Shop OBEX P1 Docs P2 Docs Learn Events
New FTDI chips — Parallax Forums

New FTDI chips

TubularTubular Posts: 4,703
edited 2014-04-13 16:57 in Propeller 1
There seem to be some new FTDI chips, such as the FT230X

There are a little cheaper than the FT232R, and available in 16 pins, about half the size of the FT232RL. This is good news because the FT232RL is largish, and FT232RQ difficult to hand solder. Cheaper too.

There are are some USB<>I2C converters in the new range too

Anyone had a look yet?
«1

Comments

  • jmgjmg Posts: 15,173
    edited 2012-03-24 04:03
    Tubular wrote: »
    There seem to be some new FTDI chips, such as the FT230X

    Nice looking parts, 50mA available at 3v3, and with MTP memory and no crystal. (claims 1667ppm)
    Also gives 24MHz 12MHz and 6MHz clock out choices

    ON the clocks it says this
    ["*When in USB suspend mode the outputs clocks are also suspended."]
    and also this

    ["Keep_Awake# CBUS0, CBUS1, CBUS2, CBUS3
    Prevents the device from entering suspend state when unplugged."]

    So maybe you can configure it to always deliver a clock ?

    Good baud rate choices too..
    Baud Rate = 3000000 / (n + m/8) { n >= 1, m = 0..7, if n=1 then m=0}

    I see they also have a 10 pin i2c slave part, and SPI variants...
    http://www.ftdichip.com/FT-X.htm

    A pity the extended Config memory cannot allow _one_ part number to do SPI and UART ?
  • LeonLeon Posts: 7,620
    edited 2012-03-24 04:54
    They are available from Farnell, D-K, Mouser, etc.

    I was going to use the FT232R in a design, but I think I'll use the new chip.
  • Cluso99Cluso99 Posts: 18,069
    edited 2012-03-24 05:30
    prices are considerably lower and the chips are smaller. However, the pin spacing is tight at 0.635mm.

    I wonder if it could be possible for someone to write a com driver that pulsed the gp pin for reset to save using the transistor reset circuit.
  • RaymanRayman Posts: 14,659
    edited 2012-03-24 05:38
    Is it 1/2 the price of FT232RL? That is a real motivator...
    I'm trying to figure out what this battery charge monitor signal is all about...
    Need to see if the default mode for the GPIO is the same... It's a real pain to reprogram every chip...
  • StefanL38StefanL38 Posts: 2,292
    edited 2012-03-24 05:44
    Hi Cluso,

    i'm using a small board with the CP2102-chip for downloading code into the prop. But - in this post - I don't want to focus on the chip. I connected the DTR-pin of the board over a 0,1 µF-capacitor to the reset-pin of the propeller-chip.

    So far I haven't experienced any problems with this "no-transistor"-circuitry. Is this just luck or do you want to aproach something different with the transistor?
    best regards
    Stefan
  • RaymanRayman Posts: 14,659
    edited 2012-03-24 06:09
    Might be hard to choose between 230 and 231... 230 is smaller and cheaper, right?
    But, 230 doesn't have DTR... So, you'd have to tell Prop Tool to use RTS instead...
  • LeonLeon Posts: 7,620
    edited 2012-03-24 07:40
    I'm working on a Propeller PCB design using it. Schematic (partly completed) is attached.
    pdf
    pdf
    QS2(Page1).pdf
    222K
  • RaymanRayman Posts: 14,659
    edited 2012-03-24 08:06
    Leon, I think your circuit has a lot of problems... So, I'd like to hear how it works for you...
  • LeonLeon Posts: 7,620
    edited 2012-03-24 08:17
    What are they? It's very much a work in progress.

    The FT230XS 3.3V supply is only 50 mA, so I might add a regulator.
  • RaymanRayman Posts: 14,659
    edited 2012-03-24 08:37
    I think the 230, 231 chips are really just repacked versions of the FT232RL chip. Might be wrong, but at first glance it looks like the pins work the same way...
    So, I think the designs of the Quickstart or other boards are still good places to start from.

    Leon, I don't think the LEDs will work the way you have it... Also, I think I'd pick RTS instead of CTS... My version of Prop Tool (old?) works with RTS and not CTS.
  • LeonLeon Posts: 7,620
    edited 2012-03-24 09:24
    It's a completely new chip. I used the circuit in the data sheet.

    I should have used RTS. Schematic has been updated.
  • RaymanRayman Posts: 14,659
    edited 2012-03-24 11:19
    Chips are new. But, I seriously doubt it's a "Completely New" design.

    Here are the differences I see:
    1. smaller and cheaper
    2. battery charging mode
    3. you now need external resistors and capacitors on the USB signal lines

    Anything else?
  • Phil Pilgrim (PhiPi)Phil Pilgrim (PhiPi) Posts: 23,514
    edited 2012-03-24 11:49
    I just made a rather astounding discovery. I was looking at the FTDI datasheet and saw a reference to the Micrel MIC2025-2BM soft-start IC. So I copied and pasted that part number into (what I thought was) the Google search box. But, instead of the typical Google page, up popped the datasheet in the PDF reader! I'd never seen that happen before, and went back to Firefox to investigate. It turns out that, instead of typing the part number into the search box, I had typed it into the address (URL) field. I'm using OpenDNS as my DNS provider, so maybe that's a service that it provides that I wasn't aware of. Or maybe Firefox intercepts the bogus URL and redirects it?

