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

   Home   Help Search  
Pages: [1] 2 3 4 Next   Go Down
  Print  
Author Topic: Screwdriver Problems - No 80m Zero Reactance?  (Read 3567 times)
K1KRV
Member

Posts: 14




Ignore
« on: August 24, 2009, 05:13:10 PM »

My installation of a 36" screwdriver antenna with 72" whip is not performing well and only seems to transmit consistently on 20m.  

(Humor) I did manage ONE contact on 80m -- a 'rag-chew' 3 miles across the river -- all parties agreed we'd do better to go to the water and shout! :-)

I suspect the antenna, but being new to HF mobile I'm not sure if it isn't something else.

One thing I notice is that this screwdriver CANNOT tune to zero (0) reactance on 80m.  However, the antenna DOES exhibit zero (0) reactance on 40m and 20m.  

Using a Palstar analyzyer connected at the antenna through 4" of RG58, the best numbers I can find on 80m:

Freq: 3905  SWR 1:4  R=38 +j  X=12

On 40m and 20m:

Freq: 7120  SWR 1:2  R=38 +j  X=0

Freq: 14200 SWR 1:2  R=38 +j  X=0

Also, the Palstar readings don't seem very 'solid' on 80m, they take time to settle and inclined to 'jump'.  Testing over the weekend, I swapped to an MFJ analyzer and got similar results.

My vehicle is a 2006 GM Sierra Crew Cab, with the screwdriver installed in a Breedlove 'stake-pocket' mount.  

Bonding: 'Electro-motion' braid, bed-rail to Breedlove mount (1), pickup bed to frame (all 4 corners), exhaust (front and back), hood to firewall (2), engine block (1 additional)

K0BG, my sincere thanks!  I tried to follow your advice as closely as possible.  Your website is a wonderful resource and an incredibly generous gift to amateur radio!

As a test, I installed a 20m 'hamstick' in the Breedlove mount.  'On air' performance seemed about the same as the 'screwdriver'.  The Palstar numbers for the 'hamstick' are:

Freq: 14300  SWR 1:7  R=29  +j  X=0

This 'screwdriver' antenna has a shunt coil integrated in the base.  It's very difficult to adjust and may be part of the problem.  

Does the poor performance and inability to tune 80m suggest the screwdriver antenna is defective?  Do I have the right data to identify the antenna as the problem?  Or, have I missed something in the install?

Any advice or suggestions will be greatly appreciated!

73,

K1KRV
Logged
K0BG
Member

Posts: 9839


WWW

Ignore
« Reply #1 on: August 25, 2009, 06:37:08 AM »

One problem you might be having is BCI. One way to tell is hook the AA to the antenna, and push the mode switch over to Frequency Measurement, and look at the reactance meter. If is jumps around, you have BCI.

And... because of the accuracy of hand held AA devices, your closeness to the antenna, and maybe a few more, the fact it won't go to X=Ø isn't too surprising.

I don't remember the antenna brand, or where it is mounted, but based solely on the numbers, it appears you have a lot of ground loss. Typically,  the input impedance of a decent-quality HF mobile antenna will be about ±25 ohms on 20 and 40. When I see 38, I wonder.

Refresh my memory about the type and mounting position.

Alan, KØBG
www.k0bg.com
Logged

K1KRV
Member

Posts: 14




Ignore
« Reply #2 on: August 25, 2009, 05:01:50 PM »

Alan, thank-you!

My Palstar doesn't have a "Frequency Measurement mode" to check BCI.  I got similar test results in locations 300miles apart, so maybe we can discount BCI, for now.

High impedence values!  Good point!  This surprised me too, considering the installation of ten (10) bonding straps.

The antenna location is 18" behind the pickup cab on the driver's side.  The Breedlove 'stake-pocket' mount wedges into the forward pocket in the bed-rail.  

The upper portion of the Breedlove mount sits on the bedrail with the mounting plate 3" above.  This provides clearance to tighten the antenna bolt. It looks like this:

http://bellsouthpwp.net/b/r/breedlove1/page18.html

I bolted a 14" bonding strap to the front of the box and attached it to the Breedlove mount.  

I might mention that my bedrails have plastic caps, so the top section of the mount sits on plastic.  Electrical connection is made by the bonding strap and the aluminum wedges inside the stake-pocket, so wouldn't think it's a problem. I realize it's not proof of RF ground, but my VOM does show continuity between the mount and the pickup box.

