Elecraft › [K2]

K2 Poor sensitivity

Classic

List

Threaded

6 messages Options

Radiobeamer

K2 Poor sensitivity

Hi group,

I bought a used K2 some time ago. I've not used it much but it seemed to work fine. I recently upgraded it with the SSB option and the 60m/transverter option. I had to go through the BFO re-alignment procedure in order to make the filters work properly after installing the SSB option, and this process seemed to work quite well. The added options appear to be working OK and I was able to make a few CW and SSB QSOs.

However, I noticed that the S meter was not moving very much so I thought to check the sensitivity. According to the spec we should have a sensitivity (MDS) of something like -130dBm without the preamp and -135dBm with the preamp. I have an HP signal generator and, without being too precise about MDS etc, I was not able to hear anything at all below about -120dBm (ie the signal was completely buried in the noise) and it would probably take -110dBm to -100dBm to be able to copy a signal. This is with the preamp on. So I am looking for reasons why my sensitivity is a bit down, say 20 to 30dB.

My K2 has, in addition to the SSB and 60m options, the noise blanker and top band options as well as the KPA100 and KDSP2, which were not in use during these checks. I was able to make the following observations:

a) The sensitivity is about the same on all bands which would seem to exclude the 60m module as a potential source of the problem.
b) The sensitivity is the same using the separate receive antenna input in the top band option.
c) The transverter option does not appear to be causing the problem since I can use it to switch between the transverter input and the normal BNC antenna connection.
d) The sensitivity seemed to be slightly better using the SSB filter than the CW filters. By the time the 200Hz crystal filter bandwidth was selected the signal was audibly weaker than using the fixed SSB filter.

I'm not all that familiar with the K2 architecture and particularly the alignment procedure. So I should like some advice about the best way to approach this issue. I see that the noise blanker has an IF filter (labelled -6dB) which I imagine could cause a problem like this if it was built incorrectly. I guess it is an easy enough step to remove the option and replace it with a wire link.

Assuming that this did not resolve the problem, would I be wise to look for a definite fault (eg in the IF amplifier chain) or to go down the path of complete re-alignment? Can poor alignment (excluding front-end alignment) of the radio cause poor sensitivity? Could this be the reason why the CW filters seem less sensitive than the SSB filter? Bear in mind that there could be an original build fault causing this issue (perhaps a wrongly wound matching transformer, for example).

Thanks in advance for your help.

Paul

John Oppenheimer

Re: K2 Poor sensitivity

Hi Paul,

There is a procedure for estimating MDS documented in the XG3 manual,
the "Measuring Receiver Sensitivity" section:
http://www.elecraft.com/manual/E740154%20XG3%20Owner%27s%20Manual%20Rev%20D2%20-%20ECO%2001887.pdf

I've performed this test and found my K2 to have a reasonable match.

John KN5L

On 09/27/2016 05:38 AM, Radiobeamer wrote:
> I noticed that the S meter was not moving very much so I thought to
> check the sensitivity. According to the spec we should have a sensitivity
> (MDS) of something like -130dBm without the preamp and -135dBm with the
> preamp. I have an HP signal generator and, without being too precise about
> MDS etc,
______________________________________________________________
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
Message delivered to [hidden email]

Don Wilhelm

Re: K2 Poor sensitivity

In reply to this post by Radiobeamer

Paul,

If the sensitivity was OK before adding the KSB2 option, then the KSB2
is the most likely cause.
Remove the KSB2 and put a bent wire jumper between pins 1 and 3 of both
J9 and J10 and see if the sensitivity improves.

If so, re-flow the soldering on the KSB2 board, check the diode
orientation and check the leads of T1 and T2 to be sure they were well
stripped and tinned.

If you need to optimize the filter alignment, take a look at the K2 Dial
Calibration article on my website www.w3fpr.com - part 3 of that article
covers filter alignment.

If it turns out that there is something else wrong, you will have to do
Receive Signal Tracing to isolate it to the failing stage. See Appendix
A in the manual. It takes a strong signal input to do the checks with
an RF Probe. If your HP signal generator is not strong enough to
provide the specified input signal level, build the signal generator
shown in the manual.

