PDA

View Full Version : XTR-434 transceiver



Toufic
- 25th June 2008, 09:10
hy everyone i am a new user here...
i have a project and i need some help plz

i am using the aurel transceiver xtr434 and i am getting on the pin 12 data out a voltage of 2.45V whatever the input of the other transceiver is (0 or 5V)..
so if anyone has a shematic for how to connect this transceiver or if anyone faced this problem send me the reply plz..


thanks in advance
regards...

mackrackit
- 26th June 2008, 09:18
http://www.aurelwireless.com/rf-wireless-modules/data-transceivers.asp

Toufic
- 27th June 2008, 07:35
10x for the reply man bt i have already downloaded the user manual and the datasheet from the aurel site and they tell u how to relay each pin and i have done this but still the output on pin 12 is 2.45 and the say in the datasheet that it is supposed to be 0 or 5V
so that is my prob..
thank u again

Ioannis
- 27th June 2008, 08:01
From the data sheets I see that your load on pin 12 should be more than 100K. At 10K load the output is 3.5 volts.

What are you driving with pin 12?

Ioannis

dhouston
- 27th June 2008, 12:03
10x for the reply man bt i have already downloaded the user manual and the datasheet from the aurel site and they tell u how to relay each pin and i have done this but still the output on pin 12 is 2.45 and the say in the datasheet that it is supposed to be 0 or 5V
so that is my prob..
thank u againAre you reading the manual correctly? The manual indicates there will be random noise whenever the RX is enabled. You must use the CD pin to determine whether it is signal or noise. The manual also indicates that you need to use manchester coding as the dataslicer is designed for a 50% duty cycle (and will not function if the duty cycle deviates too far). This means that you cannot set the transmitter High or Low (steady-state) and expect the RX pin to reflect that. If you send a series of pulses (50% duty cycle), you should see the same at the receiver. The manual specifically notes that you cannot use the module for RS232 data streams without using some scheme to achieve DC balance (e.g. manchester coding). All in all, this is not a module for the inexperienced.

Toufic
- 28th June 2008, 10:06
hey 10x for ur reply
but let me ask u something if i am using his transceiver as receiver all the time unless i am sending data and the data input for the transmitter is from the USART of pic16f877a i am using a baud rate of 9.6kbps but if i changed the rate to 56.7kbps would it work...
i am not familiar with this manchester coding thing so that is why i skipped this part..
thanks again

dhouston
- 28th June 2008, 10:56
Page 3 of the manual says...Bit ON/Bit OFF relation: the Data Slicer is optimized for a 50:50 duty cycle. It will continue to operate,
even with bigger distortion and less tolerance to interferences, till a 30:70 or 70:30 duty-cycle.
Therefore there is no possibility to directly transmit an RS232 sequence if no action is done to balance the 1
and 0 bits ratio, as the duty-cycle could reach up to 90% preventing the proper performance of receiver. It
will continue to operate, even with bigger distortion and less tolerance to interferences, till a 30:70 or 70:30
duty cycle.XTR-434
While modulation is applied to TX, conditioning input pin [pin 14] with a logic signal, it is reccomended not
to exceed 200 μs continuous time with no transition from ON to OFF or OFF to ON. This not to downgrade
the RX sensitivity. It is requested that modultation is carried on with techniques that allow a low duty-cycle,
such as Manchester coding, 8 to 12 bit coding or other available technique. If no bit balancing technique is
used, if it is requested to work with RS232 protocol, a minimum speed of 57.600 bps is required to assure
maximum performance, transmitting, for example, one byte followed by the complementary byte (byte
balancing).The Data Slicer is a comparator with a capacitor on one input. The data goes to both inputs but the capacitor charges to an average of the data. When the data is higher than the average level, the output is high and when the data is lower than the average level, the output is low. The higher baud rate allows less time for the capacitor to discharge which may explain why it appears to work at higher rates but you should not count on this being reliable unless you use byte balancing and rates of 57.600 bps or higher.

Follow the manufacturer's recommendation and use manchester coding (or byte balancing) or find another transceiver. You can find thousands of web pages that describe manchester coding and a search of the forums will turn up Melanie's example code.

Byte balancing means you send each byte twice - once normally but followed immediately by its bitwise complement (i.e. each 1 becomes 0 and vice versa). The sequence is Byte1 ~Byte1 Byte2 ~Byte2...ByteN ~ByteN. On the receiving end, each pair must sum to $FF. If not, there was an error in transmission or reception.

