Subject:
|
RF transmitters
|
Newsgroups:
|
lugnet.robotics.handyboard
|
Date:
|
Thu, 8 May 1997 18:12:48 GMT
|
Original-From:
|
Randy Sargent <rsargent@&ihatespam&newtonlabs.com>
|
Viewed:
|
1799 times
|
| |
|
|
> Seems like I remember reading in NUTS and VOLTS, or somewhere, about
> Randy Sargent and his partners involved in a competiton, using a Radio
> Shack AM transmitter on their robot. It gave details on how they
> interfaced it. Sounded pretty simplistic. In retrospect, they commented
> that they would convert to FM because of some interference problems.
>
> Maybe someone else recalls reading about that. Or perhaps Randy could
> comment.
>
> Terry King
> Terry.King@fmr.com
We did use 900 Mhz Radio Shack RF headphones (cat # 33-1145), which gave us
a reasonable analog transmission (the transmitter took L+R RCA audio in,
and the we connected in place of the speaker on the receiver). We could
get 9600 baud through the connection, although since the connection was AC
coupled, we had to keep the "DC" content of our transmission pretty
constant.
On the receiver side, we just turned up the volume all the way, and put a
comparator on the output.
Keeping the "DC" content pretty much in the middle means we restricted
ourselves to sending bytes that had between 3 and 5 "1" bits set.
Surprisingly (to me), this still gave us 182 values per byte (out of the
original 256), and more suprisingly, meant that we had only a 6% loss of
bandwidth as compared to a channel with no restrictions (log base 2 of 256
is 8 bits, log base 2 of 182 is 7.5 bits)
As fun as the hacking was, I don't really recommend this approach.
Instead, we found that several folks were using modules from Radiometrix
(http://www.radiometrix.co.uk/). Abacom has been advertising these in Nuts
and Volts and Circuit Cellar. I went and the pages from Abacom's fax-back
service online back before Radiometrix got a web site:
http://www.newtonlabs.com/datasheets/abacom/:
Abacom's fax-back data for Radiometrix's radio transceivers
It includes a (possibly old) price sheet.
Reading the specs, it seems like you still need to do some encoding to keep
the DC in the middle to run at least some of these parts at their full
speed. It's pretty easy, though, to construct a lookup table to do this.
-- Randy
----------------------------------------------------------------
Randy Sargent Newton Research Labs
Senior Design Engineer Robotic Systems and Software
rsargent@newtonlabs.com http://www.newtonlabs.com/
|
|
Message has 1 Reply:
 | | Re: RF transmitters
|
| (...) A simple way to get rid of the DC component is to run the data through a scrambler on the TX side, and descrambler on the RX side. This has been done successfully in 9600 baud amateur radio TNC's. 2 or 3 chips on each end of the link is all it (...) (27 years ago, 8-May-97, to lugnet.robotics.handyboard)
|
2 Messages in This Thread:
 
- Entire Thread on One Page:
- Nested:
All | Brief | Compact | Dots
Linear:
All | Brief | Compact
|
|
|
|
