Noisy Input Problem

[k3gau Dave]

GM,

I had noticed what seemed to be "a noisy input" problem even before I had Steve repair my SARK 110.  So his repair had NOTHING to do with the problem.  

It doesn't seem to act right then it is looking at an the open circuit.  It does seem to read a short, 50 ohm, 100 ohm and 200 ohm loads across its bandwidth just fine.  But looking at an open, it sometimes reads out some negative resistances, etc. when measuring 'RS' and plots them on the screen as well.  The problem seems worse at the higher end of the spectrum when scanning the full 230 MHz range.  As I recall, when I first got my unit, open circuit 'RS' was always full scale, a high value (in the kOhms) and never a negative.  I have gone through the detector check and calibration procedures.  They seemed to be about normal. 

Any idea what could be causing the problem?  Bad component, bad solder joint?

Dave  

[melchor...@gmail.com]

Hello Dave,

 

Could you please save a scan for the open circuit condition (connecting the same open load as the used for calibration) and send to me for analysis? Also, it will be helpful if you send me the calibration files.

 

Thanks,

Melchor EA4FRB

--
You received this message because you are subscribed to the Google Groups "sark110" group.
To unsubscribe from this group and stop receiving emails from it, send an email to sark110+u...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/sark110/1eda21ed-7150-417d-91a1-7739a5253e53%40googlegroups.com.

[k3gau Dave]

Melcor,

What files are looking for? (name)

Dave 

To unsubscribe from this group and stop receiving emails from it, send an email to sar...@googlegroups.com.

[melchor...@gmail.com]

The calibration .dat files, e.g. detcalib.dat, oslcalib.dat, oslprof1.dat; and a saved .csv file for the sweep scan with the open load (select <File> <Save Data File> after completing the scan).

 

Thanks,

Melchor

To unsubscribe from this group and stop receiving emails from it, send an email to sark110+u...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/sark110/b9940772-5794-4697-a133-b87d1a8a1649%40googlegroups.com.

[k3gau Dave]

On Monday, April 27, 2020 at 12:33:41 PM UTC-4, EA4FRB wrote:

To unsubscribe from this group and stop receiving emails from it, send an email to sar...@googlegroups.com.

[k3gau Dave]

Melchor,

I think I sent all files you ask for and probably a couple you don't need.   SARK 1, SARK 2 and SARK 3 are three different shots of my screen taken within a few minutes.

Dave

[melchor...@gmail.com]

Thank you for the files. I don’t see a problem with these readings and the cal files. When there is no load connected, it is possible to get negative Rs readings influenced by noise because the error is high when the load is close to the infinity.  Also, there is the influence of the used open load for the OSL calibration.   

 

Best regards,

Melchor EA4FRB

 

De: sar...@googlegroups.com <sar...@googlegroups.com> En nombre de k3gau Dave
Enviado el: lunes, 27 de abril de 2020 19:40
Para: sark110 <sar...@googlegroups.com>
Asunto: Re: Noisy Input Problem

 

Melchor,

--

You received this message because you are subscribed to the Google Groups "sark110" group.

To unsubscribe from this group and stop receiving emails from it, send an email to sark110+u...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/sark110/3eeb8866-8bca-4938-a40b-ced41b7dea9b%40googlegroups.com.

[k3gau Dave]

GE Melchor,

Have you plotted those Sark files?   I would like to send you a picture of the SARK screen but my cellphone is not working.  Maybe I can use my other camera. I can save a file with the 'open' removed and just the open connector on the input.  

In any case, I don't remember my screen looking that when it was new.  

I got them with my other camera.  Does that look normal for an 'OPEN' condition?

Dave K3GAU 

To unsubscribe from this group and stop receiving emails from it, send an email to sar...@googlegroups.com.

[melchor...@gmail.com]

Hello Dave,

 

Yes, I have plotted those files. Note that the ripples are more or less visible depending on the selected scale: fixed or automatic; or logarithmic (default) or linear. Please find below some pics for comparing.

 

The fact is that for the open condition it is possible to get negative readings of Rs due to the increased error close to infinity and influenced by the noise. This effect can be reduced by using averaging and/or double sampling, and calibrating w/o a computer connected.  

 

Figure 1, sark__01.csv, logarithmic scale, auto-scale

 

Figure 2,  sark__01.csv, linear scale, auto-scale

 

Figure 3, sark__01.csv, linear scale, auto-scale (SARK_Plots)

Figure 4, sark__01.csv, smith chart

To unsubscribe from this group and stop receiving emails from it, send an email to sark110+u...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/sark110/8efc52e0-673c-4f7d-b853-729814e04e1c%40googlegroups.com.