You can see some oscilloscope screenshots showing the effects of the data slicer here...http://davehouston.org/rf-noise.htm

Ioannis
- 28th June 2008, 14:55
What Dave is trying to tell you is that you cannot raise the transmit line to 1 and expect that at the receivers output you will get a steady 1 for long time. This is NOT possible with the XTR module. It is possible with other tranceivers that are capable of direct RS-232 transmission.

RF is very different than direct cable link.

As for Manchester do a search and you shall find. Alot of infos. With 2 words works like this:

to send 1 0 0 0 0 0 0 1 the PIC should send the 1 0 0 1 0 1 0 1 0 1 0 1 0 1 0 1 1 0, that is the 1 is translated to 1 0 and the 0 to 0 1. So even if you send a steady 1 or 0 the XTR wil get a changing 1 0 or 0 1 at the input, cancelling out the DC component that a steady 1 or 0 carries.

Hope it makes some sense now.


Ioannis

Toufic
- 29th June 2008, 07:33
thank you guys...
i think now i am going to change the transmitter...
but the problem is that we are a small country here and cann't find all kind of transceivers here..so if u know a transceiver that could work in my situation please send its number to me and i will check it out..
thank you again

Ioannis
- 29th June 2008, 11:03
If the encoding can and it sure can be done in software there is no need to change the tranceiver. It is no difficult to do it. But then, it is up to you...

Ioannis

dhouston
- 29th June 2008, 11:41
Byte balancing is super simple. I've provided an example in the Code Examples forum. So, it's not absolutely necessary that you change the hardware.

However, there may be other factors (e.g. cost) that would make a change more attractive but, without more detailed knowledge of the application, it's impossible to suggest other hardware. Is this a one time project or do you plan to manufacture a product? If it's a product, you'll need to meet FCC and/or CE requirements, depending on where you sell. What is the application? What range is needed between transmitter and receiver? Is a transceiver necessary or can you use separate transmitters and receivers? ASK modules are usually 20-30% as costly as FSK. There are some very good RF modules (made in China) available through an EBay store for extremely attractive prices (cheaper than dirt - even with collapsing real estate prices).

Chrike89
- 21st January 2009, 11:28
hy everybody....

i've recently bought a xtr-434 for a project
i want to use it for a one-way serial transmission

i've spent much time on it but it still does not work

below there are the schematics of my transmitter and receiver circuit and a link to the user manual of the xtr 434 and a max232 converter

i hope somebody can help me

XTR 434 Manual (http://www.aurelwireless.com/wireless/uk/manuals/650200588_um.pdf)

Max 232 Manual (http://focus.ti.com/lit/ds/symlink/max232.pdf)


===========================
Transmitter schematic
===========================
http://www.wasserrettung-faak.at/Transmitter.jpg





=============================
Receiver schematic
=============================
http://www.wasserrettung-faak.at/Receiver.jpg

Ioannis
- 22nd January 2009, 07:10
What about your non-working code?

Ioannis

Chrike89
- 22nd January 2009, 10:28
...i've found some failures in my schematic...so i have redrawn it

here is the link
Transmitter Schematic (http://www.wasserrettung-faak.at/TransmitterNew.jpg)

at the time i have no code because i am using Windows terminal but i think i must write a program because i've found out that a preamble (of 01010101...) is needed

do you see any other failures??

Ioannis
- 22nd January 2009, 11:28
You need not only a 2ms at least preamble but also some kind of data encoding to keep the balance of the 1 and 0 to 50%. So Terminal will not help much except if you can transmit the character with this binary number: 10101010 or 01010101.

That means definetly Manchester or Biphase encoding as it is the easiest to implement.

Do a search on the forum and many answers on the topic Manchester will be found (mine too!).

Please re-read the data sheets of the XTR. Answers included!

By the way, you new schematic, I believe will ton work! XTR is supplied through a capacitor and data are fed through 10K and 100K resistors. Are sure about the values? I'd rather choose your first one.

Ioannis

Chrike89
- 23rd January 2009, 08:31
i have asked a friend who corrected my schematic like this....
what do you think

http://www.mikrocontroller.net/attachment/45312/print.PNG

Ioannis
- 23rd January 2009, 09:21
Good. Now see the other issues as described previously.

Check this also:

http://www.picbasic.co.uk/forum/showthread.php?t=198

Ioannis