EMRFD Message Archive 13592

Message Date From Subject
13592 2017-02-16 04:25:35 Baris Resistive bridge
Hello,

I'm trying to build the resistive bridge in Fig 7.43 on page 7.24. I've seen couple of references to this circuit on the web (like http://www.vk2zay.net/article/97) but the value for R1 and R2 confused me, are they really 62 ohms? Anyone used this bridge up in the 144 MHz region?

Regards,

Baris
13593 2017-02-16 07:36:39 John Marshall Re: Resistive bridge
Boris,

The bridge works by comparing the impedance you are measuring to a "Known Termination". Ideally, R1 and R2 should be the same as the Known Termination, but 62 ohms allows good performance with either 50 or 75 ohm Knowns. Rather than a specific value, it's much more important for these two resistors to be matched as closely as possible and for the Known and Unknown arms of the bridge to be symmetrical. I built one years ago using 1/4 watt resistors that works at 432 MHz.

John, KU4AF
Pittsboro, NC

13594 2017-02-16 12:19:58 E-P Mänd VS: [emrfd] Resistive bridge

Hello Baris: It is a bridge, resistor ratios are important, not values. So = R1 / Known = must be the same as R2 / Unknown. R1, R2 = 68 ohms is good for either 50 ohms system or 75 ohms system, don’t worry. I’ve constructed similar bridges, goes up to 1 GHz with short connections and proper lines (CoAx, CoPlanar or stripline) to connectors (BNC or SMA), but use SMD resistor and capasitors only.  Even better, if you are interested in only 50 ohms system, use R1 = R2 = 50 ohms and add third 50 ohms resistor to make a three-port divider to keep all impedances same. – Kindly, EePee = OH2NFI, member of ARRL, too.    

 

13595 2017-02-16 12:54:11 Tayloe, Dan (Noki... Re: Resistive bridge

In the bridge, there are three resistors in 3 out of the four legs, with the 4th leg being the antenna under test.

 

If the three resistors are 50 ohms and the antenna is 50 ohms, then one side of the “H” bridge looks like 100 ohms (50 + 50) and the other side looks like 100 ohms (also 50+50).  Two 100 ohm resistors in parallel is 50 ohms.

 

Thus, the transmitter sees 50 ohms if the antenna is 50 ohms.  If the antenna is very high impedance (i.e., “open), then the transmitter just sees (50 + 50) in parallel with (50 + very large) or just 100 ohms.

 

If the antenna is shorted, the transmitter sees (50+50) in one leg and (50+0) in the other.  The net impedance is 33.3 ohm.

 

If you are measuring 50 ohm antennas, use all three resistors as 50 Ohm, and that will serve to protect the transmitter by limiting the load variation seen (33.3 to 100 ohm worst case).

 

From a measurement point of view, all you need is to have the two resistors in the one leg be the same, and the resistor in the second leg be the same as your target antenna impedance (i.e., the forth “resistor” in the bridge).

 

-          Dan, N7VE

 

13596 2017-02-16 14:37:01 E-P Mänd VS: [emrfd] Resistive bridge

Quite wright. I forgot the word ”almost” in my phrase “to keep all the impedances the same”. Hopefully, this does not confuse the original question! – Kindly: EePee.

 

Lähettäjä: emrfd@yahoogroups.com [mailto:emrfd@yahoogroups.com]
Lähetetty: 16. helmikuuta 2017 22:39
Vastaanottaja: emrfd@yahoogroups.com
Aihe: RE: [emrfd] Resistive bridge

 

 

In the bridge, there are three resistors in 3 out of the four legs, with the 4th leg being the antenna under test.

 

If the three resistors are 50 ohms and the antenna is 50 ohms, then one side of the “H” bridge looks like 100 ohms (50 + 50) and the other side looks like 100 ohms (also 50+50).  Two 100 ohm resistors in parallel is 50 ohms.

 

Thus, the transmitter sees 50 ohms if the antenna is 50 ohms.  If the antenna is very high impedance (i.e., “open), then the transmitter just sees (50 + 50) in parallel with (50 + very large) or just 100 ohms.

 

If the antenna is shorted, the transmitter sees (50+50) in one leg and (50+0) in the other.  The net impedance is 33.3 ohm.

 

If you are measuring 50 ohm antennas, use all three resistors as 50 Ohm, and that will serve to protect the transmitter by limiting the load variation seen (33.3 to 100 ohm worst case).

