[Elecraft] RFI, KUSB and MFJ-4225

[dave]

I used to use an MFJ-4125 switching supply, which I believe is the 
same circuit as the 4225 only sans the meters. I quit using it mostly 
due to the fan noise (despite trying various circuits to reduce fan 
noise none that I tried worked acceptably on this unit). But while 
using it I detected noise on 80m.

The noise was coming from the supply and being radiated via the power 
leads to the rig or antenna. Touching the power leads across a 125 AHr 
battery would kill the noise. Removing the leads from the battery, so 
that only the supply was in line, would allow the noise to come 
through. The battery was acting as a large filter cap filtering out 
the noise.

But it was not broad band noise. Typical of a switcher it only 
occurred at discrete intervals. I you listened between the birdies 
there was no noise. If you want to check for switcher noise be sure to 
scan the band and find the birdies.

The observation I would make about the ARRL tests is that while they 
are OK for comparison of one supply to another, they tell us nothing 
about whether or not we might expect to hear noise from any of these 
supplies. The spectrum analyzer charts have a minimum of -80 dB but we 
know not what the reference is - at least I could not find it in the 
article.

The top of the chart is 0 dB, but we don't know if this is dBw or dBm 
or something else. If the reference is either dBw or dBm then -80 is 
still god awful noisy. Remember that the noise floor of a typical RX 
is around -130 dBm so a supply producing -80 dBm of noise is +50 above 
the noise floor - or about S9. If the reference is dBw then add 30 dB 
to that. Plenty loud!

They point out in the article that the spectrum analyzer measurements 
were made under load and that the load on RX is lighter so there will 
be less noise generated. They ultimately give all but one, the Icom, a 
passing grade as OK for most installations. But there is no comment 
about important considerations such as type of antenna and feedline or 
antenna location.

This seems to me to be another case of caveat emptor! One guy could 
experience bad noise in some situation but might completely clear it 
with a different antenna or feedline (if there is feedline radiation 
and pickup). Another might never be able to detect any noise if the 
antenna is further away and there is no close-by pickup. YMMV.


73 de dave
ab9ca/4






On 10/6/10 2:19 PM, Nick-WA5BDU wrote:
>    I was surprised to hear of the MFJ-4225 generating noise on 160, since
> I use a MFJ-4225MV and work 160 a fair amount, yet hadn't noticed any
> noise.  OTOH, I could I tell if I'm hearing P/S generated noise as
> opposed to "other" 160 meter band noise?
>
> I dug out my homebrew linear supply and decided to use the AFV reading
> on the K3 for comparisons.  With AGC off, antenna connected, K3 tuned to
> an empty spot on the band with no signals present (daytime).  The
> readings jump around a fair amount with random bursts of noise and
> possibly QRN crashes, so I jotted down ten or so over a ten second
> period, first with the MFJ powering the K3, then with the linear P/S,
> and then back to the MFJ to close the loop.  I know this isn't highly
> rigorous, but it should tell me something.
>
> The average readings were MFJ, 146; linear, 168; MFJ, 141.  I don't
> think the differences are significant, that is, I don't think the linear
> power supply is noisy.  So I'd say there's no significant difference.  I
> turned on the MFJ while listening with the K3 powered from the linear
> P/S and didn't hear any noise increase.  The MFJ had long leads
> attached, going to my IC-706, but not loaded.
>
> Looking at the QST product review (January, 2000), they found that the
> MFJ-4225 was very quiet with some increase in noise between 7.5 and 10 MHz.
>
> Maybe there's something wrong with your specific power supply.  I guess
> it's also possible that a change in manufacturing techniques or design
> has caused the MFJ-4225 to degrade.
>
> 73-
>
> Nick, WA5BDU
>
>
>
> On 10/5/2010 7:38 PM, Ignacy wrote:
>> MFJ-4225 generates a 10db noise on 160m. Can a simple filter cure the
>> problem or is it better to get a linear PS?
>>
>> Ignacy, NO9E
>>
>>
> ______________________________________________________________
> Elecraft mailing list
> Home: http://mailman.qth.net/mailman/listinfo/elecraft
> Help: http://mailman.qth.net/mmfaq.htm
> Post: mailto:Elecraft at mailman.qth.net
>
> This list hosted by: http://www.qsl.net
> Please help support this email list: http://www.qsl.net/donate.html
>
>