How to Make a 75 to 300 Ohm Matching Transformer
75 to 300 ohm Balun - why can't we use .....
| Author | Message |
|---|---|
| Post subject: 75 to 300 ohm Balun - why can't we use ..... | |
| Member Joined: Jul Wed 22, 2009 8:32 am Posts: 3668 | Edit: The SG165 signal gen can produce: The SG165 comes with a 39G43 matching pad, that takes the 75 Question: So why can't we use these cheap 75/300 Baluns? To inject a signal gen 50 or 75 ohm (unbal) to 300 ohm (bal) However what about those cheap TV/FM Baluns. If working Question: Help me analyze this circuit for SG165's matching C24 (20uF) cap w/ the R401 resistor (120 ohm) is 60 hz filter? Would/could you not use a transformer to change impedance? _________________ Tube: AM/FM Zeniths, RCA, TrueTone table tops; Transistor: Kaito KA1103, TenTec RX320D, Pioneer SX780 Last edited by gmcjetpilot on Jan Sun 16, 2011 8:27 pm, edited 4 times in total.
|
| Top | |
 | |
Posted: Jan Wed 12, 2011 8:13 pm
| Chris108 | Post subject: |
| Joined: Jun Fri 19, 2009 6:34 pm Posts: 11161 Location: Long Island NY | You certainly can use a 75-Ohm to 300-Ohm TV matching transformer to match a 50-Ohm signal generator output to a 300-Ohm balanced input. Most of those little transformers will work fine anywhere from about 5 MHz to 800 MHz. As for the difference between the 50-Ohm signal generator output and the 75-Ohm transformer, it's almost totally irrelevant in a situation like this. The power lost in the mismatch isn't going to be missed at the antenna terminals of an FM tuner that is supposed to be fully quieted at 50 to 100 uV, and the SG-165 does not have a precision output level control or attenuator to begin with! The schematic of the matching network appears to contain a misprint. Unless they were expecting to pass audio through this gadget, the 20 uF capacitor is probably a typo. Chances are it is supposed to be 20 pF. It's a variation on what is known as a "minimum loss pad," which is a way of matching two different impedances using a voltage divider. The basic concept is that all of the "stuff" to the right of the 82-Ohm resistor (including the 300-Ohm receiver), forms a circuit that is effectively in parallel with the resistor. The parallel impedance "looks like" 50 Ohms to the signal generator. Going the other way, the combined impedance (including the generator) "looks like" 300 Ohms to the receiver. The 0.001-uF cap passes the lower frequencies, like AM broadcast, and the 20 pF--assuming that is the correct value--passes the higher FM frequencies. _________________ "Hell, there are no rules here--we're trying to accomplish something!" Thomas A. Edison |
| Top | |
| | |
| radiotechnician | Post subject: |
| Joined: Sep Thu 23, 2010 6:37 am Posts: 12437 Location: Powell River BC Canada | I think the 20 uFd capacitor was to inject a signal into a _________________ de VE7ASO VE7ZSO Amateur Radio Literacy Club. May we help you read better. Steve Dow ve7aso@rac.ca |
| Top | |
| | |
| gmcjetpilot | Post subject: |
| Joined: Jul Wed 22, 2009 8:32 am Posts: 3668 | Chris108 wrote: You certainly can use a 75-Ohm to 300-Ohm TV matching transformer to match a 50-Ohm signal generator output to a 300-Ohm balanced input. Most of those little transformers will work fine anywhere from about 5 MHz to 800 MHz. Not so good for MW work but OK for SW, HF, VHF, UHF. Quote: the SG-165 does not have a precision output level control or attenuator to begin with! It is precision to me! Quote: The schematic of the matching network appears to contain a misprint. Unless they were expecting to pass audio through this gadget, the 20 uF capacitor is probably a typo. Chances are it is supposed to be 20 pF. Thanks! I will check that out. The matching pad module is a good size, not light, so I wounder what is in there. The pop on end caps for the matching pad module moved and twisted, so I glued them down. I don't want to pry them off to look. I got the el cheapo- LCR capacitance meter out and the VTVM for grins. (1) Red to F-input 4.94uF (1 Mohm) It appears the 20 uF cap is 4.92 uF? I don't have high confidence in my cheap capacitance meter. Not sure how the resistors are affecting the reading. Quote: The basic concept is that all of the "stuff" to the right of the 82-Ohm resistor (including the 300-Ohm receiver), forms a circuit that is effectively in parallel with the resistor. The parallel impedance "looks like" 50 Ohms to the signal generator. Going the other way, the combined impedance (including the generator) "looks like" 300 Ohms to the receiver. The 0.001-uF cap passes the lower frequencies, like AM broadcast, and the 20 pF--assuming that is the correct value--passes the higher FM frequencies. _________________ Tube: AM/FM Zeniths, RCA, TrueTone table tops; Transistor: Kaito KA1103, TenTec RX320D, Pioneer SX780 |
| Top | |
| | |
Ha ha, I know, but it us kind of a calibrated output. It actually is calibrated in micro-volts with course and fine control. It works pretty well for eye ball work. When I calibrated it per the instructions (FM and AM are two different steps) it was close.| Peter Bertini | Post subject: | ||
