Good morning,
first of all excuse me for my broken english. I will try to be as understandable as possible. I have just started detecting with Echoes from France.
I have a Yagi antenna pointed towards the GRAVES source, an RTL-sdr dongle and i can hear FM radio so it works :D. And the installation of Echoes v.0.50 is OK.
I tried to follow the startup procedure described in the manual and here is my first batch of "results" along with the associated settings.
There is some questions i have :
The antenna is placed on the floor and inclined at 45° near a metal barrier, is this annoying?
Do you have any tips for improving capture? I work in periodic mode until i'm sure my settings are good before switch to automatic. I chek manually each capture, so i can miss good takes and keep false signals.
Do these detections correspond to meteors?
Thank you for your answers and thank you for this software and its possibilities. I can't wait to be able to implement Echoes during an evening of discovery at our amateur observatory
Max
Hi Max,
sorry for the delay in responding but I was on vacation and without a PC. I read
your message on the smartphone but I could not quite understand the attached pictures.
Meanwhile, I see from the "settings" image that you have tuned a frequency just 3khz away from the carrier of the Graves. There is nothing wrong with that, except the aesthetic problem of seeing the vertical line of the direct component of the signal at the frequency of 143.047.000, for this reason I usually put 10khz of difference, just to hide that line.
Question 1
For the antenna you should keep it at least 50cm away from the ground and walls, the metal barrier can give problems if it is a magnetic material, iron, steel... less problems with aluminum.
Question 2:
you should increase the power zoom so you can easily read, on the echoes you capture, the S-N difference they have. For example in photo #337 the difference (orange trace) reaches a peak of 33..36 dBfs on an average of 9 dBfs, thus by setting an upper threshold of, say, 13..14 dBfs and a lower one of 1 dBfs lower, you can already capture automatically with the differential thresholds. However you should adjust other parameters like "shot after..." to achieve good results.
If you look at other posts here, you should find some where I attached the rts file that I used in my station with the version 0.50 (now I'm experimenting with 0.51 and it has differences so I can't give you what I use now) the file is named AUTO_GRAVES.rts
Question 3:
Photo #68 and #126 do not look like meteors, but other artificial signals. You can recognize them because they have a regular pattern, like a pulse or a serie of regular pulses.
The others seem instead genuine echoes of meteors, in my opinion you should raise the brightness more (>80) in order not to lose some detail of the echo, even if the background becomes no longer black.
Greetings
Giuseppe
👍
1
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
thanks for your support. I found your auto_graves file and i'm trying it.
I modified the placement of the antenna and put it on a tripod
The issues i have is with the brigthness, when i raise it (80+) its all white. So i now use a value of 52, its not completely black. Or i adjust with the contrast but i do not find the correct setting for now.
The automatic mode seems to run correctly, but at some moment the upper and lower threshold change to , for exemple, 15-12 to -0.9/-0.8. and the mention CAPTURING appears. i'll update screenshots later.
I think i must work on the dump data and the delay for the screenshot (actually 80%). Can you tell me if those setting can be correct ? I have some difficulties understanding the manual and the explanaitions.
Waterfall coverage display 26000 ms, 26sec. i assume it's the time between the signal appears and when he get out of the screen (scroll down)
can i use dump data 52sec ?
and set the delay at 50% (screenshot at 13sec, when the signal reach the middle of the waterfall)
Another question (stupid) about the antenna, is there a way to set it ? like a top and a bottom ? And does the length of the cable influence the signal ? i use a 5 meter cable and if i want to get away of the metal barrier i must purchase a longer one.
I set my parameters back on periodic scans, trying to define a good upper and lower trheshold (because of the CAPTURING issues).
I don't know why my color VS dbs scale is graduating from -170 to +25 and i still have the top region of the color scale unused (no number superior to +25 in the scale). In the diffents screenshot in other topic the graduation at the top is +25.