73,
Don W3FPR

On 9/27/2016 6:38 AM, Radiobeamer wrote:

> Hi group,
>
> I bought a used K2 some time ago. I've not used it much but it seemed to
> work fine. I recently upgraded it with the SSB option and the
> 60m/transverter option. I had to go through the BFO re-alignment procedure
> in order to make the filters work properly after installing the SSB option,
> and this process seemed to work quite well. The added options appear to be
> working OK and I was able to make a few CW and SSB QSOs.
>
>

______________________________________________________________
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
Message delivered to [hidden email]

Radiobeamer

Re: K2 Poor sensitivity

Hi John and Don,

Thanks for your responses. Very useful information.

for John:
Thanks for the link. I'm not sure that my ears really understand what a 30dB S/N ratio sounds like but for sure with well over 1uV input (which is -107dBm) I was not getting what I think of as 30+ dB S/N. With the signal generator down at -120dBm the signal was lost in the noise so I'm quite sure that there is some kind of issue here. I'll hook up the audio to a PC with Spectrum lab and do some measurements.

for Don:
I used the radio for a few CW QSOs before adding the SSB and 60m/transverter options and I was happy enough with it but I'm not sure if the sensitivity was OK before adding the SSB option. It sounds OK when connected to a decent antenna and you can make contacts well enough (on 40m and 20m). With the noise in a domestic environment and a good antenna you don't need a super-sensitive receiver on those bands!

I completed the SSB board last week and I'm confident of my abilities so I'm thinking that (hopefully) the SSB module is good. As you mentioned, it is easy enough to reverse the options mods so I can easily check that. My HP sig gen goes from -136dBm to +7dBm so it can get pretty strong! One bad thing about buying used is that the test probes (frequency and RF signal) didn't come with the radio so I'll have to build some up - I already did the frequency probe to allow the BFO to be aligned. Just need to find a detector diode now ...

Also - thanks for all of the great information on your web site Don. I already downloaded your alignment guide. I'll spend some more time troubleshooting and let you know what I find.

73,

Paul

Radiobeamer

Re: K2 Poor sensitivity

Hi All,

Firstly, thanks for all of your help. I checked the RFCs on the SSB board as per Steve's suggestion and they were OK, although my problem was quite similar (see below). Also, thanks to Frank for your suggestion. As I'm in the UK it is quite cost prohibitive to send the radio to Don. Also, being able to fault-find, align and repair the K2 is part of its attraction.

Secondly, I managed to find the fault!. I noticed that the sensitivity on the transverter input was better than on the antenna input. Probing about on the TR switch led me to the conclusion that RFC3 was not connected properly, leading to excessive loss in D4. Problems with poor soldering on enamelled wire are well known! After re-making the joints of RFC3 I can hear signals down to -130dBm on all bands (with the preamp on).

I've still not measured the sensitivity accurately but it does seem to work properly now. I'll re-align everything now for 100% performance.

Thanks again for a helpful and friendly group.

73,

Paul
G4KZY

bdenley

Re: K2 Poor sensitivity

Nice work Paul.

Brian
KB1VBF
Sent from my iPad

> On Sep 30, 2016, at 10:41 AM, Radiobeamer <[hidden email]> wrote:
>
> Hi All,
>
> Firstly, thanks for all of your help. I checked the RFCs on the SSB board
> as per Steve's suggestion and they were OK, although my problem was quite
> similar (see below). Also, thanks to Frank for your suggestion. As I'm in
> the UK it is quite cost prohibitive to send the radio to Don. Also, being
> able to fault-find, align and repair the K2 is part of its attraction.
>
> Secondly, I managed to find the fault!. I noticed that the sensitivity on
> the transverter input was better than on the antenna input. Probing about
> on the TR switch led me to the conclusion that RFC3 was not connected
> properly, leading to excessive loss in D4. Problems with poor soldering on
> enamelled wire are well known! After re-making the joints of RFC3 I can
> hear signals down to -130dBm on all bands (with the preamp on).
>
> I've still not measured the sensitivity accurately but it does seem to work
> properly now. I'll re-align everything now for 100% performance.
>
> Thanks again for a helpful and friendly group.
>
> 73,
>
> Paul
> G4KZY
>
>
>
> --
> View this message in context: http://elecraft.365791.n2.nabble.com/K2-Poor-sensitivity-tp7622862p7622968.html
> Sent from the Elecraft mailing list archive at Nabble.com.
> ______________________________________________________________
> 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
> Message delivered to [hidden email]