Joined: Jan Thu 01, 1970 1:00 am Posts: 22523 Location: Somers, CT | So why can't we use these cheap 75/300 Baluns? You can. But you need to understand why the probe was made this way. One mistake you made was labelling the generator as being 50 ohms output, and assuming that the probe provides a 75 ohm match. The presence of the 120 ohm resistor in series tells us the output impedence will be in excess of 120 ohms. You would go direct from the 165 RF output for 75 ohms. To inject a signal gen 50 or 75 ohm (unbal) to 300 ohm (bal) Yes, but unlike the transformer, this is a lossy matching network. The resistors lose power, the transformer doesn't. This is the beauty of your HP8640B in having both dBm and voltage scales. The dBm measurements are always correct when using matching transformers. But, if you are working in micrvolts, and using a 50 ohm to 300 ohm transformer, you'd have to multiply the output voltage by the turns ratio to determine the RF voltage at the load. However what about those cheap TV/FM Baluns. If working Most good CATV TV baluns will go down to 5MHz. Unfortunately many of the cheap ones flooding the market don't even work well for their intended purpose. Help me analyze this circuit for SG165's matching C24 (20uF) cap w/ the R401 resistor (120 ohm) is 60 hz filter? Simple. The probe is also used "single ended" to inject signals in IF stages, and you want to provide some DC blocking measures so a tech doesn't end up backfeeding 250 VDC to an expensive 75 ohm attenuator. Look at the other applications in the manual. R402 is simply a discharge path for the caps when the probe is removed from the circuit. I also doubt that C24 is that high a value. 0.2 or 2.0uF might be more reasonable. The engineer probably thought the .001 was more effective at coupling at FM frequencies. He was out to lunch. One cap would have sufficed over the operating range of the 165. Those caps only provide DC blocking, not impedance matching. The 82 ohm resistor is part of the termination for the 75 ohm output of the generator with the probe is used. The actual impedance seen by the generator is this resistor, the two 120 ohm resistors, and whatever impedance is seen at the antenna jack on the receiver. By now it should be obvious that the uV readings on the calibrator need to be taken with a grain of salt. Use a balun, and you are stepping up the voltage. Use that probe, and the voltage is reduced. This why engineers deal in dBm power measurements for measuring stage gain or receiver specs instead of uV, which are generally meaningless when we start adding impedance matching devices. If we have two FM receivers, one 75 ohm input and the other 300 ohm input, and they both require 20uV for 20dB quieting, does that mean they have the same sensitivity? Would/could you not use a transformer to change impedance? Ah, but the probe is used for other purposes besides matching FM receivers. You could use a balun for FM, but see the caveats above. | ||
| Top | | ||
| | |||
| gmcjetpilot | Post subject: |
| Joined: Jul Wed 22, 2009 8:32 am Posts: 3668 | Peter Bertini wrote: So why can't we use these cheap 75/300 Baluns? You can. But you need to understand why the probe was made this way. One mistake you made was labelling the generator as being 50 ohms output, and assuming that the probe provides a 75 ohm match. The presence of the 120 ohm resistor in series tells us the output impedance will be in excess of 120 ohms. You would go direct from the 165 RF output for 75 ohms. Look at the picture, I labeled it 75 ohms to 300 ohms. I mention 50 ohms only because that is the another common impedance of signal gens. I always thought the SG165 was 75 ohms. I can't actually measure that can I? I don't have an "impedance meter". Nothing seem very precision about this "matching pad" in that it seems like it would work with 50 ohms as well. 20uF and 120 ohm is a RC of 66 Hz. I was thinking it was a filter. My measurement of 5 uF does not jive with 20 uF. For a blocking capacitor I use typically 0.05 uF. I am assuming my 5 uF is off, because it is in series with the 120 resistor. Plus it has that other capacitor in parallel with the 1 Meg resistor. The 5 uF does not makes sense either. I can imagine it needing to be more than 0.1 uF for a blocking capacitor. The 0.001 uF and 120 ohm is 1.36 Khz RC wise. I'm probably using Quote: Yes, but unlike the transformer, this is a lossy matching network. The resistors lose power, the transformer doesn't. This is the beauty of your HP8640B in having both dBm and voltage scales. The dBm measurements are always correct when using matching transformers. But, if you are working in micrvolts, and using a 50 ohm to 300 ohm transformer, you'd have to multiply the output voltage by the turns ratio to determine the RF voltage at the load. microvolts only means something when you know the impedance. However does dB (power) not depend on impedance? dBm means milliWatt I assume. Quote: Most good CATV TV baluns will go down to 5MHz. Unfortunately many of the cheap ones flooding the market don't even work well for their intended purpose. I took one apart to look at, nice binocular ferrite core with magnetic wire coils. It looked well made. But it's an older one I got with some electronics. Quote: Simple. The probe is also used "single ended" to inject signals in IF stages, and you want to provide some DC blocking measures so a tech doesn't end up backfeeding 250 VDC to an expensive 75 ohm attenuator. That makes sense. The attenuator does not have a problem it is that FET transistor at the end of that line that does not like monkey business. Quote: Look at the other applications in the manual. R402 is simply a discharge path for the caps when the probe is removed from the circuit. I also doubt that C24 is that high a value. 