-
DIY 10 MHz Distribution Amplifier - Requirements Gathering
Posted by Gandalf_Sr on 17 Mar, 2018 11:07 -
I just did a quick count and I have 10 peices of test gear with external 10 MHz inputs, Agilent SA, VNA, Sig-gene, Rigol scopes and AWGs. I also have a Nortel (Trimble) GPSDO like this one.
There's only 1 x 10 MHz output on my GPSDO so I'm looking to design a PCB that will go inside, probably with SMA PCB edge-mount connectors down one edge so I can drill a line of holes in the case and mount it inside. I've looked online at various designs and this one by Brian Dorey seems well thought out with a front end filter to remove harmonics plus sine and square wave outputs.
So the questions are:
1. What are the needs input-wise of the range of test gear that has 10 MHz external inputs and would the design above cover most stuff?
2. Does something like this already exist that covers my requirement? If I can buy for $20 from eBay I won't bother.
I'll share any design I come up with on the forum.
[EDIT] Corrected typo -
Brian Dorey's design looks nice but it may be a little under-powered for some instruments. For example my TTi TF930 counter requires 1Vrms for its external reference input which I find a constant source of annoyance as most other systems will make do with less. But it may be worthwhile checking your instruments.
I don't think you'll get anything decent on ebay for $20.
The one I'll probably get is the one by BG7TB (of GPSDO fame), there is a review of it here (it is 8 rather than 10 output but you may be able to use a couple of passive splitters for some instruments):
http://www.reeve.com/Documents/Articles%20Papers/Reeve_10MHzDist.pdf
but it seems to be more like $100 than $20.
The other option is to daisy chain as most instruments have a reference out as well as a reference in but then you need to have them on all the time and noise may be added.
-
Video distribution amplifiers are probably the easiest thing to use. 1 75R input, with generally a 75R resistor to define input impedance, and then a set of amplifiers to give a 75R output. AC coupled in most cases, and with enough gain to give a usable output for almost all equipment as it has to drive more than 1V into 75R for video in any case, and generally will work up to around 14MHz without issue.
If I ever get something that needs it I already have a 30 output video amplifier with modules from Opamp Labs doing all the work, including power supply, each in an octal plug in module. -
Video distribution amplifiers are probably the easiest thing to use. 1 75R input, with generally a 75R resistor to define input impedance, and then a set of amplifiers to give a 75R output. AC coupled in most cases, and with enough gain to give a usable output for almost all equipment as it has to drive more than 1V into 75R for video in any case, and generally will work up to around 14MHz without issue.
If I ever get something that needs it I already have a 30 output video amplifier with modules from Opamp Labs doing all the work, including power supply, each in an octal plug in module.
I agree. Many are easy to modify for 50 ohm as well if desired by simply stacking a 150 ohm resistor on top. I did that with an Extron ADA 6 300 HV for my GPSDO and it works fine. See Gerry Sweeney's Distribution Amplifier video on youtube (he actually replaces the 75ohm resistors with 50ohm). The Extron ADA series has at least 300MHz bandwidth. -
The Extron distribution amps are an easy mod and they aren't all that expensive I paid about $10. I expect supply far outstrips demand. The Extron ADA 6 300MX HV is a high bandwidth total overkill model. The ADA 4 model has plenty of outputs as well.
-
I just used a VDA .........1 in 50 out , a change of resistors and it was happy driving all the outputs into 50 ohms
-
I just used a VDA .........1 in 50 out , a change of resistors and it was happy driving all the outputs into 50 ohms
What he said. I have one installed in my rack. Did the resistor change to 50 ohm.
Gary -
Bought an ESE ES-208, 12 output 10MHz bandwidth VDA on eBay for $19 shipped the other day. The input is a high impedance loop, so a 50Ohm term is all that's needed on the input. Ill put it apart and once it gets here;
http://www.ese-web.com/208.htm -
Something like this: https://circuitmaker.com/Projects/Details/DaJMasta/Seven-Output-10MHz-Distribution-Amplifier ?
It's a bit of a pain to get higher power output out of these things, since the power requirements means you likely have to go to single amps per package, but up to about 1Vpp, you can use quad opamps. This has been enough for my externally referenced stuff. I've also got a square wave output version floating around on the circuitmaker, and since you can just use logic gates for everything, you can get a full TTL swing easily enough and then if you need a purer tone, can get decent results with a simple filter on the output. -
Thanks, I'll check that out although I'm going to have to install Circuitmaker to look at the schematic. I like the idea that it has optional gain setting / low pass filter on the front end. Any chance of a .pdf?
