EMRFD Message Archive 8571
Message Date From Subject 8571 2013-04-21 20:38:20 Jerry Haigwood RLB from the book
I am about to build a RLB from the emrfd book. I plan to use this RLB
on HF frequencies. I figure I can always build one later especially for VHF
frequencies if I need one. So, the book mentions the toroid core should
have a high permability so as to have a high impedance. I have some FT50-75
cores. These have an AL value of 4200. That means 10 turns of wire through
the core will have an impedance of 2640 ohms at 1 MHz and 79168 ohms at 30
MHz. These seem really high. I also have some FT50-43 cores that have an
AL value of 440. Ten turns at 1 MHz equals 277 ohms at 1 MHz and 8294 ohms
at 30 MHz.. Which one of these cores would be best for use at HF? Is there
a better mix to try? Is there such a thing as too high of an impedance for
this RLB? At VHF, I am guessing a FT50-61 might be better. Do any of you
guys and gals agree?
Jerry W5JH
"building something without experimenting is just solder practice"
[Non-text portions of this message have been removed]8573 2013-04-22 18:06:00 Jerry Haigwood Re: RLB from the book
Below is an email I sent to the group. I have not received any replies.
I am finding it hard to believe that no one on this list has an opinion
about which core might work best. If I had a known standard 50 ohm
terminator, maybe I could measure which core in a RLB would be best (as in
try one, measure, and then the other). However, I don't have a standard 50
ohm load that I know has a certain return loss. I have noticed and I am
sure you have too that my last few post seem to go unanswered. Kerry, my
good friend in Australia, is the only person who takes any interest in my
questions. Now, I realize that I am not anywhere near as learned in RF
electronics as most of you are. But, how in the H..L am I going learn
unless someone points out to me where I might find an answer? Are you guys
really all that busy that you cannot give an opinion or point me in the
general direction. I am quickly losing faith in this group. It appears to
me that this group is for an elite few to exchange ideas but who don't want
to be bothered answering some newbie's questions. If that is the case,
please just let me know and I will go away!
Very sincerely and totally disappointed,
Jerry W5JH
"building something without experimenting is just solder practice"
8574 2013-04-22 19:19:06 Chris Howard Re: RLB from the book
I am not an RF engineer.
I wonder if a latter chapter has a "project" with more detailed directions. I would look in my copy but it is in the guestroom and that room is occupied this evening by a guest.
Sorry to disappoint.
Chris Howard. W0EP
Jerry Haigwood <jerry@w5jh.net> wrote:
>I have something important to say.,
>
>
>
> Below is an email I sent to the group. I have not received any replies.
>I am finding it hard to believe that no one on this list has an opinion
>about which core might work best. If I had a known standard 50 ohm
>terminator, maybe I could measure which core in a RLB would be best (as in
>try one, measure, and then the other). However, I don't have a standard 50
>ohm load that I know has a certain return loss. I have noticed and I am
>sure you have too that my last few post seem to go unanswered. Kerry, my
>good friend in Australia, is the only person who takes any interest in my
>questions. Now, I realize that I am not anywhere near as learned in RF
>electronics as most of you are. But, how in the H..L am I going learn
>unless someone points out to me where I might find an answer? Are you guys
>really all that busy that you cannot give an opinion or point me in the
>general direction. I am quickly losing faith in this group. It appears to
>me that this group is for an elite few to exchange ideas but who don't want
>to be bothered answering some newbie's questions. If that is the case,
>please just let me know and I will go away!
>
>Very sincerely and totally disappointed,
>
>Jerry W5JH
>
>"building something without experimenting is just solder practice"
>
>
>
>
>
>8575 2013-04-22 19:22:31 Chris Howard Re: RLB from the book
Chris
Chris Howard <w0ep@w0ep.us> wrote:
>
>I somehow managed to make an RLB from that book
>I am not an RF engineer.
>I wonder if a latter chapter has a "project" with more detailed directions. I would look in my copy but it is in the guestroom and that room is occupied this evening by a guest.
>
>Sorry to disappoint.
>Chris Howard. W0EP
>
>
>
>Jerry Haigwood <jerry@w5jh.net> wrote:
>
>>I have something important to say.,
>>
>>
>>
>> Below is an email I sent to the group. I have not received any replies.
