Call Search
     

New to Ham Radio?
My Profile

Community
Articles
Forums
News
Reviews
Friends Remembered
Strays
Survey Question

Operating
Contesting
DX Cluster Spots
Propagation

Resources
Calendar
Classifieds
Ham Exams
Ham Links
List Archives
News Articles
Product Reviews
QSL Managers

Site Info
eHam Help (FAQ)
Support the site
The eHam Team
Advertising Info
Vision Statement
About eHam.net

donate to eham
   Home   Help Search  
Pages: [1] 2 Next   Go Down
  Print  
Author Topic: 'Magic coax length'  (Read 13452 times)
G3RZP
Member

Posts: 1313




Ignore
« on: May 29, 2010, 02:27:10 PM »

The 30L1 handbook specifies 20.5 feet of RG58 between exciter an amplifier. Apparently later changes suggest 4 feet. Pappenfus etc (Single Sideband Fundamentals and Principles) suggest it should be such that added to the phase shift through the driver tank, it should be a multiple of 180 degrees.

When we're talking of a GG stage with a tuned input, I've always figured that 'magic' length of coax is determined as follows:

Place exciter on desk in desired position.

Place amp on desk in desired position.

Take a piece of string and run  from exciter RF output to amp RF input, leaving enough of a loop as if it was coax, and needed some flexibility.

Add 2 feet each end so the exciter and/or amp can pulled forward on desk for trouble shooting.

Measure length of string, and make the coax about the same length, or a few inches longer.


The basis for this is that the tuned input and low VSWR should make the length relatively immaterial.

Comments?
Logged
AA4PB
Member

Posts: 15065




Ignore
« Reply #1 on: May 29, 2010, 03:26:59 PM »

I agree with your thinking. If the length of the coax was important then you'd have to change it for different bands. When I used a 30L1 I guess I didn't know any better and used a random length (whatever fits) and the poor 30L1 didn't seem to know either :-)
Logged

Bob  AA4PB
Garrisonville, VA
AA4HA
Member

Posts: 2630




Ignore
« Reply #2 on: May 29, 2010, 04:49:31 PM »

That magical length seems to be another urban legend. Somewhere along the line someone through it was a good idea or whatever changeout they did to a poor system suddenly made it better. Then it was stuck in a magazine or written on a bathroom wall and became a de-facto standard.

For 99.9% of the time coaxial length should be kept as short as possible (your rule is a good one). The only time I have ever worried about cutting a coax to a specific length was for coaxial stub filters and phasing lines. We all know that those are only good for a very specific frequency.

Tisha
Logged

Ms. Tisha Hayes, AA4HA
Lookout Mountain, Alabama
N3OX
Member

Posts: 8918


WWW

Ignore
« Reply #3 on: May 29, 2010, 04:56:18 PM »

Then it was stuck in a magazine or written on a bathroom wall and became a de-facto standard.

 Grin
Logged

73,
Dan
http://www.n3ox.net

Monkey/silicon cyborg, beeping at rocks since 1995.
K7ZRZ
Member

Posts: 279




Ignore
« Reply #4 on: May 30, 2010, 10:09:57 AM »

I'm sure this is a debatable subject - for every reason that a discussion can prompt the debate.  But, I have a Clipperton L with a tuned input board, comprised of fixed components (fooey).  I have discovered that there are DEFINATELY certain lengths of coax connector cables between the amp and exciter that do not work well on ALL bands (75, 40, 20, 15, for me).  I haven't had the ambition to try to determine a length which satisfies the amp on all bands, but I have come close by trial and error with differing lengths of connecting cables.

My point being that I really have encountered a connection scheme which caused a significant change in the ability of the exciter to drive the amplifier to full power on a particular band - when it was just fine on other bands. Changing to a different length cable made all the difference in the world in my case.

There is a complicated set of conditions between the amp and the exciter for some of us. I happen to have a monitor scope, AND a coax selector switching unit between my sets. I have two radios that I drive the amp with, and need the switching for that.  If it was a simple short cable between the two, perhaps there would be a different outcome, but to define it in the manner as above, I think my results vary from that.

Brian K7ZRZ
Logged

Brian K7ZRZ
K0BG
Member

Posts: 10248


WWW

Ignore
« Reply #5 on: May 30, 2010, 03:13:07 PM »

I didn't try to find the link, but Tom, W8JI, answered this question a few weeks ago. His explanation, I believe, was correct, in that, it was an Art Collins flub!
Logged

AA4PB
Member

Posts: 15065




Ignore
« Reply #6 on: May 30, 2010, 04:03:58 PM »

One thing is for sure. If changing the cable length (within reason) makes any difference then the amp input is not 50-j0 as it should be.
Logged

Bob  AA4PB
Garrisonville, VA
G3RZP
Member

Posts: 1313




Ignore
« Reply #7 on: May 30, 2010, 11:51:50 PM »

Suppose the amp input IS 50+j0, but the source impedance is high? What then?

