It's a combination of things. When you are in the passband, the maximum signal that you can pump in without compression is determined by the hardware AGC, all else being equal. It might be possible to calibrate out some of the HAGC compression, and extend the range on the top end. I think this can be done with a careful application of the attenuator, pre-amp and manual gain control and the DBV audio output.
I don't know when I'll have time to tinker with it more, as I have some higher priority projects to clean up, so perhaps there is an opportunity for others to pitch in with their experiments. Jack K8ZOA Joe Subich, W4TV wrote: > >> The reason for first determining the appropriate manual gain >> setting is to set the gain at the point where the test signal >> just starts to hit non-linearity in the filter passband. >> > > This appears to be more non-linearity in the DSP since the > measurement is being done with detected audio than with the > IF/Filter. > > I suggest talking with Wayne ... there is an undocumented RF > level reading that might make for a quicker and more reliable > measurement. > > >> Mike and I tried a number of experiments with stitching >> together pieces of the plot taken at different gain settings >> with the thought that the dynamic range could be extended to >> 100 dB or more. While I have not given up on this thought, >> the K3 has some quirks that make it difficult to implement >> without much more thought and tinkering. >> > > Since the RF Gain is already being set using the attenuator, > the same thing should be possible with the preamp. The > measurements could be started with attenuator on until the > detected level gets to - 20 dBV or so ... at that point > the attenuator switched out (you have already calibrated > its exact level in the RF Gain/dBV calibration) and values > taken/adjusted until the measurements again reach -20 to > -30 dBV. At that time the preamp could be switched in ... > > While it might not make 100 dB, it should be possible to > see at least 85 dB. > > >>> What's the most vertical skirt db's anyone has gotten and >>> what were your input and RF gain settings to get it. >>> > > 65 dB with an XG-2 and the RF Gain at 185. > > 73, > > ... Joe, W4TV > > > > >> -----Original Message----- >> From: elecraft-boun...@mailman.qth.net >> [mailto:elecraft-boun...@mailman.qth.net] On Behalf Of Jack Smith >> Sent: Sunday, January 18, 2009 2:31 PM >> To: Guy Olinger, K2AV >> Cc: k...@verison.net; elecraft@mailman.qth.net >> Subject: Re: [Elecraft] K3 Filter Plots >> >> >> Because of the K3's range between noise contamination of the >> reading and >> non-linearity due to hardware AGC, you will see between 60 >> and 70 dB of >> usable dynamic range in the plot. I see around 70 dB on my K3 >> with the >> hardware AGC modification, whilst Mike sees around 60 dB on >> his K3 built >> before the hardware AGC modification. >> >> The reason for first determining the appropriate manual gain >> setting is >> to set the gain at the point where the test signal just starts to hit >> non-linearity in the filter passband. This allows one to use a test >> signal of arbitrary strength, instead of saying "apply a test >> signal of >> -50 dBm" which requires a calibrated signal generator or test >> oscillator. By adjusting the manual gain using the linearity >> program you >> can use a test signal anywhere between -50 dBm or so up to -20 dBm or >> even greater. >> >> Mike and I tried a number of experiments with stitching >> together pieces >> of the plot taken at different gain settings with the thought >> that the >> dynamic range could be extended to 100 dB or more. While I have not >> given up on this thought, the K3 has some quirks that make it >> difficult >> to implement without much more thought and tinkering. >> >> Jack K8ZOA >> >> >> Guy Olinger, K2AV wrote: >> >>> I too had the serial port lockup, and had to reboot windows >>> >> to release >> >>> the hold on it. The program explictly should leave the RTS/DTR >>> alone. Once I started a scan I was never able to stop it. Doc >>> doesn't give any help about how to set RF gain, and what >>> >> the numbers >> >>> mean. That would be really helpful. >>> >>> I think it would be helpful to have the scan automatically add or >>> subtract the preamp to extend the vertical range of the >>> >> scan off the >> >>> passband. >>> >>> Note that I'm not deleting the program. I will get good at using it >>> however it works. But I notice more about the program when >>> >> I'm trying >> >>> to learn it. >>> >>> What's the most vertical skirt db's anyone has gotten and what were >>> your input and RF gain settings to get it. >>> >>> 73, Guy. >>> >>> ----- Original Message ----- From: "Jack Smith" >>> <jack.sm...@cliftonlaboratories.com> >>> To: "Jerry Flanders" <jefland...@comcast.net> >>> Cc: <k...