 

From a measurement point of view, all you need is to have the two resistors in the one leg be the same, and the resistor in the second leg be the same as your target antenna impedance (i.e., the forth “resistor” in the bridge).

 

-          Dan, N7VE

 

13602 2017-02-18 07:00:31 Baris Re: Resistive bridge
Thanks all, I had used a 68 and a 680 ohm resistor in parallel to get the correct value, now I can replace them with two parallel 100 ohm parts. In place of 1N5177 I used a 1n5819, is this OK? I also replaced the 10K pot with a fixed 10K resistor and connected another BNC detector port that goes to my multimeter or oscilloscope. I noticed there's a lot of the original AC signal added to the DC offset at the detector port, even after adding another 100n cap at the detector output, is this normal too?

Regards,
Baris

2017-02-17 1:15 GMT+03:00 E-P Mänd e-p.mand@pp.inet.fi [emrfd] <emrfd@yahoogroups.com>:
 

Quite wright. I forgot the word ”almost” in my phrase “to keep all the impedances the same”. Hopefully, this does not confuse the original question! – Kindly: EePee.

 

Lähettäjä: emrfd@yahoogroups.com [mailto:emrfd@yahoogroups.com]
Lähetetty: 16. helmikuuta 2017 22:39
Vastaanottaja: emrfd@yahoogroups.com
Aihe: RE: [emrfd] Resistive bridge

 

 

In the bridge, there are three resistors in 3 out of the four legs, with the 4th leg being the antenna under test.

 

If the three resistors are 50 ohms and the antenna is 50 ohms, then one side of the “H” bridge looks like 100 ohms (50 + 50) and the other side looks like 100 ohms (also 50+50).  Two 100 ohm resistors in parallel is 50 ohms.

 

Thus, the transmitter sees 50 ohms if the antenna is 50 ohms.  If the antenna is very high impedance (i.e., “open), then the transmitter just sees (50 + 50) in parallel with (50 + very large) or just 100 ohms.

 

If the antenna is shorted, the transmitter sees (50+50) in one leg and (50+0) in the other.  The net impedance is 33.3 ohm.

 

If you are measuring 50 ohm antennas, use all three resistors as 50 Ohm, and that will serve to protect the transmitter by limiting the load variation seen (33.3 to 100 ohm worst case).

 

From a measurement point of view, all you need is to have the two resistors in the one leg be the same, and the resistor in the second leg be the same as your target antenna impedance (i.e., the forth “resistor” in the bridge).

 

-          Dan, N7VE

 

13603 2017-02-18 09:26:05 Harold Smith Re: Resistive bridge
The 1N5819 is a power rectifier, and will look like a fairly large capacitance in parallel with a diode at RF.  You would be better off using a true small signal diode, even if it's not a Schottky type.  If you can't find the 1N5177 (surface mount) there are lots of others.

de KE6TI, Harold

13611 2017-02-20 06:08:27 Baris Re: Resistive bridge
Thanks Harold, I replaced the diode with a 1N4148 for the time being, I'll put in a 1n5711 soon. This is a picture of my completed circuit (before changing the diode): http://imgur.com/YtlP3a3 .

Regards,
Baris

2017-02-18 20:03 GMT+03:00 Harold Smith harold.smith1@gmail.com [emrfd] <emrfd@yahoogroups.com>:
 

The 1N5819 is a power rectifier, and will look like a fairly large capacitance in parallel with a diode at RF.  You would be better off using a true small signal diode, even if it's not a Schottky type.  If you can't find the 1N5177 (surface mount) there are lots of others.

de KE6TI, Harold

13612 2017-02-20 06:38:46 Tayloe, Dan (Noki... Re: Resistive bridge
The resistor on the bottom right is across the "antenna" port, right?  It seems like that ought to be large like 5k.  It looks like 100 ohms.

Likewise, I would expect the bottom left to be 50 ohms and not 100.

- Dan

Sent from my Galaxy Tab® A


-------- Original message --------
13614 2017-02-20 10:33:44 Baris Re: Resistive bridge
Hi Dan, those are both 1K resistors (1%), as it's indicated in Fig 7.43.

Baris

2017-02-20 17:22 GMT+03:00 'Tayloe, Dan (Nokia - US/Irving)' dan.tayloe@nokia.com [emrfd] <emrfd@yahoogroups.com>:
 

The resistor on the bottom right is across the "antenna" port, right?  It seems like that ought to be large like 5k.  It looks like 100 ohms.

