10MHz distribution amplifier

[ChrisLX200]

Looking at options for distributing a 10MHz reference signal, just curious whether anyone has knowledge of this Chinese offering: https://www.ebay.co.uk/itm/10MHz-Distribution-amplifier-frequency-standard-8-port-output/121677559563?ssPageName=STRK%3AMEBIDX%3AIT&_trksid=p2060353.m1438.l2649

I have yet to search for possible DIY solutions but doubt they would be much cheaper than the above unit. I could be wrong though

[edpalmer42]

The other popular option is to use a video distribution amplifier.  Extron is one frequently mentioned brand.  You should probably change the impedance from 75 ohm to 50 ohm by (typically) changing one resistor per channel.  Look for one with BNC jacks and maybe a few channels.  Make sure it's for analog video.  Bandwidths can go up to 100 MHz, but you probably won't need it.  Look for maximum outputs of +7 to +10 dBm.  That gives you room to attenuate as necessary.

I scored a Leitch 682 shelf full of cards so that's what I use.  It gives me 10 channels with 4 outputs per channel.  I can daisy-chain the inputs if I need more outputs.

Ed

[cdev]

I think they likely would be much cheaper if you wait until you see a decent price on one.

I recently bought an Extron video distribution amp mostly for the metal case, which should be large enough for my GPSDO and then some The video distribution amp has three channels plus sync which is a TTL level device. And a bunch of BNC connectors for them

It hasn't arrived yet, but this looks like a good way to go, despite the fact that it appears the amp is too broadband as it comes to work optimally for 10 MHz but, it should be easy to tweak it to have less bandwidth. Also, people are supposed to isolate the outputs which can be done with transformers so I plan to use toroids and wind my own baluns.

It is "for parts or not working" so I am not expecting it to work but I figure the major cost of this device is in the case and connectors and power supply so, I think the chances are good that I should be able to get something workable out of it, even if I have to replace the actual amplifier. 10 MHz @ 5 volts at unity gain  is not difficult to achieve using op amps that I already have.

It probably will need a coat of paint too.

[glarsson]

Looking at options for distributing a 10MHz reference signal, just curious whether anyone has knowledge of this Chinese offering: https://www.ebay.co.uk/itm/10MHz-Distribution-amplifier-frequency-standard-8-port-output/121677559563?ssPageName=STRK%3AMEBIDX%3AIT&_trksid=p2060353.m1438.l2649

I have one of these. The sine input is filtered and turned into a logic level square wave. The logic level is buffered by eight gates and filtered using passive 10MHz low pass or bandpass filters to give eight sine wave outputs. The level of the outputs are therefore independent of the level of the input signal (as long as the input signal is within the acceptable range). Note that a variant of this amplifier is available without the filters, i.e. the outputs are square wave. One odd thing about this distribution amplifier is that the outputs 3, 4, 7 and 8 are inverted compared to outputs 1, 2, 5 and 6.

Anything more you need to know?

[ChrisLX200]

Thanks all! I've seen a few options in terms of video distribution boxes - including some that are switchable between 50 and 75 Ohm. Cheap enough to take a punt on and not worry too much if it doesn't work. If that happens I could always shell out for the purpose designed unit.
Thanks again

[cdev]

See attached photos.

This is my Extron unit, which just arrived!

The box is almost empty.  Great! It also has four mounting posts for a PCB which aren't occupied by anything. Perfect.

I'm going to investigate if it works and if so, see what kind of gain it has with my signal generator now. One potential gotcha.. it appears from the label that it may only output one volt. (Frankly, I likely am misunderstanding something, because others have been successful with these units). Need to consult the manufacturers site and Time-nuts a bit more.

Look at it this way, Even if when we buy them, assuming we get a good price. even if we have to replace the entire amplifier, with a homebuilt one, I think I will still come out way ahead just based on the typical cost of a case and BNC connectors.

Update: It appears to work well, however, I need to build some proper terminations for my test instruments, clearly, because without terminators, the measurements the scope is seeing arent really valid- however I can see that the output of the video amplifier is a nice looking sine wave its not clipping at 10 MHz within the low voltage range. From what I can tell, its looking quite good on all three video channels.

Additionally there is another channel for TTL sync which I just verified worked.. I just tried a quick and dirty test to see what happened when I provided it with a square wave in the several Hz range.. offsetting the voltage so it was positive going, I didn't provide it with a 1PPS signal, nothing like that. What did I get? it looked like half of a square wave, with some ringing consistent with not having any termination. it looked reasonable. Even at 10 MHz what came out looked fairly reasonable considering the sync is likely much much lower than that.  So its looking good.

[ChrisLX200]

Yes it is do-able - Gerry Sweeney used one of those Extron boxes for that very purpose:

[cdev]

Thank you for this! 

I'm at about 20 minutes into this video and yes, this looks like exactly what I need to do. I'll end up with nine 10 MHz outputs and three 1 pps outputs.

My sync waveform looks exactly like his. Its basically the same situation.

There is more info here too..

http://www.ko4bb.com/getsimple/index.php?id=extron-ada-3-80-video-distribution-amplifier

[f1rmb]

I've also built one for my GPSDO, based on the DF9NP schematics:

http://www.df9np.de/Contents/FCT3807.pdf

Cheers.
---

[edpalmer42]

One comment on Gerry Sweeney's build - if you can hold your hand on the unit, you don't need - and shouldn't use - a heatsink.  A Rb standard has two or three ovens in it.  If you go crazy with the heatsinking, all you're doing is forcing the ovens to work harder to keep things hot.

Ed

[dnessett]

Looking at options for distributing a 10MHz reference signal, just curious whether anyone has knowledge of this Chinese offering: https://www.ebay.co.uk/itm/10MHz-Distribution-amplifier-frequency-standard-8-port-output/121677559563?ssPageName=STRK%3AMEBIDX%3AIT&_trksid=p2060353.m1438.l2649

I have one of these. The sine input is filtered and turned into a logic level square wave. The logic level is buffered by eight gates and filtered using passive 10MHz low pass or bandpass filters to give eight sine wave outputs. The level of the outputs are therefore independent of the level of the input signal (as long as the input signal is within the acceptable range). Note that a variant of this amplifier is available without the filters, i.e. the outputs are square wave. One odd thing about this distribution amplifier is that the outputs 3, 4, 7 and 8 are inverted compared to outputs 1, 2, 5 and 6.

Anything more you need to know?

Serendipitously, I was preparing some information on the unit you bought when I noticed this thread. It would have been inappropriate for me to use this thread for presenting that information, so I created a new one. For those interested, it is here.

[cdev]

ChrisLX200,

Read the threads on the Time-Nuts mailing list on the Extron units. It seems the unit I have may ADA 3 180, I think not be as good as the one pictured in the video (which also has more outputs) 

My GPSDO puts out a full 5 volt 10 MHz signal and 5 V 1PPS signal, the 1PPS is okay, and will have to use a variable resistor to fine tune the input for least distortion. I think the other similar looking Extron unit is capable of more gain with a different input system.

But, its still v. nice, the distortion is nil from input to output on 10 MHz with a lower input. So getting nine outputs should be a piece of cake.