#61: Basics of RF Samplers, Sampling-Tees, RF-taps, etc.
Вставка
- Опубліковано 29 вер 2012
- This video discusses the basics of what RF Samplers are, why/when they are used, and the design variations available. Samplers are used most often to "pick-off" a small sample of an RF signal (usually from a transmitter) to be measured with a scope, spectrum analyzer, deviation meter, etc.
- Наука та технологія
Nice Tutorial. I've been a technician in the two-way radio industry since 1972 and am still carrying on. My first boss, and mentor, came by the same type of sampler at a hamfest many years ago. This was in the 70s and our primary use of his adjustable sampler was to check desensitization in repeaters. It was great having his tap and using my Singer FM10 to ensure the customer that they had a "real" (conventional) repeater system. After 6 years I left that job when a Motorola shop came looking for me. They had a lot of old equipment but at least one "state of the art" (1972) service monitor (R-1200?). Once again I was the repeater guy. I had learned to tune duplexers and had come up with some hybrid ways of testing them. Back to the sampler. The owner of the Motorola shop didn't know what a sampler was. So, I made my own by taking a UHF "T" connector with two female and one male connector. The male connector became the sampling port after I removed the center pin and then screwed a UHF female barrel connector. After four more years I left the Motorola shop to start my own business. Two-way radio was only supposed to be a "temporary" job after my return from Vietnam. Life happened and since I had a following and ten years experience I set up my own company. Today that company has been around for 36 years and I'm still the prez. These days I find either the Bird 43 or the Telewave watt meter with sampling port to supply about 40 db of attenuation at the sampling port, much more than enough to protect monitors and signal generators. Once I activate a repeater with a signal generator/with CTCSS if necessary and then activate the repeater's transmitter any sign of desense is easily noticed and then can be tracked down. I haven't had an adjustable sampler in years. All I've had, and used, is my home made sampler. I haven't had to buy a watt meter in years but all the units in my company have sampling ports. It would be nice to be able to adjust the attenuation but in my situation it just isn't necessary.
You are an RF GOD ! Seriously.... I am sitting here trying to regain decades worth of info that kinda got lost post-stroke. I was able to get my spectrum analyzer up and running for the first time in years and just wanted to say THANKS !
I just wanted to share with you how much I enjoy your videos Alan! I'm a swedish meteorology grad-student so I am interested in electronics as a hobby. I've constructed some of your experiments and circuits myself. Thanks for sharing your knowledge!
Kind regards from a future HAM
Thank you - so glad to hear that my videos have inspired you to build and learn!
I've been looking for a video on this topic, so the information was very useful. Thanks for explaining so clearly. 👍
Excellent, as usual, and extremely useful. Thank you Alan!
Excellent, very useful tips as always.
Hi Alan
Another great video.Thanks for your time and knowledge.
Regards
Francisco
This is exactly the video i needed. Thanks so much! You're the best. Thanks sir
This is superb, connecting DUTs always seems to be a bit of a shady area when going thru the equipment manuals - i have a RSA3408B on the way and am very much looking forward to having a play, but even this has very little info in the manual on the actual safe DUT connections, and parts to use to connect it, and has the old DC 5v max warning on the front
now i feel a little more comfortable with whats good to have lying around the machine ready for action!
Keep up the good work!!!
great video man, i need something like that to cut down noise on an automotive oscilloscope when doing high current amp draws from the starter when doing a relative compression test
Hi there you can make an RF sampler by getting a BNC T Piece and breaking out the centre pin, that gives you a lot of attenuation, its in the Heathkit manual...from Fred in UK.
great video.
built one of those setups years ago and still use it.
One thing that occurred to me while watching was that you didn't talk much at all about trying to make actual measurements based on the sample.
I wonder if an idea for another video might be to show what can and can't be accurately measured using the homebrew samplers. I think it would be cool to see you show how one piece of test equipment can be used to check the performance of another. thanks for all the great content so far!
Very Interesting! Thank you!
Late night.
Thanks for the video
Thanks for spreading the fever!
