Author Topic: mV Threshold  (Read 1811 times)

0 Members and 1 Guest are viewing this topic.

Offline 92merc

  • BismarckWeather.net
  • Forecaster
  • *****
  • Posts: 934
  • BismarckWeather.net
    • BismarckWeather.net
mV Threshold
« on: June 06, 2014, 06:22:04 PM »
I'm still running 6.8 firmware in manual mode.  Someone had recommended running in manual and start with 10*10 and 100mv setting.  I've been doing that for awhile now and I have a fairly good feel for when I have to back the gain down.

What I'm not totally understanding is the function of the mV Threshold number.  As I'm changing the amplification, should I be adjusting that threshold as well?  If there are local storms, I have to back off my amp values to as low as 10*2 to stay out of interference mode.  I'm just wondering if I should be changing the mv setting too.

I've also been reading about some changes in the 7 beta version with the threshold.  It looks like the beta version can auto adjust some values depending on conditions.  Is the beta version better with these auto adjustments?

TIA
https://www.BismarckWeather.net
Davis VP2, Cumulus, WeatherDisplay, Blitzortung, Bloomsky, Saratoga Scripts, NOAA Stream via PI

Offline Cutty Sark Sailor

  • WxElement panel
  • Forecaster
  • *****
  • Posts: 3127
    • Frankfort Weather - TwinHollies WeatherCenter
Re: mV Threshold
« Reply #1 on: June 06, 2014, 06:37:21 PM »
The threshold determines the amplitude required to 'trigger' a signal...
but it gets a bit confusing when you think of interference,...
"interference" occurs when the number of signals exceeds a certain amount...
and there are two types:
"Burst"... over 2 seconds the system averages 30 signals per second.
"Normal... over a period of 60 seconds, the system averages 15 signals per second.

The threshold determines the trigger point and should be set high enough to only trigger on stronger signals.

For some reason, I tend to think of it in terms of an 'interference' adjustment, since, in effect it
determines the number of triggers, for me, more obviously than gains do. I do not remember what the suggested thresholds were,
but I believe it was originally 120.  For the H amp mine are now typically 100-120, and for E
typically 90.  More discussion in the "Adding E-field" thread... http://www.wxforum.net/index.php?topic=22710.0
 

Offline 92merc

  • BismarckWeather.net
  • Forecaster
  • *****
  • Posts: 934
  • BismarckWeather.net
    • BismarckWeather.net
Re: mV Threshold
« Reply #2 on: June 06, 2014, 06:53:57 PM »
That 90mv threshold jives with what I found too.  After figuring out my amp numbers, I found I could run 90 quite easily.  But even dropping it to 85mv, I start going into interference quite easily.

I'll read that other thread.   Thanks.
https://www.BismarckWeather.net
Davis VP2, Cumulus, WeatherDisplay, Blitzortung, Bloomsky, Saratoga Scripts, NOAA Stream via PI

Offline JonathanW

  • Engineer
  • Forecaster
  • *****
  • Posts: 323
Re: mV Threshold
« Reply #3 on: June 06, 2014, 07:41:27 PM »
The appropriate Gain and threshold, it seems to me, are likely complex functions of 1) the noise floor (too close to the threshold, and you'll get false triggers, 2) the strength of the incoming lightning signals, 3) any interfering signals you're getting, 4) overhead between the signals you're getting and the maximum signal the ADC can translate (in other words, is the gain set so high that lightning waveforms are being clipped?  Better to turn down both the gain and the threshold) and 5) the minimum signal the ADC can translate (if the signal is too low, even if it's above the system noise, the lightning impulse may have poor resolution in bits).

So, any manual or automatic system must make sure that the highest lightning signals you receive don't drive the ADC into clipping, while at the same time making sure the gain is high enough that system noise doesn't swamp low-level signals (to the greatest extent possible).  The threshold must be set high enough that even low-level lightning signals trigger a detection, yet false triggers due to noise or interference are minimized.

Where's the sweet spot?  Depends on the system, including the antenna types and ambient RF environment, and you guys know all of that stuff better than I do :)

I'm slowly learning the in's and out's of my system, and yes, Mike, I've been running manual mode for experimentation :)

I agree with those who say with running both E- and H-field antennas that the alternate channel mapping seems to make more sense...
« Last Edit: June 06, 2014, 07:45:56 PM by n0ym »

Offline dfroula

  • Forecaster
  • *****
  • Posts: 543
Re: mV Threshold
« Reply #4 on: June 06, 2014, 08:33:41 PM »
The threshold determines the "relative gain" of your system, using 100mV as the reference point. The "relative gain" can be adjusted higher by increasing the gains of a channel or by reducing the threshold. It is better to run as near your noise floor as possible. However, lower thresholds will increase the effect that transient noise has on the channel, increasing the probability of false strike indications.

