RF Spectrum analyzer for computer - recommendations?

[Steve - K4HX]

I think I'm correct in saying that Tom's SA requirement is to measure IMD. Instantaneous wide band coverage is not required for this close in measurement.

[KF1Z]

HPSDR is going away from USB2 to gain more horsepower to monitor broader spectrum. New board will be sold next month. I read the Hermes transceiver will have the same new interface.
QSR1 I think Phil is overloaded supporting what he has. This could be the reason he has not produced more hardware.  Too bad he he went off on his own rather than stick with HPSDR. I think he wanted to make some money for all his hard work.

He told me 2 years ago that the RFFE1 front end board was ready to go....
He was waiting until he thought he could sell enough of them to gain back the cost....
I think he waited too long.

Though I  wouldn't want to foot the bill for 100 units and hope for my money back either ( right around $27k )

I think the same is true for the transmitter.

I'm going to look at HPSDR again... I wouldn't be supprised if I get going with it in a couple months...

[KF1Z]

I think I'm correct in saying that Tom's SA requirement is to measure IMD. Instantaneous wide band coverage is not required for this close in measurement.

I think Tom can do what he wants with the softrock..


There are also USB PC dual channel oscilloscopes that are pretty cool too that do SA as well for under $400.

[KF1Z]

Hey Tom....

Got any telescope stuff you want to trade for a QS1R  ?

[WA1GFZ]

Bruce,
Wait for the Hermes transceiver. The set up with the Atlas mother board adds up to more than a QSR1.

[K1JJ]

Yo Bruce,

I have a set of 10" Newtonian binoculars, 2" eyepieces, aluminum tubes, rollaway alum mount, with pristine, matched mirrors. They are worth at least $2.5K or more.  I wud probably trade them for a Flex... :-)  The big 22" scope is in an observatory and not movable without mucho effort.

How much $ for the QS1R?

In the meantime, I will be trying a converter off the SP-600 to get me going for a while.

I'll take a look at the Herpes transceiver, Frank. I think eventually I'll at least do a dual diversity RX next to the FT-1000D TX.


** BTW, Brandon/KF5IIA tells me he is kicking major ass down there with his new pair of phased dipoles. Seeing 25-30db f-b and big reports.   Best antenna anyone can put up short of a rotary Yagi for 75M.

T

[KF1Z]

Bruce,
Wait for the Hermes transceiver. The set up with the Atlas mother board adds up to more than a QSR1.

Well, when you include Penny it does...
But I think it was a few hundred LESS than QS1R without the transmitter...maybe not

Hermes looks like it's a LONG way off from what I read this morning...
The comment " ... looks like Hermes for 2011 is a good possibility.." ( or words very similar to that)
Lead me to believe there's a good chance it won't make it until 2012 ...

Maybe wrong there...

[Tom WA3KLR]

In regards to the Sherwood receiver dynamic range list:

I've always been amazed to see the humble stock Atlas 350XL mid-to-late 1970's transceiver has held up near the top of the list.

[WA1GFZ]

Atlas 350 is a very good rig and I don't think it has a synthesizer.
TR7 is another but it has a bit much phase noise.
my last homebrew had 96 dB dynamic range but the synthesizer was not that great. I would love to drop a HP8640B cavity in with a string of dividers as an LO.

[Steve - K4HX]

I also see the FT5000 now tops the list.

In regards to the Sherwood receiver dynamic range list:

I've always been amazed to see the humble stock Atlas 350XL mid-to-late 1970's transceiver has held up near the top of the list.

[WA1GFZ]

Steve they do it with multiple first IF filter bandwidths and very high dynamic range mixers. Yup kicks butt but takes a lot of hardware. It seems to have the best of both worlds. Tight filters in the first IF help a lot since the selectivity is closer to the antenna. I trust Sherwood's numbers.
I wonder how it would stack up against the Cubic R3150 A similar design but 20 years old.

[Steve - K4HX]

Along with the Atlas, the old Kenwood TS-830 is right up there. Seems like those non-general coverage down-conversion (usually to a 8 or 9 MHz IF) type receivers do better than the general coverage up-conversion rigs. I'd bet with a better front-end and a tight first filter in the 8/9 MHz IF, rigs like the 830, FT-102,etc would be most rigs 10-20 years newer. You could spend a lot of money modifying a cheap old TS-830 before you would get to the price tag of most of these newer radios. Inrad sells 250 Hz filters for the FT-102. Hmmm.

[Tom WA3KLR]

Yes the 350-XL, TS-830 and very similar TS-820 being ham-band-only rigs were able to rely on single-conversion, a big help.  The 350-XL used a 5595 kHz i.f. and the -820 was 8830 kHz.  The 350-XL used a double-balanced diode mixer.

For the l.o., the 350-XL used a permeability–tuned oscillator mixed with a crystal oscillator.  The TS-820/830 used a capacitor–tuned (I think) oscillator mixed within a phase-locked loop with a 500 kHz reference, the vco output for the l.o.

[WA1GFZ]

Steve, I think many of those old rigs had a 500 KHz wide first IF to fit the whole band through. I built a tracking filter tuned with VVC diodes for my old SB303 I geared a 10 turn pot off the main tuning to tune it. I was able to increase the dynamic range of that radio to 85 dB up from the 60s. The synthesizer degraded many a receiver design.
Up conversion eliminated the need for a tuned preselector.