Likewise, I would expect the bottom left to be 50 ohms and not 100.

- Dan

Sent from my Galaxy Tab® A


-------- Original message --------
13617 2017-02-20 12:12:17 Tayloe, Dan (Noki... Re: Resistive bridge

For the bridge to “null” for a 50 ohm antenna, the left bottom resistor needs to be 50 ohms.  You can test the set up “as is” by placing a 50 ohm dummy load to see if the detected power drops to a minimum.  If it does not do that, it is not configured right.

 

As it is, I think it will not null with a 50 ohm antenna.

 

-          Dan

 

13621 2017-02-20 14:46:11 Baris Re: Resistive bridge
Oh, I'm connecting a 50 ohm coaxial terminator at the left port, I believe that's the 50 ohm resistance you said that should be there. It seems to be working as intended now, I'm seeing only a few microvolts with a 50 ohm load.

Baris

2017-02-20 21:38 GMT+03:00 'Tayloe, Dan (Nokia - US/Irving)' dan.tayloe@nokia.com [emrfd] <emrfd@yahoogroups.com>:
 

For the bridge to “null” for a 50 ohm antenna, the left bottom resistor needs to be 50 ohms.  You can test the set up “as is” by placing a 50 ohm dummy load to see if the detected power drops to a minimum.  If it does not do that, it is not configured right.

 

As it is, I think it will not null with a 50 ohm antenna.

 

-          Dan

 

13622 2017-02-20 15:13:00 Tayloe, Dan (Noki... Re: Resistive bridge

I guess that is a more flexible arrangement.  If you wanted to test 75 ohm antennas, you could simply switch to a 75 ohm termination on the left port and you would be good to go.

 

-          Dan

 

13623 2017-02-20 15:20:22 Phil Sittner Re: Resistive bridge

Baris-

If you want real insight into your antennas it's suggested that you look at the VIA kit offered at http://www.qsl.net/k5bcq/Kits/Kits.html. This thing sells for under $80.00 and allows you to measure the real impedance, including sign, at frequencies up to 150Mhz. Not only that but you can calibrate the device at the end of your coax so the measurement is at the actual antenna termination. You also can measure coax impedance, reactive component values and so on. It's the absolute best deal I've found for dealing with real impedance issues. You'll need to provide your own enclosure and a few other pieces but it's still an unbelievable deal.


Phil, KD6RM


13698 2017-03-13 07:13:08 Baris Re: Resistive bridge
Hi Phil, I actually just built an NP2K VNA last month. It still has a few touches left to do but seems to be working well for the moment. I noticed that the input VSWR of the bridge goes up to 1.2 at around 60 MHz, which brings me to my next question, how does the input VSWR of the bridge (at the RF source port) affect its measurement capability? If we're seeing 1.2 VSWR at a certain frequency, does that mean we can't match the DUT to a better VSWR than that using the bridge, or all that matters is the symmetry of the bridge? How can we determine its usable frequency range?

Thanks,
Baris




2017-02-21 2:20 GMT+03:00 Phil Sittner sittners@sbcglobal.net [emrfd] <emrfd@yahoogroups.com>:
 

Baris-

If you want real insight into your antennas it's suggested that you look at the VIA kit offered at http://www.qsl.net/k5bcq/Kits/ Kits.html. This thing sells for under $80.00 and allows you to measure the real impedance, including sign, at frequencies up to 150Mhz. Not only that but you can calibrate the device at the end of your coax so the measurement is at the actual antenna termination. You also can measure coax impedance, reactive component values and so on. It's the absolute best deal I've found for dealing with real impedance issues. You'll need to provide your own enclosure and a few other pieces but it's still an unbelievable deal.


Phil, KD6RM


13699 2017-03-13 10:12:42 Phil Sittner Re: Resistive bridge

Baris-

The N2PK VNA is the most accurate, home built network analyzer that I have ever come across. That being said I'm not sure what bridge you are measuring but there are a couple of things that you should be aware of.

First, all network analyzers must be calibrated before you can have any meaningful measurements. If the bridge you are talking about is the "reflection bridge" part of the N2PK system I suspect that you have not done an "OSL" (open, short, load) calibration. The calibration routine tells the computer how to interpret the reflection data in order to present accurate, adjusted measurements.