0.2 or 2.0uF might be more reasonable. The engineer probably thought the .001 was more effective at coupling at FM frequencies. He was out to lunch. One cap would have sufficed over the operating range of the 165. Those caps only provide DC blocking, not impedance matching. Oh so the one Meg is a discharge. Interesting. The value is a mystery. The probe can be taken apart but I glued my together with some silicone to keep the end caps from twisting and coming off. So basically a blocking or coupling capacitors. Quote: What is the R403 (82 ohm) resistor for? So this is pretty approximate stuff here. From the manuals for aligning FM radios they say just use two 120 ohm resistor (assuming a 50 ohm signal gen I am sure). I am thinking a better precision matching pad could (should) be made, like a L-pad. Brian Beezley has an interesting write up on 50 to 75 ohm matching. I'm going to make one of those. (The HP 3336C has both 50 and 75 ohm out with precision dB power.) http://ham-radio.com/k6sti/match.htm He makes a point about mismatch: Quote: By now it should be obvious that the uV readings on the calibrator need to be taken with a grain of salt. Use a balun, and you are stepping up the voltage. Use that probe, and the voltage is reduced. This why engineers deal in dBm power measurements for measuring stage gain or receiver specs instead of uV, which are generally meaningless when we start adding impedance matching devices. If we have two FM receivers, one 75 ohm input and the other 300 ohm input, and they both require 20uV for 20dB quieting, does that mean they have the same sensitivity? I got you, it's very load dependant. I personally don't use it at all. Most of the time I am connecting direct to the radios grids or just inductively coupling. For grid signal injection I just use my own coaxial to clip test cable, way less bulky than this matching pad. I use my own blocking capacitor and that is it. For signal into the antenna terminals I might use the matching pad or just two 120 ohm resistors. For AM I inductively couple with my coaxial test cable and test lead just draped near by the loop antenna. _________________ Tube: AM/FM Zeniths, RCA, TrueTone table tops; Transistor: Kaito KA1103, TenTec RX320D, Pioneer SX780 |
| Top | |
| | |
| Peter Bertini | Post subject: | ||
Joined: Jan Thu 01, 1970 1:00 am Posts: 22523 Location: Somers, CT | I always thought the SG165 was 75 ohms. I can't actually measure that can I? I don't have an "impedance meter". Nothing seem very precision about this "matching pad" in that it seems like it would work with 50 ohms as well. Yes, you can measure it. Modern signal generators look like a source resistance. If the generator is 50 ohms, the output acts like a 50 ohm resistor in series with the output. If there are no DC blocking or protection devices in line, an ohmmeter will read the source impedance. Another way is to read the open circuit voltage on the generator output. When you have a load resistance equal to the generator impedance, the output voltage will be exactly one half the unloaded output. The attenuator is calibrated when the generator is correctly loaded. No load, open circuit, the generator output voltage will be twice what is expected. If you have a step attenuator you will see a resistance that is equal to the impedance of the attenuator on either port. The other port must be terminated. microvolts only means something when you know the impedance. However does dB (power) not depend on impedance? dBm means milliWatt I assume. Once you know the power, that is all that is important. Radio and TV stations are licensed for power, not antenna impedances or antenna voltages--unless you want to get into uV per meter readings field strength readings If we have two FM receivers, one 75 ohm input and the other 300 ohm input, and they both require 20uV for 20dB quieting, does that mean they have the same sensitivity? I got you, it's very load dependant. Converting the impedances and voltages to dBm power levels would show which RX needs the least amount of signal energy for a given quieting. Another pitfall is it is easy to assume that the output level at the generator has any validity when the load impedance is wrong, or when we are transforming impedances or voltages between the generator and device being tested. But I'm rambling away, for whatever it is worth. Probably not much....but knowing how things works can be helpful. | ||