I think i got some meteor shots but it's still hard to see them clearly in the waterfall. How can i make them bigger ? Or maybe set a slower waterfall covering ? But even if i change the FFT points from 32768 to a higher value the waterfall covering is still around 26sec.
On the dedicated PC i'm running the 32bit version of echoes v0.50
I answer to myself about some questions after reading the manual one more time :) RTFM !
I think i got some meteor shots but it's still hard to see them clearly in the waterfall. How can i make them bigger ?
Changing "Downsampled BW" from 2 048 000 to a lower one allow me to set a frequency ruller more precise.
Or maybe set a slower waterfall covering ? But even if i change the FFT points from 32768 to a higher value the waterfall covering is still around 26sec
I change the "Refresh interval requesting" to 50ms (100ms initially) and now the waterfall covering is faster : 13sec (26sec initially).
i'll check the results later, hope to have some catch !
Max
Last edit: Maxdelm 2023-07-20
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
I currently don't use the downsampling, but while developing 0.51 I found a bug in 0.50: while loading rts files, the bandwidth and offset sliders are not restored in the original positions so they remain in a default position. With a donwnsampled bw 2048000, I zoomed 100% so you get a display of 20 khz width and set offset +10khz with a tuning of 143.040.000 in order to get 143.050.000 in the center of the view, see screenshot attached.
Thanks for your help, I will check that after work.
Please can you tell me where I can change de color VS dbf scale ? In your screenshot its from -105 to +75. In my screenshots the range si from -170 to +25 and crushed. I think that if I change the scale it will be easier to define brightness and contrast settings.
Another question about the antenna, does the length of the cable influence the signal ? I use a 5 meter cable and if I want to get away of the metal barrier (and recover some space on my terrace) I must purchase a longer one.
Max
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Hi,
the brightness slider is bad scaled, in practice nobody will never set it below 80%, I fixed this too in 0.51. The dbfs are db not calibrated, so 0dbfs in my station can look different than yours. You could "calibrate"the dbfs in "preferences tab", with "scale gain" and "scale offset" controls. The first is a multiplier, the second is added to the resulting dbfs.
Your cable is long about the same of mine, I used a TV cable without care too much of impedency matching.
The antenna is a 3 elements yagi 50 ohm, the TV cable is 72ohm, the receiver is 50 ohm, I don't think it's so meaningful while receiving signals, I could be wrong, but I have no way to compare the my results with a pure 50 ohm signal chain.
Giuseppe
👍
1
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Automatic mode is now running for about 2 days and i think i found "good" parameters. Set the peak behavior in diffenrential mode, with a upper treshold at 16.5 dbfs and lower at 14 dbfs, based on the peaks i got.
I got some good shots, but I also get a lot of fake shots. From time to time my peak power curve is getting erratic and trigger shots.
Also my SDR dongle is very hot and this morning the peak power curve was completely flat. This is the model I'm currently using : This model
I'll update screenshots later, and investigate the dongle.
Max
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Hi,
about dongle hot, it seems to be normal. Mine too is hot even if it is a different model (Thumbnet Nongles N3). In the picture you see how I arranged the aluminium leg of my table as a heatsink.
The computer with Echoes is on the table - the dongle, a preamplifier and a splitter (to connect another dongle to the same antenna) are under the table (!)
About the false positives, do they all look similar? or are there several causes? During this time I receive many lightnings. In the past it happened that someone close to me transmitted packet radio on the same frequency as GRAVES, fortunately one day he suddenly stopped, because otherwise I wouldn't know how to discriminate such disturbances automatically, an ad hoc filter would be needed for that type of disturbance and I don't want to customize the program too much on my needs, if they are only mine
Giuseppe
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
num.0081 is harder for me to explain but happens more often
num.325 happens when the power curve is getting erratic, and it can trigger multiples shots, those are annoying and most of my fakes shots are like these.
I plugged the SDR, its still working. I dont know why it was disabled this morning.
For the good shots here are some (but I'm not shure if its meteors scatters...)
num37-73-91.
the 505 seems a normal underdense event, most of echoes look like this. Unless it happens at regular intervals, it should be a good one.
