Elecraft › [K3]

[K3] High internal noise level

Classic

List

Threaded

7 messages Options

Pete Smith N4ZR

[K3] High internal noise level

Ever since I got my K3, I have been a little surprised at what seems to
be a relatively high noise level with no antenna connected (or on a
dummy load). Very unscientifically described, on 20m it registers 1-2
bars on the S meter, and the background is quite loud on the speaker
with the audio gain set at 12 o'clock (RF gain full, AGC - fast,
bandwidth 200Hz with a 500 Hz roofing filter in, no noise blanker or
noise reduction).

The radio seems to perform fine, and I have not had the feeling that I
was missing many signals because they were being masked by receiver
noise. Nonetheless, I wonder how this compares with others' experience,
or with the factory expectation?

73, Pete N4ZR

The World Contest Station Database, updated daily at www.conteststations.com
The Reverse Beacon Network at http://reversebeacon.net, blog at
reversebeacon.blogspot.com,
spots at telnet.reversebeacon.net, port 7000

______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html

Bill W4ZV

Re: [K3] High internal noise level

Pete Smith N4ZR wrote

Ever since I got my K3, I have been a little surprised at what seems to
be a relatively high noise level with no antenna connected (or on a
dummy load). Very unscientifically described, on 20m it registers 1-2
bars on the S meter, and the background is quite loud on the speaker
with the audio gain set at 12 o'clock (RF gain full, AGC - fast,
bandwidth 200Hz with a 500 Hz roofing filter in, no noise blanker or
noise reduction).

Pete this is a common issue for new users. It's actually not unique to the K3...Orion had the identical issue for the same reasons.

1. Your gain settings are incorrect. If AF Gain is at 12 o'clock it's too high, and RF Gain should almost never be at maximum. My settings are typically AF Gain at 9-10 o'clock with RF Gain around 1-3 o'clock (but that depends on the specific antenna and band). This has been covered many times and Barry N1EU has a nice explanation including a summary of how to set both gains here:

http://n1eu.com/K3/K3_agcgain.htm (scroll to Gain Adjustment)

2. The K3's S-meter is more accurate than most rigs, especially at the low end of the meter scale. Typical FT-1000 meters will read S1 at a -104 dBm level.

http://www.n6rk.com/FT1000_S_meter.doc

By contrast, a properly calibrated (and accurate) K3 meter will read S4 at the same level!

http://www.n6rk.com/S_unit_definitions.doc

The combination of #1 above (setting RF Gain to max) and #2 (an accurate meter which indicates higher than many rigs at low levels) leads many new users to think the K3 is noisy. In fact it isn't but we need to learn how to properly set gain to take maximum advantage of the K3's dynamic range. K3NA's version is very detailed, but N1EU's version above is much simpler.

BTW for those of you with a P3, setting it's Ref Lvl just at or slightly above the noise floor is the equivalent of what you're doing in the K3 above. When done properly with the K3, you'll hear weak signals at the noise floor and still have the full ~100 dB of dynamic range above that level. If you set RF Gain too high, the rig will sound noisy and you'll be wasting some dynamic range by extending the K3's low end below band noise levels.

73, Bill

Don Wilhelm-4

Re: [K3] High internal noise level

In reply to this post by Pete Smith N4ZR

Pete,

Don't tell me we are going to get into this "the K3 is noisy" discussion
again - it is one of those that comes around once in a while. The fact
is that the K3 has adequate gain to enable the user to hear its internal
noise floor - it can also hear everything above that noise floor. If
connecting an antenna results in an increase in the noise level (you are
then hearing the atmospheric noise picked up by the antenna), then the
internal noise floor will not get in the way of hearing weak signals.

The AGC settings will influence this effect. When you disconnect the
antenna, the AGC sets the K3 to full gain. If you have the pre-amp on,
you will be hearing the output noise level of the preamp too. The MDS
of a proper functioning K3 is very good, and that is the indicator of
the signal level which can be discerned above the receiver noise floor.
If you believe your K3 may be sub-par, I encourage you to use an
Elecraft XG2 signal generator or a suitable low level generator to
measure the MDS of your K3.

A 12 o'clock setting on the AF Gain is higher than usual, so yes, in
that condition, you will begin to hear the internal noise floor of the
receiver - turn back the AF Gain until the internal noise is just barely
audible.

As far as your concern about the S-meter reading - the S-meter response
can be altered by changing the AGC characteristics on your K3. Jack
Smith (www.cliftonlaboratories.com) has done measurements of the K3 AGC
settings, so if you want to customize your AGC response, I suggest you
use his charts for reference. S-meter ABS setting can also influence
your S-meter response and make it 'imune' to the ATT and PRE in/out
condition.

73,
Don W3FPR

On 8/28/2010 6:18 AM, Pete Smith wrote:

______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html

Joe Subich, W4TV-4

Re: [K3] High internal noise level

In reply to this post by Pete Smith N4ZR

Pete,

The first thing I would do is get a calibrated signal generator (the
XG-2 is fine if you don't have something more complex) and perform
the RF gain calibration in the K3 Utility. Follow that with an S-
Meter calibration. After calibration with S9 = 50 uV (-73 dBm), I
see no S-meter activity with a dummy load or antenna on a "dead band."
The P3 shows the noise floor between -140 and -145 dBm (2 to 3 S units
BELOW S0) with a dummy load and around -130 dBm (still less than S0)
with my low Windom and no preamp on 20 meters.

Second, your AF Gain may be a bit high ... mine is normally around
9 - 10 o'clock with the Sony bookshelf speakers or the CM-500 and
around 10 - 11 o'clock with the Heil Proset. That represents a
comfortable listening level for most signals and provides almost
no noise with the antenna on 20 (160/80/40 are a different story)
and quiet with the dummy load/no antenna. As a reference point, the
"no antenna" noise is noticeable at full RF/AF gain and the 20 meter
antenna noise is at the threshold of becoming uncomfortable at full
AF/RF gain. Note, I generally do not need the preamp even with a
poor antenna except on 6/10/12 meters and always use ATT on 160/80.

I am not using the default settings for ACG threshold and slope ...
I prefer setting both to "8" which provides for a higher threshold
before the onset of AGC and less flatness after AGC begins. For an
excellent look at the K3 AGC behavior see Jack Smith's (W8OZA) work:
http://www.cliftonlaboratories.com/elecraft_k3_agc_and_s-meter.htm .

73,

... Joe, W4TV

On 8/28/2010 6:18 AM, Pete Smith wrote:

> Ever since I got my K3, I have been a little surprised at what seems to
> be a relatively high noise level with no antenna connected (or on a
> dummy load). Very unscientifically described, on 20m it registers 1-2
> bars on the S meter, and the background is quite loud on the speaker
> with the audio gain set at 12 o'clock (RF gain full, AGC - fast,
> bandwidth 200Hz with a 500 Hz roofing filter in, no noise blanker or
> noise reduction).
>
> The radio seems to perform fine, and I have not had the feeling that I
> was missing many signals because they were being masked by receiver
> noise. Nonetheless, I wonder how this compares with others' experience,
> or with the factory expectation?
>
> 73, Pete N4ZR
>
> The World Contest Station Database, updated daily at www.conteststations.com
> The Reverse Beacon Network at http://reversebeacon.net, blog at
> reversebeacon.blogspot.com,
> spots at telnet.reversebeacon.net, port 7000
>
> ______________________________________________________________
> Elecraft mailing list
> Home: http://mailman.qth.net/mailman/listinfo/elecraft
> Help: http://mailman.qth.net/mmfaq.htm
> Post: mailto:[hidden email]
>
> This list hosted by: http://www.qsl.net
> Please help support this email list: http://www.qsl.net/donate.html
>

Jim Brown-10

Re: [K3] High internal noise level

In reply to this post by Pete Smith N4ZR

On Sat, 28 Aug 2010 06:18:06 -0400, Pete Smith wrote:

>Nonetheless, I wonder how this compares with others' experience,
>or with the factory expectation?

What I think you are observing is the AGC bringing up the system gain
so that you hear circuit noise in the front end. As W8JI has often
observed, if the noise level increases at least 10dB when you connect
the antenna, the circuit noise in the receiver is not limiting weak
signal performance. That condition is met on all bands except 6M, where
an external preamp is needed in any reasonably quiet location.

73, Jim Brown K9YC

______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html

Pete Smith N4ZR

Re: [K3] High internal noise level

In reply to this post by Bill W4ZV

Thanks to Bill and everyone else for their suggestions. I implemented
N1EU's AGC settings - not much different from the defaults, but a
little, and the receiver seems quieter. Also switched off the
peak-reading S meter, and the S-meter indication on a dummy load is now
zero. That may be the biggest psychological factor in the "quieter
RX." But boy, the S-meter seems scotch by comparison - I have done the
calibration routine from the manual, though, and I'm satisfied that it's
just honest - hi.

I guess my analog roots are showing - I have grown up used to running RF
gain full on, and turning off AGC when I had to ride the RF gain. But I
notice that, as in an analog radio, the S meter is effectively disabled
when you turn the RF gain down very far. I guess I can live without it.

