LPC5536 HSCMP dont work Hello everyone, We are using the HSCMP peripheral of the LPC5536 primarily to monitor a slow voltage. We configured HSCMP2 for "Nano Power Comparison," along with the DAC sub-module. However, we are experiencing instability with the output of HSCMP2, which oscillates. The duty cycle of the oscillation appears to depend on the voltage of the negative input, leading me to suspect that the positive input—the DAC signal—is not stable. When changing the power mode to "Low Speed Comparison," the oscillation issue is resolved. Additionally, my colleague has encountered similar problems with HSCMP0 and HSCMP1. He requires switching to "High Speed" mode for stable operation, but another PCB does not function properly with this setting. Has anyone else faced similar issues? Is there a reliability concern with the internal HSCMP? What steps can we take to achieve stability? Thank you for your input! Re: LPC5536 HSCMP dont work Hi, it happens with 3 out of 3 devices on Rev. 1B Chip. I compare a DC voltage with 8bit DAC (11/256 VCC (3.5V)) ~150mV. It triggered just an error flag in Application. I also checked it with Lab supply as input voltage. It always oscillate in Nano power mode changing the mode to a higher level will fix it. Re: LPC5536 HSCMP dont work Hello @nobodyKnows, In order to support you better, could you provide me more details about when the oscillation occurs? For example: -Which signals are you comparing? -Does it happen in 100% of the cases? -Do you perform any specific actions in your code when it occurs?
Additionally, I highly recommend reviewing these two application notes (AN13540 and AN13714), which provide detailed explanations of the HSCMP and may be helpful. BR Habib Re: LPC5536 HSCMP dont work Hello @nobodyKnows, I am investigating into this further, in the meantime. Could you provide me the error flag that you mentioned? BR Habib Re: LPC5536 HSCMP dont work it is really nothing special. I just test for COUT in CSR Register and set a Errorflag in my application. if(HSCMP2->CSR & HSCMP_CSR_COUT_MASK){
errorFlg = 1;
} Re: LPC5536 HSCMP dont work Hello @nobodyKnows,
It is expected that, in low‑power modes, the overall performance and responsiveness of the peripheral will be reduced. For example, for the HSCMP, in low‑power comparator mode the propagation delay can increase by up to a factor of 60 compared to the high‑speed mode, as shown in Table 49 of the datasheet. For this reason, you observe reduced performance across the different power modes.
BR Habib Re: LPC5536 HSCMP dont work Hello, I currently have another device where the HSCMP2 peripheral is not functioning. I have two identical devices on my table with the same firmware, one of which works and the other does not. To facilitate testing, I connected my laboratory power supply to the positive input of the CMP, allowing me to easily adjust the voltage. I am comparing the output at 163 mV generated with the DAC sub-module, which creates the reference voltage at the negative input from the 3.5 V supply. The output of the CMP is inverted. The working device sets the HSCMP_CSR_COUT bit when the voltage is below 160 mV; otherwise, the bit is 0. The non-functioning device sets the HSCMP_CSR_COUT bit below 210 mV and above 780 mV; in between, it remains 0. I directly readout the status with debugger. What I would also like to mention is that in the datasheet, the bit fields INPSEL (bits 29-28) and INPSEL (bits 25-24) in the CCR2 register cannot be written to and are always read back as 0. I think that the fields do not exist and that only MSEL and PSEL are used for the input mux. Is anyone else experiencing a similar problem? Is there a workaround? I'm starting to have a very, very bad feeling about using this mcu in mass-produced devices. Best regards, nobodyKnows Re: LPC5536 HSCMP dont work Hello, next device with defect on HSCMP2. Now it faulty triggers ~10sec after init and a false trigger above 2,9V. Can I assume that I will receive help here, or do I need to open a ticket regarding this matter? Re: LPC5536 HSCMP dont work Hi @nobodyKnows, I apologize for the delayed response. Since the issue is different to the original question, could you please open a ticket of this matter? This also help to bring a better personal support.
Inside this new ticket, could you please provide the next information? -The detailed steps you followed when the issue occurred. -If you are able to reproduce this issue on any of our EVKs. -To verify whether the issue follows the MCU, could you swap the MCU that is presenting the problem onto a board that is not showing the issue, and share the results? You can open a support ticket on this page.
BR Habib
記事全体を表示