The 81 seems to be the GRAVES carrier reflected due to atmospheric phenomena. It happens sometimes, specially if you are located not too far from Dijon. To get rid of them, you should try by lowering the lower threshold so it would result in a too long event (lasting more than 60s) that will be discarded. In this case, the automatic thresholds work better than differential ones because in case the noise N varies along the day the thresholds will be automatically adapted while with differential the noise increment would deceive the capture algorithm resulting in a continuous discard.
325 - this seems artificial ESD due for instance to electric engines like hair dryers, vacuum cleaners or similar or even industrial ones. My ESD filter works better with natural ESD (lightings). Such noise could come from the air (antenna) or through the electric line. If they are regular, you should try to power your receiver via battery, make Echoes run on a laptop with battery, unplugged from electric line and see if you still get such noises. I have no such annoyances anymore since I installed an UPS
others for me are genuine meteor echoes.
Giuseppe
👍
1
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
I'm 350km away from GRAVES radar, I will try automatic thresholds.
The laptop is old and discharged quickly without his battery, but I will do some tests.
For the excessive heat, what do you think about a fanless heatsink ? Like this on the outside of the dongle
For now I think I can run Echoes during the "Nuit des étoiles" gathering scheduled the 12th and 13th august. Once again thanks for your advices and patience.
A new question that might deserve a dedicated topic, is that possible to run Echoes for meteor automatic detection and another software like SDR# or SDR++ for recording "audio" in the same time ? I saw some youtube videos with both softwares running. Do I need an additionnal SDR dongle ?
Max
Last edit: Maxdelm 2023-07-28
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Hi
that heatsink is for raspberry, it's sized about a 1€cent coin, you'll need a longer one I think, assuming your dongle is like a Nooelec RTL-SDR. Maybe there are ready made ones tailored for it.
About the second question, the same device cannot be opened by two different softwares simultaneously, so in the video you've seen, they probably used 2 different dongles.
With Echoes you could capture the audio output from SDR# from a mic or line inputs and open it with the audio device. You'll obtain an audio spectra where you can capture events as usual.
However, this is an option that, although implemented, has not been tested very much, so I don't have a ready rts file, I've read of someone who has tried it but I don't know with what results.
Giuseppe
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Good morning,
first of all excuse me for my broken english. I will try to be as understandable as possible. I have just started detecting with Echoes from France.
I have a Yagi antenna pointed towards the GRAVES source, an RTL-sdr dongle and i can hear FM radio so it works :D. And the installation of Echoes v.0.50 is OK.
I tried to follow the startup procedure described in the manual and here is my first batch of "results" along with the associated settings.
There is some questions i have :
Thank you for your answers and thank you for this software and its possibilities. I can't wait to be able to implement Echoes during an evening of discovery at our amateur observatory
Max
Last edit: Maxdelm 2023-07-10
Hi Max,
sorry for the delay in responding but I was on vacation and without a PC. I read
your message on the smartphone but I could not quite understand the attached pictures.
Meanwhile, I see from the "settings" image that you have tuned a frequency just 3khz away from the carrier of the Graves. There is nothing wrong with that, except the aesthetic problem of seeing the vertical line of the direct component of the signal at the frequency of 143.047.000, for this reason I usually put 10khz of difference, just to hide that line.
Question 1
For the antenna you should keep it at least 50cm away from the ground and walls, the metal barrier can give problems if it is a magnetic material, iron, steel... less problems with aluminum.
Question 2:
you should increase the power zoom so you can easily read, on the echoes you capture, the S-N difference they have. For example in photo #337 the difference (orange trace) reaches a peak of 33..36 dBfs on an average of 9 dBfs, thus by setting an upper threshold of, say, 13..14 dBfs and a lower one of 1 dBfs lower, you can already capture automatically with the differential thresholds. However you should adjust other parameters like "shot after..." to achieve good results.
