XBee Pro Series One, but I think it is appropriate for any radio.

The question again, is why would you want any rep rate other than as fast as possible?

You have a limited throughput but you want to sample at a high rate. The idea behind the TX before samples is to allow you to sample at a higher rate and get the data. It just will require you to have a few completed samples sent all at once. That is to reduce the overall throughput required.

Thank you Mr. MVUT.
And do you increase the range with less throughput?
Is that range increase significant?

Yes you do increase range with lower data rates. How much more, I do not recall.

Thank you Mr. MVUT,
It is good enough to know the purpose of slower data sending extends the range. Thanks for that answer. I can now handle that question when it comes up again.
Another question that I couldn’t answer in a talk I gave, was about the data rate. I spent quite a while on the limitation of the “physical layer” and the pulse rate of 50 bursts of data per second. The challenge from the audience was that I must be wrong because the stated data rate is 250,000 bps which is a very long way from 50 per second times whatever number of bits in the analog channel plus a mere 11 digital inputs and a 64 bit address. I was stumped. His claim that 50 bursts of RF I showed on the o-scope was because of something I was doing wrong. There must be a way to configure the XBee faster than one burst every 20 msec.

I am not sure I understand where you are getting your figure of one burst of data every 20ms either. The module is capable of streaming data at 80kbps with bursts of up to 115.2kbps.

The 20 ms number is what I observe when I put my microwave field strength meter near the sending XBee. Normally one pulse, every 20 msec with a 4% duty-cycle. Actually, it’s 18 msec. When the receiving XBee is turned off, there’s four pulses every 18 msec. But even without an RF detector, you can see in the output of the analog data, that the update rate is every 18 msec or slower. You can make it slower by increasing the delay before Tx. I’ve not been able to make it faster.