Page 1 of 2

Results of Testing on 2 Metres - Not good.

Posted: Thu Sep 17, 2015 6:14 pm
by M0LDZ
There is no easy way to say this as I know it is a disappointment, but it has been concluded that the recent Beta firmware, which included a widening of DTX1 lower frequency range to cover 2m operation, is to be withdrawn. We have hit an insoluble issue with the hardware, and it is that the modulator IC is giving an unacceptably poor MER at those frequencies, as low as 7dB, due to IQ imbalance, and we have no way to correct it externally. This is of course a great pity as the spectrum itself looks terrific, but to be fair it was a brave experiment - but it has failed. Accordingly the next formal supported release will exclude operation in that band. We will however continue suport the TS Dock add-on board, which is RF frequency agnostic.

Meanwhile we are looking to see whether we can find a pin-for-pin drop-in alternative IC for the modulator which would perform OK at the low end, but it is a pretty remote possibility at present.

73

Laurence

Re: Results of Testing on 2 Metres - Not good.

Posted: Sun Sep 20, 2015 12:57 pm
by g8gtz
Several people have suggested the way to go would be do a mix down to 146 from somewhere in L band where the constellation is good. So a small PCB with an oscillator running between 1 and 2 Ghz and a simple mixer with filters and maybe an amp is what we need.

But hang on if we turn it around, that’s the rx up converter we also need for 146 MHz - so anyone got any ideas on a simple oscillator?

Re: Results of Testing on 2 Metres - Not good.

Posted: Sun Sep 20, 2015 1:58 pm
by G4GUO
Define simple?

There are plenty of synthesiser chips with integrated VCOs available
but they are all surface mount.

- Charles

Re: Results of Testing on 2 Metres - Not good.

Posted: Sun Sep 20, 2015 3:34 pm
by g8gtz
Relatively low cost, stable, reproducible and easy to construct - and I would say the wider pin spacing SMD devices can be included in that.

73

Noel

Re: Results of Testing on 2 Metres - Not good.

Posted: Sun Sep 20, 2015 4:43 pm
by G4EWJ
You could possibly run an Si570 synth at 180MHz and pick off the harmonic at 900MHz with a Toko 1586F-535R helical to use as an LO and use a Toko 1989-535R at 1046MHz for an RX converter

Are Tokos still manufactured?

Brian

Re: Results of Testing on 2 Metres - Not good.

Posted: Sun Sep 20, 2015 5:16 pm
by g0mjw
Or use one of those ABRACON 1 GHz LO modules from Mouser? Not wonderful on phase noise, only moderate stability, but maybe good enough. They are SMT (which is a good thing) but fairly big 5x7mm and no worse than an si570 to fit.

Mike

Re: Results of Testing on 2 Metres - Not good.

Posted: Mon Sep 21, 2015 6:06 pm
by M0LDZ
Well first of all may I say how good it is that people are so inventive on ways to crack this nut!

Second, there is no need to set the DTX1 output at an amateur frequencey so long as the selected mixed product is at the desired frequency. You could just as easily set it to 647MHz and beat it with 500MHz. In fact, there is a possible advantage to doing something like that, because the absolute frequency error at the DTX1 output (as a result of reference crystal never being exactly accurate) is less as the synthesized frequency falls. When you consider the final 2m externally mixed output is so low in this case, that could be significant. (The same applies to the external mixer osc freq.)

Worth a thought?

73

Laurence M0LDZ

Re: Results of Testing on 2 Metres - Not good.

Posted: Tue Sep 22, 2015 9:57 am
by G4EWJ
For an rx converter, it would be advantageous to keep the output above 950MHz so that all MiniTiouners can receive it.

For transmit, anything goes. Perhaps 437MHz tx with an LO of 584MHz to encompass fixed frequency 70cm DigiThins.

Brian

Re: Results of Testing on 2 Metres - Not good.

Posted: Sun Sep 27, 2015 11:24 am
by M0LDZ
Thanks Brian

A very good suggestion, I think. That way such a board would not be restricted to DTX1 users, and so the volumes would be greater with attendant cost reduction. As you know PCB's are dreadfuly pricey in small volume to cover the tooling! A little bit of RF gain on the transmit side would be beneficial too.

Is anyone up for doing such a design? Should we move this to a more upbeat general heading (!), in the forum under some new title - what does everyone think?

Cheers

Laurence M0LDZ

Re: Results of Testing on 2 Metres - Not good.

Posted: Sun Sep 27, 2015 12:27 pm
by G4EWJ
Yes, a new topic is a good idea.

Brian