[DSP-10] How about 432?

Sat Nov 24 17:15:22 EST 2012

Hi Courtney and All,

I was just inside a DSP-10, and it got me thinking about your 432 
question.  I  think your outline is about on.  It certainly could be 
done.  But, a direct substitution might leave you annoyed with the 
performance.

It would make sense to run the VCO at 410 or 450 MHz, rather than use a 
tripler, just to minimize stray responses.  The current LO noise that 
Joe referred to I think is limited by the basic VCO performance.  Even 
in the 2-meter version phase noise could be improved by increasing the 
VCO inductor Q.  This would suggest some development work on a VCO for 
432, probably a separate little box.

The RF filtering requirements are getting more severe with a 20 MHz I-F 
and 3 times the RF  frequency.  I'm not sure what adequate image 
rejection would be, but if you use high side LO, the image is in the 
mobile UHF band, I think.  I'm not sure what is at the image for low 
side LO.  In any event, it would be worth some study time for filter 
design to see what the trade-offs would be.  The current 2-meter RF 
filtering uses a "free" zero to improve the rejection on the low side.  
That might be of value in the 432 filter design.

It might be best to construct all of this in a larger open box. Possibly 
a box with two sides to increase the flat, non-covered space.  This 
would allow for the fact that this would be a development project, 
needing good mechanical access.  That would also leave room for several 
shielded connectorized boxes, again making testing easier.

But, then as Mike says, if you just want to get on 432, the transverter 
approach is relatively easy.  Along those lines, Mike has done a lot of 
work on parts for the transverter.  Such as
http://home.comcast.net/~kd7ts/html/SI4133/SI4112.html   I have pieces 
of a KD7TS 432 transverter and plans to get that on the air.

Another easier conversion for the DSP-10 is down in frequency.  I have, 
on occasion, thought about a 75-m DSP-10 that would be fun (and pretty 
simple).  The trouble there is that most of the software does not add 
anything for ragchewing!  50 MHz might have some value.  But again, for 
any of these bands it is reasonable to just use transverters to 144 
MHz.  BTW, if anyone wants to do 50 to 144, use a high-side LO.

Let us know what you come up with, and how  we can support you.