Great demo
I am going to make the one with the tee and the slide with bnc connector wondering whats inside the slide part just center conductor going from disk to connector or a resistor does it have any resistance or maybe a cap inside ?
thanks
Great video as always - although I'm missing inductive taps..
Hello Alan. Your videos are the cream of the crop. Crazy question; can I carefully plug an electric guitar into the mike jack of a tranciever and see similar results on the scope?
Great info, thank you.
73
Carl
w2aew, I always love your videos and learn so much. I recently bought and repaired a Tek 5S14N sampler, PG506 Calibration Generator and TG501 Time Mark. In the process I realized I have no other equipment that generates 1+GHz signals. Can you recommend any old Tek generators for 1GHz and greater signals? My budget is low and on the fix-it end. It doesn't have to be Tek unit, but I'm partial. Thanks
Interesting that a simple voltage divider influences the signal a lot in some negative way, while bringing some "metal rod" close to the hot wire seems not to do it that much.
@w2aew Thanks, another great video. Very well explained making it very informative and good learning experience. I'm interested in learning more about antenna theory and didn't find a video, perhaps future one? Can a coil with open end be tuned as antenna for an HF frequency? I'm very interested in using a tuned coil as antenna for compactness. I really appreciate your efforts and would truly appreciate help with this, I've also tried to find a good book about tuning coils for this application.
instablaster.
this may seem a simple question, but to measure upto 1kw on my Bird 43, what would the values for the R's and the C have to be? tthanks
that Bird sampler have nominal attenuation 40db ,and plus 20 db with regulation ,so you can monitor sampled signal from very low to over 5kw of nominal signal. They working great,but they are expensive.
Very informative video Alan.... I have Bird attenuator models 4273 -020 & 4275-020 both adjustable. How would I go about adjusting either one to give me an attenuation of 40db down from the main power level ( ie 100 watts ). Just got a spectrum analyzer and want to keep it happy. Keep the videos coming...
The coupling of these samplers is frequency dependent, so can't adjust them to be a fixed 40dB. In fact, at the lower operating frequencies you can't 40dB coupling, it will have more attenuation. Here is the manual:
www.birdrf.com/~/media/Bird/Files/PDF/Products/manuals/921-4273_75S.ashx
Would there be a difference in using a tee in your signal line to an attenuator that then goes to your scope? 50-ohm attenuator, scope input 10M. Same thing, except for the variable control? So I guess the RF samplers are rated for attenuation factor of the sample as well as the power it can pass? Thanks Alan!
Those samplers are cool but I prefer using a (big) good quality attenuator as it's theoretically aperiodic and linear. With a capacitive sampler you have more coupling on the higher frequencies and when you try to test the bandwidth of an amplifier or compare harmonics amplitude to F0, the results are distorted : +8dB from 28 to 432 on the -40dB sampler port of my Telewave 43A for instance.
Yes, that's always an excellent choice, if you have one available. A decent broadband directional coupler will also provide good flatness in response.
How would you verify the attenuation of the air gap before connecting your scope?
For instance; when using the 4431, how would you know the power out of the RF sample port? Would you need another bird? Or, would you max attenuate before connection and hope for the best?
You'll get a feel for it for the common devices you work on. Typical attenuation factors at HF frequencies will be 40, 50dB or more, so there's really no problem even up to 1000W or more. At VHF/UHF the attenuation will be a bit less, but still at least 30dB typically at best sensitivity. When in doubt, crank in the most attenuation (least sensitive) and setup the instrument for a high level signal and start there. Never had a problem for anything I've worked on up to 1500W. For the 4431, there is a table in the manual that indicates the attenuation level of the sampling port.
I have build restive dividers that are flat +- 0.5db from 1-1200Mhz.It can be done. but is is a pain to compensate.
You're probably tired of reading my thesis on samplers. :-) Another great use for a sampler is determining "effective sensitivity" of a receiver.
Hi! Can you expand on that? thanks!
Thank you for your great videos!!!!
