New FW release: Number of bus solved and normalize ask for reference level
New FW released
Version v1.0-62
Changes:
- Bug solved: trace store command not working
- Bug solved: Overload flashes in normalized mode
- Normalize activation now asks for reflevel and sets reflevel to manual
- save/recall [id] command added, does store/load of preset (yes, confusing....)
|
Re: A possible probpem with the Noramlize and Stor subtract function
May better to ask for ref level after normalize action.
If you want the normalized level at the bottom you enter 80
if you want it at the top you enter 10
|
I think the Wiki has bloomed since I last looked at it.
I did not realise the sa could do all that. Especially such a range of attenuation. Another 40db (from an external attenuator and one is approaching noise floor (after the attenuator has been validated by the vna).
I think that when the probable next version comes out hopefully with a bigger screen ( I wish I could solder the flexible ribbon with confidence, as a 3.2” screen makes for easier viewing and hardly overlaps the existing case), I will be getting it so that the current sa can be used as a tracking generator and sig gen.
Steve L. G7PSZ
|
Re: A possible probpem with the Noramlize and Stor subtract function
So, after hitting the normalize button a keypad screen to ask for which level to normalize?
And should the reflevel be set such that the normalized level is in the middle of the screen? So reflevel = normalized level + 5 * scale?
|
Re: A possible probpem with the Noramlize and Stor subtract function
Hi Erik Thank you, indeed that would cure the problem. Have you considered to include normalization to a used defined level ? to allow a better view for level changes going below the 0 dB line at the bottom of the screen ? Kind regards Kurt ? ? On Fri, Aug 28, 2020 at 06:13 PM, hwalker wrote: The red overload warning is triggered because the tinySA can currently account for external attenuation and amplification, but not for level offsets due to the normalization or trace math routines.
This is indeed a bug as the overload warning should trigger on the actual input signal and not on the normalized level
|
Re: A possible probpem with the Noramlize and Stor subtract function
Hi Herb Thank you for the comment. As Erik respond it is a bug as 0dBm is a warning level. In my video I demonstrate the Normalization by Normalize to no input signal (the noise floor) and then apply a small signal just 10dB above noise level. Then I can measure the low pass filter drop off directly as the 0 dB reference line is flat over the entire frequency range. Another way of testing the calibration and filter correction constants incl. level offset. Kind regards Kurt ??? ? ? On Fri, Aug 28, 2020 at 03:35 PM, Kurt Poulsen wrote: When using the DISPLAY NORMALIZE the overflow warning is enabled and will continuously flash red color. Same applies for When storing a trace and the use the Subtract mode. It is a matter of internal calculation which has no bearing on the actual usage as such.
Kurt,
? ?Erik added the normalize menu option when I was trying to duplicate his tracking generator set-up during beta testing.? In the standalone mode, I use it as a quick check of any attenuator or pre-amp that I introduce into my test set-up - as follows (the sequence is the same as I would use for normalizing the output of a tracking generator) :
?30MHz calibrator output at zero span? ? ? ? ? ? ? ? ? ? ? ? 30MHz output normalized to 0dBm? ? ? ? ? ? ? ? ? ? ? ? ? ? ?40dB attenuator, value can be read directly
? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? that level triggers the compression
? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? warning (red colored letter) display.
 ?? ? 
?
60MHz, 2nd harm. calibrator output at zero span? ? ? ? 60MHz, 2nd harm. output normalized to 0dBm? ? ? ? ? ?40dB attenuator, value can be read directly
? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? that level triggers the compression
? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? warning (red colored letter) display.
 ?? ? 
? At both of the test frequencies above, the 40dB attenuator is measured correctly after normalization.? The red overload warning is triggered because the tinySA can currently account for external attenuation and amplification, but not for level offsets due to the normalization or trace math routines. The actual measured levels are none the less correct.
? Normalization is a one step STORE/SUBTRACT procedure so the same applies to those menu options.
? - Herb
|
Re: A possible probpem with the Noramlize and Stor subtract function
On Fri, Aug 28, 2020 at 06:13 PM, hwalker wrote:
The red overload warning is triggered because the tinySA can currently account for external attenuation and amplification, but not for level offsets due to the normalization or trace math routines.
This is indeed a bug as the overload warning should trigger on the actual input signal and not on the normalized level
|
Re: A possible probpem with the Noramlize and Stor subtract function
On Fri, Aug 28, 2020 at 03:35 PM, Kurt Poulsen wrote:
When using the DISPLAY NORMALIZE the overflow warning is enabled and will continuously flash red color. Same applies for When storing a trace and the use the Subtract mode.
It is a matter of internal calculation which has no bearing on the actual usage as such.
