PDA

View Full Version : Coaxial dB Loss



W0AJA
07-12-2014, 01:31 AM
I picked up 2 new rolls of LMR-400 solid center for a good price, factory installed PL-259's.

When I roll the m out on the ground, I get a straight 1.4 dB loss on 146.000

However when I send it up the pole or in the tree I get about 7-11dB of loss. Factory has no idea, and I figure I would ask the only nice group for some elmering. Am I missing some thing with coax?

W3WN: Ron, I can hit the 146.640 and the 147.090 now but wish the loss was lower, doesn't make real sense.

Feel free to ask your questions to help answer my question!

Thanks!

PA5COR
07-12-2014, 03:11 AM
Some questions arise, length of coax?
How did you measure loss on the ground and when the coax was up to the tower?

N2CHX
07-12-2014, 07:11 AM
Some questions arise, length of coax?
How did you measure loss on the ground and when the coax was up to the tower?

Yup, this is the first question I have. What method did you use for arriving at the numbers?

W0AJA
07-12-2014, 01:13 PM
1.) 100 Feet (Solid Core)
2.) I have a MFJ 259C and it was in the settings to measure coax loss.

K7SGJ
07-12-2014, 01:17 PM
Are you measuring the coax terminated (or shorted), or open at the end? If you are measuring it unterminated, put a 50 ohm resistor or dummy load on the end it, and measure again and see if it changes radically.

ETA I heard of several people having weird readings using the coax loss function on the 259B. It appears to be very erratic over small frequency excursions. However, that was with the B series, which is what I have. I'm not familiar with the C series as I've never used it.

W0AJA
07-12-2014, 01:25 PM
Yes, it was unterminated, just a PL-259 on it, no dummy load or antenna.

But when it goes vertical it is connected to a antenna.

K7SGJ
07-12-2014, 01:31 PM
Yes, it was unterminated, just a PL-259 on it, no dummy load or antenna.

But when it goes vertical it is connected to a antenna.

Let me see if I have this right, you're reading it when it is vertical AND hooked up to the antenna? Or is the coax just supported vertically and open at the end when you measure it?

N2CHX
07-12-2014, 01:51 PM
So when measuring on the ground you leave it open and unconnected, but when measuring it in the air you have the antenna connected to it?

https://forums.hamisland.net/attachment.php?attachmentid=12678&stc=1

W0AJA
07-12-2014, 02:14 PM
Yes?....

Well, now that I think about it you would get a higher coaxial loss because of the antenna element and radials. I just thought of that now that I re-read these bottom 2. So, to get a true loss measurement, it should be unterminated or terminated with a dummy load. Correct?

I haven't played much with coax measurements and im just getting my feet wet.

W2NAP
07-12-2014, 02:15 PM
LMR-400 about 1.4 loss per 100 @ 146.

ignore the connected to the antenna..

W0AJA
07-12-2014, 02:26 PM
LMR-400 about 1.4 loss per 100 @ 146.

ignore the connected to the antenna..

Yes it does, and that's what I had flat and unterminated. Its definitely nice on 2 Meters, the loss up in the air had me turning my head but now it makes sense. It even helped bring my SWR down to 1.0/1.1 on 146 compared to the 75-Ohm (Yes, that will start controversy but I know some other's who use it) with a 1.5 SWR

N2CHX
07-12-2014, 02:29 PM
LMR-400 about 1.4 loss per 100 @ 146.

ignore the connected to the antenna..

Bingo.

KC2UGV
07-12-2014, 02:29 PM
Yes it does, and that's what I had flat and unterminated. Its definitely nice on 2 Meters, the loss up in the air had me turning my head but now it makes sense. It even helped bring my SWR down to 1.0/1.1 on 146 compared to the 75-Ohm (Yes, that will start controversy but I know some other's who use it) with a 1.5 SWR

An SWR of 1.5 into a 75 ohm coax, with a 50 ohm load is expected, and acceptable.

50 Ohm signal --> 75 ohm line == 1.5 SWR. Totes normal, and meets the math.

W0AJA
07-12-2014, 02:31 PM
An SWR of 1.5 into a 75 ohm coax, with a 50 ohm load is expected, and acceptable.

