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.

Max3223 powered with 3.3V - Voh/Vol levels typically below 5V/-5V

Other Parts Discussed in Thread: MAX3223, MAX3232

Have several CCAs designed with Max3223 ICs powered with 3.3Vdc, all with 4 X 0.1uF external caps. None of them have we seen typical voltage level of 5.4V, per datasheet. Half of them we see less than min levels of 5V/-5V. We have disconnected all test equipment, and measured voltage with scope probe directly on CCA output - and see approx 5.1V. What voltages have been observed @ TI during testing ?  Do we need to use different cap values to see the 5.4V typical output ?

CCAs with these ICs were all tested on the Takaya flying prober - with voltages less than 0.2V on each pin. Is it possible Takaya testing is damaging these parts ?

 

  • Hi Mark,

    Unfortunately, I am not the person that characterized this device, but I will try and reach out and contact the team that did.

    Would you be able to share any more information in regards to the set-up? A schematic showing the loading on the output and the bypass cap would be helpful. The datasheet is characterized with a 3kΩ load on the output. Therefore, if your load impedance is lower this could cause the output to fall out of spec.

    Thanks,

    John

  • Mark,

    The typical value of 5.4V will may be the average of  3.3V and 5V VCC results. The VOH specs covers both VCC ranges.
    However The minimum +/-5V applies to both VCC.

    The charge pump voltage responsible for driver VOH/VOL has a narrow valid range. If it is set too high then the charge pump will run continuously with 3.0V VCC and fail ICC. If it is set too low then the loaded output voltage will fall below 5.0V. In summary, most loaded results will be less than 5.4V when VCC is 3.3V.

    What is the load when the driver output is less than 5V? Was this measured with an oscilloscope or voltage meter?

  • Ron, Attached to the original post is a screen shot from O-scope, when we probed device output with no load. You can see that at some points we are above 5V....but then it ramps down to below 5V....we are trying to determine if this is indicitive of a damaged part - or just normal for 3.3V power application. Do you have any actual measurements of output voltage with this part powered with 3.3V that we could compare this to ?

  • Mark,

    I looked closely at the waveform. VOL holds steady and VOH droops. The droop on VOH is 0.6V over 0.6mS, with a 0.1uF storage capacitor yields a 10uA (0.1uF*0.6V/0.6mS) drain on V+ capacitor. The internal circuits have a logic using a V+ to V- path that should draw very little static current and no droop on V- shows this is the case. There is an internal voltage translator using V+ to ground for DIN logic that appears to be using the 10uA calculated current flow. I will run tests this week to see if that waveform is repeatable.

    With the data at hand, I do not believe there is any damage to the MAX3223.

  • Mark,

    I had to ordered samples so the data won't be available until next week. However, testing with MAX3232 (similar charge pump and drivers) supports that MAX3323 waveforms are normal except for low VOH,VOL.

  • Ron,

    Just wondering if you got those samples ?

    Also - did you power the MAX3232 with 3.3v...and use same cap values 0.1uf ? What levels did you see ?    Can you post a scope screen shot ?

  • Mark,

    I have the MAX3223 samples. I have output captures for MAX3223 with no load and with 3k load. Channel 1 is VOH and Channel 2 is VOL. Yes, VCC is 3.3V and the capacitors are 0.1uF.

    MAX3223_DOUT.zip
  • Mark,

    If needed, a couple devices could be submitted for failure analysis.

     

  • Ronald,

    My scope shot from initial post - was of pin 17.....what device pins are captured in your plot ?

    You show VOL w/3K load that does not meet VOL min in spec....it is above -5V....so is the spec too tight ?

    We have 2 parts pulled from our stores, and we have another part we removed from a CCA - that we only saw maybe +/- 4.5V on....and we have a 4th part - where the output can't get above 0V (I assume a blown part)....should we send all 4 parts to you ?   To what address ??

    We are trying to determine if parts that can only drive 4.8V, or 4.9V....are they bad....did we do something in the build process of the CCAs that exposed the ICs to excess heat.....or is our Takaya flying prober that applies 0.2V.....to all pins including V+ and V-....is that weakening the parts ?    or with the load in our test equipment - is it normal that we might see voltages as low as 4.5v....and the parts are all good.

    Thanks,

    Mark

     

     

  • Mark,

    I measured DOUT1 (CH1) and DOUT2 (CH2). The 3k load is passing, the informational -4.8V value includes the switching noise. I purposely used a 1V/div scale and set to zero on a major line so that it would be visually obvious that VOL(rms) is more than -5V.

    200mV will not damage any pins, ESD of type HBM, MM, and CDM can damage devices. 4.8V open and 4.5 loaded is probably not damaged in any way, but data sheet says that is not good.

    Normally local sales returns will go to local FA. However, for this special case it will be best to send them to me first. Afterwards, we can open a QTS job if needed. be sure the return device are stored ESD safe.

     

  • Ron,

    We delivered 4 parts to you - they were on your dock on Friday. (see tracking info below) Two of the parts were virgin parts from our stores, 1 part - blown? - fails to get above 0V......and 1 part - ok? - fails to get above +4V and below -4V. There should have been data in the box - but I also attached the 2 scope plots to this post. Let us know if you see same behavior there, and what your failure analysis group sees.

    Thanks,

    Mark

     

  • Mark,

    I tested the units and will send them to FA next week.
    I will post my wave forms on Monday.