@verison.net>; <elecraft@mailman.qth.net> >>> Sent: Sunday, January 18, 2009 12:47 PM >>> Subject: Re: [Elecraft] K3 Filter Plots >>> >>> >>> >>>> Jerry: >>>> >>>> I've not seen any setup corruption from the K3 Filter >>>> >> program, but I >> >>>> have not tried all my K3's functions after running the filter >>>> program. >>>> >>>> I hope to add a web page with some details on how to use >>>> >> the filter >> >>>> program, but my project list is expanding faster than the time >>>> available to complete them, so I can't promise it will be >>>> >> done soon. >> >>>> Jack K8ZOA >>>> www.cliftonlaboratories.com >>>> >>>> >>>> Jerry Flanders wrote: >>>> >>>>> After I used it, my K3 had been changed. Before, it was a 100 W >>>>> unit. After, only 12. I had to restore an earlier K3 >>>>> >> configuration >> >>>>> file to restore it. There might have been other K3 >>>>> >> changes also that >> >>>>> I corrected by this config restore without even knowing they had >>>>> occurred, so other users should watch for them. >>>>> >>>>> Also, I had to reboot the computer to get the serial port back - >>>>> K3FilterTools apparently did not release it when it exited. >>>>> >>>>> OTOH, I recall having some weirdness occur while trying >>>>> >> to get the >> >>>>> filter scan and might have upset the program and caused these >>>>> problems myself as I tried to recover from the weirdness - don't >>>>> remember details of it. >>>>> >>>>> Thankfully, my K3 did not transmit into the signal generator - >>>>> probably would have ruined the little Elecraft XG-2 and >>>>> >> w8diz 10 MHz >> >>>>> oscillator I was using.. >>>>> >>>>> I think this will be a really useful program when we >>>>> >> understand how >> >>>>> to safely use it. I read all the help files that came with the >>>>> program, but have not read Jack Smith's writeup on the method yet. >>>>> >>>>> Jerry W4UK >>>>> >>>>> At 10:19 1/18/2009, Ian White GM3SEK wrote: >>>>> >>>>> >>>>>> Joe Subich, W4TV wrote: >>>>>> >>>>>> >>>>>> >>>>>>>>> The program is called K3 Filter Tools and can be downloaded >>>>>>>>> >>>>>>>>> >>>>>>>> free from: http://mysite.verizon.net/ks7d >>>>>>>> >>>>>>>> >>>>>>>>> Mike, KS7D >>>>>>>>> >>>>>>>>> >>>>>>> Note: when using Mike's software, AUTOINF must be set to NOR. >>>>>>> Anyone who normally has AUTOINF=1 (e.g. for a SteppIR antenna) >>>>>>> will need to manually reset it before using the >>>>>>> >> software in order >> >>>>>>> to get usable data. >>>>>>> >>>>>>> >>>>>>> >>>>>> Mike's program is an excellent addition to the K3 >>>>>> >> toolkit... but as >> >>>>>> Joe >>>>>> points out, be aware that it takes control of the K3 in >>>>>> >> ways we might >> >>>>>> not expect. >>>>>> >>>>>> If the K3 is configured for PTT or CW keying using the >>>>>> >> CTS and/or >> >>>>>> DTR lines, the Filter Tools program may cause the rig to >>>>>> >> transmit. >> >>>>>> (No harm done, as the signal generator was connected to >>>>>> >> the RX ANT >> >>>>>> port, but definitely an unexpected event.) >>>>>> >>>>>> Before using *any* new program that takes control of the >>>>>> >> rig, it is >> >>>>>> also >>>>>> a good idea to use Elecraft's K3 Utilities program to >>>>>> >> grab a config >> >>>>>> file, and then reload the original rig configuration afterwards. >>>>>> >>>>>> >>>>>> >>>>>> -- >>>>>> >>>>>> 73 from Ian GM3SEK >>>>>> http://www.ifwtech.co.uk/g3sek >>>>>> _______________________________________________ >>>>>> Elecraft mailing list >>>>>> Post to: Elecraft@mailman.qth.net >>>>>> You must be a subscriber to post to the list. >>>>>> Subscriber Info (Addr. Change, sub, unsub etc.): >>>>>> http://mailman.qth.net/mailman/listinfo/elecraft >>>>>> >>>>>> Help: http://mailman.qth.net/subscribers.htm >>>>>> Elecraft web page: http://www.elecraft.com >>>>>> >>>>>> >>>>> _______________________________________________ >>>>> Elecraft mailing list >>>>> Post to: Elecraft@mailman.qth.net >>>>> You must be a subscriber to post to the list. >>>>> Subscriber Info (Addr. Change, sub, unsub etc.): >>>>> http://mailman.qth.net/mailman/listinfo/elecraft >>>>> >>>>> Help: http://mailman.qth.net/subscribers.htm >>>>> Elecraft web page: http://www.elecraft.com >>>>> >>>>> >>>>> >>>> _______________________________________________ >>>> Elecraft mailing list >>>> Post to: Elecraft@mailman.qth.net >>>> You must be a subscriber to post to the list. >>>> Subscriber Info (Addr. Change, sub, unsub etc.): >>>> http://mailman.qth.net/mailman/listinfo/elecraft >>>> >>>> Help: http://mailman.qth.net/subscribers.htm >>>> Elecraft web page: http://www.elecraft.com >>>> >>>> >>> >>> >> _______________________________________________ >> Elecraft mailing list >> Post to: Elecraft@mailman.qth.net >> You must be a subscriber to post to the list. >> Subscriber Info (Addr. Change, sub, unsub etc.): >> http://mailman.qth.net/mailman/listinfo/elecraft >> >> Help: http://mailman.qth.net/subscribers.htm >> Elecraft web page: http://www.elecraft.com >> > > _______________________________________________ > Elecraft mailing list > Post to: Elecraft@mailman.qth.net > You must be a subscriber to post to the list. > Subscriber Info (Addr. Change, sub, unsub etc.): > http://mailman.qth.net/mailman/listinfo/elecraft > > Help: http://mailman.qth.net/subscribers.htm > Elecraft web page: http://www.elecraft.com > > _______________________________________________ Elecraft mailing list Post to: Elecraft@mailman.qth.net You must be a subscriber to post to the list. Subscriber Info (Addr. Change, sub, unsub etc.): http://mailman.qth.net/mailman/listinfo/elecraft Help: http://mailman.qth.net/subscribers.htm Elecraft web page: http://www.elecraft.com
