That actually looks very interesting and doable. I've ordered some samples!
-- Failure does not prove something is impossible, failure simply indicates you are not using the right tools... nico@nctdevpuntnl (punt=.) --------------------------------------------------------------
Well the output is a square wave (with loads of harmonics).
I'll order a couple of those then :-)
The fastest digital scope I have goes to 500 MHz so thats a no-go. I think I'll build an RF detector around an ADI RF detector chip. That would turn the entire gadget into a much more versatile device.
If you reduce the function to (out * divider) /refin=... the math becomes quite easy. A few lines of code does the job if you can live with a few kHz resolution.
-- Failure does not prove something is impossible, failure simply indicates you are not using the right tools... nico@nctdevpuntnl (punt=.) --------------------------------------------------------------
er
e.
g-multi...
I get around 51 ohms, so with thicker metallization from the tinning it seems right. But 14 mils is an odd 42 mil thick board? Please tell me you DIDN'T go 14, 32, 14 ?? You'd be far better off with a 32 mil board and go 10, 10, 10 than do that. If you need to stay at 62, add two layers and interlace power and GND with 8 mils each, be a lot better off.
14 mil thick and 26 mils wide moves a LOT of field up into the air! changes the 4.2 to 3.2 effective. Not sure it is good to have that floating around above your circuitry.What do you mean by 'serious mismatches in circuits'?
=3D =3D =3D not even going to ask: from URL
You have to program it with every frequency in your chosen sweep. i.e. you have to program every step manually.
You can alter the output power via one of the control registers and this might help flatten the output power but my guess is that your output stage from the 4350 will contribute a lot of the unequal power delivery.
From memory you have to reprogram registers 0 and 4 each time you update for a new frequency.
You could run it into a limiter amplifier and then compensate for the rolloff in the limiter with a simple lumped network to get a very flat output power across the full range.
Note: You will get fairly poor spectral purity from the ADF4350 because of the way it operates via internal dividers. i.e. you will get fairly high harmonic content from it. If you use a limiter to level the output then the purity will get worse but it might still meet your requirements.
Your sweep speed will be limited by the speed of your SPI interface to the ADF4350 and the number of programmed step points you choose.. =3D =3D =3D
I may be wrong, but it you programmed output+ and output- to be the same, you have a MUCH lower transmission line impedance right at the terminals of the IC. More like 43 ohm differential 57 ohm com mode yield around 31 ohm per line! pretty bad mismatch.for about 100 mils which is approx only 1/6 of a 1/4 wave on the PCB. May not even relate to your dip in amplitude, just not very 'controllable' layout. What I mean is you can't count on the layout to 'be' what you think it is.
Again, thinner dielectric and it gets easier, plus go to a material with a dielectric constant that is constant from 1+GHz upwards, not FR4 Those are short runs, less than inch(?) so losses won't matter.
I'd be tempted to get in there and start modeling using LTspice and some lossy transmission lines to see what you get.
er
e.
g-multi...
Rough calculation 0.6 inch is 1/4 wavelength along that S curve which means about 1/2 wavelength to SMA? Are these traces launching any significant energy? Note a lot of field above the trace which has one speed in air and energy below trace which has another speed, can't 'noodle' that one out right now, keep going back to make these boards thinner.
a
her
re.
g epg
.highlandtechnology.com=A0 jlarkin at highlandtechnology dot com
Not very perfect right at the IC and the run to the SMA is into the wavelength size, which is significant length.
wavelength INSIDE dielectric is approx 300e6/2.4GHz/.0254/sqrt(4.2)/
0.0254 =3D .2.4 inchesthat means the 1/4 wavelength where all the 'magic' rotations occur, is around 0.6 inches!
Use your fingers!
when you touch the part that is resonant, it will shift frequency lower and de Q.
Someday you can attain RF wizard status like me.
I can fix things simply by touching them. :-)
Mark
Its a 62 mil board.
How much improvement would that get me? The 'problem' is that when it comes to these kind of low volume tinker projects there is not much choice in board stackup. I have to go with what a pooling service offers. I could choose a 6 layer stackup for the next version but it is serious overkill for the rest of the board which could easely be on a 2 layer board :-)
A slightly different subject: AFAIK there are at least two ways to determine mismatch. TDR or by using a directional coupler and see what is reflected back. I'd like to get a setup with a directional coupler.
Thats another problem I'm going to think about later on. I'm not quite happy with the amount of harmonics produced by the 4350.
The SPI interface can go up to tens of MHz. Thats not really a problem :-)
Why is that? The outputs are open collectors. More or less current sinks with infinite (tm) impedance. They outputs are terminated into
50 Ohms from the power supply.Anyway, I have removed one of the output traces with a dremel-like tool to make sure but the problem persists.