Kurt, ? ?Erik added the normalize menu option when I was trying to duplicate his tracking generator set-up during beta testing.? In the standalone mode, I use it as a quick check of any attenuator or pre-amp that I introduce into my test set-up - as follows (the sequence is the same as I would use for normalizing the output of a tracking generator) : ?30MHz calibrator output at zero span? ? ? ? ? ? ? ? ? ? ? ? 30MHz output normalized to 0dBm? ? ? ? ? ? ? ? ? ? ? ? ? ? ?40dB attenuator, value can be read directly ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? that level triggers the compression ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? warning (red colored letter) display.  ??  ?  ? 60MHz, 2nd harm. calibrator output at zero span? ? ? ? 60MHz, 2nd harm. output normalized to 0dBm? ? ? ? ? ?40dB attenuator, value can be read directly ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? that level triggers the compression ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? warning (red colored letter) display.  ??  ?  ? At both of the test frequencies above, the 40dB attenuator is measured correctly after normalization.? The red overload warning is triggered because the tinySA can currently account for external attenuation and amplification, but not for level offsets due to the normalization or trace math routines. The actual measured levels are none the less correct. ? Normalization is a one step STORE/SUBTRACT procedure so the same applies to those menu options. ? - Herb
|
A possible probpem with the Noramlize and Stor subtract function
Hi Erik When using the DISPLAY NORMALIZE the overflow warning is enabled and will continuously flash red color. Same applies for When storing a trace and the use the Subtract mode. It is a matter of internal calculation which has no bearing on the actual usage as such. Kind regards Kurt
|
On Fri, Aug 28, 2020 at 01:45 PM, Stephen Laurence wrote: There was an Arduino based 0-30mhz signal generator briefly discussed on the Homebrew group here a while ago. I wonder if any of the design wizards who have done so much to support the beginner engineer/ amateur enthusiast had ever considered a signal generator in the tinysa /nanovna class? A signal range up to 100mhz or so, a signal output from 1v down to 1 microvolt and of necessity, minimal signal leakage outside the instrument. I imagine various modulation types and a sweep function could be included but are not essential. ============================= Steve, ? The LOW output signal generator mode of the tinySA meets many of your requirements. ?? Low output mode spec:
?
- Sinus output with harmonics below -40dB of fundamental
- Output frequency range from 100kHz to 350MHz
- Output level selectable in 1dB steps between -76dBm and -6dBm
- Optional AM, narrow FM and wide FM modulation or slow sweep over selectable frequency span
- Optional output level sweep over maximum the entire output level range
? Though not a self contained solution, adding an external 50ohm step attenuator and rf amplifier will cover your high and low level requirements.? Both frequency and level sweep features are a surprise in this class instrument.? Members of my ham club have purchased two tinySA's just so they could dedicate one as a programmable rf signal generator (Yes, both manual and USB programmable control are available).? Terrific performance to cost ratio at cost of $100 for two units before shipping and taxes. See this page for additional information on the tinSA LOW output mode: ?  ? ?There is also a HIGH output mode (240-960MHz) but it does not have the filtering and programmable attenuation of the LOW output mode. It has a square wave output versus the sinus output of the LOW output mode, and is basically a throw-in feature. If you add your own external filtering and step attenuator it increases the utility of the HIGH output mode. - Herb ? ? ?
|
toggle quoted message
Show quoted text
On 28 Aug 2020, at 20:20, Kurt Poulsen < kurt@...> wrote: Hi Erik Thank you very much for this info. I will print this info and give I a trial. Thank you once again Indeed it is a complex instrument, and Auto is indeed a huge benefit to the novice and eventhe ?more experienced users Kind regards Kurt ? ? ? On Fri, Aug 28, 2020 at 06:38 AM, Kurt Poulsen wrote: The problem I had was that is did not realize the Ref level and level/division is fixed and frozen by a factor of 10 and no other ratio can be chosen.
Its a bit more complex.
Normally linear modes work in auto reflevel/scale mode where the bottom of the screen is at zero and the reflevel is 10*scale. This because you can not have negative values in linear scale.
When you change the scale the reflevel setting moves to manual!!!!!
You can set the scale and reflevel to anything as long as the bottom of the screen is zero or larger.
If you set reflevel back to automatic it works with the selected scale and the reflevel can only be a multiple of the scale AFTER the first sweep in auto reflevel mode.
So if the scale is very small (as can be in a linear scale) it will take many sweeps before the signal can become visible again. This is something I may improve in the future. If all signals are off screen and reflevel is automatic I jump after one sweep to a visible signal.
This is all kind of complex but I did not know how to do it better. Any suggestions?
|
There was an Arduino based 0-30mhz signal generator briefly discussed on the Homebrew group here a while ago.
