Why don't more distribution amplifiers have lowpass filters on the output?

[matthuszagh]

Consider the Wenzel LNDA-3 (https://www.quanticwenzel.com/wp-content/uploads/LNDA-3.pdf). The harmonic performance is specified as -30 dBc. It should be easy to cut this dramatically by placing a lowpass filter at the output considering the input frequency is known and fixed (eg Mini-Circuits SCLF-10+ or really anything else). That seems like an easy way to improve that spec. Why don't they do that, especially for such a premium amplifier?

Is the reason cost, or board real estate, or something else? Or, is there some performance reason why it can actually be detrimental in certain cases? The filter will have a little insertion loss at the target frequency, but it's not that large. For instance, for the SCLF-10+, it's only about 0.5 dB at 10 MHz.

[coppercone2]

why not filter it at the input of what it is distributing to?

that way you catch leakage from the coax maybe

bit of a VCR clock IMO

[matthuszagh]

why not filter it at the input of what it is distributing to?

that way you catch leakage from the coax maybe

bit of a VCR clock IMO

How would you pick up harmonic distortion from coax runs? I see how filtering at the end would be useful for other or broadband noise, but I'm not sure how I see that as useful for harmonic suppression.

[coppercone2]

if your gonna put 1 filter its better to put it at the end then the beginning because it all leaks signal meaning there might be signals entering in the transmission line between the amp and the receiver. Its better. You get something extra if you do it that way. Extra resilience. You might have a reallly long cables connected to distribution amplifier, so it might not be trivial, especially when it starts getting damaged. Then you need a filter ANYWAY to handle that, so you just paid for filter thats not doing all it can do simply because of its position, since generally LPF have a pretty long stop band.

And half a db can be alot. It can be the difference between a cable run getting the correct signal across or not getting the correct signal across. In many cases the cable installation can cost more then all the equipment, particularly commercial  buildings. Not having a filter there give you the best chance of success without having to buy extra equipment like a amplifier.

[ejeffrey]

Why do you care about harmonic distortion in a clock?

Filters, if they are usefully narrow have large group delay and usually high temperature coefficients of delay. Adding filters you might not need can compromise phase noise/stability that you can't get back.  If you need a harmonic filter it's better to put it at the other end of the cable in any case.

[fourfathom]

In my uses I would rather have a clean square wave with minimal filtering.  This way I get minimum phase noise at the clock edges.  If the distributed clock is filtered down to a sine wave then any coupled noise, or even the thermal noise of the receiving device, will add to the clock jitter.  If you can't stand the harmonics, then filter at the receive end.

And if I am distributing a 1pps pulse then I definitely don't want that narrow pulse getting filtered.

[David Hess]

... Why don't they do that, especially for such a premium amplifier?

Is the reason cost, or board real estate, or something else? Or, is there some performance reason why it can actually be detrimental in certain cases? The filter will have a little insertion loss at the target frequency, but it's not that large. For instance, for the SCLF-10+, it's only about 0.5 dB at 10 MHz.

The reason they do not include a low pass filter on the outputs to reduce noise and distortion is that the filter would have to have a low enough cutoff frequency to be useful, but the low cutoff frequency would introduce variable phase distortion, so now the phase would vary over time and temperature.

This is also why a bandpass filter is a terrible idea if phase is to be preserved.

As a compromise, a higher cutoff frequency could be used to minimize the phase distortion, and notch filters added at the low harmonics.

In my uses I would rather have a clean square wave with minimal filtering.  This way I get minimum phase noise at the clock edges.  If the distributed clock is filtered down to a sine wave then any coupled noise, or even the thermal noise of the receiving device, will add to the clock jitter.  If you can't stand the harmonics, then filter at the receive end.

Fast enough edges create their own problems.  Transmission line loss increases at higher frequencies, and there is a phase distortion component as well from a change in propagation velocity versus frequency.  Combined these lead to transmission line "dribble-up" when fast edges are used and the rise/fall time becomes slower and non-linear.  Frequency components beyond the TEM propagation mode for the cable also get mangled arguing for smaller cable which further increases dribble-up.

Using the fastest possible comparator on the receive side also increases jitter because of increased sensitivity to high frequency noise.

[fourfathom]

In my uses I would rather have a clean square wave with minimal filtering.  This way I get minimum phase noise at the clock edges.  If the distributed clock is filtered down to a sine wave then any coupled noise, or even the thermal noise of the receiving device, will add to the clock jitter.  If you can't stand the harmonics, then filter at the receive end.

Fast enough edges create their own problems.  Transmission line loss increases at higher frequencies, and there is a phase distortion component as well from a change in propagation velocity versus frequency.  Combined these lead to "dribble-up" when fast edges are used and the rise/fall time becomes non-linear.

Using the fastest possible comparator on the receive side also increases jitter because of increased sensitivity to high frequency noise.

Yes, agreed.  Let me amend my comment to read "reasonably clean square wave, but not so fast as to cause problems".  As for the too-fast comparator being sensitive to high-frequency noise, that's one reason why I want a square (-ish) wave at the comparator input.  This gives the noise less time inside the comparator threshold.  But all things in moderation...

[ejeffrey]

Overly sharp square waves have problems, including dispersion and reflections.  Sometimes the faster slew rate is worth it other times not.  But I don't see why anyone would care about moderate harmonic distortion  in a clock distributior.

[Gerhard_dk4xp]

The reason they do not include a low pass filter on the outputs to reduce noise and distortion is that the filter would have to have a low enough cutoff frequency to be useful, but the low cutoff frequency would introduce variable phase distortion, so now the phase would vary over time and temperature.

This is also why a bandpass filter is a terrible idea if phase is to be preserved.

As a compromise, a higher cutoff frequency could be used to minimize the phase distortion, and notch filters added at the low harmonics.

In my uses I would rather have a clean square wave with minimal filtering.  This way I get minimum phase noise at the clock edges.  If the distributed clock is filtered down to a sine wave then any coupled noise, or even the thermal noise of the receiving device, will add to the clock jitter.  If you can't stand the harmonics, then filter at the receive end.

Fast enough edges create their own problems.  Transmission line loss increases at higher frequencies, and there is a phase distortion component as well from a change in propagation velocity versus frequency.  Combined these lead to transmission line "dribble-up" when fast edges are used and the rise/fall time becomes slower and non-linear.  Frequency components beyond the TEM propagation mode for the cable also get mangled arguing for smaller cable which further increases dribble-up.

Using the fastest possible comparator on the receive side also increases jitter because of increased sensitivity to high frequency noise.

Exactly.
I have a Lucent GPS that delivers 5 MHz from a MTI-260 crystal oven, but I needed 10 MHz
distributed. So I built a push-pull doubler and notched away at least the close-in multiples
at 5, 15, 20, 25 MHz. It showed that crystal notches are the best and easiest.

<     http://www.hoffmann-hochfrequenz.de/downloads/DoubDist.pdf     >

Cheers, Gerhard