| Top | | ||
| | |||
| gmcjetpilot | Post subject: |
| Joined: Jul Wed 22, 2009 8:32 am Posts: 3668 | Peter Bertini wrote: But I'm rambling away, for whatever it is worth. Probably not much....but knowing how things works can be helpful. Oh no Peter, I appreciate the info, I'm getting it. I have always been confused about this. I think my brain made a few new neuron connections, which isn't easy in my hard head. I'm going to play around with it, to "actualize" this new info. _________________ Tube: AM/FM Zeniths, RCA, TrueTone table tops; Transistor: Kaito KA1103, TenTec RX320D, Pioneer SX780 |
| Top | |
| | |
| majoco | Post subject: | ||
Joined: Mar Mon 17, 2008 5:05 am Posts: 5673 Location: Ashhurst, New Zealand | If you go back to the basic need for a 300ohm input, it was to match the incoming 300ohm twinlead from a 300ohm folded dipole. There was usually a mechanical grounding at the centre of the folded dipole, and also at the centre tap of the tuner's 300ohm input transformer. Now given that the impedance goes up as the square of the turns, then from one end of the input transformer to the centre tap there is only a 75ohm impedance, which is why your matching pad only puts signal into one leg of the input, the other is effectively at signal ground. IMHO a 75 to 300ohm balun would do a better job as you present the input transformer with a correct anti-phased input signal. When it was found that coaxial cable got rid of the interfering signals picked up on 300ohm twinlead and an unfolded (?) dipole matched the 75ohm coax nicely, then you joined the centre conductor to either of the old twinlead screws and grounded the screen. Hey presto, a 75 ohm match! Or am I teaching Granny to suck eggs here....... _________________ Cheers - Martin ZL2MC | ||
| Top | | ||
| | |||
| Chris108 | Post subject: |
| Joined: Jun Fri 19, 2009 6:34 pm Posts: 11161 Location: Long Island NY | Quote: This why engineers deal in dBm power measurements for measuring stage gain or receiver specs instead of uV, which are generally meaningless when we start adding impedance matching devices. If we have two FM receivers, one 75 ohm input and the other 300 ohm input, and they both require 20uV for 20dB quieting, does that mean they have the same sensitivity? There's a lot of confusion about dB's, uV's and other measurements of signal levels. It's not as easy as figuring out which size battery fits your flashlight, but here's an attempt to shed a little light on the subject. At last count, there are something around 200 different kinds of decibel notations in use, and not all of them have anything to do with electronics. A decibel is just a convenient kind of mathematical notation which uses logarithms for comparing two numbers that would be miles apart if they were on a linear scale. Since it is a logarithmic notation, other advantages accrue when it comes time to multiply and divide such numbers. The thing is, a decibel starts out as a ratio of what you are measuring to a reference value. This way, if the measured signal is equal to the reference, the ratio is 1, and the log is 0. For example, 0 dBm = 1 milliwatt across 50 ohms in RF, or 1 milliwatt across 600 ohms in audio. The letter(s) that come after the 'dB' tell you what the reference is. Unless it is obvious from the context of the discussion, this needs to be stated if you expect anybody to know what you are talking about. Decibel numbers without a reference are just blather! Note that the two definitions of dBm given above also mention impedances. This is because they are power levels. In the more familiar world of household power wiring, the AC power system has a very low impedance from the user's point of view. The goal is to maintain a more-or-less constant voltage at the outlets, regardless of load. It doesn't matter if you've got one tiny night light turned on, or every light and appliance in the house. The voltage should remain within close limits. Now in electronics--particularly where small signals are involved--it's neither necessary nor practical to design circuits that way. Such circuits are designed to deliver a rated amount of power into a rated load impedance. If the load impedance is either higher or lower than expected, less than the rated power will flow. So when dealing with decibel power notations, the impedance has to be stated or implied if the measurement is to have any validity. Of course it is also possible to express things besides power in decibel notation. Specifically, dBuV (microvolts), dBmV (millivolts), and dBV (volts). Being the ratio of a measured voltage to a reference voltage, these measurements do not imply or require any particular impedance, exactly like uV, mV, or V. This can be of great value in situations where the impedance is unknown or difficult to determine. The reason why the sensitivity of FM receivers in particular, and most others besides, is typically given in uV (or sometimes dBuV) is because it gives you a way of comparing the electronic sensitivity of the RF stages in receivers. Of course, the assumption is that you are feeding the