Author Topic: CMT2210  (Read 1494 times)

0 Members and 1 Guest are viewing this topic.

Offline hpmaximTopic starter

  • Regular Contributor
  • *
  • Posts: 134
CMT2210
« on: April 02, 2019, 07:07:06 am »
I am trying to hack up a remote controlled device by hooking a Raspberry Pi to it to emulate the data from the remote -- and running into a bizarre problem.  The device uses a CMT2210LD (I'm not absolutely positive about the LD part, the chip silk screen was next to impossible to read even under a microscope).  According to the only datasheet I could find: https://www.maritex.com.pl/product/attachment/38884/CMT2210L.pdf the pinout is as follows:

1) Ground
2) VDD
3) Data Out
4) Xtal Connection 1
5) Xtal Connection 2
6) VCO Connection 1
7) VCO Connection 2
8) RF Input

Pins 5 and 6 are supposed to be tied together through an inductor, and pins 4 and 5 go to a crystal...

Pretty straightforward.

Here's the weird part, and something I only noticed after I couldn't get it to work.  I soldered a 30 gauge wire to pin 1 (or rather the pin with the embossed dot in the upper left corner).  I clipped my probe's ground lead to the wire and then probed pin 3.  Sure enough, I got data out, and I could even figure out the protocol... but here's the weird thing.  Pin 3 seemed to go switch between GND and NEGATIVE VDD.  And, no, it wasn't something screwed up in my scope setup.  When I clipped the ground lead to the ground on my Raspberry Pi the output from the Pi swung from GND to VDD.

The other weird thing, and its hard for me to know what to make of it is there is a 10k ohm resistor attached to the ground pin.  I'm not clear if it's in series with the chip ground, or in parallel, but neither really make much sense to me.  There is no other active device on this part, and lest you think that the 10k might be a pull-down on the data-out pin...  Nope... it's not.

The negative VDD thing may not be the end of the world, I might be able to simply tie the Pi's 3.3V VDD to the ground pin to simulate the reversed swing, but I'm not keen on doing that until I understand what I'm seeing better.  Is it possible the datasheet is wrong about power and ground?  Given the PCB layout, pin 1 had a much beefier connection to the PCB than pin 2, which would suggest pin 1 is in fact ground.
 

Offline hpmaximTopic starter

  • Regular Contributor
  • *
  • Posts: 134
Re: CMT2210
« Reply #1 on: April 02, 2019, 02:04:05 pm »
I just tacked a 30 gauge wire to pin 2 and... sure enough, it appears VDD = -3.2V...  Is it really possible the datasheet is wrong about something so... important as VDD vs ground?
 

Offline JackJones

  • Regular Contributor
  • *
  • Posts: 229
  • Country: fi
Re: CMT2210
« Reply #2 on: April 02, 2019, 02:48:41 pm »
I don't think the datasheet is wrong. They would have had to mess up both the pinout and the example schematic. I also found a picture of the chip which shows pin 1 as GND.

Did you measure the VDD with a multimeter? Are you sure it's not just a silly mistake like accidentally having inverting on on your scope?
 

Offline hpmaximTopic starter

  • Regular Contributor
  • *
  • Posts: 134
Re: CMT2210
« Reply #3 on: April 03, 2019, 05:11:14 am »
So, I'm thoroughly confused and yet at peace...

Yes, I was measuring the voltage correctly.  I have no explanation whatsoever for this apparent goofiness, only more mysteries.  The board had 1 inductor, 1 resistor, and 4 caps, whereas the schematic has 4 caps and 2 inductors.  I can't possibly imagine what the resistor (10k) was doing or supposed to be doing.  But I swapped my ground from pin1 to pin 2 and it OCCASIONALLY worked.  Meaning it did what it was supposed to, but rarely.  I thought this was kind of odd.  Something had to be floating because I was confident the data stream was correct.

So therein lies the rub.  There was a 4 pin header on the board.  I was initially thinking this was some sort of probe points, but on more careful inspection, I realized this is how the radio board communicates with the main board.  It appeared pin 1 was ground, pins 2 and 3 were shorted together and were data, and pin 4 was VDD.  So I was probing the VDD with my scope probe and found that it was noisy.  My ground lead wasn't hooked up, but I figured the Pi is grounded through USB to my computer which is on the same receptacle as my scope, so they should have a common ground and the noise shouldn't be bad.  But regardless, I touched the ground clip of the probe to pin 1 of the header and magically everything worked.  So... there you have it.  I ran my wires directly from the Pi to the header and now it magically works perfectly.  I don't know what goofy stuff they did with that resistor and the CMT2210 which seems to be wired wrong, but yet somehow still works, but when I just connect to the header it's all good.  I'm super psyched...

Now, if I can just figure out how to route traffic to the Pi's webserver through the Wifi tether on my phone I'll be happy.  Somehow, I suspect this may not be trivial.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf