Saturday, January 31, 2015

The Ultralight dBµ Mystery, S-Meters, And Field Strength

On our little ultralight DSP receivers like the Tecsun PL-380 and PL-310, etc. you may have noticed the dial face has some numbers which vary with the received signal strength. Next to them are marked the letters "dBµ". It seems like a field strength meter of some kind. Just what is that? Well, it is indeed an indicator of signal strength. Let's dig into what it means and see how it is directly related to the S-meter of old, the field strength meter.

National Radio Company

Tecsun's use of dBµ (a funny-looking, backwards 'u', which is the Greek letter µ 'mu', meaning micro, or one-millionth) is really an improperly-used, shortened version of the term dBµV. Warning! You may have also seen the term "dBu" (lowercase "u") written in various publications, associated with field strength also. It refers to something different (actually the E-field of the passing wave). We can handle that one in a separate article, so be sure not to confuse Tecsun's dBµ with dBu!

Back to Tecsun's dBµ. Let's break it down:

dB = decibels, or simply a way of expressing magnitudes of a value, like voltage, logarithmically
µV = microvolts, or millionths of a volt

Consequently, dBµV is a voltage expressed in dB above (or below) one microvolt. This is measured across a specific load impedance, commonly 50 ohms. Important! Here we have a real received voltage measured across a specific load impedance like a tuned circuit!

The 'dB' or decibel measurement is a logarithmic ratio as you may know. In terms of voltage, an increase of 6 dB is a doubling of voltage. So, if our little Tecsun receives a signal at 28 dBµ and it increases to 34 dBµ, the received voltage has doubled. Coincidentally, this is also an increase of one S-unit! Now we are getting somewhere.

Let's translate our received dBµV into actual received voltage:

dBµV   µV(millionths of a volt)
 94    50000.0
 84    15810.0
 74     5000.0
 64     1581.0
 54      500.0
 44      158.1
 34       50.0 (the S-9 of old!)
 28       25.0
 22       12.5
 16        6.3
 10        3.2
  4        1.6
 -2        0.8 (less than 1 µV sends the dB ratio to a negative value!)
 -8        0.4
-14        0.2

The following formula is used to convert dBµV to millionths of a volt:

µV = (10 ^ (dBµV/ 20))

To convert millionths of a volt back to its decibel representation:

dBµV = 20 * Log(µV)

(Log is the common logarithm, or base 10).

The modern DSP receivers like the Tecsun PL-380, 310, etc. which employ the Silicon Labs chips, measure and display dBµV as received at the tuned front end across a load. They call it the RSSI indicator. Our radio's antenna, the iron core ferrite rod, is basically a signal concentrator. The longer the rod and thus the more iron ferrite, the more the concentration, and the greater the signal voltage transferred to the radio's tuned input.

So what exactly is this so-called dBµ indicator on our DSP radios telling us?

Some time ago, more than a year ago, I posed this question to Scott Willingham, who was on the design team for the SiLabs DSP receiver chips used in these radios.

He stated:

"The RSSI (dBµ) readings are referred to the pins of the chip, which are the inputs to the LNA. In the Tecsun radios operating in the MW band, this is also the voltage across the loopstick. In SW bands, the Tecsun ULRs use a preamp/LNA on the circuit board between the whip antenna and the Si4734. In that case, the RSSI readings reflect the signal at the output of Tecsun's external LNA."

Essentially for mediumwave, the received signal is measured in microvolts right off the loopstick and then converted to dBµ, which is decibels above a base of one microvolt. Remember again, dB is just a logarithmic ratio. Of course a PL-380 is not going to read the same dBµ as a PL-310 or a PL-398, etc., because the antenna setups (loopstick lengths, whip extension, tuned circuit efficiency) are different and each will induce different received voltage levels to the radio.

A curious measurement, yes, but there is also some meaningful information here in comparing signal strengths within the same radio just like an S-meter did, and in fact there is a direct correlation to the S-meter.

The analog S-meter us old guys remember in now ancient receivers was based on S-9 indicating a 50 µV (microvolt) input signal to the antenna circuitry, at a load impedance of 50 ohms. That is, the S-meter read S-9 if the receiver S-meter was calibrated right, as the meter was further down the IF chain, and usually responded to the AGC (automatic gain control) level. Each S-unit is 6 dB apart, meaning a signal reading S-9 is 6 dB stronger than a signal reading S-8. S-9 +10dB is 10 dB greater than S-9, or one S-unit plus 4 more dB.

What does this mean? An S-9 signal is twice as strong as an S-8 signal. The received voltage is double. An S-9 signal is four times as strong as an S-7 signal. The received voltage is doubled twice.

Some direct correlation can be attempted with the SiLabs DSP chip dBµ readings used in the Tecsun radios.

S-unit        µV  dBµV  dBm
S9+60dB  50000.0   94   -13
S9+50dB  15810.0   84   -23
S9+40dB   5000.0   74   -33
S9+30dB   1581.0   64   -43
S9+20dB    500.0   54   -53
S9+10dB    158.1   44   -63
S9          50.0   34   -73
S8          25.0   28   -79
S7          12.5   22   -85
S6           6.3   16   -91
S5+4.9dB     5.6   15   -92
S5           3.2   10   -97
S4           1.6    4  -103
S3+1.9dB     1.0    0  -107
S3           0.8   -2  -109
S2           0.4   -8  -115
S1           0.2  -14  -121
Look at the S-unit and dBµV columns. As can be seen, a 34 dBµV signal (again, the Tecsun DSP radios label it dBµ) is essentially equivalent to an S-9 signal on the old S-meter setup. The 25 dBµ signal shown in the picture below represents a signal halfway between S-7 and S-8.

On the Tecsun PL-380 (at least the version I own, which registers from 15 dBµ - 63 dBµ), somewhere around 15 dBµ seems to be the signal detection threshold which translates to just below the old S-6, at 6.3 microvolts of signal. As noted elsewhere, these modern drug store consumer radios are not as sensitive as the old communications receivers we remember. S-6 on an old vacuum tube receiver was virtually "arm chair" copy. This is where an FSL or passive loop brings up the weak received signal to similar levels in the DSP radios.

So there you have it. Keep this chart handy and you can convert between Tecsun's dBµ and S-units.

Stay tuned for the second article in this series: The dBµ Versus dBu Mystery: Signal Strength vs. Field Strength?

25 dBµ, or between S-7 and S-8


Stephen said...

Hi Bill,

This is an interesting post. :)

I've found that the dBu readings aren't exactly linear related to actual signal strength. For example, here's a pipe-delimited table of signal strengths at my location, as referenced on a few radios. This is for daytime reception.
Frequency & Call are self-explanatory.
The RX mV/m is from the rec-mV/m column, and the RX dBu from the rec-dBu column, in the custom file you gave me last September for reception at my location.
PL-398 dBu (can be either 398mp or 398bt, they're essentially the same radio) is what the dBu indicates on that radio after I've entered a desense-modification mode I stumbled upon some time ago. To do this, turn on the radio, go to SW, hold VF until it starts scanning then give the tuning knob a quick jab downward. It should show 00/00 or similar (if you stopped on a blank channel), with fairly loud noise. Then you can go back to AM, and it'll show lower signal levels on weak stations, and not show the high readings on blank channels near strong signals. (I've even noticed that the noise in blank channels in this mode is as loud as or louder than the program content on the local blowtorches!)
Offset is how far off the PL-398's reading is compared to what the received dBu should be. (The PL-398 always reads lower than it "should".)
SW20 S# is based on a qualitative to-my-ears assessment based on hearing those signals on my Panasonic RQ-SW20. More on that later.
PL-606 readings are just what it says. A "xx" after the / means the station isn't heard, whereas "00" means it is heard maybe faintly, or at least there's the sound of beating carriers from two fringe interfering stations, but the muting/desense is preventing the audio from coming up.
Not shown are readings on my PL-380. Like yours, mine tops out at 63 dBu, and mine gives similar desense characteristics to my PL-606. Also, readings on stronger signals are about 3 dB or so lower on my PL-380 than on my PL-606.

Freq | Call | RX mV/m | RX dBu | PL-398 dBu | dBu - PL-398 offset | SW20 S# | PL-606 readings

560 | KBLU | 0.102 mV/m | 40.1 dBu | PL-398 “09” | dBu - 398 = 31.1 | SW20 s-0 | PL-606 “24/xx”
570 | KLAC | 0.794 mV/m | 58.0 dBu | PL-398 “33” | dBu - 398 = 25.0 | SW20 s-3 | PL-606 “32/22”
600 | KOGO | 35.49 mV/m | 91 dBu | PL-398 “68” | dBu - 398 = 23.0 | SW20 s-7 | PL-606 “67/25”
640 | KFI | 3.813 mV/m | 71.6 dBu | PL-398 “46” | dBu - 398 = 25.6 | SW20 s-5 | PL-606 “45/25”
670 | KIRN | 0.799 mV/m | 58.1 dBu | PL-398 “24” | dBu - 398 = 34.1 | SW20 s-2 | PL-606 “27/14”
710 | KSPN | 1.566 mV/m | 63.9 dBu | PL-398 “28” | dBu - 398 = 35.9 | SW20 s-2 | PL-606 “29/22”
740 | KBRT | 2.095 mV/m | 66.4 dBu | PL-398 “32” | dBu - 398 = 34.4 | SW20 s-3 | PL-606 “32/??”
760 | KFMB | 46.912 mV/m | 93.4 dBu | PL-398 “71” | dBu - 398 = 22.4 | SW20 s-7 | PL-606 “70/25”
790 | KABC | 0.829 mV/m | 58.4 dBu | PL-398 “32” | dBu - 398 = 26.4 | SW20 s-3 | PL-606 “32/23”
830 | KLAA | 0.905 mV/m | 59.1 dBu | PL-398 “34” | dBu - 398 = 25.1 | SW20 s-3 | PL-606 “32/25”
870 | KRLA | 0.196 mV/m | 45.9 dBu | PL-398 “19” | dBu - 398 = 26.9 | SW20 s-1 | PL-606 “25/00”
910 | KECR | 27.129 mV/m | 88.7 dBu | PL-398 “67” | dBu - 398 = 21.7 | SW20 s-7 | PL-606 “66/25”
930 | KHJ | 0.209 mV/m | 46.4 dBu | PL-398 “18” | dBu - 398 = 28.4 | SW20 s-1 | PL-606 “??/??”
960 | KIXW | 0.059 mV/m | 35.5 dBu | PL-398 “13” | dBu - 398 = 22.5 | SW20 s-0 | PL-606 “32/00”
970 | KNWZ | 0.196 mV/m | 45.9 dBu | PL-398 “19” | dBu - 398 = 26.9 | SW20 s-1 | PL-606 “32/00”
980 | KFWB | 0.224 mV/m | 47.0 dBu | PL-398 “21” | dBu - 398 = 26.0 | SW20 s-1 | PL-606 “30/02”
1000 | KCEO | 2.64 mV/m | 68.4 dBu | PL-398 “43” | dBu - 398 = 25.4 | SW20 s-4 | PL-606 “42/25”

Stephen said...

1020 | KTNQ | 0.215 mV/m | 46.7 dBu | PL-398 “21” | dBu - 398 = 25.7 | SW20 s-1 | PL-606 “34/00”
1040 | KURS | 8.063 mV/m | 78.1 dBu | PL-398 “54” | dBu - 398 = 24.1 | SW20 s-6 | PL-606 “53/25”
1070 | KNX | 3.29 mV/m | 70.3 dBu | PL-398 “48” | dBu - 398 = 22.3 | SW20 s-5 | PL-606 “47/25”
1110 | KDIS | 0.469 mV/m | 53.4 dBu | PL-398 “25” | dBu - 398 = 28.4 | SW20 s-1 | PL-606 “43/00”
1130 | KSDO | 63.838 mV/m | 96.1 dBu | PL-398 “75” | dBu - 398 = 21.1 | SW20 s-8 | PL-606 “75/25”
1150 | KEIB | 0.093 mV/m | 39.4 dBu | PL-398 “14” | dBu - 398 = 25.4 | SW20 s-0 | PL-606 “45/xx”
1170 | KCBQ | 123.032 mV/m | 101.8 dBu | PL-398 “80” | dBu - 398 = 21.8 | SW20 s-8 | PL-606 “80/25”

note: before KCBQ's move out of Santee, they would have been rated S-9 on the RQ-SW20. IIRC, someone on a forum post somewhere estimated that their field at my location would have been around 300-400 mV/m or so, OR, their power radiated toward me was around 300-350 kW or so.

1190 | KGBN | 0.142 mV/m | 43.0 dBu | PL-398 “18” | dBu - 398 = 25.0 | SW20 s-0 | PL-606 “43/00”
1210 | KPRZ | 11.344 mV/m | 81.1 dBu | PL-398 “54” | dBu - 398 = 27.1 | SW20 s-6 | PL-606 “54/25”
1230 | KXO | 0.042 mV/m | 32.5 dBu | PL-398 “16” | dBu - 398 = 16.5 | SW20 s-1 | PL-606 “32/00”
1240 | KNSN | 9.636 mV/m | 79.7 dBu | PL-398 “55” | dBu - 398 = 24.7 | SW20 s-6 | PL-606 “55/25”
1250 | KZER | 0.225 mV/m | 47.1 dBu | PL-398 “21” | dBu - 398 = 26.1 | SW20 s-2 | PL-606 “32/08”
1260 | KMZT | 0.108 mV/m | 40.7 dBu | PL-398 “13” | dBu - 398 = 27.7 | SW20 s-1 | PL-606 “35/xx”
1280 | KFRN | 0.412 mV/m | 52.3 dBu | PL-398 “27” | dBu - 398 = 27.3 | SW20 s-3 | PL-606 “30/21”
1290 | KZSB | 0.106 mV/m | 40.5 dBu | PL-398 “15” | dBu - 398 = 25.5 | SW20 s-1 | PL-606 “30/00”
1300 | KROP | 0.042 mV/m | 32.5 dBu | PL-398 “16” | dBu - 398 = 16.5 | SW20 s-1 | PL-606 “30/00”
1320 | KKSM | 0.371 mV/m | 51.4 dBu | PL-398 “18” | dBu - 398 = 33.4 | SW20 s-1 | PL-606 “30/00”
1330 | KWKW | 0.287 mV/m | 49.2 dBu | PL-398 “16” | dBu - 398 = 33.2 | SW20 s-1 | PL-606 “30/00”
1340 | KCLU | 0.118 mV/m | 41.5 dBu | PL-398 “15” | dBu - 398 = 26.5 | SW20 s-1 | PL-606 “32/00”
1360 | KLSD | 31.007 mV/m | 89.8 dBu | PL-398 “67” | dBu - 398 = 22.8 | SW20 s-7 | PL-606 “66/25”
1370 | KWRM | 0.108 mV/m | 40.6 dBu | PL-398 “13” | dBu - 398 = 27.6 | SW20 s-1 | PL-606 “32/00”
1430 | KWST | 0.033 mV/m | 30.4 dBu | PL-398 “14” | dBu - 398 = 16.4 | SW20 s-1 | PL-606 “27/00”
1450 | KFSD | 0.626 mV/m | 55.9 dBu | PL-398 “30” | dBu - 398 = 25.9 | SW20 s-3 | PL-606 “31/25”
1460 | KTYM | 0.083 mV/m | 38.3 dBu | PL-398 “11” | dBu - 398 = 27.3 | SW20 s-0 | PL-606 “25/xx”
1490 | KSPE | 0.138 mV/m | 42.8 dBu | PL-398 “16” | dBu - 398 = 26.8 | SW20 s-1 | PL-606 “25/07”
1510 | KSPA | 0.094 mV/m | 39.4 dBu | PL-398 “14” | dBu - 398 = 25.4 | SW20 s-1 | PL-606 “29/00”
1520 | KUNX | 0.392 mV/m | 51.9 dBu | PL-398 “14” | dBu - 398 = 37.9 | SW20 s-1 | PL-606 “25/00”
1540 | KMPC | 0.078 mV/m | 37.8 dBu | PL-398 “12” | dBu - 398 = 25.8 | SW20 s-0 | PL-606 “27/00”
1580 | KBLA | 0.279 mV/m | 48.9 dBu | PL-398 “10” | dBu - 398 = 38.9 | SW20 s-1 | PL-606 “27/00”
1590 | KVTA | 0.053 mV/m | 34.4 dBu | PL-398 “09” | dBu - 398 = 25.4 | SW20 s-1 | PL-606 “27/00”
1600 | KAHZ | 0.046 mV/m | 33.3 dBu | PL-398 “12” | dBu - 398 = 21.3 | SW20 s-1 | PL-606 “25/00”
1650 | KFOX | 0.277 mV/m | 48.9 dBu | PL-398 “06” | dBu - 398 = 42.9 | SW20 s-1 | PL-606 “25/00”
1670 | KHPY | 0.095 mV/m | 39.6 dBu | PL-398 “10” | dBu - 398 = 29.6 | SW20 s-1 | PL-606 “27/00”

Stephen said...

I mentioned the Panasonic RQ-SW20 earlier, and my qualitative "signal" reports. This, or its twin the RQ-SW10, was one of the first radios I had as a kid, at least the first one that I thought at the time was halfway usable, compared to another freebie that was essentially junk. This radio doesn't even have an S-meter, only a tuning indicator light that lights up when the received signal is over a certain level. Also, back then, I learned about the RST system of reporting signals - - probably from a book my brother (who's a ham but never goes on the air afaik) had.

So, my signal reports on that radio are based on the "S" part of RST, based on signal quality, and on stronger signals, splatter width. (I find the RQ-SW20 has selectivity comparable to the Sony SRF-M37W, with a little better falloff as you get farther off frequency from strong signals.)

What those signal strengths mean to me:
1 - Faint signal, barely perceptible - either a beating carrier is faintly heard, or some words can be distinguished in the noise if you're listening on headphones and don't have other noise around (ambient audio like a dishwasher, birds singing, etc)
2 - Very weak
3 - Weak
4 - Fair - The tuning LED lights on-channel at this level. Also here is where the perceived volume of the audio levels off; above here the noise volume decreases. The audio is also faintly heard +/- 10 kHz.
5 - Fairly good
6 - Good - station's audio still faintly heard +/- 20 kHz
7 - Moderately strong - a full-quieting signal; tuning indicator also lights up +/- 10 kHz, audio can be heard +/- 30 kHz
8 - Strong - tuning indicator also lights up +/- 20 kHz
9 - Very strong signals - tuning indicator lights up at least +30 kHz; weaker audio to +90 kHz or so, also the sound of the splatter closer in has a different character than on weaker signals. (In some extreme cases, like near transmitter sites, the audio close to the channel will be at a much reduced volume compared to on-channel or farther off-channel.)

At the time I came up with those ideas, I didn't even know such a thing as S9 + X dB even existed, nor did I know there was such a thing as S-meters calibrated to a reference microvolt level. If I was going to go outside the range, then an S-0 means the signal is undetectable, and an S-10 would mean the radio is overloading so that the audio on-channel is distorted. (This usually only happens, for example, within like 500-1000 feet of a 50 kW AM station's transmitter site.)

Also, right now my PL-398bt has a broken AM antenna, so the only signal that reads greater than 15/00 in the daytime is 1170 KCBQ. The only signals receivable at all daytime on it right now, from the above list, are 600 KOGO, 760 KFMB, 910 KECR, 1130 KSDO, 1170 KCBQ and 1360 KLSD.

So, I guess I interpret people's "s" reports a bit differently. For me, it's impossible to have an S-1 be an undetectable signal, and an S-9 means the front end of the radio is being overloaded, splattering & imaging across part of the band.

Also as you can probably see from the table, there's a bit of an overlap on the S-readings, and especially on weaker signals, there's a huge variation in offsets to what the PL-398bt says versus what the chart says the signal should be.


Hi Stephen,

Glad you liked the post. It's one that has been on my list to do for some time. And more coming in part 2.

Yes, I am in AZ right now. Most of those stations you mentioned in your other message I have received here. WWL-870 is pretty tough for some reason, but I have heard it. I'm only about 250 miles east of L.A. here, so I'm not so far from San Diego either. maybe 160 miles northeast?

Your 1-9 signal strength system is a fair assessment of strength. Hams used the old RST system which is still used today, though the "T" part is really only for Morse signals.

The Ham's definition:

1 -- Unreadable
2 -- Barely readable, occasional words distinguishable
3 -- Readable with considerable difficulty
4 -- Readable with practically no difficulty
5 -- Perfectly readable

1 -- Faint signals, barely perceptible
2 -- Very weak signals
3 -- Weak signals
4 -- Fair signals
5 -- Fairly good signals
6 -- Good signals
7 -- Moderately strong signals
8 -- Strong signals
9 -- Extremely strong signals

You have been talking about dBu levels. 'Also, re: my strong night skywave signals, I forgot to mention something. Signal levels of "48 - 54 dBu" (3.16 - 6.3 mV/m) are frequently/consistently seen'. Plus some extensive lists of received levels.

Not sure where you're getting these dBu figures, though I suspect it is off the Tecsun's display. I fear you are confusing "engineering" dBu (derived from mV/m) with Tecsun's dBµ displayed on the ultralight receivers. They are not the same. Also, mV/m is linear. dB... in any form is logarithmic. They will not follow each other in a linear fashion. In my latest blog post I'm trying to spell out the difference. A second part post will follow soon explaining "engineering" dBu.

The dBu in the list I sent you has no correspondence with what you are reading off your Tecsun display. They are different values, apples and oranges. One is voltage across a load (input to the tuned circuit - the Tecsun - they call it dBµ - Greek letter mu), the other (the list I sent) is field intensity at the receive location (dBu - lowercase 'u') and directly related to mV/m. I know it's confusing, but the difference is important.

Hope this helps.


Stephen said...

Hi Bill,

I look forward to reading part 2. :)

I have a friend who's getting married the last Sunday in March near Tucson, AZ, so my parents and I are planning to go to the wedding. I don't know how much opportunity I'll have for attempting DX, but there may be a little bit of time. The wedding is in the late afternoon that Sunday, then dinner afterwards. Then we're staying in a motel and coming home the next day. (I'd be okay with driving through the night some, but my parents would rather stay overnight.) I think we'll be taking I-8.
I'm not yet sure what radios or antennas I'm taking with me, but I do plan to take a few. Almost all of my Tecsun radios are currently in some form of disrepair, but I do also have an SRF-59, filter-modified SRF-M37W, CC Pocket, Superadio III, and a few others, plus a Select-A-Tenna. (I may also consider buying another radio or two but I'm not sure which ones yet. Hopefully I can find a good deal at Rat Shack on the Traveller III, for example, before they're gone.)

I've heard WWL a couple times here too, although KRLA is usually the more common one. It's no blowtorch though, of course, and on lesser radios XEMO covers it up. I guess 160 miles, maybe a bit more, sounds about right. IIRC there's a sign on I-8 as you're going around El Cajon that says about 160 or 162 or so miles to Yuma, and that's taking into account curves on the freeway.

Ahh, well it's basically how I interpret them. Although, sometimes I think people interpret them a bit differently than I do. :) For example, more recently (as in the last few years or so) I've gotten to thinking that quite a few hams/DXer's scales reach S9 at a much lower level than mine does. Also, sometimes, I see posts where someone mentions the noise level is S6, and that's foreign to me.
To me, S0 *IS* the noise level, no matter what the actual received field intensity is (or voltage across the antenna).
Below S4 is where the AGC usually is no longer able to keep the station's audio "loud", also sometimes this is the threshold where a tuning indicator lights up, or a scan stops.
Around S7 is where you've maxed out your S/N ratio.
Anything over S9 means you're overloading your radio, like what happens in these videos:

Also, I'm thinking maybe I need to adjust my scale UPWARD a bit. As it is now, there's a pretty big gap between the minimum threshold for S9, and how high S9 can go. Maybe I should do some shuffling so that S7-S9 become S6-S8, then S9 would be a stronger signal like what you'd have a mile or two from a 50kW blowtorch like KFI across a saltwater path. :) But, there's still the issue of having them conform to the descriptions. I wouldn't call a signal strong enough to stop a scan a "weak" signal, necessarily, and I normally call a full-quieting signal a "moderately strong" signal. (An "extremely strong" signal of course means the station is splattering or desensing or imaging across a significant portion of the band.) Hmm....

Stephen said...

Yes, those are off the Tecsun's display. I do realize they're not the same, and I'm aware of the relation between mV/m and dB. For a very simplified version, 0 dBu (I can't type the other character on this computer) is 0.001 mV/m), and for every 20 dBu you add you multiply the mV/m by 10, and every 6 dBu you add is a doubling of the mV/m. 60 dBu is 1 mV/m, and 120 dBu is 1 V/m. (Around 180 dBu is about where I'd think a well designed radio would start to overload, like start to splatter the signal onto the first-adjacents.) Did I get that close to right?

And, yes, I've pretty much found that out, that the Tecsun display can be pretty far off compared to reality. :) At first I was thinking that there might be a conversion factor, and I've found one in my situation, but it only works across a limited range. Basically you'd add like 21 to the Tecsun's number to get the real number, but it really only works in situations like at home (where my strongest signal is around 80 dBu displayed or 102 dBu received or so), and only down to about medium strength or so. I think somewhere around KNX's or KFI's strength going down, if not before, it starts to fall apart, and when you get down to fringe stations like KFWB, or weak signals near blowtorches like KEIB, that "formula" falls flat on its face. I don't have the chart open in front of me, so I can't remember if the gap widens or closes, or more likely goes completely erratic. And of course if I was close to a somewhat strong signal (like several tens of volts/meter), there'd be a lot of skewing there.

I look forward to that second post. :)



Hi Stephen,

120 dBu is indeed 1 volt per meter. 180 dBu is 1000 volts per meter. I should say the receiver would overload, it might be fried.

The Tecsun display is measuring a voltage across its receiver input and converting its magnitude to a logarithmic (dB) value. It is totally a different animal than the mV/m-dBu discussed above. I would expect Tecsun's voltage reading is fairly accurate. In the practical sense, you can't convert between Tecsun's dBµ and the dBu derived from millivolts per meter. Too many unknowns. One is a voltage across a load, and one is a electrical field. Another difference is something called Antenna Factor, or the ability (actually efficiency) of the antenna to convert the passing E-field to an electrical voltage which can then be received. As each antenna (loopstick in this case) is different, each will transfer a different signal voltage to a radio's input.


Stephen said...

Ahh, I wouldn't be surprised to see a receiver overload at 1 kilovolt per meter. :)

I think I've mentioned something before about inductively coupling a Select-A-Tenna to a utility groundwire. :)
Well, here are a couple examples caught on camera. :) -- PL-398mp -- @ 32-45-43.6 N, 116-56-45.2 W -- receiving 760 KFMB nighttime, both barefoot and with "big" antenna, about 7.2 miles from KFMB -- PL-398mp -- @ 32-50-33.8 N, 117-1-32.0 W -- receiving 760 KFMB nighttime, barefoot

The signal at the location with the "big" antenna is about the same strength as the barefoot signal much closer to the station. -- PL-606 -- @ 32-45-43.6 N, 116-56-45.2 W -- receiving 1170 KCBQ daytime, both barefoot and with "big" antenna. about 9.2 miles from KCBQ -- PL-606 -- @ 32-53-37.9 N, 116-55-40.5 W -- receiving 1170 KCBQ daytime, both barefoot and with "big" antenna

The "big antenna" signal at 9.2 miles distant was actually stronger than the barefoot signal at the closer location. :)
Also, the signal on that last one near the end, when coupled with the longwire antenna, was SO strong that the radio actually quit responding to my keypresses. I had to remove the batteries and leave them out for several minutes before I could get it to work again.

Something I'm wondering about ... is it possible to guesstimate the antenna gain based on that info somehow? (For example, calculate the difference in field intensity at the two locations.) I've sometimes seen upwards of 50-60 dBu difference in lesser signal environments (I think I linked a 1550 XEBG video in another comment), but I still think those utility groundwires have much more gain. And what might the field intensity be at the location where I'd get that same effect that I got on the PL-606 near KCBQ, if I was using the radio barefoot with the loopstick antenna desoldered? :o


Hi Stephen,

You asked:

'Something I'm wondering about ... is it possible to guesstimate the antenna gain based on that info somehow? (For example, calculate the difference in field intensity at the two locations.) I've sometimes seen upwards of 50-60 dBu difference in lesser signal environments (I think I linked a 1550 XEBG video in another comment), but I still think those utility groundwires have much more gain.'

You may be increasing your overall signal by attzching or coupling to a utility ground wire, but you are introducing a tremendous amount of garbage, noise, etc. to your antenna. It woulbe be impossible to calculate any gain, etc. under that circumstance.

You asked:

And what might the field intensity be at the location where I'd get that same effect that I got on the PL-606 near KCBQ, if I was using the radio barefoot with the loopstick antenna desoldered?'

If you desolder the loopstick you have removed the tuned circuit from the front end of the radio. Technically, you should get no signal at all.

Under normal circumstances, you could calculate the voltage gain by subtracting the two different dBµ readings on the Tecsun's display. The resultant dB figure is the voltage gain.


Stephen said...

Hi Bill,

Yes, I do know I get a ton of noise when I do that. :) One of these days I'd like to learn how to build a portable antenna that has the same dB gain without the noise, by not being something hooked up to utility services. One idea I'd like to try is an easily-deployable tunable beverage-on-ground antenna. I would have the antenna wire stored on one of those orange cord reels. (The ones that store up to a 100-foot electrical cord, although with the much thinner gauge antenna wire you might fit closer to a half or full mile of wire on it, but idk yet.) I would roll it out on the ground (up to however much room I have wherever I'm DXing), then inductively couple it to a portable radio's loopstick somehow. I'd also like to build a large (between 4 and 10 foot diameter or so) tuned loop (preferably with a "shape factor" similar to the SiLabs DSP chips - flat tops and near-clifflike sides) that's collapsible to fit inside a backpack. I kind-of like the idea of Gary Debock's PVC loops that he's mentioned on the ultralightdx group, but I'm wondering if there might be other better ideas. (A crate loop or that type wouldn't work for me, as something that's a foot or two wide just wouldn't work for me, it'd be nigh impossible to store it.)

Well, I have done the desoldered loopstick experiment on one of my Tecsuns - the PL-398mp. I somehow managed to do something wrong when reattaching the leads one time, and zapped the chip I think, but before that, I took it to CA-52 at Mast Blvd, on the onramp from Mast to 52 east, parked on the side, and was able to hear KFMB with a fairly good signal (iirc reading somewhere between 55-75 dBu or so), even with the loopstick disconnected. And yes, it still would tune (so KFMB didn't leak across the entire band), and true, I didn't pick up any other signals at that site, even though KSDO is only a couple miles away.

Well, in my case, subtracting the dBu readings doesn't quite work as well as I'd like. My radio desenses, which I think is compressing the range. For example, I don't have a video of it, but there was one time I was able to get 1550 XEBG to temporarily spike up to 98 dBu (from about 30 barefoot), and it immediately dropped down into the 80s due to the desense.
I've noticed that if I use the strongest signal as my antenna gain check, it doesn't desense the same way as others do, although it too does. If I was to use a weaker signal, especially one close in frequency, then the strong one would desense the other signal, reducing the gain. This can be seen on some other videos that I've tried the utility ground gain trick. At Pauma Valley, CA, I got around 50-52 dB gain on several stations. (The strongest one there, 690 XEWW, is in the low 40s barefoot.) OTOH, at home, stations like 740 KBRT, 990 XECL/KTMS, 1070 KNX, 1280 KFRN only got like 30-35 dB or so.

Also when I've tried my tests of comparing big antenna reception near home vs barefoot near the transmitter site, it completely pegs the meter on the radio. (Maybe that'd be just a weee little bit outside the scope of "normal circumstances"? ;) especially considering that in one case, at a site near my house, 1170 KCBQ shows 97-98 dBu on the 2nd harmonic!) That's why I was thinking that if I could figure out the field intensity at the two sites and calculate the difference between them, I'd have some basic idea of the antenna gain.

Stephen said...

And I'm guessing that if I was close enough to one of KCBQ's towers to have a 98 dBu reading on the 2nd harmonic with my antenna disconnected, I'd have a lot more to worry about than frying my radio. ;) Although, a few months ago, I was with my dad and an on-site KECR engineer in their transmitter shack attemping to troubleshoot an issue where KCBQ's audio was getting into the background of KECR's program, and I was close enough to the transmitters that I could touch KECR's TX with one hand and KCBQ's with the other, and both were on the air. I forgot to pull my radio out and do a selectivity test, although I had it with me. I did notice that in the studio building about 50 feet away (southeast), the signal for KCBQ was considerably weaker, even though I was only maybe 100 feet from a tower or so, than it was out at the street at least 300+ feet from a tower. Also when I was outside between the buildings, KCBQ wasn't as strong there as out at the street.



Hi Stephen,

One good way I've found to couple a very long wire to a pocket radio is through a Q-stick type device. Take the end of the longwire and wrap a few turns around the end of the Q-stick ferrite if you have room. Place the Q-stick near the radio's internal ferrite and tune the Q-stick to the frequency the radio is tuned to.

Now, the Q-stick will pick up signal on its own, so make the Q-stick small. If you have a way of enclosing the radio and Q-stick in an RF proof enclosure, even better. I'm in a metal skinned trailer here in Arizona, and it works perfectly as a shielding device. I bring the longwire through a window and the Q-stick and radio sit on the kitchen table. The radio and Q-stick are pretty well dead on their own because of being inside the metal trailer, and the signal from the longwire is then coupled to the radio without the additional signal pickup.

I have experimented with loops as big as 4 ft. They are amazing in the amount of signal gain. I have a little 12 inch loop made out of PVC that I carry with me for traveling. I would guess it's gain is about the same as your Select-A-Tenna.

In order for the dBu subtraction for gain calculation to work properly, you really must not be in desense or have extranneous garbage coming in like from that pole groundwire. :-)

Here's a formula you might be interested in. You are always wanting to know the field strength right up close to a tower. OSHA has a formula they use for what they call "IMMUNITY TESTING". It only works for close up distances in terms of meters.

For Immunity measurements, the generated electric field strength can be calculated by:

FS(V/m) = SQRoot(30 * P * g) / d


FS = Approximate Field Strength in (V/m)

P = Power in watts

g = Numeric Gain

d = Distance in meters

Use a value of 1 for the gain.

For example, a 50 KW tower at a distance of three feet results in a field strength of 408 volts per meter. Not millivolts per meter but V/m!


Stephen said...

Hi Bill,

That might be interesting to try. :) I’d also need to find some decent, but not too expensive wire that can withstand various kinds of weather that I might be deploying it in, and a cord reel to store it and make it a little easier to string it out. I was thinking something like one of those Home Depot (or wherever you get them) electrical cord reels, the kind that fit a 100-foot electrical cord, although if I wanted a particularly long antenna maybe I might need to consider those crank things that you store garden hoses on. :) And then, where am I gonna find several miles in an urban area (like a city over 1 million population) to string up an antenna :p

As for the ferrite stick with which to make the Q-stick, maybe I could sacrifice the one out of my Coby CX-70? It was pretty cheap at Fry’s, IIRC - only a few bucks or something like that, and as you can probably see from the video, there isn’t much room for the internal antenna. :)

Also, maybe now that I’ve gotten a car (an ‘02 Honda Civic with just over 60k miles on it), I could use it as an RF shield for my radio & Q-stick. :) Also the car will open up possibilites for selecting DXpedition sites, compared to relying on the local transit bus system. :) (More on that, and a suggestion/request, in a reply to another post.)

I’d like to do some experimenting with loops too, just having varying plans of how to build them, etc, kinda confuses me as to what type I should try. :/ I’d like something that’s fairly easily storable, and isn’t too huge on its width if possible, maybe something with similar proportions to a Select-A-Tenna or Terk loop or similar, just scaled up somewhat. I like the relative proportions of Gary DeBock’s PVC loops, for example, but my dad’s a bit concerned that the PVC wouldn’t hold up all that well in the 100+ degree temperatures we often have in summer - in the shade. (and I might be using them out in the open for mid-day DXing in May-September, too.)

How would I set it up to calculate the gain properly, then, without going to the mid-Pacific? It seems based on my preliminary experience / calculations, that the SAT + Powerline antennas have the potential to bring a signal up from nearly unreadable to moderate to severe desense.

That’s an interesting formula :) although it does seem to give different values than what I’ve been figuring.

One method I’ve been using in a pinch for close-in distances is the inverse field rule, where:

Field_Strength (mV/m) = Field_at_1_km * (1000 / distance_in_meters)

For example, a 50 kW station with a half-wave tower may have a field of approximately 2,700 mV/m at 1 km. To figure the strength 5 meters from the tower, I calculate as follows:

2700 * (1000 / 5) = 540,000 mV/m or 540 V/m

The maximum exposure limit in most of the AM band is 614 V/m, per FCC 1.1310

Also, OET Bulletin 65 gives more guidelines on fields near a tower, & supplement A is particularly applicable to broadcast stations.

In skimming through those, there appear to be several formulas in there. I wonder if any of them may help figure out how to implement close-in fields in your program?

Referencing the inverse field rule example above (or whatever it’s called), and the OET Bulletin, the predicted distance for compliance with the FCC limits for a 50 kW station with a half-wave tower is 4 meters. My example above gave a field intensity of 540 Volts/meter at 5 meters, which is a little below the limit.

The closest I’ve been to a live AM tower is probably about 2 meters, from one of 590 KTIE’s 3 towers, a 2.5 kW AM station.



Hi Stephen,

Long lengths of copper wire can be pretty pricey. Long lengths of iron wire can be had for fairly cheap. Hardware stores sell rolls of these they call "tie wire", approximately 250 ft. rolls and about 6 inches in diameter. It's cheap, not as good as copper, but doable. Not insulated though. I've used it before. If you are careful not to kink it when rolling it up, it should work. I have also stripped down 4 conductor telephone wire and pieced it together. The gauge is fairly small but workable, and it's insulated. That way you get 4 times the length you buy. It's pretty cheap too.

I've extracted numerous ferrite sticks out of old radios. You can even superglue them together, end to end, to make long ones. It works quite well!

Congrats on the new car! I feel a DXpedition coming on.

I've had good luck with PVC loops. They can get a little flimsy if larger than 48 inches if you don't use a large enough diameter pipe. I've had no trouble with them in the desert heat. My 12 inch one is made from 1/2 inch PVC. Square.

The only way I would attempt to calculate gain is in a very quiet and somewhat distant location in order to maintain signal levels that were not too high. High signal levels would skew the results. Even so, you are only going to get an approximate gain. Be careful not to be in desense, or anywhere near it. Calculating the gain off of the Tecsun dBu display, you could arrive at a dB gain and from that calculate the voltage gain for the loop over the internal ferrite.

I checked the maximum exposure formulas used in the FCC document links you posted. Looks like they arrive at their result using power density (m/W per cm^2). The basic end-result is pretty close, and very much in the same ballpark. These type of formulas are only supposed to be used out to a few meters. I don't have plans on including them. They are easy to figure in any event since they don't need any engineering info more than transmitter power and antenna gain...though array gain really doesn't matter if you are standing 3 meters from a 50 KW tower.


Stephen said...

Hi Bill,

Yeah, I can guess they'd be quite pricey. :) For example, if I remember last I checked a few years ago, bass strings in pianos, just the string not the installation, were like 15 or 20 bucks each! (A set, at least at the place I bought the strings for a piano I was working on several years ago, was around $250-300 or a bit more, which did work out to somewhat less per string.) So, yeah, it can be expensive. :)

Ahh maybe I might want to look into one of those other types of wires. :) The vast majority of the time, I'd be only setting it up (rolling it out), DXing, then when the session is done, rolling it up & putting it away. (One place I might use it is the elementary school playground adjacent to my house, where there's close to 1000 feet of room east and west.
What kind of prices might I be looking at for those wires, maybe? (Of course regional differences may apply but what might be a ballpark range?)

Well, my next ferritectomy performed will be my first. :)

Thanks! :) I haven't decided where to go yet for my first one. :) I want to find a place that's not too far away where even the strongest signals are pretty weak, and not in some cave somewhere. :) One place I'm considering is Borrego Springs, but from what I remember, 970 KNWZ had a fairly potent signal there. It was like 10 or so years ago, but from what I remember it doing on a radio I had then and still have (and comparing it to a Tecsun now), I'd be guessing if I had a Tecsun at Borrego, KNWZ would be up around 45 to 50 dBu on the built-in antenna. :/ As for other places I've been, at Pauma Valley, 690 XEWW clocks in at around 42 dBu, and at Cameron Corners, 1390 XEKT was around 47 to 48 dBu or so. :( I'd prefer to find a place where the strongest barefoot signals are in the upper teens to mid 20s, and with a 12" loop like my SAT, maybe a couple hit 39 or 40.
One thing I might even consider doing, is plot out a range of where I might go, then in the mid morning (like between 8:30 and 10 am) on that day I go, drive around monitoring several projected pest frequencies. Where I find a place where the pests are weakest (without it being some condition that knocks out the desired DX), that might be where I'd sit. :)

Ahh that's good to know. :) Yeah, I think I remember either Gary's article or something like that. I'm not sure what size I'd start with, yet. I think the largest I'd want to use in the house would be 6 feet per side (because of the 8 foot ceilings and the 80-inch doorways, so it'd fit.) If I was using it on a tabletop, I'd limit it to about the size of a box fan or a little larger, like 24 or 36 inches. For an outdoor loop, I'd go larger like 8 or 10 or maybe even 12 feet, maybe more, idk yet. :) I'd like to build one sometime that has the gain like the powerline has (I'm guessing minimum 50 dB, possibly 60 or 70, maybe 80 dB if it wasn't desensed so much?) But I'd also need it to be fairly high Q, and preferably a good shape factor so that good selectivity, within reason, doesn't come at the cost of super muddy audio. For example, if I could hear KSL at armchair copy in the middle of the day in summer (if XEQIN was off the air) from near KCBQ's transmitter site, that would be awesome. :)

Stephen said...

Yeah, I've been wanting to find that quiet location to calculate gain. I find that even if I don't use the strongest signal to calculate my gain, it will still affect my results, often more so than if I used it. For example, here at home, 1130 KSDO and 1170 KCBQ are only 5 dB apart, but KCBQ overloads fairly severely, and KSDO barely has much gain. Even in rural areas, the strong signals (like XEKT in Cameron Corners or XEWW in Pauma Valley) can be pests, although I have recorded a few videos at PV that show around 50 dB of gain.

I've generally found that once signals get up around 80 dBu, the radio is starting to desense fairly severely, and in the low 80s it starts really compressing the RF dynamic range. For example, I might be able to be several miles from a 50 kW site reading about 80 dBu. As I go closer, it might go up...84...86...88... then snap back down 83...86 ... snap down again...82...85....89....86...88....83...87... then within maybe a mile or so it starts running up in the 90s but stil snapping down, like 93...88....91...87..84...87...92...88 and so on... until I'm within, say, 1/4 mile of the site or so, then finally it gets to its limit of turning down the gain and just continues rising...89...91...93... at around 95 the audio starts to distort on-channel, and the meter pegs at 98 dBu.
The most severe overload I've ever had is using the powerline & SAT outside KCBQ's site, and I think I've posted a video of that in another comment. :)

Ahh okay. :) So I guess my simplified formula I mentioned could be close to the ballpark up close? :) (Also idk if you've seen my list of suggested sites for a few abridged station list in another post, but a couple of them, marked for "selectivity testing" are within several feet of a tower. :) )



Hi Stephen,

Actually I think they sell that iron tie wire by the pound. 3.5 lbs. will run you about $7.50. I think that's about 250 ft. worth. Check Home Depot or Lowes or even Ace.

I've been to Borrego Springs. Nice little town. The desert national monument there is pretty cool. It should be a quiet location, and far enough away from L.A. and San Diego to be in a fringe zone. The closest station is KNWZ (5 KW), 32 miles away at Coachella. Closest 50 KW station is KCBQ at 40 miles.

You might be surprised at how much gain a 48 inch loop has. Never tried anything larger. Remember that the signal capture area is a function of the area of the loop. 12 inch loop = 1 sq. ft. 48 inch loop = 16 sq ft., 16 times the capture area. The Q on loops is usually pretty high. Keep the windings evenly spaced for best results. I also like to keep the winding width small, i.e., approximately one wire diameter between coil turns.

I did see your list of DXing sites. I'll work something up for you on one of them at least.


Stephen said...

Hi Bill,

Thanks for the suggestion on the wire. :)

I'm thinking I might go to Borrego again, possibly as soon as Friday but idk yet. (I'm repairing one of my radios right now.) (I'm a little worried about KNWZ's blowtorch strength, though. From what I remember, last time I was there it was strong enough to brightly light a tuning indicator on a not-so-sensitive radio. I have a bit of a fear the strongest signal there, probably KNWZ, would exceed 0.5 - 0.7 mV/m by a large margin.)

I think maybe I should start with a 48" loop and see what happens, then. :) I guess keeping the width small would help the Q, so maybe I could get KSL or KERN next to KCBQ without much splash in the daytime? And would the iron wire work, or should I take another look at Gary's PVC loop idea? (And there's how to construct it too, so many different loop plans, yours, his, others... information overload in my brain :o )

One idea I had for my first mini-DXpedition (like maybe as soon as Friday), was to go on a route, periodically checking the strength of stronger stations along the way, in hopes to find a suitable site without a too-much-blowtorch signal. I was thinking starting east on I-8 to Pine Valley, then north on CA-79 to Julian, east on CA-78 then north to Borrego Springs, west through Ranchita to Warner Springs, south by Lake Henshaw to Santa Ysabel on CA-79, southwest on CA-78 and 67 through Ramona, east of Poway, by KCBQ's site, down into El Cajon then home. I'm guessing the driving should take about 3 or 4 hours, and stopping to check the "beacons" at each site could maybe be done in a couple minutes each (especially if I video the check instead of trying to write down each signal strength). Hopefully I'd be able to complete the circuit in about 6 hours or so. Once I found a suitable site, I'm thinking I'd plan to go back again another day and spend a few hours there with my radios & antennas. :)

73, Stephen


Hi Stephen,

Your Borrego Springs DXpedition sounds interesting. KNWZ should put in about a 2.44 mV/m signal there in the daytime at 5 KW. Local, but not too bad. I'll send you a couple of charts based on that location.


Stephen said...

Hi Bill,

Thanks for the Borrego Springs list. :) Btw I can import them into an office suite (in my case LibreOffice, although I think Google Docs/Drive may also work, and MS Office also works on my dad's computer), and sort them any way I like, so I only really “needed” one version. Thanks anyway though. :)

So yesterday (still feels like today since I haven't gone to bed yet) I went around to a few places to check radio signals there. Unfortunately I got a late start, so I was only able to check two places – Pine Valley and Borrego Springs. I did a quick check in Pine Valley (32.822153, -116.528829) around 2pm
I did a little local shopping in Julian, and decided to skip my radio check there so I could get to Borrego, arriving there (33.256999, -116.375139) and starting the bandscan with the barefoot PL-606 around 4pm. Some skywave, especially in the upper end of the band, was already starting to come in, but I was at least able to check the approximate strengths of the stronger pests.

After Borrego, I just headed toward home as it was getting to be too late to do any checks. In spite of my touching 80-85 a few times on a couple rural 2-lane highways when there was almost no traffic, it was already nearing sunset by the time I got to Santa Ysabel, and starting to get dark when I was going through Ramona. I wanted to check a few things around the KCBQ/KECR site, but it was starting to rain and one of the roads is a dirt road that I'd have to navigate to get to one of the spots, so I figured it'd be better to wait for drier weather.

Anyway, here's somewhat of a list from the Borrego Springs bandscan. In cases where the display was fluttering due to co-channel, I'm generally only listing one of the readings per station.
A 12-minute video (all the time I was able to take) is posted at:

540 – XESURF – ? mV/m – “23/07”
560 – KBLU – 0.34 mV/m (50.6 dBu) - “18/12”
590 – KTIE – 0.1 mV/m (40.0) - “15/02”
600 – KOGO – 1.12 mV/m (61.0) – “28/22”
620 – XESS – ? mV/m – “18/13” - co-channel beating (KTAR suspected but not ID'd, BUT, KFYI was a no-show on 550)
640 – KFI – 1.08 mV/m (60.7) – “21/15”
690 – XEWW - ? mV/m - “28/24”
(The absence of KALL is due to not having time to use the SAT; also being so late in the afternoon it would have been skywave anyway.)
710 – KSPN – 0.355 mV/m (51.0) - “15/02”
731 – internal het – n/a mV/m - “26/19”
740 – KBRT – 0.45 mV/m (53.0) - “15/03”
760 – KFMB – 0.96 mV/m (59.6) - “27/21”
790 – XESU - ? mV/m - “35/25”
800 – XESPN - ? mV/m - “15/08”
820 – XEABCA - ? mV/m - “20/16”
830 – KLAA – 0.63 mV/m (56.0) - “22/18”
840 – KXNT – 0.11 mV/m (40.5) - “15/10” - suspect skywave, as 720 KDWN was absent
(may have accidentally skipped 850 XEZF, or at least forgotten to un-null the antenna rotation)
860 – XEMO - ? mV/m - “21/15”
910 – KECR – 0.11 mV/m (41.0) - “19/12”
910 – XEAO - ? mV/m - “26/21”
920 – KPSI – 0.16 mV/m - “15/08” (43.8) - co-channel (XESDA?)
940 – XEMMM(?) - ? mV/m - “20/15”
950 – XEKAM - ? mV/m - “15/03”
960 – KIXW – 0.073 mV/m (37.3) - “15/01”
970 – KNWZ – 2.44 mV/m (67.8) - “41/25” - a bit too big of a pest, but similar to what I remember from several years ago.
990 – XECL - ? mV/m - “36/25”
1000 – KCEO – 0.16 mV/m (44.1) - “15/02”
1010 – KXPS – 0.124 mV/m (41.8) - “25/09”
1030 – XESDD - ? mV/m - “15/02”
1050 – XED - ? mV/m - “43/25” - turns out 970 KNWZ got edged out by this station as being the biggest pest
1070 – KNX – 0.3 mV/m (49.4) - “15/06”
1090 – XEPRS - ? mV/m - “19/15”
1110 – KDIS – 0.29 mV/m (49.1) - “15/01”
1120 – XEMX - ? mV/m - “22/17”
1130 – KSDO – 0.27 mV/m (48.5) - “15/08”
1140 – KNWQ – 0.35 mV/m (50.9) - “24/18”
1150 – XERM - ? mV/m - “25/14”
1170 – KCBQ – 0.34 mV/m (50.5) - “17/12”
1190 – XEMBC - ? mV/m - “22/17” - co-channel
1200 – KPSF – 0.22 mV/m (47.0) - “26/22”
1210 – KPRZ – 0.43 mV/m (52.7) - “17/11” - co-channel

Stephen said...

skywave interference is getting more obvious here, so I'll only list a few stronger/closer signals to finish the Borrego Springs list…
1230 – KXO – 0.167 mV/m (44.5) - “30/21” - co-channel
1300 – KROP – 0.227 mV/m (47.1) - “39/25” - co-channel
1400 – KESQ – 0.133 mV/m (42.5) - “23/08”
1430 – KWST – 0.13 mV/m (42.2) - “28/10”
1470 – XERCN - ? mV/m - “22/19” - co-channel
1490 – KGBA – 0.065 mV/m - “23/21” - co-channel
1560 – KNZR(?) - ? mV/m - “23/21” - skywave suspected
1630 – XEUT - ? mV/m - “22/15” - co-channel
1670 – KHPY – 0.06 mV/m (35.6) - “28/25” - co-channel
1700 – XEPE - ? mV/m - “28/22” - skywave suspected

I hope to return, maybe to Borrego, or maybe to another site that's farther from some of the pests, in the not too distant future. Also hopefully I'll be able to get an early start in the day and spend a few hours there instead of just a few minutes. :)

Also, here's the Pine Valley quick scan info (32.822153, -116.528829, ~1:55pm) …

600 KOGO – 34/25
640 KFI – 28/18
690 XEWW – 36/25
740 KBRT(?) - 17/06
760 KFMB – 41/25
800 XESPN – 28/21
830 KLAA – 24/18
860 XEMO – 35/25
910 XEAO/KECR – 18/08
950 XEKAM – 25/21
970 KNWZ – 25/9
1030 XESDD – 25/18
620 XESS – 29/24 (a couple times I realized I forgot to check a station so went back to check it)
1040 KURS – 16/11
1050 XED – 25/13
990 XECL – 16/06
1070 KNX – 23/18
1090 XEPRS – 28/23
1110 KDIS(?) - 15/03
1130 KSDO – 23/14
1150 XERM – 15/01 (co-channel?)
1170 KCBQ – 31/25
1190 XEMBC – 15/02 (co-channel?)
1210 KPRZ – 25/19
1240 KNSN – 23/19
1270 XEAZ – 24/19
1280 (KFRN?) - 15/00 – I was under powerlines and behind a restaurant in a parking lot, so I suspect my noisy location blocked out KFRN. I've heard it in Cameron Corners which is farther south and a little east, though, so I suspect it could be heard in Pine Valley if I'd picked a quieter spot.
1300 KROP – 15/04 (co-channel?)
1310 XEC – 20/16
1360 KLSD – 29/24
1390 XEKT – 35/25
1420 XEXX – 31/17
1450 KFSD(?) - 15/01 (co-channel?)
1470 XERCN – 34/25
1550 XEBG – 20/14 (co-channel?)
1580 KBLA(?) - 15/02 (co-channel – KMIK?)
1630 XEUT – 30/21
1670 KHPY(?) - 15/02 (slight early-afternoon skywave?)
1700 XEPE – 30/25

Also, re: checking a few spots around KCBQ/KECR's transmitter that I mentioned earlier … I had another idea of something I've wanted to try. (I've kind-of done it, but need to do it again due to insufficient/unreliable data or something.)
That would be to look up the monitor points and field strengths specified in the licenses of some AM stations, then go to those spots at those times and see what readings I get on my Tecsun radios, and see how much they differ from the actual strength the FCC specifies for that spot.

I think it'll be kind-of hard to come up with some conversion formula to convert indicated signal strength to received field strength, though. I've discovered so far that there's a somewhat small range in which a conversion factor works for *my* radio, but it's definitely not linear. For example, it gets off-kilter when I'm near a really strong signal (indicating well into the 80s dBu), it also seems to be a bit wider gap when I'm in a somewhat weaker signal area (for example Borrego Springs), and wherever I am, any attempted “conversion/calibration” completely falls apart on weaker stations.

73, Stephen

Stephen said...

Forgot to post the link for the video of the Borrego bandscan. (Actually, it's STILL uploading - I think it hung up, says like 1000 hours remaining and it didn't take long to get to 10%, so I'm restarting it.)

Stephen said...

Ok I had to try a couple different things, but the Borrego PL-606 12-minute video is posted now -



Very interesting, and thanks for the bandscan list and video. The signal strengths correspond pretty closely in relative signal strength to the calculated values.

The one I have conflicts with here is Yuma's KBLU-560. The calculated value is fairly high (1.19 mV/m), but it's actual strength on the PL-380 here is fairly weak to moderate at 19/20 on the meter. Some pretty rough and mineral-laden terrain is between here and Yuma (am still in QTZ).


Stephen said...

Ahh, that's interesting, Bill. IIRC, a ~1.2 mV/m signal would theoretically indicate somewhere around 36-40 dBu or so on the meter, it seems, assuming other factors like you mention didn't skew the results.

I had a couple other ideas for maps, and idk how hard it'd be to generate them....

One would be fairly close into some transmitter sites in SoCal, showing the day/night groundwave plots, to get an idea of how much of a change there should be in signal when the pattern changes. I'm guessing that there's spots east of, say, 910 KECR, and especially 1130 KSDO and 1170 KCBQ, where if I sat there with my radio even just a few miles from the site at local sunset, the signal would go from fairly severe front-end overload to only a medium strength signal. I'd like to go to one of those sites (for example a particular direction/distance east of KSDO, haven't determined yet), and monitor a second-adjacent signal, preferably one whose sunset is at a different time so their change doesn't skew my experiment. (In KSDO's case I'd likely pick a spot where I can null KCBQ.) I bet that the difference in desense could make a signal go from undetectable to armchair copy! :)
TBH I think the best spot for that experiment would be next to 1550 KUAZ's transmitter :) It's a bit far of a drive for me, though. Although... my parents and I are planning to go there for a wedding the end of March and are staying overnight, so maybe the next morning I might want to see about going by the site and doing the experiment there. (Now what station to monitor... KFBK? KMIK? XERF?...)

My other idea is a general contour map - different than the one I mentioned on another comment, but more like the ones you've done, with a twist. Instead of just showing the X.YZ mV/m contour for each station, draw the areas around each one where at night they should pretty much dominate the channel, and shade areas that would suffer co-channel interference. For daytime, something similar, although maybe adding some kind of shading where the signal is below a few certain thresholds, like ranging from clear in an urban area on a portable radio, to ID'able but faint in a wilderness with a good radio and antenna.

73, Stephen

Stephen said...

Oh, I just thought of something re: a possible desense comparison experiment near KUAZ, that might help me a little. :)

Would it be possible to come up with a list showing...

signal strengths (maybe from 1400 to 1700 or just 1490 to 1600 where the nearest-frequency locals either side are; minimum maybe 0.1 mV/m - although I'd mostly be interested in like 2 mV/m or higher, the lower number would be to check how "crowded" the channel is, I prefer picking one where one station dominates; if skywave applicable then 50%),

at around sunrise (6:30 AM MST at KUAZ in March, I'd be there on the 30th),

from the KUAZ tower site (32.3727034, -111.0986553 - I obviously won't be "right there" but plan on trying to be close enough so that KUAZ would be reading between 85 to 93 dBµ on-channel and desense would be 50/00 within a few channels or farther either side of it, like somewhere between 700 mV/m to 7 V/m or so)?

I think it could help me decide which station to monitor near the site to do an experiment I'd like to do, and would be appreciated if possible. :) You know how these Tecsun radios tend to desense near strong signals, right? Well KUAZ is 50 kW, daytime only, so it'd be a pretty big contrast in desense when they power up. :) (I think a better time would be when they power down the previous evening, but I'll be busy somewhere else then.)

I was just looking at some FCC data, and a few stations I'm thinking might be possibilities may be 1510 KCKK, 1520 XEEH, 1520 KOKC, 1530 KFBK, 1540 XEHOS, 1540 KZMP, 1560 KNZR (although I'd rather not do first-adjacents cause I want a frequency that has little or no splatter from KUAZ when it's on), 1570 XERF, 1570 KPIO, 1580 KMIK, 1580 XEDM or 1580 KREL.

73, Stephen


Hi Stephen,

Thanks for that mock-up graphic of of the signal contours, and the color codes. I tore the program apart over the last week and came up with something similar, using your general color scheme. It now has the ability to plot up to 7 predetermined signal levels and 10 additional custom plot levels (17 in all). That should be enough.. more than 4 or 5 plots on one map gets confusing anyway.

The program has a local, non-Google sandbox mapping mode, which of course I've never mentioned. It provides for comparing different stations at different levels, even adjacent channels. Actually, the Google map driver can do about the same thing.

I could send you a text chart for the Tucson location if you'd like. It would be based on thresholds as before. I can't go through it and hand pick stations, it would be a lot of work. That chart is your best bet for comparing adjacent channel signal levels.


Stephen said...