[WA1GFZ]

TR7 did the same thing Tom.
I've always wondered how far you could hot rod an R390A

[Steve - K4HX]

The first IF filter in the FT-102 (a roofing filter before they were called that) is 15 kHz. They used this to accommodate the FM mode. Since it's at 8.2 MHz, I don't see why it couldn't be narrowed down to 3 kHz (if you want to  accommodate SSB, less for CW). Inrad makes a more narrow roofing filter (70 MHz) for the FT-1000. It improves the close-in (less than 20 kHz) up to 10 dB. Of course, once you get the spacing inside the new filter width (4-5 kHz), you're right back where you started. But for most SSB operation, you probably won't have two strong signals (in this case about 10 dB-over-9) within that 4-5 kHz range (except for contests). If you did have two stations that close, you're more likely to get interference directly from their signal (splatter or just legitimate signal bandwidth) that is stronger than the IMD.

[KF1Z]

The 350-XL... great rig...  NO TUBES !
 

[WA1GFZ]

I thought the FT102 used dual gate mixers? Yes with the option to switch in a tighter filter performance would go up. I don't think it has a synthesizer so you could get lucky. I yanked the dual gate mixers out of my SB303 and build discrete DBMs with 1N5711s.

[Tom WA3KLR]

Yep the TR-7/R-7 l.o. scheme is essentially the same as the TS-820, however the TR-7 is up-conversion/double-conversion.  

The HW-101 and the SB-series used a crystal 1st l.o. and the 500 kHz wide 1st i.f., like the R390 but did not have the tracking filters like the -390.  The HW-101 is not on the Sherwood list but if it was would probably be at the bottom, fairing worse than the SB-303 that is on the list.  The scheme was very good for increased frequency stability for SSB.

(Yikes we’ve hi-jacked the thread.  Happy New Year everyone.)

[Pete, WA2CWA]

Yes the 350-XL, TS-830 and very similar TS-820 being ham-band-only rigs were able to rely on single-conversion, a big help.  The 350-XL used a 5595 kHz i.f. and the -820 was 8830 kHz.  The 350-XL used a double-balanced diode mixer.

For the l.o., the 350-XL used a permeability–tuned oscillator mixed with a crystal oscillator.  The TS-820/830 used a permeability–tuned oscillator mixed within a phase-locked loop with a 500 kHz reference, the vco output for the l.o.

TS-830 is dual conversion. 820 and 820S were single conversion.

[K1JJ]

There's no such thing as hijacking a thread, in my book. Info is info and activity is a good thang, no matter who's thread it gets posted in. Post away, even about the type of electric toothbrush you like...

Yes, I remember the Atlas 350 well from the late 70's. Worked on a few. I didn't realize they were that good until seeing the listing.

The FT-102 has very similar circuitry to the FT-1000D. Much of the same actual parts too.

Steve, I tried a roofing filter for the FT-1000D back about 6 years ago. At the time they only had 2.4 kc 70 mhz roofing filters and it killed the 2.8khz filter bandwidth in the last IF.  Both Chuck and I took them out. Evidently  they have them wider now, but the poor FT-1000D dynamic range is close in, (2kc away) so maybe the 4-5kc wide is too wide to help that.  The wider dynamic range is already pretty good, at 90db, IIRC.

T

[Steve - K4HX]

I'm pretty sure they never had a 2.4 kHz filter for the FT-1000. To build such a filter at 70 MHz would be difficult and quite expensive.

[K1JJ]

Steve,

Whatever Chuck and I had in there was too narrow and brought the final ssb bandwidth back down to 2.4kc again. I sent it back for that reason and they agreed - and took it back. So dunno what to say.  Or wait a second - maybe the filter was an 8.2mhz  back to back second filter in series with the first? I'd have to check.
But it was called a "roofing filter."

Anyway, another idea:  I once met a Japanese ham at Dayton who told me the bands were so congested with loud signals in Tokyo, it was bedlam for RX IMD even with good close in dynamic range specs. The solution was to put in a pair of CRYSTALS as a bandpass filter at the receiver RF input. For example, on 3.790 they would use a 3.788.5 and 3.791.5 set of xtals to give a brick wall 3kc bandwidth before the RF even got to the mixer or RF stage, if used.   He said they had many pairs and covered the various favorite freqs with a switching scheme. Or use +- 1 kc for more narrow work.

I could see that working on the small 10kc 75M DX window and even on 40M if kept to a small band of freqs.


Does this sound viable?   Some kind of class A pig device could be used in the matching 50 ohm interface to keep IMD way down in the xtal circuit.  I'd be game to try somewthing like that ahead of the TR relay to the RX in the FT-1000D.  I wonder what dynamic range could be attained with a simple add-on like that?

See ya on 75M later.

T

[WA1GFZ]

My Hot Rod Racal does about 85 dB at 2 KHz. The 6830 synthesizer goes through -90 dBc at around 2 KHz. I had to replace both first IF filters with 8 KHz. BW units from a 6790 VLF radio. Then change out the second mixer module with one from a D.F. radio. They used a RAY 6 DBM mixer in that module with plus 13 dBM LO.
Then had to add a PAS3 attenuator to get AGC action in the first IF. I cascaded the 455 KHz filter module with modules using collins mechanical filters. They are much shorter than the crystal filters so could stack them for better skirts and ultimate rejection of around 100 dB. Dynamic range at 10 KHz is around 100 dB
The HPSDR and hot rod run neck and neck when conditions suck with high noise. The Hot Rod always beat the Softrock style IF set up under high noise conditions.

[WA1GFZ]

Wow Tom, that is hard core. You could do that if you had a box of crystals.
Filter loss would be 2 to 3 dB for a pair of crystals. There are tons of filter design articles out there.