receiver with a signal generator that either matches the impedance of the receiver's input terminals, or you've got a proper matching device between the generator and the receiver. For this reason, it is a common fallacy to think that a receiver which shows a lower uV number for a particular result on a 75-ohm input is somehow more sensitive than one which requires more uV's of signal on a 300-ohm input to get the same result. As Marty points out, there's a 4:1 difference in impedance between a 300-ohm system and a 75-ohm one, so for identical antennas the signal voltage will be double on the 300-ohm input and the current will be halved. Same amount of signal power in either case. Lesson here is that when comparing sensitivities between receivers, you have to do it at the same impedances! _________________ "Hell, there are no rules here--we're trying to accomplish something!" Thomas A. Edison |
| Top | |
| | |
| Peter Bertini | Post subject: | ||
Joined: Jan Thu 01, 1970 1:00 am Posts: 22523 Location: Somers, CT | Chris108 wrote: This why engineers deal in dBm power measurements for measuring stage gain or receiver specs instead of uV, which are generally meaningless when we start adding impedance matching devices. If we have two FM receivers, one 75 ohm input and the other 300 ohm input, and they both require 20uV for 20dB quieting, does that mean they have the same sensitivity? Of course, the assumption is that you are feeding the receiver with a signal generator that either matches the impedance of the receiver's input terminals, or you've got a proper matching device between the generator and the receiver. Assumptions are what get us into trouble. There was no mention of a proper matching device. The information could have been taken directly from each receiver's specification sheets! It was a trick question to give one pause to think about the answer. For this reason, it is a common fallacy to think that a receiver which shows a lower uV number for a particular result on a 75-ohm input is somehow more sensitive than one which requires more uV's of signal on a 300-ohm input to get the same result. It is a common fallacy for many beginners to believe a balun and loss matching pad will deliver the same power, or voltage, to a load because impedances are matched. As Marty points out, there's a 4:1 difference in impedance between a 300-ohm system and a 75-ohm one, so for identical antennas the signal voltage will be double on the 300-ohm input and the current will be halved. Same amount of signal power in either case. Lesson here is that when comparing sensitivities between receivers, you have to do it at the same impedances! I did state that we can transformer voltages and impedance, but for a lossless network the power is constant at both ports. I would have to reread my earlier comments to see if what I meant to say could have been taken incorrectly, that is a fault of this sort of medium where self editing is often lacking due to writing in haste. Signal power (dBm) was the basis for my arguments. You are completely missing the points I was trying to make. Many beginners do not understand that the an impedance matching device can deliver a different voltage to the input of a receiver than is what shown on the generator's output level. As I mentioned earlier, power is a constant for lossless matching devices, like baluns. 0 dBm is defined as being 1mW, regardless of impedances or voltages. Impedance does matter, but most RF accessories like signal generators, power meters, fixed pads, attenuators, etc. are designed for specific impedances. Most of us only deal with 50, 75 or 300 ohm devices. Most modern generators use step attenuators and have well defined output impedances; the output is a source resistance and the voltage is load dependent. This increases the chances for errors, because the generator output is only accurate when properly terminated. The matching section used in the SG-175 probe (loss match pad) an impedance matching balun are both proper matching devices. George asked about using a balun instead of the probe supplied with the SG-175. I think it was important that he should understand that why both will provide impedance matching; they would give completely different sensitivity measurements. As an aside, the SG-175 uses a 75-ohm step type attenuator, and is calibrated in uV. Oddly, Sencore's calibration procedure sets the attenuator voltage using an unterminated RF voltmeter. The generator's calibrated output might only be valid with the SG-175 probe is used. Pete Last edited by Peter Bertini on Jan Thu 13, 2011 6:15 pm, edited 1 time in total. | ||
| Top | | ||
| | |||
| Burnt Fingers | Post subject: |
