EMRFD Message Archive 305

Message Date From Subject
305 2007-01-06 09:25:43 Brian George Multiband DC Receivers
Hi all,

I am trying to build a direct conversion receiver covering all amateur bands
up to 30 MHz. The DDS module is finished and the rest of the design will be
based on circuits in EMRFD with a diode ring mixer.

In order to avoid the complexity associated with multi-banding, would
performance be seriously degraded if I used the front-end from figure 8.22
(page 8.14) of EMRFD but with 30 MHz low pass filters before and after the
fet rf amp, to cover all bands without any switching?

Any guidance would be greatly appreciated.

Brian G3ZOH
306 2007-01-06 10:15:18 Luiz Amaral Re: Multiband DC Receivers
Remember that, with a square wave LO (or saturated sinusoidal) that are the
best ones, the direct conversion radio will receive also all the odd
harmonics frequencies, that is, if you tune it on 3 Mhz, it will also
receive 9, 15, 21 and 27 MHz simultaneously, if you use a simple 30 MHz
low-pass filter. The sensitivity for these harmonics decreases with 1/N,
where N is the odd harmonic order for a square wave LO. If you choose LSB to
receive the 3 MHz signal, the 15 and 27 Mhz signals will be received as LSB
too, but the 9 and 21 MHz ones will be received as USB and vice-cersa.
The input filter for a direct conversion reception for all HF continuous
band may be of low-pass type, but must be switched in such a manner that
the 3rd harmonic is always rejected, or the receiver performance will be
degraded by the presence of those harmonic spurious detections.
There are also some other cares that must be taken to guarantee the
performance.

Luiz - PY1LL

----- Original Message -----
307 2007-01-06 10:33:40 Lasse Re: Multiband DC Receivers
Brian,
yes the performance will suffer a lot!
You will have whole of HF spectra hit first your pre-amp and the with a even higher amplitude the mixer.

You will hear a lot of ghost signals wich really are IM2 products. I do not know what type of DDS you are using, but it will have atleast some spurs, and thes will mix with all the signals too.

If you can get a DDS going and build a good DC receiver you should have NO problems designing a band-pass filter bank.
Why not get the receiver done and then add the front end filters, thus having a chance to compare how big improvement you will have :)

Cheers
Lasse SM5GLC

On Sat, 6 Jan 2007 16:57:50 -0000
"Brian George" wrote:
> Hi all,
>
> I am trying to build a direct conversion receiver
>covering all amateur bands
> up to 30 MHz. The DDS module is finished and the rest of
>the design will be
> based on circuits in EMRFD with a diode ring mixer.
>
> In order to avoid the complexity associated with
>multi-banding, would
> performance be seriously degraded if I used the
>front-end from figure 8.22
> (page 8.14) of EMRFD but with 30 MHz low pass filters
>before and after the
> fet rf amp, to cover all bands without any switching?
>
> Any guidance would be greatly appreciated.
>
> Brian G3ZOH
>
>
>
>
> Yahoo! Groups Links
>
> <*> To visit your group on the web, go to:
> http://groups.yahoo.com/group/emrfd/
>
> <*> Your email settings:
> Individual Email
308 2007-01-06 10:53:55 Joe Rocci Re: Multiband DC Receivers
No matter how you drive it, the diode ring mixer will have significant odd-order harmonic responses. This type of mixer works by diode-switching the input phase by +1 or -1,depending on whether the positive or negative half-cycle of the LO is driving the diodes at any given instant. If you do the math on this process, you'll find inherent odd-order responses. Driving the diodes with a square-wave (or driving them hard with a sine wave, which accomplishes the same thing) makes the switching process more efficient and less distortion-prone because the diodes spend less time in the non-linear & lossy transition range, and more time in the ON or OFF range.
 
Joe
W3JDR
 
 
 
----- Original Message -----
309 2007-01-06 12:10:24 dwarden2@suddenli... Re: Multiband DC Receivers

It seemed to me that I heard a lot of birdies on my homebrew (BITX 20) SSB radio when I used a square wave LO as compare to using a fair sinewave (with 2nd harmonic 30 dB down).

I’ve decided that I will be using a sinewave shaped LO with a LPF.

 

Thanks,

Dave AD5JO

 


310 2007-01-06 13:01:31 Joe Rocci Re: Multiband DC Receivers
Maybe your square wave wasn't 'square'. This might happen if you started off with a sine wave that wasn't pure and squared it up in a gate. This could result in an output rectangular wave that didn't have a 50% duty cycle, If the LO drive signal has significant even-order content (either because it wasn't squared up properly or because you used a non-perfect sine wave) you can introduce even-order products in the mixer output. A non-square rectangular wave can have more even-order harmonics than a distorted sine, resulting in more mixer responses to these harmonics.
 
Joe
W3JDR
 
----- Original Message -----
311 2007-01-06 17:41:12 Luiz Amaral Re: Multiband DC Receivers
You are right....the birds must be just due the odd harmonics on LO. As the saturated/square wave LO is more efficient, the best solution for me is to use double-balanced mixer with suitable input filter.

Luiz - PY1LL
 
----- Original Message -----
345 2007-01-14 19:09:27 Luiz Amaral Re: Multiband DC Receivers
Ok, but I'm not talking about the inherent converter distortion, but only about the odd harmonics content of the LO, and so, the direct conversion receiver will get all signals of those harmonic frequencies.
Another problem with these receivers, that can be minimized by the use of double balanced mixers, is the direct feedthrough, specially with strong AM broadcastings that are detected just by the non-linearities of the mixer and are very annoying.

Luiz - PY1LL
 
----- Original Message -----