Yes, definitely. This really came home to me when I directly multiplied the output of an AD9852 DDS at 72MHz up to 10GHz, then listened to it on an SSB receiver. It sounded absolutely disgusting - like a ball of noise several tens or even 100s of kHz wide. I could only just about discern the carrier at the centre frequency. All those millions of close in spurii at -80dB, when passed though a times144 multiplier, intermodulated and in-bred massively. Simple theory says they should have increased by 20.LOG(144) = 43dB, but seemed worse than that even. Hence the crystall oscillator PLL approach described last time, quickly put together when I'd recovered from the horror of the direct multiplication result :-)
The deliberately added noise sounds a fascinating idea - do you know if Analog Devices (or anyone else) do, or intend doing, something like this in any of their products? Andy G4JNT www.scrbg.org/g4jnt/ -----Original Message----- From: KY1K <[EMAIL PROTECTED]> To: [email protected] <[email protected]> Date: 2006/01/26 19:41 Subject: Re: [soft_radio] Re: DDS + PLL to clean up spurs? >It seems to me that a real world receiver will have it's performance >significantly degraded when running with a DDS LO..... > >We aren't talking a few spurs here that we can steer clear of (by >design), are we? > >Let's say we have a conventional 7 Mhz double conversion receiver. >The first IF is 10.7 Mhz, the second IF is 455 KHz. The front end is >tuned, but not overly sharp, but lets say it admits out of band >signals from 5 to 15 Mhz. Our DDS operates at 17.7 Mhz (10.7 + 7 >Mhz). For the sake of argument, let's assume our mixer is a 'good' mixer. > >The DDS operates at 17.7 MHz, but it has hundreds (or thousands) of >sprurs, in the 70 to 90 db down range (relative to the desired 17.7 >MHz desired output). Some of these spurs happen every few hundred >Hz, some happen every few KHz (on average). While the spurs are 70 >to 90 db down, there are many of them. Each spur allows the receiver >to receive a small amount of interference, which mixes with many 30 >to 60 over S9 signals within the range of the front end tuned circuit >(5 to 15 MHz). > >While spurs down 70 to 90 db sounds rather harmless, the end result >is a disaster because there are so many of them and because out of >band signals within the range of the front end tuned circuit are very >strong and numerous as well. You end up with hash, all across the >desired listening frequency that really isn't there.....and your >ability to copy weaker signals is gone! > >Clearly, any DDS used as a local oscillator should be de-spurr'd (is >that a word)? > >Is my assessment of the overall impact on the receivers ability to >copy weak signals based on the example above accurate or am I missing >something? If the receiver's ability to copy weak signals is >compromised by using DDS technology, shouldn't we (as a group) be >incorporating de-spuring countermeasures every time we hook up a DDS >for use as a LO in a receiver? > >The point I'm trying to make here is that we aren't just concerned >with the spurs that happen near the desired listening frequency that >cause us grief!!!! It's ALL the spurs mixing with ALL the signals >within the bandpass of the front end filter that we need to be concerned with. > >Correct, or incorrect? > >Regards, > >Art > > > > > >Yahoo! Groups Links > > > > > > > Yahoo! Groups Links <*> To visit your group on the web, go to: http://groups.yahoo.com/group/soft_radio/ <*> To unsubscribe from this group, send an email to: [EMAIL PROTECTED] <*> Your use of Yahoo! Groups is subject to: http://docs.yahoo.com/info/terms/