Is it possible to have NOT ENOUGH connection between the mount and the pickup box?  Would insufficient connection between the bed and the mount explain high impedence?

Bonding presents a point of diminishing returns, if I understand it correctly.  My ten (10) bonding straps, seem like a reasonable installation.  Would doubling the quantity of bonding straps substantially improve the ground plane and reduce the impedence from 38 to 25?  

Notes on the antenna:

The antenna is an Alpine and a bit different from traditional 'screwdriver' designs.  The mast is 'sliced' near the bottom to create an integral matching coil.  A length of grey PVC pipe is inserted inside the mast to provide support.  The integral shunt coil 'wraps' around this PVC section.  

There is no fingerstock section in the Alpine antenna.  Cut-outs near the top of the tube expose the wire coil.  A SS steel spring coil wraps around the mast tube (like a rubber band) and seats in these cut-outs.  The spring loop maintains pressure on the antenna coil, connecting it electrically to the mast.  

The Alpine antenna coil is formed around a grey material that looks like PVC, so the Q may be lower than some.  The wire size appears to be #14.  

The motor works just fine and I did wrap the wires through a ferrite choke (several times).  Also, I did my antenna testing with the motor wires disconnected.  

That's everything I can think of.  

I'm coming back from a 'ham hiatus' and having fun with my 'first fling' with HF mobile.  

I really appreciate your help!!

73,

Kurt
K1KRV
Logged
K3GM
Member

Posts: 1766




Ignore
« Reply #3 on: August 26, 2009, 05:21:02 AM »

Hi Kurt.  You didn't mention what brand of antenna you're using.  An antenna like the High Sierra wants the ground point to be right at the antenna's base, and provide a stout mounting lug for it.  I would try replacing the braided strap between the antenna and ground, with a 1.5"-2" wide copper foil strap.  It made a world of difference in my installation which was a mess while using the braid.
Logged
K0BG
Member

Posts: 9839


WWW

Ignore
« Reply #4 on: August 26, 2009, 06:20:18 AM »

Let's look at the "grounding" issue again.

Ground straps, a ground plane don't make! While this seems like double talk, it is not. Any quarter wave vertical antenna (it doesn't matter who made it), loaded or not, requires a ground plane. It does not require a ground! (Call it an earthing point if you please.) In any case, ground planes don't need to be grounded to anything.

The shield of the coax should be connected at, or very near, the base of the antenna, and that connection needs to be coincident to the antenna's ground plane, whatever that may be. In this case, that's the body of the vehicle.

That body should be as RF consistent as we can make it. That is to say, at a minimum impedance to the flow of RF (pun intended). It is this requirement that we satisfy with bonding straps.

If the shield is properly attached, no further ground straps to the mount, near the mount, or coax connection are required. If installing one cured, fixed, or made matching possible, then something else was amiss in the first place.

Remember too, the best vehicle in the world, is still an inadequate ground plane at HF and low VHF frequencies. The reason to place the antenna as high as possible on the vehicle, is an attempt to cause as much RF current to flow through the body, rather than the very lossy surface under the vehicle.

If everything is done properly, and the input impedance is still higher than say 25 ohms from 80 through 20 meters, one should look directly at the antenna (coil Q) for the answer, and not necessarily at a grounding problem.

Alan, KØBG
www.k0bg.com
Logged

K1KRV
Member

Posts: 14




Ignore
« Reply #5 on: August 26, 2009, 10:16:17 AM »

Thanks, Alan!  That's an excellent summary on "grounding"!  I've constructed several ground mounted verticals, but mobile HF is a little different.  It's a big help to have you emphasize the fundamentals.

KOBG wrote:

"If everything is done properly, and the input impedance is still higher than say 25 ohms from 80 through 20 meters, one should look directly at the antenna (coil Q) for the answer, and not necessarily at a grounding problem."

Could you expand on this, please?  Are you saying that when the ground plane is 'as good as we can make it' -- that is, mechanically (lots of straps) -- then the coil 'Q' is likely to be the 'sticking point' creating high impedences?  I'm not entirely clear on how the coil 'Q' contributes to the Resistive/Reactive components in the impedance value.

Since the Hamstick demonstrates a lower impedence in the same installation (R=29 vs R=38) does that suggest the 'Q' of the 'screwdriver' may be problematic?