| Joined: Oct Sat 20, 2007 3:36 am Posts: 11481 Location: Southern NH, 03076 | This has been covered many times on here and elsewhere and many fall into believing the statements on one rather prolific self inflated writer on his web page. Pete knows who I mean! Of the over 2 dozen baluns I tested here with a MFJ 259B as a basic tool a few weeks ago only 5 would even make it to 8 MHz without falling apart. The majority wouldn even get below 40MHz before the VSWR and reactance went sky high. These are the small cylindrical versions with an F connector on one end and a small length of 300 Ohm twinlead on the other all from Taiwan or China. I also tested a few real old flat case or rectangular plastic baluns and their performance never made it belo about 35MHz. If you take any of these apart and see the tiny bit of ferrite used as well as very low value caps its easy to see these are strictly for VHF. Its so easy to wind any ratio you want with a FT-50 or FT82-43 or -77 toroid I dont understand the long drawn out constant yapping over such a simple subject Carl |
| Top | |
| | |
| gmcjetpilot | Post subject: |
| Joined: Jul Wed 22, 2009 8:32 am Posts: 3668 | Burnt Fingers wrote: This has been covered many times on here and elsewhere and Its so easy to wind any ratio you want with a FT-50 or FT82-43 or -77 toroid I dont understand the long drawn out constant yapping over such a simple subject Are you cranky today Carl? You say IMPEDANCE IS EASY? I disagree Carl. It's complex http://cp.literature.agilent.com/litweb ... 0-3000.pdf I cut and paste this from the web: Simply stated, it's the obstacle to current flow provided by an The law governing the relationship between DC resistance, E = I * R where E is the impressed voltage in volts across the resistance E = I * Z where, now, E is the impressed voltage magnitude in volts Z = E/I Now, I use the terms like "impedance magnitude" here. The AC |Zm| = sqrt ( R^2 + X^2 ) Math symbology gets complex polar or cartesian math notation, so see this ref for math where R is the resistive portion and X is the reactive portion. Because of the energy storage properties of the reactive portion, omega = arctan (X/R) Math symbology gets complex so see this ref for math where P (more properly, the Greek letter Phi) is the phase Basically, all we need to do is then to put a voltage across the Z = E/I And out pops the impedance, Z. In principle, this is absolutely correct, but in practice, it Imagine putting a voltage of 10 volts across an 8 ohm I = 10/8 = 1.25 amps While 1.25 amps is a convenient current to measure (it's large MEASUREMENT SCHEMES A common assumption is that one needs two meters: one to Z = E/I However, this poses some problems. As mentioned above, it Another method that seems to have escaped many peoples' TEST INSTRUMENTS Very little is required for a properly working impedometer. 1. AC sine wave generator This can either a function generator (usually meaning an There are a lot of new instruments that are acceptable, On the other hand, you can often find used equipment that is One of the all-time best sine generators is the venerable HP 2. AC voltmeter This can either be an analog or digital unit. Ideally, it Unfortunately, the sensitivity requirement eliminates most Thanks for the input. I am easily amused. _________________ Tube: AM/FM Zeniths, RCA, TrueTone table tops; Transistor: Kaito KA1103, TenTec RX320D, Pioneer SX780 Last edited by gmcjetpilot on Jan Thu 13, 2011 8:52 pm, edited 6 times in total. |
| Top | |
| | |
| majoco | Post subject: | ||
Joined: Mar Mon 17, 2008 5:05 am Posts: 5673 Location: Ashhurst, New Zealand | My Dad used to say "Don't believe everything you read in the newspaper" I guess that later on he would have said the same about radio and television. I would say the same thing about the things I read on the internet. Don't forget that anyone can post anything into Wikipedia. ...now where's my large grain of salt.... _________________ Cheers - Martin ZL2MC | ||
| Top | | ||
| | |||
| MarkPalmer | Post subject: | ||
Joined: May Mon 18, 2009 5:55 pm Posts: 3081 Location: Erie, PA | Impedance has been getting like the G word here lately (ground). Say the word and out come the six-shooters at high noon! Very informative thread though -Mark- _________________ KC3PKK | ||
| Top | | ||
| | |||
| gmcjetpilot | Post subject: |
| Joined: Jul Wed 22, 2009 8:32 am Posts: 3668 | MarkPalmer wrote: Impedance has been getting like the G word here lately (ground). Say the word and out come the six-shooters at high noon! Very informative thread though Right now I know that if something has a placard or If you are an old hand this is old news, but in my Impedance gets complicated, at least for me. I don't _________________ Tube: AM/FM Zeniths, RCA, TrueTone table tops; Transistor: Kaito KA1103, TenTec RX320D, Pioneer SX780 |
| Top | |
| | |
| majoco | Post subject: | ||
Joined: Mar Mon 17, 2008 5:05 am Posts: 5673 Location: Ashhurst, New Zealand | It was put to the class 50years ago that for maximum transfer of power from source to load, the load impedance must be the same as the source impedance. Then the class tutor gave us a quick homework task - prove the theory by calculation using simple DC and resistors - don't ignore the power losses in the source. Make the source resistor 100ohms and prove for values in 10ohm steps from 10 to 200 ohms. Kirchoff's Laws will apply! A modern spreadsheet should do this in seconds. _________________ Cheers - Martin ZL2MC | ||
| Top | | ||
| | |||
