• State
  • Locked Locked
  • Replies 6 replies
  • Subscribers 33 subscribers
  • Views 493 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.

MAX202I failing at temps > 50 degC

Sasha Marks
Prodigy 40 points
Other Parts Discussed in Thread: MAX202

Hi, 

I am using a MAX202IDG4 in the exact same configuration shown in the datasheet.  I am using 100nF X7R 50V capacitors.

The application is in an LED light.  I have noticed that if the ambient get to over around 50 deg C, then I try to send a command, the command is received correctly however the reply (DOUT2) is not within spec.

It seems to only happen if the unit is hot *before* sending a command.  If the unit is cool, then the serial port accessed, there are no problems - even if the unit heats up in the meantime!

It's not isolated to one or two boards, but the whole batch (20+ units).

I've put a scope on the RS232 level RX and TX lines, and when it fails, it shows the TX 'mirroring' the RX line, but very badly deformed and only peak to peak of around 4V.  When temp drops below 45degC the TX returns to normal looking signal.

Any thoughts?  I've tried putting in 1uF caps and the same thing happens.  Its really strange that once it's working, it keeps working - even over 60degC!

Warm Regards, 

Sasha.

  • 0
    Mastermind 7395 points

    I don't understand several bits of your message.

    For example, what does it mean that it keeps working? An RS232 will usually send very small packets and it is difficult to see how it will change temperature in the middle of such a small packet.

    Also I didn't understand the bit about Rx and Tx, and the "mirroring". I would check, for example, the Rx in and Rx out, Tx in and Tx out. Could it be that whatever is sitting on your board is getting sometimes confused and replying with bad data, and the driver is only reproducing the bad data?

  • 0 in reply to Avi Chami1
    Prodigy 40 points

    Sorry about confusion - to clarify:

    The ambient temperature of the electronics increases as the device (an LED light in this case) is left on, until it stabilises at around 55 deg C.  

    1) If packets are regularly sent from the device as it warms up, then no problems with the Transmit waveform are noticed.

    2) If however the first packet is sent from the device, when it is already > 50deg C Ambient, then the waveform is distorted.

    About the 'mirroring':  The Tx line is 'mirroring' what is on the Rx line (only it case 2 above) with no delay at all.  The cause of this I discovered is the charge pump circuit is failing to start (see below)

    I've done a bit more investigations and found the following:

    * If the ambient temperature gets to > 50degC, the charge pumps do not run (No square wave)

    * If the temperature drops to <45, they start running again

    * The charge pumps produce a square wave at approx 600kHz.  Interestingly, other MAX202 chips I've used in the past (From TI, exactly same part number) all run at around 120kHz)

    Ideas?

  • 0 in reply to Sasha Marks
    Mastermind 7395 points

    Did you check your P.S? Could it be that the P.S. is not delivering enough voltage at high temp.?

    Or, could it be that when the pump is started at high temperature, your P.S. drops?

  • 0 in reply to Avi Chami1
    Prodigy 40 points

    I checked the PSU, it is delivering exactly 5 volts.  I also put a scope on the rails and checked for ripple, there is approximately 40mV Pk-Pk ripple, which does not change as the temperature changes.

    I tried swapping the MAX202 chip for another one we have in stock and the problem went away.  Could it be possible that I've been supplied a bad batch of MAX202 chips? (the problem is happening on multiple boards).  The PCB's were assembled in China (this is the 3rd time they've made them for us though, and never seen this problem before).

    The markings on the MAX202 chip read: 15CPF5K  G4  MAX202I.  It looks like a genuine part, and I've been assured by our supplier that it is genuine (for what thats worth...)

    Has anyone experienced with counterfeit  TI parts before?

  • 0 in reply to Sasha Marks
    Mastermind 7395 points

    http://www.designchainassociates.com/counterfeit.html

    Sadly, I have heard many reports for transistors and even DRAMs. I have not personally heard about TI, but you can see in the link above. You should contact your local representative to see how to pursue the issue and to check the markings.

    All the best

    Albert

  • 0 in reply to Sasha Marks
    Prodigy 20 points

    I was going through some older forums and noticed you were having issues with a MAX202 chip.

    if i can help you - we have a lab that does testing on IC's to see if the parts meet the Mfg standards and is specifically designed to lok for counterfiets.

    the industry is plagued with substandard parts coming from China--  So if there is anything ican do to help- please call me.

    i can have your chip checked for you if you like.  call me 631-580-9500 x215   

    Thanks, Chris Joseph   

    Lintech