KG3M, thanks for the 2" copper suggestion!  Yesterday, I added a 2" aluminum bar for mechanical support bar from the box to the mount.  It didn't alter the impedence value, so I suspect the problem may be elsewhere.

Kurt
K1KRV
Logged
K5END
Member

Posts: 1309




Ignore
« Reply #6 on: August 26, 2009, 12:16:20 PM »

K0BG said,

"Typically, the input impedance of a decent-quality HF mobile antenna will be about ±25 ohms on 20 and 40. When I see 38, I wonder."

Alan, on what do you base this (I assume you mean "25 +/-") statement? Can you support this claim?
There are problems with that generalization. A given antenna's impedance is a sum of its radiation resistance, its other L & C properties and its losses.

For example:
A low loss 1/4 wave (spatial, not electrical) vertical antenna's impedance is about 36 ohms.
An 1/8 wave (spatial, not electrical) antenna may have a radiation resistance of 7 ohms.
Or a 1/16 wave antenna's radiation resistance may be as low as 2 ohms.

LC properties are incorporated so the antenna impedance will electrically match the transmitter and/or line impedance, which is added to the radiation resistance, and the remainder is the sum of the losses.

So how can you generalize this across the board to 25 ohms?


K0BG said,

>" The reason to place the antenna as high as possible on the vehicle, is an attempt to cause as much RF current to flow through the body, rather than the very lossy surface under the vehicle."

Not only is that incorrect, it makes no sense. There are many reasons to place the antenna high on the vehicle.

ONE of these is to reduce unwanted coupling between the antenna mast and its coil with the body's sheet metal. An excellent example of how NOT to do this is the Ridgeline antenna installation on your own web page. You have the mast mounted in the corner of the bed. That is quite possibly the worst location one could choose. Yes, lots of guys do it that way, but the physics will not change according to the practice of humans.

Consult sources who are knowledgeable about antennas, such as W8JI and the information on his web page.

My sources are my college physics textbooks ( such as Lorrain/Corson and Reitz, Milford et. al) so I'm reasonably confident about their accuracy.  

The misinformation has to stop. Please.
Logged
K1KRV
Member

Posts: 14




Ignore
« Reply #7 on: August 26, 2009, 04:03:14 PM »

Hi K5END,

Interesting comments.  Since my original question pertains to a 'screwdriver' antenna, I took Alan's "25 ohms" reference to mean 'typical for this installation'.  I've read "17 ohms" in other articles about 'screwdrivers'.   I may be incorrect, but I understood these numbers to be 'ball-park' estimates, similar to your writing "36 ohms" for a 1/4 w vertical (btw: my 160m 'inverted L' is 34 ohms).  

I think Alan means "25 ohms" at a frequency where the reactance components are zero (0).  That's how he describes it at the K0BG website.

Regarding antenna height, I understood the emphasis to mean raising the antenna high enough to maximize the capacitive function of the entire vehicle.  As Allen puts it elsewhere, 'getting as much metal under the antenna as possible'.  

I picture the vehicle as (sorta like) ONE big radial helping the ground serve as the 'bottom half of the dipole'.  However, I'm new to this subject, not an engineer, and may very well be wrong!  

If so, I hope someone will correct me!  

73,

Kurt
K1KRV
Logged
K5END
Member

Posts: 1309




Ignore
« Reply #8 on: August 26, 2009, 04:39:58 PM »

>"That's how he describes it at the K0BG website. "

There are a few technical errors on his website too. I will say that he has been cleaning it up recently. His content is far more accurate than it was in March of this year. Whatever happened in March to start this is unknown to me.

And I am not saying Alan doesn't have a wealth of information to offer. He does, especially on the power distribution and physical installation on the vehicle.

He's a smart guy, and my understanding he is a retired engineer.

Fact is, we probably agree on 95% of the subject matter.

But there is that other 5% that shows up now and then that I simply cannot abide.

There is a lot of good knowledge on the internet but, unfortunately, there is a lot of misinformation and misguided people with good intentions. Choose the advice you heed carefully.


Regarding the vehicle:

At HF spectra the metal vehicle may be described as one plate of a capacitor coupled to the soil beneath it and letting Mother Earth be the "ground plane," and the Earth is one of a very few real world examples of a true "ground plane."