If you look at other posts here, you should find some where I attached the rts file that I used in my station with the version 0.50 (now I'm experimenting with 0.51 and it has differences so I can't give you what I use now) the file is named AUTO_GRAVES.rts
Question 3:
Photo #68 and #126 do not look like meteors, but other artificial signals. You can recognize them because they have a regular pattern, like a pulse or a serie of regular pulses.
The others seem instead genuine echoes of meteors, in my opinion you should raise the brightness more (>80) in order not to lose some detail of the echo, even if the background becomes no longer black.
Greetings
Giuseppe
thanks for your support. I found your auto_graves file and i'm trying it.
I modified the placement of the antenna and put it on a tripod
The issues i have is with the brigthness, when i raise it (80+) its all white. So i now use a value of 52, its not completely black. Or i adjust with the contrast but i do not find the correct setting for now.
The automatic mode seems to run correctly, but at some moment the upper and lower threshold change to , for exemple, 15-12 to -0.9/-0.8. and the mention CAPTURING appears. i'll update screenshots later.
I think i must work on the dump data and the delay for the screenshot (actually 80%). Can you tell me if those setting can be correct ? I have some difficulties understanding the manual and the explanaitions.
Another question (stupid) about the antenna, is there a way to set it ? like a top and a bottom ? And does the length of the cable influence the signal ? i use a 5 meter cable and if i want to get away of the metal barrier i must purchase a longer one.
Thanks,
Max
Last edit: Maxdelm 2023-07-13
Back after some vacation.
I set my parameters back on periodic scans, trying to define a good upper and lower trheshold (because of the CAPTURING issues).
I don't know why my color VS dbs scale is graduating from -170 to +25 and i still have the top region of the color scale unused (no number superior to +25 in the scale). In the diffents screenshot in other topic the graduation at the top is +25.
I think i got some meteor shots but it's still hard to see them clearly in the waterfall. How can i make them bigger ? Or maybe set a slower waterfall covering ? But even if i change the FFT points from 32768 to a higher value the waterfall covering is still around 26sec.
On the dedicated PC i'm running the 32bit version of echoes v0.50
Keep on trying.
Max
Last edit: Maxdelm 2023-07-20
I answer to myself about some questions after reading the manual one more time :) RTFM !
Changing "Downsampled BW" from 2 048 000 to a lower one allow me to set a frequency ruller more precise.
I change the "Refresh interval requesting" to 50ms (100ms initially) and now the waterfall covering is faster : 13sec (26sec initially).
i'll check the results later, hope to have some catch !
Max
Last edit: Maxdelm 2023-07-20
Hi Max,
I currently don't use the downsampling, but while developing 0.51 I found a bug in 0.50: while loading rts files, the bandwidth and offset sliders are not restored in the original positions so they remain in a default position. With a donwnsampled bw 2048000, I zoomed 100% so you get a display of 20 khz width and set offset +10khz with a tuning of 143.040.000 in order to get 143.050.000 in the center of the view, see screenshot attached.
Giuseppe
Thanks for your help, I will check that after work.
Please can you tell me where I can change de color VS dbf scale ? In your screenshot its from -105 to +75. In my screenshots the range si from -170 to +25 and crushed. I think that if I change the scale it will be easier to define brightness and contrast settings.
Another question about the antenna, does the length of the cable influence the signal ? I use a 5 meter cable and if I want to get away of the metal barrier (and recover some space on my terrace) I must purchase a longer one.
Max
Hi,
the brightness slider is bad scaled, in practice nobody will never set it below 80%, I fixed this too in 0.51. The dbfs are db not calibrated, so 0dbfs in my station can look different than yours. You could "calibrate"the dbfs in "preferences tab", with "scale gain" and "scale offset" controls. The first is a multiplier, the second is added to the resulting dbfs.
Your cable is long about the same of mine, I used a TV cable without care too much of impedency matching.
The antenna is a 3 elements yagi 50 ohm, the TV cable is 72ohm, the receiver is 50 ohm, I don't think it's so meaningful while receiving signals, I could be wrong, but I have no way to compare the my results with a pure 50 ohm signal chain.