Every RF measurement changes with the corresponding change in frequency. This is because every resistor, capacitor, inductor and feedline has parasitic components; a resistor also has both inductance and capacitance, a capacitor has resistance and inductance, an inductor has capacitance and resistance and a piece of feedline has all three components as well. The reactance of inductors and capacitors changes with frequency.

You should also know that no one has ever successfully built a bridge whose parameters did not change with frequency.

Use of the N2PK VNA requires familiarizing oneself with the calibration routine and the nature of reflected RF energy. There are a number of tutorials available in addition to the information regarding the excellent N2PK device. You ma wish to join one of the topic forums like the excellent DG8SAQ VNWA forum on Yahoo Groups.

Keep at it and you will find that you have the tools to really understand how all of this RF stuff works. Good Luck.


Phil, KD6RM


13706 2017-03-14 03:42:30 Baris Re: Resistive bridge
Thanks Phil, I agree this VNA is really amazing. I recently managed to complete the 5-term calibration which allows one to measure the reflection and transmission measurements simultaneously, and so far it seems to be very accurate. The only thing I would like in the N2PK VNA would be, of course, a wider frequency range. I might also start looking into the VNWA you mentioned for that reason.

The bridge I'm talking about was the resistive bridge in EMRFD figure 7.43, I swept it with the VNA (after calibrating it) and saw the input VSWR steadily rises to about 1.2 before 60 MHz, which is the upper limit of this VNA, so I was wondering whether this would limit its measurements at these frequencies, or at higher bands like 2m.

Best regards,
Baris

2017-03-13 20:12 GMT+03:00 Phil Sittner sittners@sbcglobal.net [emrfd] <emrfd@yahoogroups.com>:
 

Baris-

The N2PK VNA is the most accurate, home built network analyzer that I have ever come across. That being said I'm not sure what bridge you are measuring but there are a couple of things that you should be aware of.

First, all network analyzers must be calibrated before you can have any meaningful measurements. If the bridge you are talking about is the "reflection bridge" part of the N2PK system I suspect that you have not done an "OSL" (open, short, load) calibration. The calibration routine tells the computer how to interpret the reflection data in order to present accurate, adjusted measurements.

Every RF measurement changes with the corresponding change in frequency. This is because every resistor, capacitor, inductor and feedline has parasitic components; a resistor also has both inductance and capacitance, a capacitor has resistance and inductance, an inductor has capacitance and resistance and a piece of feedline has all three components as well. The reactance of inductors and capacitors changes with frequency.

You should also know that no one has ever successfully built a bridge whose parameters did not change with frequency.

Use of the N2PK VNA requires familiarizing oneself with the calibration routine and the nature of reflected RF energy. There are a number of tutorials available in addition to the information regarding the excellent N2PK device. You ma wish to join one of the topic forums like the excellent DG8SAQ VNWA forum on Yahoo Groups.

Keep at it and you will find that you have the tools to really understand how all of this RF stuff works. Good Luck.


Phil, KD6RM


13707 2017-03-14 07:39:55 Phil Sittner Re: Resistive bridge

Baris-

As mentioned in my earlier response all components have parasitic components that become apparent as frequency rises. An interesting exercise is to measure a 50 ohm (49.9 or 51) resistor on your VNA; first make the measurement with long leads on the resistor, then trim the leads as short as possible. It's a very telling exercise.

I have four home built VNA's. The first is the DG8SAQ, built about 8 years ago when it was a parallel port controlled device, the second is the N2PK, the third is Scotty's spectrum analyzer and the final one is the device I reference earlier, is the Vector Impedance Analyzer designed by K5BCQ.

All devices connected to a VNA will display frequency dependent characteristics. The good news is that you do not need an external bridge to make measurements with a VNA--the bridge is part of the instrument and the calibration routine corrects for the reactances within the bridge.

The device shown in figure 7.43 is no different. There are reactances present in your bridge circuit that compromise the accuracy of the measurements. You can minimize their effect by using surface mount components, using a very small PCB with short traces, building the bridge with 50 ohm microstrip designs and paying careful attention to layout etc. But there is still a frequency dependency because of parasitics.

The most accurate measurements you can make are those done with a VNA. If your concerns are at VHF or UHF frequencies you need a VNA designed for those frequencies. There is a design for a converter that allows the N2PK to work at those frequencies that you may wish to consider. I believe you will find it at the Makarov site and also on N2PK's. Keep up the experiments.

Phil, KD6RM