[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [linrad] Time Machine....?

Hi, Jim, and all!

I agree QEX has been superb this past year, as have been Leif's 
Articles.  I can't wait for the next installment!

I had thought that there had been a thread on The Time Machine (TTM) on 
this archive...but maybe not.  Thanks for bringing the QEX ad up; I had 
missed it, and it provides a good reason to bring up TTM again for those 
who aren't aware of it's existence.

I had gotten two TTM's at Dayton last year and have used them on 
everything from LF thru HF to 144 MHz dual polarity receive with Linrad, 
both using their xtals and using the daughterboard and the PTS 
synthesizers for LO.  There is a bit of discussion on them in my 
DSP/Linrad papers included in the EME2002 syllabus, as well as the 
microwave update  2002 syllabus.  THe MUD paper is also on my website at
http://www.qsl.net/w3sz/w3szdspnew.htm#linrad ; this URL will take you 
right to the Linrad section, which contains the information on TTM.

I had previously put some information on TTM on my website, last updated 
back in October.  It should still be reasonably current:

I also discussed it's use with Linrad in the DUBUS article that is 
coming out in the next issue, but I don't think there is much there 
regarding TTM that is not in the sources just given..

>It is an I Q receiver with 80khz bandwidth.
The bandwidth is really not well defined (or written in stone), in that 
it is dependent on the characteristics of the LC filter that you use 
with the unit.  The 80 KHz they mention in the add is just a ballpark 
number.  I run them with Linrad and the Delta44 and get the full 90 KHz 
or so bandwidth available with the Delta44.  I am sure I could get more 
if I had one of the super soundcards.

> It is crystal controlled and can tune the popular
>HF bands
It is really a wideband device.  You select bands by (1) changing the LO 
and (2) sticking an LC filter on the front.  The main limit on the 
frequency range currently is the difficulty of finding a chip that will 
do 4*f for the LO as is needed for the I/Q mixer.  They are working on 
this and higher frequencies than 28 MHz should be do-able.  I don't use 
the crystals at all, just an external (GPS locked) synthesizer for the LO.

ESS has made daughter boards available for TTM that allow you to easily 
connect an external LO using SMA connectors.  These work very well and I 
highly recommend them as an accessory for TTM.

>, so it seems the 10 Mhz version might be the logical choice to fit Leifs mixing 
>plan for down conversion. As well, the 28 mhz version could be built and used with a 
>standard 2m/10m converter.
Yes!  It works fine with both the LT2S here and also homebrew front end 
to go from 144 to 28.

> The target use is for HF reception which aparently is less demanding then EME, as Leif 
>has defined it. It uses TUF-1 mixers and ordinary crystal oscillators. There is no mention of 
>any attempt at reducing phase noise or any mention of using an advanced sound card such 
>as the Delta 44.  I Q detection is balanced on board. Since a 6db NF at HF is perfectly
>ok and most signals tend to be similar in signal level, the TUF-1 mixers probably have 
>sufficient dynamic range most of the time on HF. A 20db attenuator (26db noise figure) on 
>the PCB is made available for setting the noise floor on HF.
Don't forget that your NF is mostly determined by the NF of the first 
stage and the NF contribution of following stages is divided by the gain 
of the intial stages.  So if you have a 0.2 dB NF preamp on the tower 
with a gain of 20-25 dB, you will still do quite well if the next stage 
has TUF-1's or whatever.  I use that here, actually with two stages 
before the TUF-1H's in the shack.  If you have 20 dB of net gain before 
TTM, its 6 dB NF becomes
the equivalent of 0.3 dB, I think.

>  I suspect that when a strong 
>local signal is present and a very weak EME signal is desired in the same passband, the 
>Time Machine would not hold a candle to the 2.5 mhz I Q detector designed by Leif. The 
>assembled price is $170 and kit form is $135, making it a tempting project for a rural 
>EMEer or perhaps an Urban EMEer such as myself that does not normally experience 
>local signals at the low end any closer then 30 km away (10db over S9) and is somewhat 
>pocket book challenged. The big question then becomes, what is the dynamic range of the 
>Time Machine?  80 db perhaps?
I think that the dynamic range of your sound card and Linrad will likely 
be the determining factor rather than TTM hardware.  The TUF-1 is not 
the limiting factor.  And if you are concerned about that, and have 
enough LO power, you can substitute a TUF-1H.

I had alot of fun using TTM for the ARRL FMT.  For those who are 
interested, this should appear on my website at 
http://www.qsl.net/w3sz/arrlfmt.htm later today.  I made up the webpage 
a while ago, but was keeping it back until the FMT results were 
released.  I am told they are now in the mail, though I haven't received 
mine yet.

I think TTM is an excellent way to get started with an IQ type of mixing scheme that allows you to get nearly the sampling rate of your sound card in bandwidth with Linrad.  IT is simple to build and I'd highly recommend the kit.

I have a bunch of other stuff on TTM that I'll put together and it will be accessed from my TTM webpage noted above.


Roger Rehr
2 Merrymount Road
Reading, PA 19609