50 Ohm signal --> 75 ohm line == 1.5 SWR. Totes normal, and meets the math.

Not bad for if your on a budget and got it free from the local cable company :)

W2NAP
07-12-2014, 02:37 PM
nothing wrong with 75ohm. specially when free

N2CHX
07-12-2014, 02:41 PM
Not bad for if your on a budget and got it free from the local cable company :)

For a few years after I moved into this house, I used 300 ohm twinlead that was part of the ancient (1950's) rooftop antenna mounted in the attic of my old Victorian. With a balun and my old MFJ tuner with its balanced output, it worked very well even on QRP. So yeah, old "crap" can be a treasure.

K7SGJ
07-12-2014, 02:49 PM
Looks like you're in business. FWIW, when you make A/B comparisons on coax, or anything for that matter, you have to make sure the conditions are the same. For example, if you want to compare two antennas to see which works better for receiving, you want to set up your receiver, take whatever measurements you are interested in, and then switch antennas without changing anything on the receiver. That way, you will get a pretty good idea of how they compare.

Glad the other worked out for you.

KC2UGV
07-12-2014, 03:15 PM
Not bad for if your on a budget and got it free from the local cable company :)

Not at all! 1.5:1 SWR is completely nominal, and within the range of what's expected by your modern transceiver.

XE1/N5AL
07-12-2014, 06:13 PM
According to the MFJ 259C manual, when making coax loss measurements:


CAUTION: Do not measure conventional transformers, or attenuators and coaxial cables that are not 50 ohms.
When making measurements, the opposite end of the device under test must have an open circuit, a short circuit,
or a pure reactance for termination. Any loss resistance will make attenuation appear worse than it actually is.

An antenna at the opposite end of your 50 Ohm coax would look like some combination of resistive and reactive components (i.e., not a pure reactance). So, the coax loss results given by your MFJ 259C would be in error if an antenna is connected.

Theory of Operation:
------------------------
I believe the theory of operation, in the MFJ-259C coax loss measurement mode, is as follows:
(I hope this makes sense, and somebody please correct me where I am wrong)

1) The MFJ-259C sends a signal of a pre-determined power level into the coax. This is called the "incident" signal.

2) This signal travels to the far end of the coax, where it is reflected back towards the MFJ-259C.

3) The MFJ-259C now measures the power of this "reflected" signal it received.

4) The coax power loss is 1/2 of the power difference between the incident and reflected signals. This is because, the signal traverses the length of the coax cable two times: the first time going out and the second time reflecting back. Thus, the measured loss is twice what it is for just one direction of signal travel.

Notes:
--------
Power is dissipated in resistive components. Purely reactive components don't dissipate power.

With an open circuit, short circuit, or pure reactance at the far end of the coax; all of the incident signal generated by the MFJ-259C will be reflected back towards the MFJ-259C, when the signal hits the far end of the coax. Any difference between the incident and reflected signals, as measured by the MFJ-259C, will be attributed to coaxial cable loss.

If anything with a resistive component is attached to the far end of the coax, some of the incident signal power will be dissipated in this resistance. Less power will be reflected back to the MFJ-259C and the coax loss will look higher than it actually is. This is because we are erroneously measuring both the power loss of the coax and the power dissipated at the far end termination.

The worst case for error: If a pure 50 Ohm resistance was connected to the far end of loss-less, 50 Ohm coaxial cable (i.e., a perfect cable that doesn't exist in the real world), all of the MFJ-259C incident signal would be dissipated in the 50 Ohm resistance. Nothing at all would be reflected back. This would cause the MFJ-259C to give a completely erroneous result for coax loss. It would say that the coax loss was infinite, because the MFJ-259C didn't receive any reflected signal (and it has no way of knowing that you stuck that resistor on the end of the coax). But remember, our cable was actually loss-less (i.e., a 0 dB loss). An infinite error is a BIG one! Therefore, MFJ says that you can't have any resistive components attached to the far end of the coax cable.

W7XF
07-12-2014, 07:02 PM
Austin, here is a link to a handy chart showing various feedlines and estimated loss per 100 ft:
Coax loss per 100 ft/30m by type (http://www.w4rp.com/ref/coax.html)

W0AJA
07-12-2014, 10:29 PM
Austin, here is a link to a handy chart showing various feedlines and estimated loss per 100 ft:
Coax loss per 100 ft/30m by type (http://www.w4rp.com/ref/coax.html)

Thanks, I also use "Coax Loss Calculator" and compare it to the times microwave site and some other sites.