| Peter Bertini | Post subject: | ||
Joined: Jan Thu 01, 1970 1:00 am Posts: 22523 Location: Somers, CT | Good exercise GMCpilot: don't get caught up complex mpedances for this discussion. Substituting Z for R works for all of the equations for http://www.kpsec.freeuk.com/imped.htm this is nice, easy to understand explanation. Pete | ||
| Top | | ||
| | |||
| Chris108 | Post subject: |
| Joined: Jun Fri 19, 2009 6:34 pm Posts: 11161 Location: Long Island NY | Hi George, The more I think about this, the more I think you'd be better off simply trying to replicate the original Sencore probe circuit. My initial answer, that you could use a cheap TV matching transformer, was a "shot from the hip." While not incorrect, it does require a lot of qualification. First off, it now appears from Burnt Fingers' post that they don't make those transformers like they used to. Either that, or I just hit a run of good ones a few years ago, which was the last time I used them this way. At this point, I wouldn't trust one of those things much below FM frequencies. AM broadcast frequencies would never work out well in any case, since even of the best of those transformers folded and departed the game above 5 MHz anyway. But even if you did find or make a decent transformer, you'd still want to put DC blocking caps on the probe end of it, and you'd have to consider the impedance of the termination presented to the signal generator, and the ratio of the transformer, in order for the output level calibrations to have any meaning. You see, the whole point to impedance matching is to maximize the transfer of power from one device to another. No practical source of power, whether it is AC, DC, or RF, is capable of supplying infinite current or voltage to an arbitrary load. This is because all sources have some internal resistance or impedance. If you connect a load that has a much higher impedance than the source, the voltage probably won't change much, but little current will flow, and little power will be transferred. Make the load much lower in impedance, and a lot of current will flow. This causes a significant portion of the voltage drop to occur across the source impedance instead of the the load, so again little power will be transferred. Only when the source impedance and the load impedance are equal to each other, will the maximum transfer of power occur. You can prove this to yourself by drawing an ideal signal source (i.e. an imaginary one with no internal resistance), with its source resistance in series with it, and with a load resistance connected. Pick some easy number like 100 volts for the signal source, and 50 ohms for the source resistance. Then using Ohm's Law, calculate the current through this simple series circuit and the voltage drop across each resistor if the load is 25 ohms, 50 ohms, and 75 ohms. Now if you calculate the power in the load at each resistance, you'll see that you get the most power into the load when its resistance is equal to the source resistance. That's what matching is all about! Now there are a couple of different ways this can be accomplished. Transformers are one of them. They have the advantage of converting power at one impedance into another with very little loss. This is important in antenna circuits, where the whole purpose of the exercise is to get as much signal as possible from the antenna into the receiver. On the down side, a transformer that is perfectly flat over a broad range of frequencies is not easy to design or make, and the impedance presented by the primary depends very much on the load connected to the secondary. Another way is to set up a voltage divider which has one impedance to match the source, and another impedance to match the load. Both the source and the load "see" the correct impedance, so they are happy. And since non-inductive resistors may be used, a lot of the trouble with frequency response over wide ranges can be avoided. The cost is that significant power is lost in the divider. My guess is that Sencore figured the loss in a matching pad was a reasonable price to pay for simplicity and stability. They simply put enough power into the output stage of the signal generator to accomplish the mission despite the loss. _________________ "Hell, there are no rules here--we're trying to accomplish something!" Thomas A. Edison |
| Top | |
| | |
| gmcjetpilot | Post subject: |