Next, the input tuning should ensure a low impedance from cathode to ground at the harmonics, especially the second. If for some reason, this isn't the case, one could perhaps see an effect when changing cable length. I just haven't done.
Logged
W8JI
Member

Posts: 9748


WWW

Ignore
« Reply #8 on: May 31, 2010, 02:29:01 PM »

The 30L1 handbook specifies 20.5 feet of RG58 between exciter an amplifier. Apparently later changes suggest 4 feet. Pappenfus etc (Single Sideband Fundamentals and Principles) suggest it should be such that added to the phase shift through the driver tank, it should be a multiple of 180 degrees.

When we're talking of a GG stage with a tuned input, I've always figured that 'magic' length of coax is determined as follows:

Place exciter on desk in desired position.

Place amp on desk in desired position.

Take a piece of string and run  from exciter RF output to amp RF input, leaving enough of a loop as if it was coax, and needed some flexibility.

Add 2 feet each end so the exciter and/or amp can pulled forward on desk for trouble shooting.

Measure length of string, and make the coax about the same length, or a few inches longer.


The basis for this is that the tuned input and low VSWR should make the length relatively immaterial.

Comments?


This was a smoke screen for the lack of neutralization in the amplifier. It is total fiction.

If you run the network phase shifts you will see they are nowhere near 180 degrees on all bands. They are all over the place.

What Collins did wrong, that even Gonset and Heathkit were too smart to do wrong, was parallel four 811 tubes and think they would be unconditionally stable without neutralization. Collins then aggravated the situation by using a small grid bypass in hopes of rolling the good circuit design from the 30S1 AB1 tetrode into the 30L1 AB2 triode amp.

What the 20 foot small cable really did was terminate the input in some loss to slightly improve stability. It would have been tough for Collins to follow Heathkit and Gonset and neutralize.

73 Tom

Logged
G3RZP
Member

Posts: 1313




Ignore
« Reply #9 on: May 31, 2010, 11:08:56 PM »

Tom,

the RCA Transmitting Tubes data book of 1962 shows a 4 times 811A gg amp. That is neutralised, so it's a little surprising that Collins didn't follow that lead I believe the previous edition also showed the same circuit.

I heard a tale that the 30L1 happened by 'accident'. The story is that a Collins engineer built a four times 811A amp in a Collins cabinet for his own use: taken into work for checking out, it was seen by the powers that were, and it was decided to make it commercially to fit in with the rest of the S line. But how true that is I do not know - but I was told it in Cedar Rapids, back in the days when I visited Collins (or rather, Rockwell) regularly.
Logged
K1BXI
Member

Posts: 821




Ignore
« Reply #10 on: June 01, 2010, 09:33:25 AM »

Is this the same as the 'Magic coax length' some swear by for a G5RV??

I've always used the string method.


The devil in me made me ask.................John 
Logged
WB6BYU
Member

Posts: 18516




Ignore
« Reply #11 on: June 01, 2010, 03:27:39 PM »

Quote from: K1BXI
Is this the same as the 'Magic coax length' some swear by for a G5RV??

No, that was to transform the 80m impedance to a higher value - 17 ohms was a bit low for many of the pi-networks
to load into.
Logged
W6OU
Member

Posts: 354




Ignore
« Reply #12 on: June 05, 2010, 02:41:28 PM »

Perhaps the magic length is related to what is discussed in the thread on the 3CX800A7:

http://www.eham.net/ehamforum/smf/index.php/topic,68615.0.html

This thread explains how the reduction of the second-harmonic component generated by the linear power amp results in improved efficiency and lower IMD. Perhaps Collins found by trial-and-error on 10 meters that the magic length with a KWM-2 driver resulted in a low impedance for the second harmonic at the power amp cathode. This would reduce the degradation in efficiency on 10 meters. By luck it might cause improvement in some of the other higher bands also. Perhaps someone with a KWM-2 and 30L-1 could check the 10-meter efficiency with the magic length vs a short length.

If there is an optimum length, then I wonder if the converse is true. Is there another length which results in a terrible case of efficiency and IMD?
Logged
N2EY
Member

Posts: 5093




Ignore
« Reply #13 on: June 06, 2010, 04:02:12 AM »

the RCA Transmitting Tubes data book of 1962 shows a 4 times 811A gg amp. That is neutralised, so it's a little surprising that Collins didn't follow that lead I believe the previous edition also showed the same circuit.

My TT-4, dated 1956, does not show the 4x811A circuit. What edition is your 1962 one?

When did the 30L1 reach the market?

73 de Jim, N2EY
Logged
G3RZP
Member

Posts: 1313




Ignore
« Reply #14 on: June 07, 2010, 12:28:15 AM »

Jim,

right now, I'm in Mainz in Germany, so I can't tell you which edition of the RCA manual it is. I know I've seen the same edition in different print runs with changes in it between print runs.

The 30L1 handbook is dated August 62, but it was out before that.
Logged
Pages: [1] 2 Next   Go Up
  Print  
 
Jump to:  

Powered by MySQL Powered by PHP Powered by SMF 1.1.11 | SMF © 2006-2009, Simple Machines LLC Valid XHTML 1.0! Valid CSS!