>>I am finding it hard to believe that no one on this list has an opinion
>>about which core might work best. If I had a known standard 50 ohm
>>terminator, maybe I could measure which core in a RLB would be best (as in
>>try one, measure, and then the other). However, I don't have a standard 50
>>ohm load that I know has a certain return loss. I have noticed and I am
>>sure you have too that my last few post seem to go unanswered. Kerry, my
>>good friend in Australia, is the only person who takes any interest in my
>>questions. Now, I realize that I am not anywhere near as learned in RF
>>electronics as most of you are. But, how in the H..L am I going learn
>>unless someone points out to me where I might find an answer? Are you guys
>>really all that busy that you cannot give an opinion or point me in the
>>general direction. I am quickly losing faith in this group. It appears to
>>me that this group is for an elite few to exchange ideas but who don't want
>>to be bothered answering some newbie's questions. If that is the case,
>>please just let me know and I will go away!
>>
>>Very sincerely and totally disappointed,
>>
>>Jerry W5JH
>>
>>"building something without experimenting is just solder practice"
>>
>>
>>
>>
>>
>>8576 2013-04-22 20:47:23 Jerry Haigwood Re: RLB from the book
I have several 50 ohm loads. However, I have no idea how good of a
load they are. Do they have a return loss of 20db, 30db, 40db (I doubt it).
If I had a 50 ohm load with a RL of say 30db, then I could build my RLB and
check to see if I get a 30db RL with the load plugged in. The bridges
accuracy will also depend on the three 50 ohm resistors used inside of the
RLB. Actually the top two resistors need to be match closely and the third
50 ohm resistor needs to be matched to a standard load. If I had a standard
50 ohm load, I would use it. I am thinking the higher AL value core would
not do as well at 30 MHz as the -43 core. So, I think I will use the
FT50-43 core and leave the RLB uncalibrated. At least I can use it for
relative measurements. BTW, you have not disappointed me.
Jerry W5JH
"building something without experimenting is just solder practice"
8579 2013-04-23 02:04:31 Roelof Bakker Re: RLB from the book
As a rule of thumb the impedance of the winding of a RF transformer should
be four times the system impedance at the lowest frequency you will use
it. For a 50 ohm system this is 200 ohms, so both a FT50-75 and a FT50-43
can be used.
I found that high mu cores work well.
You could try both types and see what the difference is.
For 50 ohm terminator, use to 100 metal film resistors in parallel.
This will work fine.
I hope this helps.
73,
Roelof bakker, pa0rdt8580 2013-04-23 03:48:28 Chris Howard Re: RLB from the book
Chris
Jerry Haigwood <jerry@w5jh.net> wrote:
>Hi Chris,
>
> I have several 50 ohm loads. However, I have no idea how good of a
>load they are. Do they have a return loss of 20db, 30db, 40db (I doubt it).
>If I had a 50 ohm load with a RL of say 30db, then I could build my RLB and
>check to see if I get a 30db RL with the load plugged in. The bridges
>accuracy will also depend on the three 50 ohm resistors used inside of the
>RLB. Actually the top two resistors need to be match closely and the third
>50 ohm resistor needs to be matched to a standard load. If I had a standard
>50 ohm load, I would use it. I am thinking the higher AL value core would
>not do as well at 30 MHz as the -43 core. So, I think I will use the
>FT50-43 core and leave the RLB uncalibrated. At least I can use it for
>relative measurements. BTW, you have not disappointed me.
>
>Jerry W5JH
>
>"building something without experimenting is just solder practice"
>
>
>
>
>
>
>
>8582 2013-04-23 09:36:39 Thomas S. Knutsen Re: RLB from the book
get answers when those who have them have time, it may be minutes or it may
be weeks. If you continue to be rude I can assure you, it would be newer.
When asking questions, try an search in the group archives first, there is
a lot of information there.
In this case there is also an chapter in EMRFD explaining the use of
transmission line transformers, starting on page 3.31. In addition there
are several books written about such transformers, so there should be a lot
of information avaible, if you look instead of expecting us to do the work
for you.
As Mr. Bakker said, you want an impedance larger than 4 times the nominal
impedance of your system, so in an 50 ohm system you need 200 ohm. With an
FT-37-43 thats about 11 turns at 1MHz. (36�H).
In my experience, I get better bandwith by using binocular BN cores than
with regular toroids.
Since the idea behind EMRFD is MEASURE - UNDERSTAND you should do just
that, build a couple RLB's and measure them. An signal generator, power
meter and some termination resistors should be all that is needed.
73 de Thomas LA3PNA.
2013/4/23 Jerry Haigwood <jerry@w5jh.net>
> **
>
>
> I have something important to say.,
>
> Below is an email I sent to the group. I have not received any replies.