73, Pete N4ZR

The World Contest Station Database, updated daily at www.conteststations.com
The Reverse Beacon Network at http://reversebeacon.net, blog at reversebeacon.blogspot.com,
spots at telnet.reversebeacon.net, port 7000

On 8/28/2010 7:21 AM, Bill W4ZV wrote:

>
> Pete Smith N4ZR wrote:
>
>> Ever since I got my K3, I have been a little surprised at what seems to
>> be a relatively high noise level with no antenna connected (or on a
>> dummy load). Very unscientifically described, on 20m it registers 1-2
>> bars on the S meter, and the background is quite loud on the speaker
>> with the audio gain set at 12 o'clock (RF gain full, AGC - fast,
>> bandwidth 200Hz with a 500 Hz roofing filter in, no noise blanker or
>> noise reduction).
>>
>>
> Pete this is a common issue for new users. It's actually not unique to the
> K3...Orion had the identical issue for the same reasons.
>
> 1. Your gain settings are incorrect. If AF Gain is at 12 o'clock it's too
> high, and RF Gain should almost never be at maximum. My settings are
> typically AF Gain at 9-10 o'clock with RF Gain around 1-3 o'clock (but that
> depends on the specific antenna and band). This has been covered many times
> and Barry N1EU has a nice explanation including a summary of how to set both
> gains here:
>
> http://n1eu.com/K3/K3_agcgain.htm (scroll to Gain Adjustment)
>
>
> 2. The K3's S-meter is more accurate than most rigs, especially at the low
> end of the meter scale. Typical FT-1000 meters will read S1 at a -104 dBm
> level.
>
> http://www.n6rk.com/FT1000_S_meter.doc
>
> By contrast, a properly calibrated (and accurate) K3 meter will read S4 at
> the same level!
>
> http://www.n6rk.com/S_unit_definitions.doc
>
>
> The combination of #1 above (setting RF Gain to max) and #2 (an accurate
> meter which indicates higher than many rigs at low levels) leads many new
> users to think the K3 is noisy. In fact it isn't but we need to learn how
> to properly set gain to take maximum advantage of the K3's dynamic range.
> K3NA's version is very detailed, but N1EU's version above is much simpler.
>
> BTW for those of you with a P3, setting it's Ref Lvl just at or slightly
> above the noise floor is the equivalent of what you're doing in the K3
> above. When done properly with the K3, you'll hear weak signals at the
> noise floor and still have the full ~100 dB of dynamic range above that
> level. If you set RF Gain too high, the rig will sound noisy and you'll be
> wasting some dynamic range by extending the K3's low end below band noise
> levels.
>
> 73, Bill
>

Bill W4ZV

Re: [K3] High internal noise level

Pete Smith N4ZR wrote

I guess my analog roots are showing - I have grown up used to running RF
gain full on, and turning off AGC when I had to ride the RF gain. But I
notice that, as in an analog radio, the S meter is effectively disabled
when you turn the RF gain down very far. I guess I can live without it.

The S-meter can actually help to set RF Gain correctly. After you determine the lowest gain setting of ATT/Off/PRE which gives band noise with an antenna connected and RF Gain max CW, then adjust RF Gain CCW until the meter holds just above visual noise peaks. The meter still works for signals above the noise floor you've just set.

Another point I forgot is to do the RF Gain Calibration on the K3 Utility Calibration page. In fact, the best order to do everything is:

1. RF Gain Calibration.
2. S-meter Calibration.
3. Set Rf Gain max CW and set ATT/Off/PRE for lowest gain option that still gives band noise.
4. Back off RF Gain CCW so meter holds just above noise peaks.
5. The K3 stores ATT/Off/PRE settings per band, so you don't need to do this every time you change bands if you set it up ahead of time (although multiple antennas per band will necessitate some adjustment).
6. Set AF Gain for a comfortable listening level on a medium signal (typically 9-10 o'clock).
7. If you then let AGC do the work, you probably won't need to touch AF or RF Gain much in normal operation.

73, Bill

P.S. The K3 hears well if you set it up correctly. Some anecdotal evidence is below (all from 160m and Diversity also deserves some credit):

http://lists.contesting.com/archives//html/3830/2009-02/msg01298.html
http://lists.contesting.com/archives//html/3830/2009-12/msg02168.html
http://lists.contesting.com/archives//html/3830/2009-12/msg02145.html
http://web.jzap.com/k7rat/2009_Stew.htm (3-way tie for most QRPs)
2010 ARRL DX SOSB160 - 5 QRPs in EU and most DX QSOs, including multiops who had the added advantage of Packet.