But, because of a number of physical reasons, terra firma not even close to being a perfect ground plane.

You've seen the little cell phone antennas that couple through the vehicle glass. That is a similar method of capacitively coupling the antenna, but in that case the radiating element is coupled and not the ground. And before someone jumps on that, they are not perfect antennas either. The surface area of the metal against the glass and the thickness (with the dielectric properties) of the glass determine the capacitance value. At sufficiently high frequencies the capacitor behaves more like a conductor (EE-0101.)
Logged
K0BG
Member

Posts: 9839


WWW

Ignore
« Reply #9 on: August 26, 2009, 05:53:51 PM »

Larry, what information are you referring to?

The input impedance of a vertical antenna, is a function of it length. Since we're talking mobile HF antennas, the length is always less than a 1/4 wave with the exception of 10, perhaps 12, and in a few cases, 17 meters. Thus the input impedance is less than the theoretical 36 ohms.

For example, an 8 foot, 80 meter mobile antenna has an input impedance (primarily radiation resistance) of under an ohm. To that, we have to add ground losses, and if we do everything correctly, that might be (?) as low as 10 ohms. We have to cancel the capacitive reactance inherent in shortened antennas with a coil, and that coil will have resistive losses at the operating frequency. If the coil Q was 300, then the loss would be 12 ohms. So, 10 + 12 + ≈1 = 23 ohms. Add in a little mounting loss due to capacitive coupling, and 25 ohms is very close.

The real truth is, very few commercial, HF mobile antennas have coil Qs anywhere near 300, especially on 80 meters. And, although ground losses are usually considered to be about 10 ohms on 80 meters, poor mounting schemes, especially ones where the antenna is close to the surface under the vehicle, increase those losses. Fact is, sometimes the input impedance ends up being more than 50 ohms!

Alan, KØBG
www.k0bg.com
Logged

K0BG
Member

Posts: 9839


WWW

Ignore
« Reply #10 on: August 26, 2009, 06:01:35 PM »

That should read:

So, 10 + 12 + (approximately) 1 ohm = 23 ohms.

I guess the server side software doesn't like the approximate sign (double ~).

Alan, KØBG
www.k0bg.com
Logged

K1KRV
Member

Posts: 14




Ignore
« Reply #11 on: August 26, 2009, 07:54:21 PM »

K5END,  very interesting, thank-you!

You wrote:  "At HF spectra the metal vehicle may be described as one plate of a capacitor coupled to the soil beneath it and letting Mother Earth be the "ground plane," and the Earth is one of a very few real world examples of a true "ground plane."
 
That's how I visualize it also.  Your 'through the glass' cell-phone antenna example is an excellent illustration!

K0BG's website has been a big help to me and I appreciate the work he's put into it!  I find it a great place to learn.  If you guys agree on 95% I'd say that's pretty good! ;-)  

I hope Allen will elaborate on his point regarding coil 'Q' and it's influence on impedence in my 'screwdriver' installation.  I installed the bonding in my truck following Allen's website recommendations pretty closely.  It could be improved, but I'm not sure by how much.  That's why I found it interesting that my 'hamstick' (which measures R=29 X=0) seems to out-perform the 'screwdriver' (measuring R=38 X=0) on 20 meters.
 
If it's fair to assume that my bonding system is mechanically 'reasonable' does that point antenna 'Q' as the reason for high impedance?  I don't know.  I do know that this antenna system is 'nearly deaf and as dumb as a stone' -- certainly on 80m.  It's just not getting out, so something's not right.
 
You properly raised the point of antenna location.  On my truck the antenna is 'in the clear'.  The base is mounted on the bed-rail and the coil is well above the roof-line of the cab.  

I certainly welcome any ideas!

73,

Kurt
K1KRV
Logged
K5END
Member

Posts: 1309




Ignore
« Reply #12 on: August 26, 2009, 08:24:26 PM »

Well, the best way to explain "Q" is to look at the definition of it and see it for what it says and nothing more...nothing less.

Like so many other topics, it can hint of mystery when not viewed in the right context.

Imagine a (remember, this is imaginary) low loss antenna that somehow retains most of its "low-lossiness" from 160 meters to a sub-millimeter wavelength.

Aside from the fact we would have to wake up from this dream, to get a pencil and make the calculations, the Q of that antenna would be low.

Now consider a LASER. It has a VERY narrow bandwidth, so it has a very high Q.