-- Failure does not prove something is impossible, failure simply indicates you are not using the right tools... nico@nctdevpuntnl (punt=.) --------------------------------------------------------------
Incwww.highlandtechnology.com jlarkin at highlandtechnology dot com
There's no way his transmission line structure can explain the huge dip he's seeing. It's very nicely matched in the DC-3 GHz sort of range. I do boards that look just about like that, on FR4, and go a lot faster.
-- John Larkin Highland Technology, Inc jlarkin at highlandtechnology dot com http://www.highlandtechnology.com Precision electronic instrumentation Picosecond-resolution Digital Delay and Pulse generators Custom laser controllers Photonics and fiberoptic TTL data links VME thermocouple, LVDT, synchro acquisition and simulation
I think its external interference. He should do a site scan out of curiosity. Or like suggested, put the board in a metal can.
Those 6 control lines on the left look lonely, for kicks I'd place a ferrite bead or core across them and see what happens.
Cheers
Measure the output without the filter. You can do that without building or buying anything. Then if you have determined that it is really the filter you can add 10-20% or so to a cap here and there and see which one moves the resonance. If your analyzer has a track gen you can also plot the filter passband. There are so many ways to peel the onion here.
If you really have to have a directional coupler, Mini-Cicuits carries those. Not sure how good, always meant to buy one but so far did all such jobs without.
If you need a wide frequency range to be convered by the chip consider a tracking filter.
On really short stretches (a centimeter or so) such mismatches don't affect things much.
-- Regards, Joerg http://www.analogconsultants.com/
Yup. Hand waving can show this already, literally. Cup the hands over the board and if the dip changes, bingo.
The usual ferrite material is no good up there anymore. Forming 1-2 loops around a chop stick would likely yield better results. And it ain't gonna cost nothing if you pull them chop sticks back out :-)
-- Regards, Joerg http://www.analogconsultants.com/
Now if the whole table starts to hover when you do that I'd say it's not an RF effect :-)
-- Regards, Joerg http://www.analogconsultants.com/
The above is a slightly different topic. Once I get this setup going I'd like to use it for testing some odd ways to connect ethernet. RJ45 is not always the most practical connector.
Actually there is no filter at the output :-)
I ordered some RF detector schottky diodes. If all goes well they'll arive tomorrow. I really need to figure out whether the problem is in the SA or in my circuit.
I've use a directional coupler from Mini-Circuits before for antenna matching. But that was only 430MHz. Fortunately I can buy direct from their UK office.
That sounds like a good plan. But I'll need to read more about that first.
As I noticed :-) There is some change (a few dB) in the output level though.
-- Failure does not prove something is impossible, failure simply indicates you are not using the right tools... nico@nctdevpuntnl (punt=.) --------------------------------------------------------------
The spectrum analyzer can easily be tested for that. Hang a preamplifier onto it with nothing connected to it's input. This will raise the noise floor. Most spectrum analyzers are sensitive enough by themselves though, you can test that by shorting the input and seeing if the noise floor changes. A dip in the noise floor at 2.45GHz should be clearly visible and if there is none the analyzer can hardly be at fault here. I still think you might have been tripped up by your or someone else's WLAN. Or by some other device that transmits around 2.45GHz (there are a lot).
That's nice. In Germany we always had to go through a distributor in the
90's. But they were quite competent. Might still be that way.
It's essentially just a filter with varicaps in there. I always liked those from the good old Philips Gloeilampenfabrieken ... ahm ... NXP :-)
Muxed individual filters work as well but not so great if this is to be continually sweeped.
A few dB? That's a bit much, almost sounds like you have a resonance in some part.
-- Regards, Joerg http://www.analogconsultants.com/
Oh, on your next iteration, consider edge-launch SMAs. We get a really nice part from ShiningStar for $1.75, and it makes a beautiful electrical transition if you get the stackup right.
-- John Larkin Highland Technology, Inc jlarkin at highlandtechnology dot com http://www.highlandtechnology.com Precision electronic instrumentation Picosecond-resolution Digital Delay and Pulse generators Custom laser controllers Photonics and fiberoptic TTL data links VME thermocouple, LVDT, synchro acquisition and simulation
Of course it is. It's the Race Freak effect...
-- Thanks, Fred.
Wild guess from the peanut gallery; so far it sounds like a layout based problem.
?-)
Am i seeing through hole parts used in the GHz region?
?-/
to
aFR4
extra
use
another
somewhere.
thinking
simple
part:
'S'
Zo=3D3D50
Incwww.highlandtechnology.com=A0 jlarkin at highlandtechnology dot com
Did you scale carefully for the nominal footprint size of an SMA = connector so visible in he layout? ?-)
Hmmm. Up to gigabit Ethernet i have to disagree. The 8P8C modular connector (mistakenly called RJ-45, YCLIU) still works just fine electrically and is adequate mechanically. =20 At 10Gb/s (Catagory 6) it gets ugly. Nothing i know if is any good, = which may be mostly my ignorance.
?-)
ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.