[time-nuts] TAPR Announces TADD-1 Distribution Amplifier

John Ackermann N8UR jra at febo.com
Tue Nov 22 08:20:37 EST 2005


David G. Andersen wrote:
> This is great, John - thanks for designing it.
> 
> Isn't that MAX477 convenient? :)  The only drawback to it, from my
> view, is that it's got too much bandwidth from all those weird "video"
> applications (who wants video when you can play with 10Mhz GPSDOs?),
> leading to the possibility of harmonics on the output.  And conveniently,
> the TADD-1 seems to have a 10Mhz bandpass filter... nice.
> 
> Curious about one thing - you provided spots for an input bandpass
> filter, but nothing on the outputs.  How do the output harmonics look
> when amplifiying a clean 10Mhz input signal?
> 
> [CC: list trimmed to just time-nuts]
> 
>   -Dave

Hi Dave --

Thanks for the comments!  The gain bandwidth of the MAX477 is a problem, 
and we had to add series decoupling resistors to stomp out a parasitic 
at about 350MHz.

The harmonic performance isn't too bad as long as you don't overdrive 
the input so that it clips.  The assembly manual has some details on my 
harmonic testing results, but I didn't have a 5 or 10MHz bandpass filter 
handy to clean up my source and as a result I had to compare the 
harmonic content of the source with that of the output.  They were 
within a couple of dB.  However, that's suspect because the  system gain 
shows a tilt toward high frequencies, so the source harmonics were being 
amplified as well.  In fact, I suspect that in my test the gain tilt 
masks any harmonic contribution of the amp.  I plan to do a more proper 
harmonic test but haven't gotten the bits together yet to do it.

The bandpass filter is really intended for something like cleaning up a 
square wave input.  I have an Efratom "brick" Rb that has a 10MHz square 
wave output, and I envisioned using the filter to get a nicer sinewave 
out of it.  The filter provides about 35dB suppression at the 
subharmonic as well as 2nd and higher harmonics.

John



More information about the time-nuts mailing list