PERSEUS SDR - PERSONAL OBSERVATIONS

UPDATED AND REWRITTEN - FEBRUARY 2008

WEB PAGES UNDER REVIEW. NICO HAS JUST ANNOUNCED THAT THE NEXT VERSION OF THE OPERATING SOFTWARE, DUE ANY DAY NOW, WILL TAKE THE RECEIVER UP TO 800kHz RECORDING CAPABILITIES!!!! THESE PAGES WILL REVIEW THE NEW SOFTWARE ASAP

INTRODUCTION: I was lucky to "grab" my receiver when I did, in November 2007. Such is the interest generated by this revolutionary piece of kit, that there seems to be a substantial waiting list. I am not surprised.

Much of my original article was hurriedly written to "get it on line" as fast as possible. Some of the text was written after only a few hours use, when I never really had a "full handle" on all the controls. In February 2008, I now feel a little more qualified to pass comment and even judgement on the PERSEUS, now that I have been actively using the receiver for about two and a half months. It does somehow seem longer.

Originally my personal observations were included along with basic operation details. I have now separated much of the text which is now mostly included in the "operation" page. My intention when I wrote the initial text, was simply to answer a lot of basic operating questions I had in my own mind before I managed to source a PERSEUS, with the hope that it may also be helpful to other dxers. It seems these pages have generated a lot of interest in the last few months. Much more than I ever imagined.

I have deliberately described my observations in "layman's terms" rather than "lab test" terms. I personally prefer to be able to test a receiver on my own known antenna set up, observing MW/ SW bands as they are at my own qth. This could be called the "ear test" rather than the "lab test".

DISCOVERY OF PERSEUS: For a long time I believed in my head that the PC and all it's associated RF noise is certainly no place for a communications receiver to be situated. I viewed the initial SDR radios with a degree of suspicion. Why would I want to use a PC to act as a radio for no other reason than because I can!! I have good communication receivers, and could see no reason to even think about an SDR "set".

A while back, I was reading an article on the then current SDR receivers on the market. My attention was captured by comments about recording great chunks of frequency spectrum, and being able to playback at a more convenient time, and still use all the receiver's functions as though it were live off air. The article in question was written by MW dxer Jack Webster, and seemed to suggest that we were maybe 10 years away from being able to record the full MW spectrum in one pass. The freq span limit was 190kHz at that time, but the impression the article portrayed was that there was much room for improvement in operation. This was only in November 2006.

In October 2007, I was reading the WWDXC on line weekly report, and saw a mention of Perseus. I had never heard of this. I visited a few web sites, and found my way onto the Perseus yahoo group, and could not believe what I was reading. A receiver that could record 400kHz of spectrum and play back at a convenient time. I think my heart missed a beat. A third of the MW band!! I had to e mail fellow dxer Guy Atkins in the States, to verify I was not mis reading the articles, and this receiver really could record and playback 400kHz of bandwidth. As it turned out, I had not misread.

That immediate attraction to the Perseus was exclusively due to the 400kHz recording capability. For a MW dxer, this is equivalent to using a bank of 40 receivers for the North American MW channels every 10kHz, and a further 40 odd for the 9kHz Euro channels in between. (Some frequencies are of course the same, 540, 630, 720, 810, 900, 990, 1080, 1170, 1260, 1350, 1440, 1530).

How the shack might look with a bank of 40 NRD515s !!!!

> Now in February 2007, we eagerly await the release of PERSEUS 1.0, which opens the receiver up to 800kHz recording bandwidth!! I shall omit the story about the bank of 80 receivers!!
The ultimate bandwidth for a MW dxer of course would be 1200kHz.Another 400kHz and we are there!! This would take in 530 - 1700 inclusive, with just a little room to spare.