| Joined: Jul Wed 22, 2009 8:32 am Posts: 3668 | Chris108 wrote: Hi George, The more I think about this, the more I think you'd be better off simply trying to replicate the original Sencore probe circuit. You see, the whole point to impedance matching is to maximize the transfer of power from one device to another. No practical source of power, whether it is AC, DC, or RF, is capable of supplying infinite current or voltage to an arbitrary load. This is because all sources have some internal resistance or impedance. Thanks a million Chris for the follow up. As far as TV baluns, I mentioned it to kick off the discussion about the SG165 matching pad. I might experiment, use a signal Gen and scope to try different things for fun. A spectrum analyzer would be nice. I also gather that mismatched impedance can cause other issues besides power loss. May be this is more critical to microwave or something, but for low Freq audio it adds distortion I read. Worse you can over heat the amp (high current) with a real low speaker impedance, eg speakers in parallel. The company Mini Circuits has some nice matching products for low power stuff. Quote: If you connect a load that has a much higher impedance than the source, the voltage probably won't change much, but little current will flow, and little power will be transferred. Make the load much lower in impedance, and a lot of current will flow. This causes a significant portion of the voltage drop to occur across the source impedance instead of the the load, so again little power will be transferred. Only when the source impedance and the load impedance are equal to each other, will the maximum transfer of power occur. Very well said. That now makes sense. I tried to memorize that and would forget, but now that I understand the theory better I get it. Quote: You can prove this to yourself by drawing an ideal signal source (i.e. an imaginary one with no internal resistance), with its source resistance in series with it, and with a load resistance connected. Pick some easy number like 100 volts for the signal source, and 50 ohms for the source resistance. Then using Ohm's Law, calculate the current through this simple series circuit and the voltage drop across each resistor if the load is 25 ohms, 50 ohms, and 75 ohms. Now if you calculate the power in the load at each resistance, you'll see that you get the most power into the load when its resistance is equal to the source resistance. That's what matching is all about! Excellent! Quote: Now there are a couple of different ways this can be accomplished. Transformers are one of them. They have the advantage of converting power at one impedance into another with very little loss. This is important in antenna circuits, where the whole purpose of the exercise is to get as much signal as possible from the antenna into the receiver. On the down side, a transformer that is perfectly flat over a broad range of frequencies is not easy to design or make, and the impedance presented by the primary depends very much on the load connected to the secondary. I have made transformers for antennas, as Carl said, wind some dang wire on a toroid and be done with it... ha ha. They are magical. With a good Balun, ground and a random wire antenna gets big low noise ears, over just a piece of wire. Another way to match impedance is active electronics, matching signal amps. That's over my head, but a broad-band active device would be a cool project. Quote: Another way is to set up a voltage divider which has one impedance to match the source, and another impedance to match the load. Both the source and the load "see" the correct impedance, so they are happy. And since non-inductive resistors may be used, a lot of the trouble with frequency response over wide ranges can be avoided. The cost is that significant power is lost in the divider. This is what Sencore's matching pad is a get the job done cheap, while having a range of 262 Khz to 120 Mhz. If I had a spectrum analyzer I could see what is up with that circuit. Quote: My guess is that Sencore figured the loss in a matching pad was a reasonable price to pay for simplicity and stability. They simply put enough power into the output stage of the signal generator to accomplish the mission despite the loss. That's why I don't use the 39G43 matching pad. First it's big and bulky, very larger for a probe/dongle to clip inside a radio. For signal input to radio, I use a test cable with clips. I add a blocking capacitor for injection into the grid (which I don't think cares about impedance). I use resistor'(s) for input to the antenna terminals, as needed. For inductive coupling a piece of wire is all I use. I recall when I had this question, I did some experiments. I compared the Sencore SG165 Matching Pad (39G43) with my jury rig test cable and coupling schemes. I didn't see any real difference. I should experiment more, but quality or accuracy of the alignment (tone or visual) wasn't different, from what I saw. To add insult to the bulky matching pad, Sencore supplies a detachable coaxial, which is nice, but the coaxial is solid core, stiff as a garden hose, unpleasant to use. As far as having power to spare, the SG-165 FM section is not very powerful, it's a 100 microvolt rated (max) output at 98 Mhz into a load, so good match is a must. At the higher freqs the power drops off even more. _________________ Tube: AM/FM Zeniths, RCA, TrueTone table tops; Transistor: Kaito KA1103, TenTec RX320D, Pioneer SX780 Last edited by gmcjetpilot on Jan Sun 16, 2011 7:33 pm, edited 2 times in total. |
| Top | |
| | |
Who is online |
| Users browsing this forum: Google [Bot], holt and 17 guests |
 |
 |
 |
 |
 |
 |
 |
 |
 |
 |
 |
 |
 |
 |
 |
 |
 |
 |
How to Make a 75 to 300 Ohm Matching Transformer
Source: https://antiqueradios.com/forums/viewtopic.php?t=154958