Author Topic: Which is better?  (Read 1130 times)

0 Members and 1 Guest are viewing this topic.

Offline Dr Obbins

  • Forecaster
  • *****
  • Posts: 1148
Which is better?
« on: July 16, 2014, 08:41:01 PM »
So which settings do you think would give a better result to the servers? Does it matter?

A. 16 * 5 * 40 with threshold at 110 for a relative gain of 2,935. Noise floor 75 - 90
B. 4 * 4 * 40 with threshold at 25 for a relative gain of 2,666. Noise floor 15 - 20

I have not seen any real difference in strike count tonight with the storms far away. Do you think option "B" will be more susceptible to interference with close storms?

Offline JonathanW

  • Engineer
  • Forecaster
  • *****
  • Posts: 323
Re: Which is better?
« Reply #1 on: July 16, 2014, 08:49:47 PM »
Personally, I'd choose the "higher gain in the first stage" one.  One of the rules I learned of amplifier design was high gain in the first stage leads to less amplification of the noise in the later stages -> less noise overall.

That's one reason why receivers typically have a high gain, low noise input amplifier stage.  The exception might be if you have a very noisy first stage (with noise increasing along with gain), followed by a bunch of low noise stages.  But I don't think we have that here.
« Last Edit: July 16, 2014, 09:10:39 PM by n0ym »

Offline Dr Obbins

  • Forecaster
  • *****
  • Posts: 1148
Re: Which is better?
« Reply #2 on: July 16, 2014, 09:11:08 PM »
So how about:
C: 32 * 1 * 40 with threshold at 45 for a relative gain of 2,909. Noise floor 27 - 35

Offline JonathanW

  • Engineer
  • Forecaster
  • *****
  • Posts: 323
Re: Which is better?
« Reply #3 on: July 16, 2014, 09:29:23 PM »
I generally run with as much gain in the first stage as possible, then raise the gain of the second stage, etc.

BUT, from what I understand, the developers recommend threshold values in the 80-120 mV range (much higher than that and you risk some signals maxing out/clipping in the A/D converters; much lower than that and you may start to run into quantization issues with very low bit resolution of the waveforms, as well as too little signal-to-noise).

If you watch the auto filters, the threshold is all over the place (including far outside the 80-120 mV range).  Even so, I generally do try to keep my thresholds around there.

What about 32*2*40 with a threshold of 90 mV (assuming you want that relative gain to be around 2900)?

Where I have my system now, I've found that I can usually run 32 * 2 * 40 with thresholds from 80-120 mV.  The noise level is usually less than +/- 20 mV peak (I'm working toward +/- 10 mV peak). If I try to run higher gain (not relative gain, absolute gain setting) with strikes anywhere near me, I start to see some signals on the scope page maxing out when the y-axis is set to 100% (I'm not sure what the actual A/D max is, but it still makes me a bit uncomfortable that I may send distorted waveforms).  PLUS, we have enough coverage in the midsection of the country now that there's not a whole lot of point in me trying to participate in signal capture from the west coast.

When strikes get closer, I reduce the relative gain accordingly, of course.
« Last Edit: July 16, 2014, 10:11:15 PM by n0ym »

Offline dfroula

  • Forecaster
  • *****
  • Posts: 543
Re: Which is better?
« Reply #4 on: July 16, 2014, 09:33:05 PM »
On the Controller web pages, thresholds are expressed in + or - millivolts, whereas noise is expressed as peak-peak voltage.

Therefore, in the first example, there is a margin of (110mV - (75mV/2)) = 72.5mV to (110mV - (90mV/2)) = 65mV between peak noise and the trigger threshold at whatever gain.

In the second example, the margin is (25mV - (15mV/2)) = 17.5mV to (25mV - (10mV/2)) = 20mV

I like to run with a target equivalent gain that offers the higher margin between noise and threshold for better noise immunity. I typically leave my thresholds at the default 120mV and just adjust the gains for optimal signal rates.

Don
WD9DMP

So which settings do you think would give a better result to the servers? Does it matter?

A. 16 * 5 * 40 with threshold at 110 for a relative gain of 2,935. Noise floor 75 - 90
B. 4 * 4 * 40 with threshold at 25 for a relative gain of 2,666. Noise floor 15 - 20

I have not seen any real difference in strike count tonight with the storms far away. Do you think option "B" will be more susceptible to interference with close storms?
« Last Edit: July 16, 2014, 09:35:02 PM by dfroula »

Offline JonathanW

  • Engineer
  • Forecaster
  • *****
  • Posts: 323
Re: Which is better?
« Reply #5 on: July 16, 2014, 09:43:09 PM »
I wish we could "like" posts -- Don's absolutely right. 

You can also watch your scope (you can check the "noise floor" checkbox and get a visual idea of what your signal noise levels are without strikes, of course) and see what kind of margins you have between your signal noise and thresholds.

No matter what you do, you'll have a certain minimum system noise.  Obviously, we want to reduce that as much as possible, but you'll still need to have some breathing room between that and the threshold to avoid sending lots of junk :)

Offline Dr Obbins

  • Forecaster
  • *****
  • Posts: 1148
Re: Which is better?
« Reply #6 on: July 16, 2014, 09:49:05 PM »
Good information guys.I am trying to find the spot where I will be leaving the settings and not changing per time of day or closeness of storms. With the settings for the lower thresholds, the noise range was less, but I understand what you are saying about the higher margin.