Trying to implement a simple RF sampler just putting an isolated electric wire parallel to and very close to the main line within my SWR meter. Testing it with my HP8640 @14.500MHz @+20dBm and injecting the sample into my 8558 SA. Very good! I get a range of 20 to 50db of attenuation just moving the wire toward or away from the main line. So far so good. But if I sweep the entire HF (1-30MHz) keeping the wire at a fixed position (let's say to get -40db), I get weird response going from 55db of attenuation @low freqs to less than 30db @higher freqs.
Is there any trick to flat the frequency response of such a sampler?
Simple samplers like this that use capacitive or inductive coupling are, by their very nature, frequency dependent. You could design a filter circuit to flatten it, or use a different coupler design.
Yep! In the mean time I discovered that"RF samplers are for pure qualitative measurements; in case you need (as I need) quantitative measurements the way to go is by calibrated fixed attenuators"... isn't it?
To calibrate my kenwood ts-430s power amp I probably need a 40db/100W fixed atten... (..and 150$ ..).
Thank you Alan!!!!
Yes, a calibrated attenuator or directional coupler is the most common way to do quantitative measurements - or simply develop a cal curve for your passive tap...
Great timing. I spent a few hours yesterday testing this very (w2aew)RF sampler I built to see if I could measure power accurately. I found the same issue, although it does seem repeatable. So maybe I could use a table, as does my old Oskerblock-200 SWR and Power meter. Thanks for confirming.
One question: Should the BNC cable enter the scope through a T fitting with a 50 ohm termination on one leg of the T?
For an uncalibrated sampler, the 50 ohm termination isn't really important, but certainly can be used.
w2aew
Thank you.
Could you not blow the spectrum analyzer by using attenuator between it and transmitter?
You could use an attenuator - but high power attenuators are much more expensive than a high power dummy load and RF tap.
Where can the small metal project boxes you show in the video be purchased?
+What's Papa doing now? Here are a couple of examples...
www.digikey.com/catalog/en/partgroup/boxes-with-connectors/3204
www.amazon.com/gp/product/B0107WUWJE
Thank you for these links. I was looking for these also.
I bought an inexpensive 30w / 300w analog watt meter---- and I have no idea if I'm using it correctly. I use a 50r / 100w thick film RF resistor as my dummy load for several of my 10w -100w transmitter projects and it gets quite warm. I decouple the scope probe by just looping it loosely around the dummy load without metal to metal contact. The scope and the dummy load seem to tell me that I'm getting a lot of power... but my new watt meter tells me otherwise. I've tried it with my dummy load and with a GP antenna, but it reads only a small fraction of what I would expect. Is this due to insertion loss, or a poorly designed antenna, or is it just a cheap meter?
Of course the dummy load resistor is going to get warm - the same reason that a 100W light bulb gets warm - it's dissipating 100W of power! Even at 10W, it will get warm, same as a little 7.5W night light bulb gets warm. Perfectly normal. Regarding your meter - it is difficult to say what it happening with it. Is it rated for the frequency that you're measuring? Is it an in-line meter, and are the connectors in good shape? Does the transmitter report a good SWR when the meter is inline?
@@w2aew Sorry for my poorly worded question. Yes, I expect the load to get warm and it does, but my watt meter does not seem to register the same level of output. The meter is rated for the the frequency & power level I'm experimenting with. I am planning to purchase a Bird meter in the future, but they are a bit pricey.
resistors and cap value in MF. OHMS AND WATTS ?
Alan - I’ve seen a lot lately on resistive taps built into DL’s (pi attenuator) or as a separate box, and I’ve also noted the non-directional inductive coupler. They seem to do the same thing - sample RF. Is there any advantage/disadvantage to using one or the other ? Some are adjustable with a pot. I want to be sure I’m not going to blow the front of my S/A ! 73, Jamie WB4YDL
The advantage of a resistive tap is that they are far less frequency dependent. An inductive or capacitive tap will couple more energy to the tap as the frequency goes up. However, it is very easy to adjust the coupling with these non-contact inductive or capacitive taps by just varying coupling distance. To be safe, always begin a test with the SA set to the highest reference level or even use an external attenuator (until you are confident that you don't need it), and start with lower transmit power if possible.