Giuseppe
Thanks for your answers.
Automatic mode is now running for about 2 days and i think i found "good" parameters. Set the peak behavior in diffenrential mode, with a upper treshold at 16.5 dbfs and lower at 14 dbfs, based on the peaks i got.
I got some good shots, but I also get a lot of fake shots. From time to time my peak power curve is getting erratic and trigger shots.
Also my SDR dongle is very hot and this morning the peak power curve was completely flat. This is the model I'm currently using : This model
I'll update screenshots later, and investigate the dongle.
Max
Hi,
about dongle hot, it seems to be normal. Mine too is hot even if it is a different model (Thumbnet Nongles N3). In the picture you see how I arranged the aluminium leg of my table as a heatsink.
The computer with Echoes is on the table - the dongle, a preamplifier and a splitter (to connect another dongle to the same antenna) are under the table (!)
About the false positives, do they all look similar? or are there several causes? During this time I receive many lightnings. In the past it happened that someone close to me transmitted packet radio on the same frequency as GRAVES, fortunately one day he suddenly stopped, because otherwise I wouldn't know how to discriminate such disturbances automatically, an ad hoc filter would be needed for that type of disturbance and I don't want to customize the program too much on my needs, if they are only mine
Giuseppe
Here are the screenshots,
For the fake signals :
I plugged the SDR, its still working. I dont know why it was disabled this morning.
For the good shots here are some (but I'm not shure if its meteors scatters...)
num37-73-91.
Max
Last edit: Maxdelm 2023-07-25
Hi Max,
the 505 seems a normal underdense event, most of echoes look like this. Unless it happens at regular intervals, it should be a good one.
The 81 seems to be the GRAVES carrier reflected due to atmospheric phenomena. It happens sometimes, specially if you are located not too far from Dijon. To get rid of them, you should try by lowering the lower threshold so it would result in a too long event (lasting more than 60s) that will be discarded. In this case, the automatic thresholds work better than differential ones because in case the noise N varies along the day the thresholds will be automatically adapted while with differential the noise increment would deceive the capture algorithm resulting in a continuous discard.
325 - this seems artificial ESD due for instance to electric engines like hair dryers, vacuum cleaners or similar or even industrial ones. My ESD filter works better with natural ESD (lightings). Such noise could come from the air (antenna) or through the electric line. If they are regular, you should try to power your receiver via battery, make Echoes run on a laptop with battery, unplugged from electric line and see if you still get such noises. I have no such annoyances anymore since I installed an UPS
others for me are genuine meteor echoes.
Giuseppe
Thanks again Giuseppe for your answers.
I'm 350km away from GRAVES radar, I will try automatic thresholds.
The laptop is old and discharged quickly without his battery, but I will do some tests.
For the excessive heat, what do you think about a fanless heatsink ? Like this on the outside of the dongle
For now I think I can run Echoes during the "Nuit des étoiles" gathering scheduled the 12th and 13th august. Once again thanks for your advices and patience.
A new question that might deserve a dedicated topic, is that possible to run Echoes for meteor automatic detection and another software like SDR# or SDR++ for recording "audio" in the same time ? I saw some youtube videos with both softwares running. Do I need an additionnal SDR dongle ?
Max
Last edit: Maxdelm 2023-07-28
Hi
that heatsink is for raspberry, it's sized about a 1€cent coin, you'll need a longer one I think, assuming your dongle is like a Nooelec RTL-SDR. Maybe there are ready made ones tailored for it.
About the second question, the same device cannot be opened by two different softwares simultaneously, so in the video you've seen, they probably used 2 different dongles.
With Echoes you could capture the audio output from SDR# from a mic or line inputs and open it with the audio device. You'll obtain an audio spectra where you can capture events as usual.
However, this is an option that, although implemented, has not been tested very much, so I don't have a ready rts file, I've read of someone who has tried it but I don't know with what results.
Giuseppe