I wonder if any of the design wizards who have done so much to support the beginner engineer/ amateur enthusiast had ever considered a signal generator in the tinysa /nanovna class?
A signal range up to 100mhz or so, a signal output from 1v down to 1 microvolt and of necessity, minimal signal leakage outside the instrument. I imagine various modulation types and a sweep function could be included but are not essential.
The ability to generate tiny amplitude signals is useful for the amateur communications enthusiast ( and professional too) but my old boatanchor generator is so cumbersome and an awful system to phase-lock the frequency every 25khz made it a pain. The output attenuators was fantastic though, down to 0.1uv.
With the addition of a tiny oscilloscope and a small digital multimeter, I could have a complete lab in my pocket. ?The soldering iron might (literally) burn a hole in my pocket, though.
Steve L. G7PSZ
|
Hi Erik Thank you very much for this info. I will print this info and give I a trial. Thank you once again Indeed it is a complex instrument, and Auto is indeed a huge benefit to the novice and eventhe ?more experienced users Kind regards Kurt ? ? ? On Fri, Aug 28, 2020 at 06:38 AM, Kurt Poulsen wrote: The problem I had was that is did not realize the Ref level and level/division is fixed and frozen by a factor of 10 and no other ratio can be chosen.
Its a bit more complex.
Normally linear modes work in auto reflevel/scale mode where the bottom of the screen is at zero and the reflevel is 10*scale. This because you can not have negative values in linear scale.
When you change the scale the reflevel setting moves to manual!!!!!
You can set the scale and reflevel to anything as long as the bottom of the screen is zero or larger.
If you set reflevel back to automatic it works with the selected scale and the reflevel can only be a multiple of the scale AFTER the first sweep in auto reflevel mode.
So if the scale is very small (as can be in a linear scale) it will take many sweeps before the signal can become visible again. This is something I may improve in the future. If all signals are off screen and reflevel is automatic I jump after one sweep to a visible signal.
This is all kind of complex but I did not know how to do it better. Any suggestions?
|
On Fri, Aug 28, 2020 at 06:38 AM, Kurt Poulsen wrote:
The problem I had was that is did not realize the Ref level and level/division is fixed and frozen by a factor of 10 and no other ratio can be chosen.
Its a bit more complex. Normally linear modes work in auto reflevel/scale mode where the bottom of the screen is at zero and the reflevel is 10*scale. This because you can not have negative values in linear scale. When you change the scale the reflevel setting moves to manual!!!!! You can set the scale and reflevel to anything as long as the bottom of the screen is zero or larger. If you set reflevel back to automatic it works with the selected scale and the reflevel can only be a multiple of the scale AFTER the first sweep in auto reflevel mode. So if the scale is very small (as can be in a linear scale) it will take many sweeps before the signal can become visible again. This is something I may improve in the future. If all signals are off screen and reflevel is automatic I jump after one sweep to a visible signal. This is all kind of complex but I did not know how to do it better. Any suggestions?
|
Perfect!? I see it now and it is working perfectly.? Thank you!
|
Hi Erik and Herb Thank you both for the info, which I was aware of. The problem I had was that is did not realize the Ref level and level/division is fixed and frozen by a factor of 10 and no other ratio can be chosen. That is a vital info to convey for a noval user and that I will include in my video clip on the linear Unit modes Kind regards Kurt ? ? On Thu, Aug 27, 2020 at 06:53 PM, hwalker wrote: Both the reference level and and attenuation can only be set in discrete steps of 1, 2 , 5, 10 and they interact to maintain a vertical scaling of 10 divisions.
There is a bug that causes strange auto reflevel behavior when in linear mode and changing the scale from small to a larger value. E.g. from 2 to 10. This causes the reflevel to reposition after one sweep after changing the scale to try to maintain a multiple of the scale
|
Peak hold is under the CALC menu
CALC moved from the LEVEL menu to the DISPLAY menu in the latest FW
|
Yes.
You can use peak hold
|
I’m testing an enclosure and need to place the tinySA inside and test received signal strength of a fundamental. ?I would think I can enter it into a peak hold type setting for when I can get it out of the enclosure. ?Is it there and I am just missing it?
Thanks in advance!
|
On Thu, Aug 27, 2020 at 06:53 PM, hwalker wrote:
Both the reference level and and attenuation can only be set in discrete steps of 1, 2 , 5, 10 and they interact to maintain a vertical scaling of 10 divisions.
There is a bug that causes strange auto reflevel behavior when in linear mode and changing the scale from small to a larger value. E.g. from 2 to 10. This causes the reflevel to reposition after one sweep after changing the scale to try to maintain a multiple of the scale
|
Kurt Poulsen
Hoshnasi