RFsyscalc (Excel) spreadsheets
This spreadsheet was done to facilitate RF cascade analysis of communications systems. There were a few reasons I migrated from using a "commercial" product to using this spreadsheet:
- The commercial product cost a fair amount of money, so if I shared the work with other folks, they all needed the program.
- The commercial product wasn't as flexible in the ways that were most important to me, such as showing the effects/outputs of each stage in the cascade. A spreadsheet is ideally suited as a general purpose analysis engine -- with a core to start from provided here, you can enhance the capabilities to meet your needs, or just get ideas about how you can create your own spreadsheet/program to do something.
- I wanted to have something that I could use on my Palm Pilot, so if I was stuck on an airplane with a gun to my head and had a life-and-death radio design to complete, I could do it ;)
If you want something more, I suggest looking at the tools offered at www.rfcafe.com as well as their links to other vendors' sites. Other options range from free to very modest registration fees up to several hundred dollars in a few cases.
Some plots
Provided to show the features of version 4.xxxx
Here's a graph of how an example Receiver's SNR varies with input signal range . The flat spot in the curve shows the effect of "delayed" AGC. This type of AGC is typically setup so that it is inactive until the signal level is high enough that the additional attenuation introduced will not degrade the SNR to below a perfectly usable value. This technique allows both good small signal performance and enhanced strong signal handling capability (especially off-channel) when implemented properly. The graph shown is from an example provided (that exaggerated the effect to show what the delayed AGC is all about). Note that with the spreadsheet you can implement the "delayed AGC" as a fixed step attenuator if you want ... just make the range that the device is active over extremely narrow -- this way it will go from one extreme of it's gain to another when the limit is reached!
This graph is of the level of two interfering tones (vs. offset from the channel center) required to cause a 3rd order IM product to fall in-band at the level of the thermal noise... effectively dropping the RX sensitivity by about 3 dB....
The features of this tool
- Cascaded gain, noise figure, IP3 (cumulative cost and power consumption are added for engineering convenience in version 3.7)
- The following are computed after each stage with a given signal
level at the input
- Pout
- OIM3 (dBc)
- IIP3/OIP3 (dBm)
- "delta" NF
- "delta" IP3
- RF SNR (with a given input bandwidth)
- simplistic frequency response model for
- RX IIP3 vs. frequency offset
- TX broadband noise (not yet really complete, just a starting point)
- new AGC graphs of parameters over input power range (RX) or output power range (TX)
- Psat sanity check
- "tx" and "rx" modes change the behavior of
the AGC.
- Along with the main signal passing through the chain and giving the above parameters, I allowed for a secondary "pair" of signals to be analyzed. This was done for IM rejection analysis... when the system is setup for a weak desired signal, you can look at the levels of the IM3 products from interfering signals at a different level. Those who have worked with tough RX IM specs can probably see the value in this.
- A crude AGC model is included, such that the output of the chain can be held constant over a range of input signal levels. As of version 4.000a/b, this now includes a delayed AGC (agc that's only active over a portion of the input signal range (RX) or output signal range (TX).
- I've used the sheet extensively for both Receivers and Transmitters.
Version information
- Version 3.8 should have fixed a noise figure calculation bug in previous versions that comes up when the system gain drops below unity under certain conditions. Please use the worksheet with caution (read "readme.txt" included in the zip file). Sorry for any inconvenience... I will work to test the spreadsheet more thoroughly when/if I get a chance. LL 12/20/2000
- Version 3.9 has only minor enhancements. More might follow this summer if I get inspired -- Lance 7/16/2001
Version 4.0f has many changes so use it at your own risk. It seems to work though and I'll be using it myself for many things in the near future so I hope to flush out any problems.
Note: I discovered a glitch in the 4.0f IIP3 vs. freq. offset curves. It is partially fixed in 4.0g, but will be updated again soon once I have the time to look at it more. Lance 11/6/2001
- Version 4.0h has mostly documentation and cosmetic improvements (8/16/2004 LL)
Download...
Most recent version of "RFsyscalc" spreadsheet
(version 4.0h uploaded 8/16/2004)
Note: all releases of RFsyscalc include copies of the old releases as well since the files are not very large.