Reply
Posts: 18
Registered: ‎03-29-2017

CP2102 3.3V output is 4.25V on CJMCU module

I ordered a CP2102 USB-to-serial module from Ebay, but I found that the output labeled "3.3V" on the silkscreen actually measures at 4.25V.  I thought it was just a bad part, but I found the same module on sale at Banggood, and the same problem is mentioned in reviews posted by three buyers.  One of them had ordered three modules, and all of them were 4.25V.  So it appears all of these modules output 4.25V instead of 3.3V.  I can't use 4.25V because the microprocessor it will connect to has a maximum voltage of 3.6V.

 

This is the module I bought:

 

http://www.ebay.com/itm/272469914995

 

On the back side it says "CJMCU" and "BQ09615H2  2025a"

 

The chip is labeled:

 

SILABS

CP2102

DCL00X

1612+

 

I don't see anything in the datasheet that indicates the output of the regulator is adjustable.  Can anyone offer an explanation for the 4.25V output?  Could all the chips, or a large number of them, be faulty?  That seems unlikely.  It seems much more likely that this was done deliberately.

 

I also ordered another CP2102 module of a different design, and its 3.3V output is correct:

 

http://www.ebay.com/itm/291874318571 

 

But it is bigger than the first module, and less suitable for my project.  So I would like to find a source for the first module with an on-spec 3.3V output, if that exists.

 

Any help or explanation would be appeciated.

 

Sherman

 

Posts: 1,845
Registered: ‎10-14-2014

Re: CP2102 3.3V output is 4.25V on CJMCU module

It was strange, do you have a schematic for this board?

WeiguoLu
Posts: 18
Registered: ‎03-29-2017

Re: CP2102 3.3V output is 4.25V on CJMCU module


delu wrote:

It was strange, do you have a schematic for this board?


I do not have a schematic, but continuity testing with my meter confirms the 3.3V output pin of the module is directly connected to pin 6 of the CP2102.  So the voltage regulator is generating the 4.25V.  I have removed the LEDs and their resistors, so the only parts left on the module are three capacitors and the CP2102.  The voltage is still high.

Is it possible that Silabs made a special run of CP2102s with the higher regulator output?  Is it possible to change the regulator output by flashing new parameters to the CP2102?

 

If not, what would be the explanation for this?  Is this a common failure mode?

 

Posts: 1,845
Registered: ‎10-14-2014

Re: CP2102 3.3V output is 4.25V on CJMCU module

Does the board still works fine?

according the datasheet the regulator should output a voltage in the range of [3.0 3.6]. refer table 6 on page 9 of the CP202-9 datasheet.

Since you don't have the schematic, it is hard to know what problem the issue come from.

Since you also have some device that seems normal, I suggest you try to measure the resistance between the VDD and VBus to see if there is some difference between the normal and problematic part.

My views are my own and do not necessarily represent the views of Silicon Labs

 

 

WeiguoLu
Posts: 2,875
Registered: ‎02-07-2002

Re: CP2102 3.3V output is 4.25V on CJMCU module

My guess is that REGIN (pin7) is not connected to VBUS on the connector, but VBUS (pin8) is. This will probably leak power through the bulk diode of the silicon to VDD (pin6) which explains the 0.7V drop.

Posts: 18
Registered: ‎03-29-2017

Re: CP2102 3.3V output is 4.25V on CJMCU module


vanmierlo wrote:

My guess is that REGIN (pin7) is not connected to VBUS on the connector, but VBUS (pin8) is. This will probably leak power through the bulk diode of the silicon to VDD (pin6) which explains the 0.7V drop.


So REGIN-pin7 should be connected to VBUS on the connector?  What should pin8 be connected to?

Well if you are right about this, then all of them are that way.  And somewhat difficult to repair in QFN land.

 

 

Posts: 1,845
Registered: ‎10-14-2014

Re: CP2102 3.3V output is 4.25V on CJMCU module

If you take look at the datasheet, for bus powered mode both PIN7 and PIN8 should be connected to VBUS.

 

WeiguoLu
Posts: 18
Registered: ‎03-29-2017

Re: CP2102 3.3V output is 4.25V on CJMCU module

I was able to check continuity, and both pin 7 and pin 8 are connected directly to the 5V pin on the USB connector. And I previously determined that the 3 3.3V output is connected properly.

 

Can you think of any other connection, or lack thereof, that would cause the regulator not to output 3.3V?

 

If not, then we either have a bad batch of CP2102s, or perhaps there was too much heat  in reflow.

 

 

Posts: 18
Registered: ‎03-29-2017

Re: CP2102 3.3V output is 4.25V on CJMCU module

Is it possible that the flash memory of this device has been modified in a way that would cause this behavior?  The ANs say there is a setting for Power Option where "Bus Powered" is the normal setting, as well as a setting for maximum milliamps.  Would selecting some non-standard options cause the regulator to be bypassed in this way?

 

The driver does recognize this device, so I think that means the identifiers are correct.  If I install the Silabs flashing software, will it tell me the current settings?

 

Just looking for an explanation that might have a solution.  Anyway, I have three more of these coming in from other sources over the next couple weeks, and will see if they have the same problem.  I just suspect they are all like this, but only those who bother to measure the voltage notice that it's wrong.  But if all are like this, then there may be nothing wrong with the chip, just some circuit design or parameter setting that was done deliberately for some purpose.

 

Does anyone know who CJMCU is?

 

 

 

Highlighted
Posts: 18
Registered: ‎03-29-2017

Re: CP2102 3.3V output is 4.25V on CJMCU module

Ok, it's not just my imagination.  Here is a Youtube video of a guy with a handful of modules exactly like mine, and they all measure 4.25V at the 3.3V output:

 

https://www.youtube.com/watch?v=8qcGdB2yVd8

 

So this was done intentionally, and all modules of this design are like this.  But how did they do it?

 

alaways check the stuff you buy from ebay! these usb to usart ttl modules output 4.2V instead of 3.3V. this is enough difference to ruin some 3.3v only devices.