I found the circuit below for the LT6550/1 (triple/quad) on page 16 of the datasheet. This driver IC is actually the same one used by Brian Dorey's design. The LT6550/1 drives into overload if the input goes over 1.25 volts on a 3 V single rail supply as shown which is exceeded by the output of my GPSDO (I measured 1.8 Vpp driven into 50
today). So maybe I need to put some kind of front end in front of it and it would be neat to allow for an adjustment so that different levels of input can be catered for.
However, if I made Vcc = 5 V, then the input of 1.8 Vpp I'm providing would be OK as long Vin max doesn't exceed 5-1.75 V = 3.25 V so I would bias the input to 3.25/2 = 1.625 which I could achieve by tweaking the input potential divider resistor values.
Any thoughts? I should wait until I see the circuitmaker.com design
-
I've been tinkering with this design:
https://www.ohwr.org/projects/pda-8ch-fda-8ch/wiki/frequency-distribution-amplifier
if you care about phase-noise (and ADEV) you need to have clean DC-rails and decemt PSRR in the amp. From a medical-rated AC/DC 9V brick I use a common-mode choke, then a BNX025 filter and then LDOs LT1936/LT3015 - seems to work for me.
There are designs that paralllell two LMH6702's and get about 3dB better phase-noise, but around -160 dBc/Hz-ish for a 10 MHz carrier is doable with op-amps. There are probably discrete transistor designs that do a bit better - if you really need it..
this page has links to some other designs:
https://www.ohwr.org/projects/pda-8ch-fda-8ch/wiki/similar-projects
(might add the ones mentioned in this thread!) -
Thanks, I'll check that out although I'm going to have to install Circuitmaker to look at the schematic. I like the idea that it has optional gain setting / low pass filter on the front end. Any chance of a .pdf?
Sort of annoying little "feature", and I don't know if there's an easy way to do .pdfs, so here's some screen captures of each:
http://www.medpants.com/pics/7%20distribution%20amp.jpg
http://www.medpants.com/pics/4%20distribution%20amp.jpg
Similar designs to what you've seen, no doubt, but the board's laid out and ready to be built up. I never made boards of the 7 way amp (I made ones with the same schematic and almost the same layout that had a few more bits to supply a GPS unit and a monitor for a custom GPSDO), but I have some spare boards of the 4 way amp I could pass along for postage. -
1. What are the needs input-wise of the range of test gear that has 10 MHz external inputs and would the design above cover most stuff?
The worst case and what I would design for is TTL level inputs which depending on exactly what is happening and amount to the 1 Vrms that jpb mentioned.Quote2. Does something like this already exist that covers my requirement? If I can buy for $20 from eBay I won't bother.
As SeanB said, video distribution amplifiers can often be modified to handle 10 MHz distribution.
-
It seems you have many devices so an Extron Video Distrubition amplifier mod is probably the way to go. (I bought a couple of these but still haven't got around to modding them... one day).
I created a 1:4 10MHz and added a pulse stretched 1PPS output for my lab equipment, based on LT6551IMS. Seems to work fine (but I'm no test nut). Another EEVBlog member built one from a PCB I had spare and seemed happy. I imagine this has too few outputs to be useful to you. -
Is it really worth the time and effort to design and build your own distribution amplifier when video distribution amps are so readily available for a reasonable price? To me, it feels like a lot of work with relatively little payoff. I can think of lots of other projects I'd rather work on.
In any case, here's another amp that has been discussed on the Time-Nuts list: http://tapr.org/kits_tadd-1.html .
Another slightly different idea is to combine a frequency divider with a distribution amp. Time-nuts often need this sort of thing. http://www.perdrix.co.uk/FrequencyDivider/index.html
Ed
-
Is it really worth the time and effort to design and build your own distribution amplifier when video distribution amps are so readily available for a reasonable price?
Probably not but if you only need <=4 outputs then a simple daughter board means you can fit it all together in the same box as your GPSDO and save space. -
Thanks for all the input guys and offers, I have a lot to chew on
I looked for old distribution amps but I design PCBs for a living so this is a very simple project for me if I can just decide what to put in it, I could also make a batch of PCBs and you guys can get them for cost if you like.
I also want a PCB that I can fit inside my GPSDO that uses SMA connectors, buying 10 BNC-BNC leads would be expensive and I have loads of cheap SMA leads from eBay (as well as some expensive ones).
Am I over-thinking the input, one of the first designs I linked to had a $6 IC to filter at a cut off of 15 MHz to suppress the harmonics, the medpants design uses a simple (optional) capacitor to form a low pass filter although I haven't checked what the -3dB point would be for that arrangement, it might be all that's needed. What do you guys think?
