using the DLPC964

Tristan Bottone — Fri, 16 Oct 2026 14:51:52 GMT

Hello JH,

Please see my response in blue below regarding the the customers DLPC964 questions.

Q1. I understand that there are firmware versions 1.0, 1.1, and 2.0 for the DLPC964, and that the choice depends on the FPGA being used. Is this correct?

Firmware versions 1.0, 1.1, and 2.0 (DLPC964-APPS-FPGA-FW) are for the front-ends Apps FPGA (ZCU102), not for the DLPC964 FPGA. TI's DLPC964 firmware is located here (DLPLCRC964EVM-SW) and are v1.0, v1.1, and v1.2.

However, please be aware that TI's DLPC964 firmware is encrypted and only runs on the security‑enabled DLPC964ZUM controller. A standard off‑the-shelf XC7VX415T cannot boot it, even on EVM-matching hardware.

The E2E forums here will explain this more in-depth -> https://e2e.ti.com/support/dlp-products-group/dlp/f/dlp-products-forum/1640904/dlpc964-i-fused-dplc964-fw-xc7vx415t-2ffg1158c-of-the-board-we-are-developing-but-there-is-no-done-signal

Q2. Should I use Version 1.0 when working with the XC7VX415T?

Same answer as Q1 above!

Q3. For Version 1.0, only the .MCS file is available. Could you please provide the .bit file as well?

Same answer as Q1 above! If you would still like me to provide the bit file, I can! Send me a friend request via E2E forum and send me a private message! I will send you the .bit file though E2E private message.

Make sure DONE_) is asserted before INIT_DONE. DONE_0 must assert before INIT_DONE! Successful flash programming alone doesn't confirm boot. If DONE_0 is high but INIT_DONE stays low, verify stable SYS_CLK100 (100 MHz) and DMD_GTREFCLK (112.5 MHz) before SYS_ARSTZ release, confirm power-rail sequencing, and check the Aurora/DMD link (HSSI_BUS_ERR).

Another great E2E reference for review -> https://e2e.ti.com/support/dlp-products-group/dlp/f/dlp-products-forum/1677158/dlpc964-dlpc964-init_done-is-always-0

Q5. If the flash memory recommended by TI is unavailable, are there alternative flash products that can be used instead? If so, please provide a complete list of suitable alternatives.

The validated flash we use is the S25FL256SAGMFI011 (Vivado device S25FL256Sxxxx1). S25FS256SAGNFI000 is a non-supported early-development option that can cause boot failure. No other specific alternate part numbers are available in current findings.

Another great E2E post to reference regarding Flash part can be found here -> https://e2e.ti.com/support/dlp-products-group/dlp/f/dlp-products-forum/1600600/dlpc964-the-dlpc964-chip-fails-to-boot-from-the-flash

Best Regards,

Tristan Bottone

DLPC900: DLPC900 video mode latency

Arik Pel — Wed, 14 Oct 2026 08:44:16 GMT

Part Number: DLPC900

When working in duel mode using the DLPC900 video mode, HDMI input, What is the video latency?

Is there any option to bypass the internal prosses to reduce the latency ?

Thanks

Arik

RE: DLPC900: DLPC900 video mode latency

Aaron Black — Thu, 15 Oct 2026 22:26:16 GMT

Hello Arik,

Could you please advise what version of firmware you have programmed here?

If you utilize a photodiode and look at the output of a white screen, you should be able to see how this is composed as you display red, blue, and green independently. Please keep in mind that this is 8-bit grayscale this means it's 3x (red, green, blue) at 8-bit totaling to a 24-bit frame.

I cannot directly say what that timing should be as this changes. The transition time for our DMDs is around 8us but that can change as well.

Best,

Aaron

DLPDLCR3010EVM-G2: DLP LightCrafter Display 3010 EVM shuts down immediately after startup

Art Mecina — Thu, 24 Sep 2026 19:19:22 GMT

Part Number: DLPDLCR3010EVM-G2

Hi Team,

Appreciate your help on our customer who are having issues with DLPDLCR3010EVM-G2 Please see below in verbatim.

"Customer File.zip I am having a startup issue with a DLP LightCrafter Display 3010 EVM (DLPDLCR3010EVM).