Hi Bill,

You're welcome. :) I look forward to seeing what you've come up with so far. Any idea when the next installment of maps may be posted that may include multiple (maybe selectable) plots?

Ahh, sounds interesting. :) I wonder if it yet has the ability to put, say, "custom" stations in? For example there's a couple Mexican stations that have quite an overlap with USA stations here (one being a semi-local first-adjacent to a local, and another being co-channel to one nearby) and I'd like to try to figure out locations of greatest "clash", so to speak. :)

Sure, that Tucson chart could be helpful. :) How hard would it be to tailor it for local Tucson sunrise, and in cases where another station's pattern change is at the same time as Tucson (like Phoenix for example), include day-pattern skywave strengths?



Hi Stephen,

Yes, have been quite busy lately. I also added a second and third skywave calculation algorithm choice. It might not have been worth the extra effort, as all three are within 3 dB or so. The latest, the ITU one, is supposed to be more accurate.

You asked:

'Any idea when the next installment of maps may be posted that may include multiple (maybe selectable) plots?'

Probably not before summer anyway. Writing program code to eject auto-generated HTML is very tedious work, at least for me. It will be a summertime project in NY when I have the time. In any event, it's not likely that an entire database of maps will be made available in one .zip due to the sheer byte size of the files. I am already near max on my account for download volume for the current daytime/nighttime .zips right now. They are currently at about 40 MB.

Increasing contour plots to five per station versus the current 'one' would increase the .zip size to about 200 MB. I can't sustain that on my download account. It might be possible to make multi-contour plots available on a station by station basis, instead of the entire database in one .zip file. A secondary problem is we also run into the possibility of choking the browser with too much code. A per-station contour .zip or let's say a map with up to 2-4 stations makes more sense.

I have toyed with custom stations but it is far from done. You'd need to modify the source of the program code to do it anyway.

I can make a chart for Tucson sunrise on whatever date you want. Stations displayed would be in realtime in accordance to the selected time. I.E., stations east of Tucson would be in their it daytime pattern, stations west in their nighttime pattern. Levels in the chart would appear accordingly. Shoot me a date before your trip.


Stephen said...

Hi Bill,

Ahh being busy is understandable. :) How's the DX been so far? And on the skywave plots, do any use other % plots besides 50 & 10, like the 6% that was mentioned in another post for example?

Hmm... I used to use box, but quit using it for some reason, I forget why now. I've been using mediafire (as has Gary DeBock), also I've used hotmail's and yahoo's file storage in the past but don't know if I have anything active there. I also have Google but haven't yet tried using it for that purpose much if at all yet, just for local/private transfers / office between myself.

So the multi-plots is still a work in progress? If possible I'd like to see a visual idea of what you mean, maybe you could post an image on a low-band Class A channel that also has a couple 50kW co-channel class B stations? Or maybe a class C, zoomed in to a particular area that has a fairly high-level grounwave contour overlap? Or is it not even up to a "preview" like that, yet?

I guess maybe I just will need to drive around the area to find the locations where 1390s XEKT and KLTX clash most in the daytime, or where the combined strength of 1030 XESDD and 1040 KURS gives the highest-level overlap?

Well, I'm thinking my only shot at Tucson will be in the morning of March 30, 2015 (which happens to be my late paternal grandpa's 100th birthday). Sign-on time for 1550 KUAZ (not counting a 39-watt PSRA) is 6:30am MST for the month of March.

To refresh re: Tucson, my plan is to sit near the KUAZ transmitter site, and compare the difference in blocking/desense on a nearby frequency when they sign on. Chances are I'll do quick bandchecks before and after sunrise, but in the few minutes surrounding the actual time, park on one specific frequency that's hopefully within a couple clicks of KUAZ, but far enough to avoid splatter/IBOC. Ideally I'd pick a frequency and location to sit so that when KUAZ is off the air, has a fairly strong signal, but when KUAZ is on, it's inaudible. I'll also need to watch out for the locals on 830 and 1030, which according to my preliminary calculations would be in the mid-upper 80s dBu on day patterns, which could spread some significant desense up into the 1500s possibly. Maybe I could try to position myself to null whichever is the bigger pest of the two, while facing KUAZ's tower and if possible facing the direction of the station I choose to monitor.



Hi Stephen,

Haven't done much DXing at all. Too busy coding :-). Actually I don't DX too much anymore anyway. More in the truck while traveling than anywhere else.

On the skywave plots, the only formulas I see existent today are the 50% and 10% varieties. I would hesitate to interpolate them for 6% or any other percent. Basically to arrive at the 10% factor, they add in 6-10 dB to the 50% dBu value dependent on the mid-point latitude between the transmitter and receiver location and its proximity to the geomagnetic pole. Good only for North America region 2 by the way. 6% would be a couple extra dB I would guess, but the exact figure is unknown to me. Even the current 6-10 dB factor is a guess.

Multi-plots is basically done at this point, as far as I'm taking it, anyway. I'll see what I can do to send you an example. Colors are so-so. Partially opaque color fills on the web in Google maps tend to blend together changing the color. Nothing much I can do about that.

Will get you a couple of charts for March 30 in Tucson soon. I'll see if I can make one just ten minutes before sunrise, and one ten minutes after sunrise. That should cover the night/day transition. I'll have to figure out how to change the date from the current day to the 30th. The program is not set up to do that. Do you want the geographic location right at KUAZ or somewhere else?


Stephen said...

Hi Bill,

Okay. :) Btw I did a bandscan in my car the other day (don't think I've uploaded it yet though), and my radio surprised me by bringing in a trace of what I think was 840 KXNT at noon. Otherwise it doesn't seem to be all that spectacular, the selectivity leaves quite a bit to be desired on it. (I'm considering putting one of the Pioneer Supertuner IIID models in, but haven't decided on which one yet, but it won't be one with a touchscreen or in-dash video playback feature.)

Well I'm not quite going to be right on top of the KUAZ tower (32.372683, -111.098667), but I do plan to try to be fairly close. I was looking at the locations of some of the nearby pests, and it looks like 830 KFLT and 1030 KVOI would be the biggest pests, with runners-up being 580 KSAZ and 990 KTKT. If I position myself east of KUAZ, like at the intersection of N Hartman Ln & W Overton Rd (32.373830, -111.080293), I would have a moderately decent chance of partially nulling KFLT (5 mi N) *and* KVOI (2.3 mi SE). But, this would have me about 1.73 km from the tower, and the signal wouldn't be quite as strong as I would like. (Ideally I could get it to where it's above "90 dBµ" on my Tecsun, without going over "94 dBµ" above which I get distorted audio and start hearing splatter go way beyond the normal boundaries.) If I sit on the I-10 WB frontage road at 32.370943, -111.101405, I could be about 322 km from the tower, but then I'd be staring right into the face of 830 KFLT, which based on my calculation (using your field strength calculator program, putting in the RMS, ground conductivity of 15, and distance of 5 miles, and using the -21 conversion factor I've discovered with my Tecsun), would put an ~ "87 dBµ" signal, which is well into the range of sending desense across the band, and could well affect even as high up as the 1500s when KUAZ is off the air.

I wonder if possible, you could run a few tests locally with the coordinates I gave, setting the minimum to 10 mV/m groundwave only to get just the locals, and see if there's more than a 1 or 2 dB difference on any station except KUAZ on those locations, like KFLT and KVOI. If so, I'll need to figure out a good place, or if not, just do right on the KUAZ tower. Also I'm thinking either this week, or maybe next Sunday when/if I go to Riverside, CA for the day), I may stop by a few stations that run omni in the daytime and try to determine how far from the tower I get to see signals beween 90 and 95 dBµ, then hopefully we could figure out the field intensity at those locations, and thereby determine approximately how far from KUAZ's tower I'd need to be. Candidates include 760 KFMB, 1450 KFSD and 830 KLAA. I'd add 640 KFI but I'm not planning to go by that way until after my Tucson trip at the earliest.

Looking at KUAZ's satellite view, it appears there may be a trail leading west from N Hartman Ln that goes almost right by KUAZ. If that proves to be true, then maybe I could get as close as about 40 meters from the tower, but then the overload (likely pegging at 98 dBµ) would throw things off. I want to be close enough to see 50/00 on the channel I end up monitoring (and ideally have it wipe out reception of a station with a 5 mV/m or better signal), without actually having KUAZ's splatter be audible when they're on the air.

73, Stephen

Stephen said...

Hi Bill,

Turns out there's something we're planning to go to Sunday morning, so we'll be going there Saturday night (the 28th) also. Would it be possible to also get a list for sunset on Saturday?

73, Stephen


Hi Stephen,

Yes, I'll get you a chart for sunset on the 28th as well. Currently I have the program all torn apart working more on the multi-plot thing. It'll be back together in a couple of days.