PERSEUS ARRIVES:With an SDR receiver, gone is the look and feel of the controls of a conventional communications receiver. It is no more than a USB box, and hence the phrase "Black Box Radio". A very clever little box as it happens. My “box” arrived late Wednesday afternoon 21st November 07. I initially tried only the software I had downloaded, BETA 0.2, but had the message “can’t find a Perseus” or similar. I checked on the CD which came with the set, and of course I needed the drivers as well. After the install, the radio burst into life, without any messing around with settings on the sound card or anywhere else. It was operational straight away, on it's 1024 screen. I had already made up a BNC cable to go to my antenna panel, and had a long USB cable to keep it away from the PC QRN. The USB extension didn't work for some reason, which didn't matter as I don’t seem to be bothered by QRN from the PC.

My first task was to excitedly record a 400kHz section of spectrum, so I recorded part of the spectrum which included Radio Beijing at 1700utc on 1521. My first DX from an SDR receiver.

I had the initial twiddle around to get used to the controls, and find out what everything does, or doesn’t do. Once I had sussed that the mouse wheel on the second smaller pane controls the bandwidth, I thought, wah.. This is a powerful piece of kit. It is not just a case of varying the bandwidth, but you can widen one sideband and keep the other narrow, depending on which sideband the QRM is worst. And at that stage I never realised there is also a right click function to drag the shaded area around.

The first night I had the receiver, the DX on MW was very poor, as it was for the following week. Nevertheless on the same antenna I did seem to have slightly less splatter on VOCM 590 than on the NRD545. Surely a good sign!!

Since the initial trials I have been able to test the receiver to it's limits. I have uploaded a substantial audio page now for the PERSEUS rx. From a MWDX slant, AM is a good mode for jumping fast up and down the band, due to the sync AM taking just that little bit longer to lock. Sync AM is also good though.

One recommendation though. Don’t spoil the receiver for the want of a three pound mouse. If you don’t have one, then go and buy a mouse with a wheel. I would judge it to be a requirement to enjoy the benefits of the Perseus SDR. "Even" laptop users would benefit from an external mouse. Maybe that should read "especially" laptop users. I can’t imagine trying to work the software using the mouse pads found on laptops.

PC SPEC: For the record, my own PC spec. 2.8g Athlon, 512m RAM, Dual monitor graphics card.

OBSERVATIONS AND TESTS

NEGATIVE

USB PEN CRASHING: Perseus doesn’t like ripping a USB pen/ memory stick in and out while the radio is on. The receiver simply locks up, and the software has to be restarted. Strange thing is, the Perseus is plugged into the front USB, and the pen, into the back of the PC. I have found a very simple solution to this so called problem. Don't unplug the "pen" when the receiver is operating!!

RECORDED FILE NAMES WEIRD: Not really negative, but an unimportant oddity. If one tries to record with the PERSEUS receiver, the software does not allow a file name of for example 080209_0758. The name would automatically drop the zero to become 080209_758. If at a later time the file is renamed in windows explorer, PERSEUS plays back and recognises the new file name no problems.

CHANGING FREQ WHILE RECORDING: Not really negative, but another unimportant oddity. Normally tuning is not possible beyond the outer frequency limits. Twice I have found that the CF step arrows below the frequency bar are lit while recording, and tuning is possible. The recording continued, but in the middle of playback, the frequency changed OK, but the frequency bar remained at the old values. So in other words, it was tuning in the dark during playback. This happened twice, but have not been able to recreate the conditions needed to light the arrows again. Probably just a glitch in the software, and would have needed a restart.

PC CRASHES WHEN A RECORDED FILE IS MANUALLY PLAYED BACK AT THE END OF THE FILE:

ANTENNA CONNECTION: A VERY MINOR POINT: BNC style connectors are not as common compared to say F connectors, which satellite receivers use. I was lucky enough to find a lead with a decent BNC kicking around, but I don’t imagine many people will have a BNC in their junk boxes. Many may appreciate a length of cable with a BNC already attached being shipped with the Perseus, or a BNC to F socket adaptor. Reminiscent of the kids Christmas day toys having no batteries in the box, and of course the shops are closed!!