Now, consider a trumpet...no, let's consider a bugle. That will be simpler.

That bugle will resonate at certain frequencies (notes) with a small amount of variance in frequency at each of those resonant notes.

Regardless of the frequency of the bugler's embouchure (vibrating lips) the bugle itself is still only resonant on the notes defined by the overall length of the bugle.

So in like manner, regardless of what frequency your radio tries to pump into the antenna, it will still resonate like the gorilla sitting on the bus...where it wants to.

At those resonant frequencies there is some variance. If not, radio communication would not be possible, for lack of modulation room--even for CW.

(yes, even CW is not truly monochromatic, and this concept applies all the way down to quantum mechanics. For a wave to be truly, perfectly monochromatic it would have to observed for a long time...infinity, I presume.)

It is the amount of variance your antenna will allow that in part defines the Q. The losses and thermodynamic efficiency are other factors.

I suggest that anyone interested in this topic set find the good books and read them.

Don't get your info from the internet, and that goes for what I've written as well.
Logged
K1KRV
Member

Posts: 14




Ignore
« Reply #13 on: August 26, 2009, 09:24:54 PM »

Larry,

Your examples are certainly thought provoking!  I particularly like the 'gorrilla radio on the bus'!  I do understand how 'Q' relates to efficiency.  Allen's article "Antenna Efficiency" describes this well.
 
I should have posed my question more from a 'troubleshooting' perspective.  That is, what's the best practical way to assess whether my problem is the bonding system or the antenna?  How do you isolate one factor from the other?

If you purchase the 'highest Q' antenna on the market and your antenna analyzer shows "R=38 X=0" do you simply 'assume' the bonding/ground-plane is inadequate?  Conversely, if the bonding system is installed in accordance with 'standard practice' and the antenna analyzer indicates "R=38 X=0" do you assume the antenna 'Q' is not high enough?  That seems right to me, but I'm looking for confirmation.

Since the antenna analyzer can measure coil 'Q' you could determine the 'Q' for a particular antenna and if the impedence measures 'high' -- make the judgement that the bonding is insufficient.  But is that a fair assumption?  Since this is my first HF mobile installation, I just don't have a 'feel' for the practical aspects.

Lastly, one of the best ways for me to understand any concept is to discuss it.  So thank you for letting me do that!

73,

Kurt
K1KRV
Logged
K0BG
Member

Posts: 9839


WWW

Ignore
« Reply #14 on: August 27, 2009, 06:11:42 AM »

There are several factors you have to consider when comparing antennas by measuring their input impedances, not the least of which is ground loss. It should be noted that ground loss is in series with the input, and stray capacitance losses are in parallel. This fact makes ground losses a much more important consideration than stray capacitance losses. It's also important to note, that these losses cannot be measured quantitatively, but can only be measured as part of the whole. That is to say, the total input impedance.

The stray capacitance in the mounting structure, or near by body parts, is in the neighborhood of a few picofarads; no more than 10 pF, perhaps a little less. This equates to a parallel loading of over 1,000 ohms even on 10 meters, so the effect is minimal, especially in light of the typical ground losses (about 2 to 4 ohms on 10 meters).

Where parallel loading gets you into trouble is when it is applied in or around the coil. The effect is so pronounced it is easily demonstrated. Just measure the input impedance of your antenna, then have someone hold their hand close to the coil (12 inches is close enough), and measure it again.

Where in the overall length the coil is mounted also has an effect. If you go through the machinations to calculate the requisite coil inductance versus its mounting position, and include the ground losses (calculated or assumed), you'll notice the optimal position is dependent on ground loss. To sum it up, the lower the ground losses, the lower the optimal position. Obviously, the Q varies; it decreases as the height increases, all else being equal.

Length matters, as the radiation resistance varies to the square of the overall length, and to a lessor effect on its diameter. Since the ground losses, and the radiation losses are in series, and cannot be separated (quantitatively measured), one factor can increase, while another decreases. As a result, comparing just the input impedance as a measure of efficiency without considering the other factors, will yield mixed results.

What you can do is measure your specific antenna as you make changes. For example, before and after bonding. In this case, the changes are rather small, and unless you have a decent quality antenna analyzer, you might not readily see the change.

Alan, KØBG
www.k0bg.com
Logged

Pages: [1] 2 3 4 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!