When I power on the unit, LEDs D2 and D3 are initially on. Then D4 turns on for approximately one second and the projector displays an image. After about one second, the projected image turns off, D4 turns off, and only D2 and D3 remain on.

The same behavior occurs whether the external display/video connection is connected or disconnected, so it does not appear to depend on the video source. The fan continues running. Power cycling the unit previously allowed it to operate normally temporarily, but the problem has returned.

Could you please help me determine what the D2, D3, and D4 LED sequence indicates and what could cause the projector to shut down approximately one second after startup? Please also let me know what measurements or diagnostic steps I should perform to identify the faulty component."

Thanks in advance,

Art

RE: DLPDLCR3010EVM-G2: DLP LightCrafter Display 3010 EVM shuts down immediately after startup

Kevin Rivera — Thu, 15 Oct 2026 15:43:00 GMT

Hello Alice,

Could you measure the following highlighted lines during bootup? If these three could be measured alongside HOST_IRQ this could help point to the potential issue.

Could you also confirm that the properly rated power supply is being used?

Additionally, the power going to the DMD would be good to capture, since proj_on is going low I do not suspect the issue is here, but it would be good to check.

Thank you!

RE: DLPC900: DLPC900 video mode latency

Arik Pel — Thu, 15 Oct 2026 13:16:05 GMT

Thanks for the replay, Aaron

One additional question:

Currently working in Video mode with the DLP9000 DMD using 2 DLPC900, 8 bits grayscale, operating at 60 frames per second and would like to obtain more detailed technical information regarding the micromirror timing characteristics.

Specifically, I am interested in understanding the timing constraints associated with mirror state transitions for the LSB and LSB +1, On & off timing:

What is the minimum micromirror switching or transition time specified for the DLP9000?
During 16.67 msec frame operation, what portion of the frame period is occupied by mirror transition and is the 8 bits has the right timing deviation? Meaning T/256, T/128 ….

Thank you for your assistance.

Arik

RE: DLPDLCR3010EVM-G2: DLP LightCrafter Display 3010 EVM shuts down immediately after startup

Alice — Thu, 15 Oct 2026 02:55:31 GMT

Hello Kevin Rivera,

Greeting for the day!

Could you help to check the customer's concern and support further ? thank you so much.

DLPC900: Gamma corection

Arik Pel — Wed, 14 Oct 2026 08:50:26 GMT

Part Number: DLPC900

Working with the DLP900 can I load a new Gamma correction table, and is there any application to build it?

RE: DLPC900: Gamma corection

Aaron Black — Wed, 14 Oct 2026 21:10:26 GMT

Hello Arik,

Thank you for your question on E2E!

I'm asking my team to come back to you.

Best,

Aaron

RE: DLPC900: DLPC900 video mode latency

Aaron Black — Wed, 14 Oct 2026 21:07:10 GMT

Hello Arik,

Thank you for coming to E2E with your question!

The DLPC900 will always have a frame buffered. If you are using the dual EVM we provide, the HDMI input is limited to 30Hz input. At this rate a frame will be at least 33.3us.

The controller is capable of more than that, but the concept is the same for having a frame of buffer.

Best,

Aaron

RE: DLP2010: Inquiry regarding replacements for DLP2010 and DLP3010

Aaron Black — Wed, 14 Oct 2026 20:46:54 GMT

Hello Grady,

We plan to have the 'LC' parts still be available for the next few years, likely 2030, at which time a LTB can be requested and stored parts can be utilized for years after.

If it is early into development and support is needed for a decade or further, I would advise to work with TI closer as we are developing alternates to this as a solution for these applications of Dental Scanning, 3D Print and Machine Vision. These parts will be available in the near future but no dates will be shared at this time.

Best,

Aaron

DLP2010: Inquiry regarding replacements for DLP2010 and DLP3010

Grady Lee — Mon, 12 Oct 2026 05:59:59 GMT

Part Number: DLP2010

Hi, TI expert

The customer has been consistently using the DLP2010AFQJ and DLP3010AFQK in their dental scanner products for a long time.

However, they recently received a "Last Buy" notification via a PCN, which lists "None" under "Recommended Replacement Products."

ti_202609030003_09072026_eol.pdf

The customer is inquiring about replacement options.
If the DLP2010 and DLP3010 are scheduled for EOL (End of Life), please let us know if there are any suitable replacement parts.

Please check. Thank you.