POSITIVE

POWER SUPPLY: The power supply which is supplied with the receiver shows no sign of strain. Heat can only barely be detected by touch, even after running for 24 hours.

FREQ BAR CRASHING REPAIRED: Like other dxers I have experienced a PERSEUS software crash when dragging the frequency scale from side to side quickly. This was a bug in beta 0.2, but fixed in beta 0.3.

AM AUDIO QUALITY IMPROVED: Beta 0.31 sound is certainly much improved over beta 0.2, but there is a real "Orban Optimod" sound to the AM audio from beta 0.1. (check my test on the PERSEUS Audio page).In BETA 0.2 the AM audio quality was a little watery, although perfectly acceptable. I know it isn’t a hi fi tuner, but I felt the set was capable of a fuller audio, because on SSB you could "drag the signal area" closer to the carrier and produce a fuller richer sounding signal, even if it was SSB. I don't know what the audio would sound like on the tinny little cheap PC speakers which are shipped by PC suppliers. Haven't tried yet. I always use my PC plugged through an amplifier, and a pair of 15 inch disco / band style speakers. I also listen to my other receivers through the same system.

RECORDING: I have no problems running a software recording programme to record the audio of the single frequency I am tuned to, even when the 400kHz recording facility is active. I can also still tune withing the 400kHz. I have line outs from my traditional receivers as well as the PC connected through an audio mixer. The PC recording software I use has no problem recording from the line in, while the line out is playing PERSEUS audio. Therefore I can give the audio a bit extra via an equaliser. This system is used to record single stations being listened to live, or IDs heard via the 400kHz recordings. I can even use the recording software to record from another receiver independently, and still record the 400kHz PERSEUS file.

PERSEUS RECORDING THROUGH WINRAD: Playing back a PERSEUS recording on WINRAD, allows the user to extend the frequency to maximum 500kHz sample rate, and not 400kHz. This is seen as an extra 50kHz either end of the PERSEUS 400kHz file. I don't know whether this should be negative or positive. It is positive in the fact that WINRAD allows this small frequency extension. Negative from the respect that PERSEUS software doesn't allow the bandwidth to be opened just a little wider.

BBC RADIO SCOTLAND 1620: I know the following test is crude, but here goes.... I have a strong signal from 100kw BBC Radio Scotland - Westerglen 810kHz (50 miles away approx). Some receivers show up a signal on 1620 if plugged into an antenna with no ATU. Oddly I do not have the same effect on 2070, twice the frequency of my local MW station, ILR AYR which is only about 12 miles from me, and rather strong as well.

A test of my receivers tuned to 1620 daytime, all using the same 1/4 wave inverted L roughly cut for the hf end of MW.

Receivers this effect shows up on:

LOWE HF225E: S8 on the tiny Lowe meter.
AOR7030: S8 on the inferior AOR signal "meter"
NRD545: S5

Receivers this effect does not show up:

NRD515: Nothing, not a trace. (bomb proof front end),
RACAL RA17: Manually able to tune image in or out with the RA17 (this set is from the 1960s!!)
PERSEUS: There is no sign of this problem what so ever on the PERSEUS. (Despite an S5 on the NRD 545).

A screenshot around 2200UTC 27th November selecting 100kHz bandwidth for the waterfall screen, to show the carriers on 590, 600, 610 etc from across the water.

COMMENTS ON RECEIVER FUNCTION

FRONT END: Preset, Preamp and Dither. Neither seem to have any noticeable effect.

Playback: Initially I found the playback progress bar to be a little sticky, and not like for example, WINAMPs bar. The Perseus seemed to go into “jam” mode, and stick. The playback worked fine if the recorded file was played from start to finish. After another attempt, I discovered the error of my ways. I was dragging the progress bar using my "winamp" brain.