> I am finding it hard to believe that no one on this list has an opinion
> about which core might work best. If I had a known standard 50 ohm
> terminator, maybe I could measure which core in a RLB would be best (as in
> try one, measure, and then the other). However, I don't have a standard 50
> ohm load that I know has a certain return loss. I have noticed and I am
> sure you have too that my last few post seem to go unanswered. Kerry, my
> good friend in Australia, is the only person who takes any interest in my
> questions. Now, I realize that I am not anywhere near as learned in RF
> electronics as most of you are. But, how in the H..L am I going learn
> unless someone points out to me where I might find an answer? Are you guys
> really all that busy that you cannot give an opinion or point me in the
> general direction. I am quickly losing faith in this group. It appears to
> me that this group is for an elite few to exchange ideas but who don't want
> to be bothered answering some newbie's questions. If that is the case,
> please just let me know and I will go away!
>
> Very sincerely and totally disappointed,
>
>
> Jerry W5JH
>
> "building something without experimenting is just solder practice"
>
>8583 2013-04-23 11:31:42 Jerry Haigwood Re: RLB from the book
Thank you for your input. It helps a lot. I remember I first saw the
"4 times rule" when reading something written by Doug DeMaw many years ago.
Both the 75 and 43 mix will give me lots of impedance. I can try both but
right now I am not sure how I would test the difference. Again, thank you
for your input.
Jerry W5JH
"building something without experimenting is just solder practice"
-----Original Message-----
8584 2013-04-23 11:32:10 Jerry Haigwood Re: RLB from the book
I am not sure but I would like to have the best accuracy I can obtain.
This list is all about making measurements and I think we should all strive
to make the most accurate measurement possible. What accuracy do you think
is obtainable with a RLB? In the power meter article by Wes Hayward and Bob
Larkin, the authors talk about making measurements using a RLB and the power
meter that were down to -60 db return loss. I really doubt if I could ever
achieve that accuracy but maybe I can build a RLB that can measure -30 db
return loss accurately. At least it is a figure I can strive for.
Jerry W5JH
"building something without experimenting is just solder practice"
8585 2013-04-23 11:32:18 Jerry Haigwood Re: RLB from the book
I am sorry I upset you and thank you for your input.
Jerry W5JH
"building something without experimenting is just solder practice"
-----Original Message-----
8586 2013-04-23 11:55:31 Chris Howard w0ep Re: RLB from the book
I understood you to be unsure of your 50 ohm internal
load because you wanted something more precise.
But I don't think that precision has much to do
with the directivity issue. Maybe I'm mistaken.
I am now able to get to my copy of EMRFD.
Page 7.23 has a schematic for
an RLB which gets directivity at HF in the mid 40 dB range.
Chris
8588 2013-04-23 19:06:36 kb1gmx Re: RLB from the book
However, If you use ft50-43 just use a few more turns of wire.
So happens I just built one a week ago and tried both the -75 and
some -43 Both worked but the 43, I used a few more turns. Why?
I had a bucket of -43 and only a few -75 (reserved for the project).
In my case I was in the 1 to 30mhz range. Hint: the core can be stacked (two -43) and glued (superglue or epoxy) for a higher effective mu. There are plenty of ways to skin that cat.
Build and test. For a load use a good 51 ohm resistor. Or go to Caddock and buy some 1% precision power resistors rated for RF.
The difference, small and only measurable if you have really good gear. Even then for calibration you need a good open circuit
or a good short.
If you read the book and chapter 3 then the answer to the questi8589 2013-04-23 19:31:57 kb1gmx Re: RLB from the book 8590 2013-04-23 19:49:21 William Carver Re: RLB from the book
W7AAZ
On Wed, 2013-04-24 at 02:31 +0000, kb1gmx wrote:
>
>
>
>8591 2013-04-23 21:37:56 Jerry Haigwood Re: RLB from the book
I have a nice Fluke meter to make reasonable resistance measurements
with. So, I can measure the internal reference resistor. But, that does
not insure that I have a good bridge. The transformer will have some effect
on the accuracy. The stray capacitance and stray inductance will make some
difference. I do not have a precision 50 ohm load - one that I know will
have a precise return loss. So, after I build my bridge, how can I test it
for accuracy? I am beginning to think that I will just accept whatever I
get and be happy with it. In the real world, it probably won't make all
that much difference.
Jerry W5JH
"building something without experimenting is just solder practice"
8592 2013-04-23 21:38:00 Jim Re: RLB from the book
Sevick recommended 10X impedance, but 4X should still work fine and as a practical matter is much, much easier to realize and implement.
The higher impedance will be problematic at higher frequency, so ... twist something up and see how it works.
73
Jim N6OTQ
>________________________________
>8593 2013-04-24 06:44:12 k6jq Re: RLB from the book
Fight your penchant to over-think/dramatize things.
73,
Dana K6JQ
----- Original Message -----
8596 2013-04-24 09:51:57 Chris Howard w0ep Re: RLB from the book
through hole 49.9 Ohm 1% 1/4 watt resistors if someone
is stuck for that part when building their RLB.
