MAX3232: Asking for the MAX3232 abnormal waveform

Hi SHH,

This looks a bit strange, so I am not immediately sure what the problem might be. Were these waveforms collected while the device was connected to another RS-232 device over a cable, or was the unit disconnected? Would it be possible to view the signals at V+ and V- (each with respect to ground) so that we can observe how the internal charge pump is behaving?

Regards,
Max

Hi Max,

The three waveform above were connected to the RS232 to the desktop PC.

I attached the waveform which was removed the PC from the CN404.

Thanks,

SHH

Hi Max,

Please refer to the V+/V- waveform.

Thanks,

SHH

SHH,

The V- waveform doesn't look right, by which the voltage level should be around -5V. It looks like there are some coupling between RXD_in and V-. To debug let's check the basics first. 1) Please make sure all capacitors are rated at least 10V and no polarity concern; 2) Can you try to connect V+ cap C3 to ground instead of Vcc? 3) Can you check the ground routing of the board to be sure it's solid; 4) Please make sure zener diode protection are bipolar and can work with -5V to 5V. I'm looking forward to your response.

Regards,

Hao

Hi Hao,

1)      Please make sure all capacitors are rated at least 10V and no polarity concern;  

all theMLCC  cap rating are 16V and no polarity.

2)      Can you try to connect V+ cap C3 to ground instead of Vcc?    

connect C3 to GND. the issue is same

3)      Can you check the ground routing of the board to be sure it's solid;

Layout as following.

4)      Please make sure zener diode protection are bipolar and can work with -5V to 5V. I'm looking forward to your response.

Zener adopt GLZ22B VZ, voltage is 21V.

BR,

SHH

SHH,

Thanks for your information. Can you please check the supply voltage and (if possible) the current? Sometimes the supply has current limit. Please make sure the power to the device is at least 3V. Another experiment you can do is to load one of the channel with some capacitors to see the situation become worse. Sorry we still cannot pinpoint the cause. But I think we will get there.

Regards,

Hao

Hi Hao,
Please provide your comment for the result.
1.VCC 5V is ok
3）change C413/C414/C415 to 1uF also can fixed the issues

Thanks,
Scott

Scott,

What's the ratio of C412/C414 before and after change?

In general, for 5V supply, C1/C3 ratio is small to mitigate ripples.

Regards,

Hao

Hi Hao,
C412:0.1UF . No modification.
C414 change from 0.1UF to 1UF

please let me know if you have comment.


BR,
SHH

Scott,

I'm sorry I didn't fully understand the situation. My understanding is that the abnormal waveforms from your first post came with Vcc=3V with C412=0.1uF and C414=0.1uF. But after you changed the supply to 5V with C412=0.1uF and C414=1uF, the waveforms look normal.

If it's the case, can you please check the supply waveform of Vcc=3V? MAX3232's operation range is 3V to 5.5V. If the supply is lower than 3V or has noise, MAX3232 might response differently.

Regards,

Hao