The "relative gain" formula is: (Stage A Gain) * (Stage B Gain) * (40) * (100/Threshold_in_mV)

So if you set your gains to 10x10 with a threshold of 100mV, the "relative gain" is 10 *10 * 40 * 100/100 = 4000

If you reduce the threshold to 90mV, the relative gain increases: 10 * 10 * 40 * 100/90 = 4444. However, the system will be more prone to false triggers on noise.

The "Auto Adapt to Noise" feature adjusts to the optimal threshold based on the noise level between strike detections. It then adjusts the A and B gains as high as possible, up to the limits you have set in the gain settings.

The percentages in the "Amplifiers" section of the Status page, with "Auto Adapt to Noise" enabled, show the percentages of the maximum gains and thresholds you have set on the Settings page.

The "Relative Gain" is always shown on the Status page, regardless of the "Auto Adapt to Noise" setting. Those are the numbers to look at to see the effect of the threshold and gain settings on the overall sensitivity of your system.

It's kind of confusing until you watch it in action for a while.

Regards,

Don


Offline JonathanW

  • Engineer
  • Forecaster
  • *****
  • Posts: 323
Re: mV Threshold
« Reply #5 on: June 06, 2014, 08:51:00 PM »
Don,
That's excellent information.  A question, though: it appears "auto-adaption" can reduce gains and thresholds from what you set, but not go over them.  BUT, right now I have my status page reporting "100%" gain on one channel, even though, obviously, the threshold could go lower.

Does the 100% just apply to the gain values?  Or does it mean the auto-adaption feature will not adjust the threshold lower than it is, for a given gain level?

In other words, does the percentage displayed apply only to the gain, or to the relative gain?

EDIT: nevermind, it appears to track the relative gain.  The controller must have a maximum relative gain value, then.

EDIT2: OK, I get it.  The controller has a maximum relative gain value as set by the gain and threshold values you enter.  The relative gain won't go over that value in any combination of gain and threshold.
« Last Edit: June 06, 2014, 09:06:33 PM by n0ym »

Offline dfroula

  • Forecaster
  • *****
  • Posts: 543
Re: mV Threshold
« Reply #6 on: June 06, 2014, 09:08:52 PM »
Exactly!

EDIT: The percentages reflect the percentage of maximum relative gain using the maximum threshold and gain you have set on the "Settings" page. All assuming "Auto Adapt to Noise" is enabled, of course. Otherwise, the percentages are not shown on the Status page. With "Auto Adapt to Noise", the controller can set any combination of gains and thresholds it wants to, but it will never go over the relative gain that your threshold and gains on the settings page compute to.
« Last Edit: June 06, 2014, 09:14:51 PM by dfroula »

Offline JonathanW

  • Engineer
  • Forecaster
  • *****
  • Posts: 323
Re: mV Threshold
« Reply #7 on: June 06, 2014, 09:17:18 PM »
So, a question I have, for those more experienced than me... obviously, it's best to site the system where there's consistently low noise and interference.  But if it's in a place where the noise and interference are variable, is it better to set the relative gain high (so you capture weaker lightning when you can) and let the system go into interference mode when interference hits, until the controller reduces the relative gain, or to set the relative gain such that interference mode is generally avoided?  This is assuming "auto adapt to noise" is on.
« Last Edit: June 06, 2014, 09:19:52 PM by n0ym »

Offline dfroula

  • Forecaster
  • *****
  • Posts: 543
Re: mV Threshold
« Reply #8 on: June 06, 2014, 09:34:34 PM »
I would put it that you should set the (max) relative gains at levels that allow the system to stay out of interference under average nationwide lighting conditions, with no medium-intensity storms with 300 miles or so.