I think I have some 100.0 also.
8597 2013-04-24 10:36:43 Ashhar Farhan Re: RLB from the book
probably something to do with their reactances. Probably, a run on
spice will reveal the reason.
The rlb will allow you to measure gamma. This will give you two
probable values of the impedance magnitude. Now, if you measure the
return loss by switching a 20 ohms resistance in series with the
unknown impedance, then you can get two more probable values for
impedance magnitude. Now, you have two equations with two unknown (one
being imaginary). Thus, you can figure out the imaginary as well as
real components of the unknown impedance (with the sign of imaginary
part missing, but that can be guessed).
The rlb is a simple yet very profound instrument capable of being used
in many configurations and applications.
- farhan
On 4/24/13, Chris Howard w0ep <w0ep@w0ep.us> wrote:
>
> I believe I have a handful of
> through hole 49.9 Ohm 1% 1/4 watt resistors if someone
> is stuck for that part when building their RLB.
>
> I think I have some 100.0 also.
>
>
>
>8598 2013-04-24 10:48:21 Chris Howard w0ep Re: RLB from the book
If it would be possible, could you
run through an example of that with the
calculations? I would like to see how
it works.
8599 2013-04-24 16:53:31 kb1gmx Re: RLB from the book 8600 2013-04-24 19:04:30 William Carver Re: RLB from the book
> and get to it.
was the closest thing to 50 ohms. Measure it with a good 4-wire ohmmeter
at DC so you are absolutely sure it is 50.0 ohms, not damaged in
assembly, etc.
Or get a load designed for calibrating a vector network analyzer (VNA).
SDR-kits sells a set for about $25. I don't know what the claimed load
resistance accuracy is, but you could ask.
> If you really want accuracy make sure you use decentOr a synchronous detector such as the N2PK or DG8SAQ VNA, which can
> connectors like SMA or N. Use the very best RF layout
> as in a printed circuit board twosided with plated through
> holes or at least through wires, lots of them.
>
> Note that if the RF source is only mediocre (harmonics or noise)
> then the accuracy will be degraded. Assuming it does 10-15dbm
> (20mW) then -60 is a mere -40dbm have you a sensitive RF power
> meter that goes down to microwatts. The 8307 based designs
> are handy for that. If your going to measure with that kind of
> accuracy then a sensitive wattmeter is needed or a calibrated
> receiver and attenuator. Also good cables with solid connectors.
ignore harmonics.
W7AAZ8602 2013-04-25 17:53:48 Kerry Re: RLB from the book
As might be expected, that clear explanation comes from Wes; see pp 152 & 153 of IRFD. This explains why "everything" in an RLB must be 50 ohms (or other Zo); any departure from that Zo produces an RF bridge, still very useful but not quite an RLB.
Wes also wrote a wonderful paper on RLBs; it's sad that it and many other gems are no longer on-line.
I have built several RLBs, each better than its predecessor as I develop my skills;
http://i42.tinypic.com/35hpatl.jpg
http://i43.tinypic.com/2rxgvfs.jpg
An RLB for HF is almost trivial; it can be built with short leaded resistors between carefully-placed RF connectors.
Things get more difficult as frequency rises. Symmetry is essential and SMD resistors and PCBs are required. Ferrite materials become critical and combinations of materials are useful so that one balun/choke "takes-over" as another runs out of impedance.
The balun/choke can be dispensed-with if a balanced detector is used; this RLB does just that;
http://i41.tinypic.com/5cjg0.jpg
with an AD8307. This is a companion to the power-measuring head and both work with the same meter unit.
This RLB works well up to the 650 MHz limit of the AD8307.
For symmetry the reference load might be external via a connector instead of internal.
If BNCs are used the cheap loads used to terminate computer lines aren't too bad at HF and would do for a start. Better loads can be made on the back of a connector using SMD resistors. A pair of 100-ohm 0805 resistors works well; I think that is because the inductance is halved (although the capacitance is doubled).
The lower bridge in the second photo is my most recent attempt; here is its inside;
http://i40.tinypic.com/34pba7t.jpg
It performs well but could be better; I have ideas for improvement but priority isn't high as 30 dB directivity is sufficient to allow errors to be calibrated out with a VNA so there is little incentive to go further.
But, whilst high directivity is now a challenge rather than a need, bridges continue to interest me, something might be done one day. I'm thinking of combining ferrite materials, perhaps 43 & 61, and putting the balun/choke on the RF input as HP did with some of their very broadband bridges. I also have some tiny (about 1mm) 50-ohm solid-outer coa-ax that might be useful.
Kerry VK2TIL.