32 * 2 * 40 with threshold at 90 for a relative gain of 2,844. Noise floor 40-60.
or
16 * 5 * 40 with threshold at 115 for a relative gain of 2,782. Noise floor 50-80.

I also notice with the higher gains, channel "A" needs about 10% higher threshold than channel "B".

Offline JonathanW

  • Engineer
  • Forecaster
  • *****
  • Posts: 323
Re: Which is better?
« Reply #7 on: July 16, 2014, 09:54:40 PM »
Good information guys.I am trying to find the spot where I will be leaving the settings and not changing per time of day or closeness of storms. With the settings for the lower thresholds, the noise range was less, but I understand what you are saying about the higher margin.

32 * 2 * 40 with threshold at 90 for a relative gain of 2,844. Noise floor 40-60.
or
16 * 5 * 40 with threshold at 115 for a relative gain of 2,782. Noise floor 50-80.

I also notice with the higher gains, channel "A" needs about 10% higher threshold than channel "B".

I haven't yet found gain settings that I can set and leave, at least if storms roll through :)  I have VPN access to my home network, and I admit I ride the settings as much as possible during the day.  Great use of my time, I'm sure...

Anyway, all good info, and I'm still tweaking my own methodology.  It appears to be an art form in many ways, and I see Mike is adding some additional considerations in another thread (I'm still working the math on TOGA myself; have gotten a bit rusty).  We'll see what kind of "auto mode magic" Egon and co. end up producing.
« Last Edit: July 16, 2014, 09:57:48 PM by n0ym »

Offline Dr Obbins

  • Forecaster
  • *****
  • Posts: 1148
Re: Which is better?
« Reply #8 on: July 16, 2014, 10:01:49 PM »
Yea.. After playing around tonight and listening to the advice, I am right back to my "normal" settings.  :???:
The only thing I am afraid of with the auto settings is that it typically cuts my lightning count in . Now if the network as a whole is getting these strikes from other stations, I guess that is not an issue. For example, in April / May about 10% of my strikes were min stations only. With the now 18 new stations since then, about 2%-3% of the strikes are min stations only. So that was the thought for tonight, find the settings that are good and don't need monitoring.

Offline JonathanW

  • Engineer
  • Forecaster
  • *****
  • Posts: 323
Re: Which is better?
« Reply #9 on: July 16, 2014, 10:04:34 PM »
Yea.. After playing around tonight and listening to the advice, I am right back to my "normal" settings.  :???:
The only thing I am afraid of with the auto settings is that it typically cuts my lightning count in . Now if the network as a whole is getting these strikes from other stations, I guess that is not an issue. For example, in April / May about 10% of my strikes were min stations only. With the now 18 new stations since then, about 2%-3% of the strikes are min stations only. So that was the thought for tonight, find the settings that are good and don't need monitoring.

Some people on the BO forum have remarked that auto mode, as it is, results in their systems going in and out of interference mode, and I've had the same thing.  I guess that's why it's a "beta" right now :)

Offline Dr Obbins

  • Forecaster
  • *****
  • Posts: 1148
Re: Which is better?
« Reply #10 on: July 16, 2014, 10:09:58 PM »
I was referring to the auto amplitude and noise filters. I have stayed away from the auto mode after trying once.  :shock:
Going with the:
16 * 5 * 40 with threshold at 115 for a relative gain of 2,782. Noise floor 50-80.
It jumped from 73% to 89% on the Effectivity L scale.
« Last Edit: July 16, 2014, 10:12:41 PM by Dr Obbins »

Offline JonathanW

  • Engineer
  • Forecaster
  • *****
  • Posts: 323
Re: Which is better?
« Reply #11 on: July 16, 2014, 10:11:43 PM »
Ah, OK.  I do make use of the auto noise filter.

Offline Dr Obbins

  • Forecaster
  • *****
  • Posts: 1148
Re: Which is better?
« Reply #12 on: July 16, 2014, 10:15:22 PM »
If any I use the Auto Amplitude. It seems to be less severe. I only use the Auto Noise if there are storms with in 100 miles.

Offline dfroula

  • Forecaster
  • *****
  • Posts: 543
Re: Which is better?
« Reply #13 on: July 16, 2014, 10:56:42 PM »
I don't think there is a "set and forget" spot when running in any variation of manual mode.

Before I relocated my antennas and got my periodic noise situation under control, I needed to enable the auto-noise threshold feature to deal with my mysterious afternoon noise bursts. The algorithm tries to keep the thresholds low and reduce the gains to keep the relative gain targets. This results in very frequent threshold changes and somewhat less frequent gain changes. It worked OK, but lowered overall system performance.

After relocating the antennas and improving the ground, I have auto-noise threshold adjust "off" and run in full-manual mode. I only need to reset the gains for near storms or unusual North American lightning density, if I care to bother. Leaving the thresholds high and boosting the gain some seems to work best for me, under very low local noise floors.

The key to good performance with minimum manual gain adjustments is to take care of all your local noise sources with a good ground, shielded ferrites, and good antenna placement. Then, run in full manual mode. Don't shoot for the insane strike rates that are now allowed. I shoot for just under 15/sec on the per-minute average under "normal" lightning conditions. This is enough to push my strike ratios into the 90%s.

Forget "Auto" mode for now.

Use that big margin between noise floor and trigger threshold, and average signal rate to the new higher maxima to your advantage! It helps minimize gain adjustments.

Don
WD9DMP

I tried the server "Auto" mode a few times and found it completely unsatisfactory for reasons mentioned by others.

 

anything