Then, keep those ideal relative gains, but adjust the thresholds and gains downward so that the thresholds are as low as possible, given your local noise/interference situation - still keeping the same computed relative gains

So, for example, you may set your gains and thresholds to 16x10 at 120mV on both channels and find that the sensitivity of the system keeps you out of interference most of time, given lightning conditions mentioned above. Those settings compute to a relative gain of 16 * 10 *40 * 100/120 = 5333. You have a low noise floor between detections and little local interference, so you decide to reduce the threshold to 65mV on both channels. To keep the same optimal relative gain sensitivity, you would also need to readjust the gains lower to maintain an equivalent relative gain, say 10x8. That would give roughly the same relative gain (10 * 8 * 40 * 100/65 = 4923.

Basically, that's part of the manual process that equates to what "auto adapt to noise" does. At some point, excessive noise or excessive lightning strikes, which look like noise, will force the "auto adapt to noise" algorithm to begin reducing the relative gains from what you have set.

Operating with lower thresholds has the advantage of requiring less gain in the amplifiers, reducing amplifier noise and operating with a better signal/noise ratio which enhances sensitivity, even though the relative gains of both sets of settings are apparently equivalent.

Easier said than done, and there is no one setting that will account for all combinations of lightning/noise conditions.

Don
WD9DMP
« Last Edit: June 06, 2014, 09:41:00 PM by dfroula »

Offline JonathanW

  • Engineer
  • Forecaster
  • *****
  • Posts: 323
Re: mV Threshold
« Reply #9 on: June 06, 2014, 09:53:37 PM »
I would put it that you should set the (max) relative gains at levels that allow the system to stay out of interference under average nationwide lighting conditions, with no medium-intensity storms with 300 miles or so.

Then, keep those ideal relative gains, but adjust the thresholds and gains downward so that the thresholds are as low as possible, given your local noise/interference situation - still keeping the same computed relative gains

So, for example, you may set your gains and thresholds to 16x10 at 120mV on both channels and find that the sensitivity of the system keeps you out of interference most of time, given lightning conditions mentioned above. Those settings compute to a relative gain of 16 * 10 *40 * 100/120 = 5333. You have a low noise floor between detections and little local interference, so you decide to reduce the threshold to 65mV on both channels. To keep the same optimal relative gain sensitivity, you would also need to readjust the gains lower to maintain an equivalent relative gain, say 10x8. That would give roughly the same relative gain (10 * 8 * 40 * 100/65 = 4923.

Basically, that's part of the manual process that equates to what "auto adapt to noise" does. At some point, excessive noise or excessive lightning strikes, which look like noise, will force the "auto adapt to noise" algorithm to begin reducing the relative gains from what you have set.

Operating with lower thresholds has the advantage of requiring less gain in the amplifiers, reducing amplifier noise and operating with a better signal/noise ratio which enhances sensitivity, even though the relative gains of both sets of settings are apparently equivalent.

Easier said than done, and there is no one setting that will account for all combinations of lightning/noise conditions.

Don
WD9DMP

Thanks again, Don.  I'll have to experiment further -- not to disagree with you, but I was always taught that higher gain in the initial amplifier stages was better for overall signal-to-noise ratio.  But then, VLF lightning detection may be a different animal :)

I went to full manual and backed my gain/threshold combination down such that I was staying out of interference mode, before returning to "semi-automatic".  I noticed that most of the burst stuff I was getting was via the e-field C-channel (50 kHz lowpass), with a very repetitive interference signal.
« Last Edit: June 06, 2014, 09:56:25 PM by n0ym »

Offline dfroula

  • Forecaster
  • *****
  • Posts: 543
Re: mV Threshold
« Reply #10 on: June 06, 2014, 10:18:50 PM »
I agree.....When setting the gains, the first stage should always have the higher gain of the two adjustable amps.

I guess the point I was making is that lower overall gain always provides better signal/noise ratio if your signal source is strong enough and your detector (threshold) is sensitive enough to allow it.

Don
WD9DMP

Offline JonathanW

  • Engineer
  • Forecaster
  • *****
  • Posts: 323
Re: mV Threshold
« Reply #11 on: June 06, 2014, 10:23:14 PM »
I agree.....When setting the gains, the first stage should always have the higher gain of the two adjustable amps.

I guess the point I was making is that lower overall gain always provides better signal/noise ratio if your signal source is strong enough and your detector (threshold) is sensitive enough to allow it.

Don
WD9DMP

I agree.  By the way, tonight's the first time I'm running E-field along with the H-field - getting strikes over 2000 km away on both types of antennas.  I'm guessing, if the E-field is intended more for close-in work, that I'll want to back the gain down/raise the threshold on that amp at some point :)