73,  Bob  W7PUA

On 11/16/2012 04:43 PM, Courtney Duncan wrote:
> Some time ago, I posted here asking about the feasibility of a DSP-10 to 432 MHz converter.  Bob suggested a high side (576) LO and others on the list had plans for boards, etc. in various stages of pre-starting.  The idea was encouraging because I already had a W1GHZ designed 1152 LO and 1152/2 is 576 and there might be something there I could hack up to get started.
>
> Of course you can't just buy a 144 to 432 converter anymore because any rig you care to buy already has both those bands in it, stock.
>
> So the other day I was thinking about this again, noting that some of the 10 GHz guys use 144 and some use 432 for IF.  I have here another DSP-10 kit that I'm about ("about") ready to build that would be the IF for that 432 transverter system and the result would be a dedicated 432 system in the shack like the existing 144 MHz DSP-10 is now.
>
> But then I thought....
>
> "Why not just build the DSP-10 up as a 432 system in the first place?"  It could be called the DSP-12 or DSP-70 or something like that.
>
> Indulge me here.  I take all risks.  I'm just asking for a little consultation.
>
>
> Hardware:
>
> On the RF schematic, all the MSA06xx parts data sheets have typical performance at 0.5 or 1.0 GHz, so they're OK.
>
> The 2N5109 final (+17 dBm) looks like it's only good to ~200 MHz, but I bet it would be easy to substitute.
>
> All the RFCs and bypasses (L3, C7/8, L18, etc.) would be OK because we're going to a higher frequency.
>
> All tuned circuits like C2/L1/C3/L2/C4 , C15/L8, etc. would have to be changed.  Getting substitute cans would be easy.  Getting the design right would be a little tricky, requiring a little help if I were going to do it myself.  That's the very sort of thing I would want to be learning about in a project like this so I would be willing to study text chapters, tool up, etc. to do those filter designs and could probably succeed given a hint or two or proofread from this list.
>
> The TUF-1 mixer goes to 600 MHz on all ports so it's OK.
>
> Turning the page....
>
> I think you'd leave everything at 19.665 MHz IF and down to the DSP and audio the way it is.
>
> Turn the page again....
>
> That means the 19.68 MHz LO is unchanged.
>
> The synthesized LO which is 124.3 - 128.3 for 144-148 MHz would need to be either 412.3-416.3 or 451.7-455.7 or thereabouts.
>
> Yes, I see that the LMX1501A synthesizer is obsolete, but I have one in the kit, and the datasheet that google finds says "RF operation up to 1.1 GHz" and has charts that go up to 1.5 GHz.  And the rest of the stuff on that board is MSA06xx again, so I think we're good there.
>
> In fact, it seems like you might be able to extend the range and get 432-438 MHz or more rather than just 4 MHz tuning range.  But the steps still need to be 5 KHz or less and that could get tricky.
>
> Q101, J310 data sheets talk about going up to 450 MHz, close to the edge but would nearly certainly work.
>
> The VCO would need some work:  C141, L102 would need a different can again.  R152, D104, D114, D113, D115, hmmmm.  Looks like the FM2201 variable cap diodes go to about 250 MHz so those would need to be upgraded.
>
> Either that or make U106 into a tripler or something.  That would probably mean adding parts, maybe another stage.
>
> No reason you wouldn't still use 10 MHz reference.
>
> Turn the page again - that's it.
>
>
> Software:
>
> The software already understands about "upside down" for the possible 451 MHz LO option.
>
> The software would have to understand about the new programming of the PLL LMX1501A.  That would be a mod, probably a 3.80a release or something.  (I'd be surprised if you could fool the configuration file sufficiently to do this without something bizarre happening in the frequency display.)
>
> Aside from that all the software is the same, so it really is just an 'a' mod.
>
> Well, OK, that part would be in the EZ-Kit SW.  There would also be things to mod in the PC software, like the understandings of transverters and things, or just ignore all that and don't use the features that would get broken.
>
>
> Summarizing what would have to be done:
>
> Rework the 144 MHz RF filter chain for 432 MHz.  Educational with a little guidance.  Probably very doable.
> Find a substitute for the 2N5109 final and check the circuit.  Pretty easy.
> Change the LO to something > 400 MHz.  Either:
>    - upgrade the VCO or
>    - triple
> Both of those are probably a little tricky but don't look intractable.  Educational, with a little guidance.  Sounds like fun.
> Either way requires some SW mods.  Pretty easy once set up, or if I can get somebody to just do it for me.
>
> (Not that I think getting set up to do SW mods would be very easy....)
>
> And then for the rest of the "rig" I'd take responsibility for some kind of 15 dBm to 35 dBm "slippers" (maybe as a personal education design project or just buy some widget somewhere -- hey maybe I'd enter the QST Homebrew contest!) to drive an old old 2-in 30-out 70 cm amplifier kit I have sitting around here somewhere.
>
> What could be simpler, right?
>
> Oh, and an antenna.  But I digress.
>
> I either have here or have access to sufficient test equipment to get something like this working.  I'm just worried that stuff like getting the PLL to work right could get dicey.  But, that's the education I'm looking for so I'm willing to go for it.
>
> This all actually sounds fun.  It's not just building another DSP-10, it's building a racing model!
>
> So what do you guys (particularly Bob and Mike and the others who really know what's going on in there) think about that?  Is there some reason you would recommend not going down such a path?  If I did go for it, would you be up for an hour here and an hour there to get me unstuck?
>
> How about 222 MHz?  (Just kidding!)
>
> Thanks and 73,
> Courtney, n5bf/6
>
>
>
> "It's science. They don't know anything; they just make a lot of educated guesses."  -- Viannah Duncan
>
> Courtney Duncan, n5bf/6
> cbduncan at earthlink dot net
> 1296.100, 1.810.
> n5bf at amsat dot org
>
>
>
>
>
>
>
>
>
> ______________________________________________________________
> DSP-10 mailing list
> Home: http://mailman.qth.net/mailman/listinfo/dsp-10
> Help: http://mailman.qth.net/mmfaq.htm
> Post: mailto:DSP-10 at mailman.qth.net
>
> This list hosted by: http://www.qsl.net
> Please help support this email list: http://www.qsl.net/donate.html
>
>