To advance the playback by a few minutes: Left click, hold and drag to make a new line which the player will immediately jump to on mouse release. The new line will be a much paler shade of yellow, when being drawn. If the user merely tries to click to advance the player, the output sound may seem to "stick". The player is probably in fact playing a very small selected area over and over again.

Trials with a couple of 1.7g files recorded on a DVD, play back from the disk flawlessly. There is no jerky behaviour when reading from the DVD

The light yellow line to the right before the mouse button is let go. As soon as the button is released, the deep yellow line jumps to the new place on the progress bar.

MODE: I haven’t quite sussed what USER setting is for in the MODE BAR..

S METER: Not convinced I like the little meter much, compared to the proper analogue meter on my on my NRD515. Has RMS and Peak settings. Haven't really found out what the intermittent display ADC CLIP really means.

MEM box: This is related to the database files which have been added since BETA 0.31. Personally I am not sure how I feel about say CHINA RADIO INTERNATIONAL automatically coming up on the screen when you select a certain freq, knowing fine and well it isn't China Radio you are listening to. May have limited use for MW dxers, but probably more helpful on the SW bands.

NB LEV control seems only to bring in unwanted mush from somewhere. Not sure of the usefulness yet. Kept off.

WHAT COULD I SUGGEST FOR FUTURE UPGRADES:

SPLATTER KILLER: If the German DXer's splatter killer feature was somehow able to be incorporated in the Perseus, this would be the second selling point. MW dxers are a breed of their own, and listen in such tight splattery situations. This looks and sounds like a wonderful idea, and the idea of any splatter reduction must surely be worth looking into.

FREQUENCY BAR AT TOP OF WATERFALL: How about the frequency being shown at the top as well as the bottom of the main waterfall / spec window. There is room for an identical frequency bar at the top of the window. The same as that which is displayed at the bottom of the screen. When you drag the frequency bar, it takes a bit of time to see where on the spectrum the carriers are in the “WATERFALL” mode. I think the direction of the waterfall is more natural, moving downwards, and a repeat of the frequency band at the top of the screen would be a welcome addition.

RECORDED FILE SIZE: Recorded automated file size is around 1.757g. I wonder why the odd choice?? Time = exactly 10m perhaps?? Wonder if there would be any harm jacking file sizes up to about 2.1g. Reason?? 2 files would fit nicely on a standard 4.38g DVD, for archiving. As it stands, two files can fit on a DVD, but with much spare/ wasted space. Maybe even user defined file sizes could be an option??

MEMORIES: Memory channels would be handy to be able to store mode, PBT when activated, freq etc all per channel, and be able to turn the mouse wheel to zip up and down the pre programme channels. This is what I currently do with the NRD 545. Channels 53 – 171 are pretuned to the North American 10kHz frequencies for MWDX. The drawback with the NRD is that the PBT is not storable per memory channel.

VERY MINOR POINT: SPAN: Maybe user defined steps in Span function would be useful for some users, or a continually variable bar option??

VERY MINOR POINT: Would it be difficult to allow user defined colours for the text. My own personal preference would be plain old WHITE on the dark background, or better still black text on a white background, like I use for my mwdx web site. (Maybe I am too plain in my outlook!!)

VERY MINOR POINT: Would be nice if the unit “remembers” / maintains the AVG main and sec settings at switch on. Always switched off on restart.

OVERALL: I can imagine me at a time in the future, recording a full hard drive or two and keeping them in storage of a sample of wideband MW spectrum. This would serve as a superb piece of broadcasting history in the future.

This is one reason to allow the large files to play without the radio connected, incase the radios are all broken in years to come!! If I had this technology in 1988 when all the Irish pirates were closing on 30th and 31st December, I could have recorded a large chunk of the spectrum at the one time…

My final words on the set for the moment… AMAZING. To Nico for his superb work, well done. PERSEUS does everything I had hoped it would do, and much more.

Ken

Scotland