G Scale Model Train Forum banner
1 - 20 of 21 Posts

flysooner9

· Registered
Joined
·
29 Posts
Discussion starter · #1 ·
Was wondering if anyone knew of a substitute to the phoenix sound usb package? 90 bucks seems absurd for what seems like a usb cord and a 2.5mm adapter. Especially since the software can be downloaded.
 
Contact me. $50.00 plus shipping and I will send you a plug and play set just as it comes from Phoenix. I ended up with a spare that came with some other things that I purchased.
 
I will give "flysooner9" a couple more days to respond to my offer. After that its open to whoever wants it for $50.00 plus shipping. I have no idea who flysooner9 is. I like real names.
 
Was wondering if anyone knew of a substitute to the phoenix sound usb package? 90 bucks seems absurd for what seems like a usb cord and a 2.5mm adapter. Especially since the software can be downloaded.
I believe Phoenix will supply a schematic. I hear that is not worth the while to build one.
 
I believe Phoenix will supply a schematic. I hear that is not worth the while to build one.
yes, and yes.
I got the list of parts and the schematic from Phoenix. 17 individual parts needed, and a very complicated schematic I didnt even begin to understand. I didnt even attempt it.

Scot
 
Here is the schematic for anyone else who is interested. The basic rundown is this:

The FT232RL is a USB-Serial converter. LMC7221 is a comparator.

There are two data lines to the sound module, plus ground: DATA and PLUS_5.
When the board is connected, it sends a continuous 5 V back to the interface. This saturates Q1, pulling the CTS (clear to send) line on the FT232RL to ground, telling the computer software that the board is connected.
The TXD pin on the FT232RL sends data directly to the DATA i/o pin.
When it's time to listen, TXD presumably goes into a high impedance mode, and RXD receives data from DATA.
The comparator provides a voltage threshold that must be exceeded by the incoming data from the sound module to improve noise immunity.

(Edited a couple times for accuracy.)

Image
 
I have been curious for a while what was actually going back and forth, but hadn't gotten around to checking. I threw a logic analyzer on it this morning, and have edited my previous post to correct an error.
The sound module always produces 5 V on the PLUS_5 line. This tells the computer that a board is connected.

The transfer of data is initialized by the computer sending a command followed by /r (carriage return). The board then responds with a packet of data. The packet contains 8 bytes, followed by "/r /n >"
It runs at 9600 baud. No wonder it takes so long to transfer a ROM file.


Here's a typical exchange:
Image
 
actually eric, when q1 conducts, doesn't it pull CTS low? when not conducting, the 4.7 pullup is high, so CTS seems to be enabled low.

Q1 is biased by the 5 volt signal that should be there when the decoder is connected.

Greg
 
Of course. I hope none of my students saw this, since we literally covered BJTs on Wednesday. It's there to invert the logic since the steady state for RS-232 is high. I'll edit my description for posterity.

The question that you and I both have, I think, is why they would waste a line on this and share the data line instead of having separate rx/tx lines. Must be some legacy reason.
 
1 - 20 of 21 Posts