    I've tried it with other part numbers. Sometimes it opens a PDF datasheet; other times it redirects to a manufacturer's product page for the part.

    Anyway, 'sorry for the hijack.

    -Phil
  • tonyp12tonyp12 Posts: 1,951
    edited 2012-03-24 12:50
    When Firefox is given an invalid URL, it doesn't attempt to determine the actual URL.
    Instead, it performs a "I'm feeling lucky" google search when confused and returns the first matching web page.

    Also some DNSservers make money on invalid ursl and show ads instead.
  • Cluso99Cluso99 Posts: 18,069
    edited 2012-03-24 12:50
    Its a new design. See the errata sheet. Guess they had to compete with the CP2102 which is apparently much cheaper (not on DigiKey though).
    Note it also requires series resistors to the USB and a few more caps :(

    Stefan: I have successfully used a CP2102 board by tying the DTR directly to the prop reset. The problem occurs when switching between windoze/linux programs when the DTR gets toggled.
  • jmgjmg Posts: 15,173
    edited 2012-03-24 15:13
    Rayman wrote: »
    3. you now need external resistors and capacitors on the USB signal lines

    That looks like rather like an 'oops', hard to imagine the designers starting saying :
    'Hey, we are making a low cost, smallest footprint chip, let's add 2xR ans 2xC, lest this gets TOO small '

    - I can see series R could help ESD figures, but surely the average cable already has that C?
    Did they fail a EMC test, and needed a RC band-aid ?
  • RaymanRayman Posts: 14,659
    edited 2012-03-24 19:44
    It is a little strange... Just as a guess, I've heard that real resistors and capacitors take up a lot of silicon space.
    So, maybe they moved them off the die to make it fit the smaller package?

    Or, you could be right and they just messed up and needed this fix...
  • WBA ConsultingWBA Consulting Posts: 2,934
    edited 2012-03-24 23:38
    In regards to the external parts, I know of other chips that did the same "improvements" because many designs using their parts already had the values in question on the board elsewhere. So, in volume, it is much easier to add 2 more resistors at $0.0015 each and a few more caps at $0.004 each than to switch to a different IC that costs $2 more to save a few parts on the board. FTDI services customers at much higher volumes than us hobbyists, so that is where this solution to compete with the CP2102 probably comes from.
  • Phil Pilgrim (PhiPi)Phil Pilgrim (PhiPi) Posts: 23,514
    edited 2012-03-25 00:06
    If you're using a service bureau to populate your boards, you also have to figure the total number of components for the pick-and-place charge and the number of different components for the P&P setup charge. That can easily erase the savings in parts cost alone, and then some.

    -Phil
  • Cluso99Cluso99 Posts: 18,069
    edited 2012-03-25 03:38
    And that also adds to real estate, taking away from any savings in space. BTW the resistors are supposed to be about 68R for the USB spec, so seems they goofed and the resistors are only about half that value!
  • LeonLeon Posts: 7,620
    edited 2012-03-25 03:54
    Perhaps it has internal resistors and they got the value wrong.
  • jazzedjazzed Posts: 11,803
    edited 2012-03-25 09:03
    Rayman wrote: »
    But, 230 doesn't have DTR... So, you'd have to tell Prop Tool to use RTS instead...

    What version of Propeller Tool has this? Where does one find the option?
  • Phil Pilgrim (PhiPi)Phil Pilgrim (PhiPi) Posts: 23,514
    edited 2012-03-25 09:07
    I know version 1.3 has it. You can find it in the options (or somesuch name) menu and can choose DTR, RTS, or both.

    -Phil
  • LeonLeon Posts: 7,620
    edited 2012-03-25 09:57
    Edit > Preferences > Operation. It took me some time to find it.
  • jazzedjazzed Posts: 11,803
    edited 2012-03-25 12:01
    Thanks Phil and Leon.
  • TubularTubular Posts: 4,703
    edited 2012-03-25 12:24
    To me those external components look like they are required for formal compliance with the USB spec, or for EMC reasons, or for marginal installations without cables.

    I suspect it is a case of "they technically should have been included on the FT232R circuit" rather than a goof on the new chip.

    The 47pF caps.... USB cable capacitance is about 20pF/foot, so the value of these 47pFs is dwarfed even by a 6 foot cable.
  • RaymanRayman Posts: 14,659
    edited 2012-03-25 16:20
    Another possiblility is that they wanted to use smaller protection diodes. Most devices I've seen like this need external protection diodes. FTDI has them internal, but they probably also take up a lot of space. Maybe adding this RC filter to the front end allows them to use smaller protection diodes...
  • User NameUser Name Posts: 1,451
    edited 2012-04-18 12:13
    For those who find PCB layout more about tedium and less about recreation (raises hand), it really is pretty difficult to beat a CP2102 board on eBay. They cost less than $3, shipping included. My experience has been that the Silicon Labs drivers (downloaded from their website) work flawlessly in Windows 7.
  • RaymanRayman Posts: 14,659
    edited 2012-04-18 12:34
    FT231X is cheaper than CP2102 at Digikey.

    Here's how the FT231X looks compared to FT232RL BTW:
    FT232R_Vs_FT231X.JPG
    FT232R_Vs_FT231X.JPG
    410 x 220 - 35K
  • Phil Pilgrim (PhiPi)Phil Pilgrim (PhiPi) Posts: 23,514
    edited 2012-04-18 12:49
    It must still be a largish die if they have to extend the package that far beyond the pin footprint.

    -Phil
Sign In or Register to comment.