If I want to measure the power output of a HF radio, I need to know how much the Sampler is attenuating. Am I need to find the gain / loss for each frequency that I will be measuring? ( I´m using the formula dB = 20 log (Vout / Vin) and PAVG = (Vout * 0.707) ² / R (50 ohm load) )
Yes, you'll have to ensure that you know the attenuation factor for the sampler at the frequencies you are interested in. the dB formula is correct. for PAVG, your formula is correct if Vout=Vpeak. If you measure Vp-p, then you have to divide by 2, then mulitply by .707, then square, then divide by 50.
***** Thank you! I´m measuring Vpp (peak-to-peak) in Scope. I´m using a 50ohm feed-through connected at sampler out and the scope input probe (10:1, BNC mini adapter - my osciloscope is an Agilent 2012A, has only 1M impedance input). Is it the correct way to measuring Power and Monitoring the frequency? Its first time that I´m using Osciloscope and Frequency counter using samplers and RF equipments. Thank you for your attention. :D
i dont understand the difference between a coupler and a sampler..... the samplers i work with have 4 pins on them , what would those pins be for?
A sampler is usually some kind of "pick-off", where a small portion of the signal is sent to the sample port. A "coupler" usually implies a "directional coupler", which have two coupled ports - one for the forward direction and one for the reverse/reflected direction.
Hey brother - I need help, 🙏💯 I got my unit called
" ENI PS 600" I Don't know what software to use?
Where do you get those little metal BNC cases?
+Alyx BioHaz Pomona is a popular manufacturer of them:
www.digikey.com/catalog/en/partgroup/boxes-with-connectors/3204
or bottom of this page:
www.surplussales.com/Cab-RacksHardware/encl_2.html
www.electronicsurplus.com/itt-pomona-electronics-2390-inline-adapter-box-bnc-female-to-bnc-female
They are not cheap ... but thank you!
+Alyx BioHaz I don't know why the Pomona connector boxes are so expensive, but I found an alternative from China. It looks very similar to the enclosure that ThorLabs uses for their lab fixture kits (www.thorlabs.com/newgrouppage9.cfm?objectgroup_id=8681, which are very cool, btw), but costs just a few bucks sans PCB. Ordering directly from China, you can get them for $3, but if you don't want to wait a few weeks, they are available on Amazon for a few bucks more:
www.amazon.com/gp/product/B0107WUWJE
They work great. The aluminum is soft enough to punch through, and it holds a PCB very snug inside.
Hi Alan, I have seen a HF radio tech on UA-cam (he is actually an Aviation tech "hobby_radio_nerd "..anyways he made his RF sampler in a rectangle dicast box, he had 2 BNC sample ports on either side , ie a total of 4 RF sample ports , he hasn't shown the internal workings / circuit. Is this actually achievable without isolation ? It seems he only uses for 11mteters/27mhz.
Thanks in advance .
You can have multiple samples/taps, using a number of techniques. Simple sniffer wire stubs will work fine, for example.
@@w2aew thank you .
Could you build me a sample like this so I could get a trapezoid pattern.
For a fee
U can't I just take the wires from the gauge and hook them up to the o scope. The slug is doing the same thing?
No, the slug is detecting the amplitude of the RF signal, so it wouldn't show you actual RF signal - only the envelope.
@@w2aew oh ok. I'm green I'm trying to learn as much as I can. Thanks for the info. I appreciate it!
Dear W2AEW ,
does this 50DB 100W attenuator do the same job ?
www.ebay.com/itm/100W-N-Attenuator-50db-male-to-female-DC-3GHZ-50ohm-RF/272899586791
because all the RF samplers I find online are quite expensive 300$ plus ..
vs this thing that is 70 $ ,
Will this suffice to connect a transmitter direct to a spectrum analyzer ?
Thank you in advance
Alex
P.S how do you calculate how much RF power will be at the other side of the attenuator given that you know the input power ? so that you can be with in the strict limits of the Spectrum maximum input dbm ?