I'm also wondering what the best way would be to allow the user to adjust for different input levels without changing components; what's the best way to do this?
I think I'll design for 9 or 12 outputs and then add a 1 pps (stretcher) and two TTL outputs, is that enough? -
If you use a large enough extron box, you can fit the GPSDO in it and still save spaceIs it really worth the time and effort to design and build your own distribution amplifier when video distribution amps are so readily available for a reasonable price?
Probably not but if you only need <=4 outputs then a simple daughter board means you can fit it all together in the same box as your GPSDO and save space.
The idea that the GPSDO is "bought" but the distribution amplifier is "made" seems backwards to me.
But hey! I'd rather read a thread about another 10MHz distribution amplifier than "what scope/multimeter/probes/soldering iron to buy"
Or yet tweaking another headphone amp if we can do some cross forum complaining
-
Just modeled 470 and 270pF in LT Spice, looks like it would be -16 dB at 10 MHz
-
I think I'll design for 9 or 12 outputs and then add a 1 pps (stretcher) and two TTL outputs, is that enough?
If you've got a PPS output from the GPS already, it's simple enough to just put it through an inverting buffer and drive some lines - even directly from a GPS unit, the adjustment step size is usually <40ns for a lot of modern ones, so you can get good raw accuracy and of course the normal stability over time from just buffering the output directly.
If you want to have TTL squarewave outputs, I would treat it as a mixed domain board, and isolate your analog ground from your digital one. I'd also put a ferrite or a small inductor between the main power rail and the opamp power supply for a little more high frequency rejection, since the digital logic will probably be fast enough switching to make a bit on the power rails even with normal decoupling.
As for whether it's enough.... how many instruments do you really have that can take the input? I've got a scope, SA, RF gen, arb gen, counter..... and 7 outputs is plenty for that, which seems like a ton of equipment already, some of which really don't benefit much from it in most situations... so 9-12 seems like more than enough. That said, it's much cheaper to build in the extras early on, so maybe going all out is nice just because it will be plenty into the future. I prefer BNCs just because the equipment you hook up is usually BNC stuff and the actual 1-2m cables can be gotten cheaply, but decent quality BNCs are definitely the most expensive part of the project if you make it yourself.
I guess one other tip if you end up using a design like the one I linked: if you use a rail to rail opamp and bias them right in the middle instead of the skewed one as I have, you should be able to squeeze a little more amplitude out of the outputs. The way it's setup, driving all four outputs into 50 ohm loads makes the chip raise 10-20C over ambient, so there's not a ton of overhead if you use quad amp chips, but there is a bit available. I just ran into an issue where a low noise quad amp was going to be more expensive if I wanted rail to rail, so instead of getting the most out of my 5V single ended supply, I adjusted the bias point down to compensate for the volt and a half or so you need to be away from the top rail, and that gives you less maximum amplitude to play with. -
Here's an early version, no power supplies and only 8 channels of sine distribution. One area I'm not sure about is the coupling from U3.2 and 3.3 into the driver amps. You'll see this is a hybrid with the front end from one of the linked circuits and the output section from the LT6551 data sheet. Any input welcome.
-
I added RF 1:1 isolation transformers onto the outputs of mine a la https://www.tapr.org/~n8ur/TADD-1_Manual.pdf
-
I'm also wondering what the best way would be to allow the user to adjust for different input levels without changing components; what's the best way to do this?
If the output stage is linear as with an operational amplifier, then an adjustable voltage divider at the input of the operational amplifier can adjust the output level. 10 MHz is well within the capability of a trimmer potentiometer.
I have never seen any distribution amplifier designs do it but a current switched output stage which drives a termination shunt resistance instead of a series termination could have an adjustable output voltage just by changing the current. Some analog function and pulse generator designs did this to control positive and negative pulse height or slew rate using a current driven diode bridge.
Another way is to use analog switches for the output stage with an adjustable reference. I know of a couple of people who built precision square wave calibrators this way.
-
I scanned through the previous posts and I didn't see this mentioned. On the video amps where the input and output have 75 ohm resistors, there is no reason to remove them and possibly damage the circuit board, just solder a 150 ohm in parallel to get 50 ohms.
-
I added RF 1:1 isolation transformers onto the outputs of mine a la https://www.tapr.org/~n8ur/TADD-1_Manual.pdf
+10 on this.
Isolation transformers have got me out of trouble so many times in the past with pro audio gear. Connecting a rack's worth of gear together with a DC path all the way through almost guarantees that it will cause you trouble at some point. Plus point is that RF isolation transformers are cheap; if sir needs to ask the price of good AF signal isolation transformers like Jensens, sir cannot afford them (they're on the close order of £50 each).