Virtual Radar Server - Trouble shooting!

[jeallen01]

Overnight I got an email from Sarah at NooElec - they have all the info they need and I don't need to send the dead unit back.

Got the new unit working but still having problems with dodgy connections - will have to more firmly mount the filter unit to the wall board to stop that happening . Currently tracking around 45-50 planes - is that about right ATM?

Still rather suspicious about the SDR unit heat issues - this one is also strapped to the SSA like the first one and the case is slightly warm, but I'm wondering if the heating is a common issue. Anyone else noticed it?

PS to anyone (Chris?) who has installed a line amp which relies on line power - you'll get virtually no signals until you power the amp up via a DC inserter (as I found when sorting the above issues!

 

[jeallen01]

And now I think this one has died as well

 

[Archive-8]

I just haven't had the time to get that sorted. But with your issues I am holding off.

My dongle gets a little warm, but not to hot, that's plugged into my raspberry pi.

It sounds to me like you are overloading the input side of the dongle.

@Vipersan @John @scopus @Llew can you shed any light on this please?

 

[jeallen01]

I just haven't had the time to get that sorted. But with your issues I am holding off.

Don't blame you!

My dongle gets a little warm, but not to hot, that's plugged into my raspberry pi.
It sounds to me like you are overloading the input side of the dongle.

The first one of mine got fairly warm even without the line amp inserted.
Can't see how I would have overloaded the I/P because the actual antenna signal would only be a few microvolts, and even with 20dB of amplification, that would only be a hundred millivolts or so at most (1V = 60dbV on 1uV IIRC).

@Vipersan @John @scopus @Llew can you shed any light on this please?

Add me to that list of who wants to know more!

PS: just done a quick continuity check on the DC inserter with a DMM - no sign of a short between the DC insertion port and the port for the receiver, which is as it should be because the case label shows there to an isolating capacitor between the two, whereas I do have continuity between the DC insertion port and the port connected to the cable from the antenna (and thus the line amp near the antenna). Also, the label shows that the unit should be capable of inserting up to 65V, and so a 15V insertion should not cause a problem to that capacitor

 

[scopus]

I just haven't had the time to get that sorted. But with your issues I am holding off.

My dongle gets a little warm, but not to hot, that's plugged into my raspberry pi.

It sounds to me like you are overloading the input side of the dongle.

@Vipersan @John @scopus @Llew can you shed any light on this please?

Not used my dongle for a while, then only with PP and Jaero, didn't notice any issues..

 

[jeallen01]

Based on what Sarah asked me to do first time around, I reinstalled Zadig (again!) and have just copied & pasted the Advanced Mode log into a WORD Document (which I can PM to someone but I can't seem to post here).
The only things that I can seem to make of it are:
- When I do "List Devices" from "Options" I see 1x Bluetooth device & 3x USB Hubs - all of which are built into the N2620 Nettop - and then two more as shown below
"Unknown USB Device (Device Descriptor Request Failed)"

NB: I think the message at the very bottom of that should be ignored because that is that last line of the full log and does not appear to relate directly to the above image

& then "FT232R USB UART" which appears to have installed correctly


but I can't really make head or tail of that - can anyone else?

 

[Archive-8]

Is the dongle plugged directly into the pc?

Not with a cable.

 

[Analoguesat]

That appears to have the windows usb driver assigned to it. The one to the right of the green arrow should change to the zadig driver.

I remember last time I rebuilt my windows machine (when the hdd died before xmas) I had a hell of a game getting the zadig driver to install correctly for the dongle. I had to do summat slightly odd to get it to assign correctly.

 

[jeallen01]

Is the dongle plugged directly into the pc?

No, I used about a 30cm long USB extension cable because the weight of the dongle puts quite a load on the port if plugged-in directly as it then hangs in mid-air. (FWIW, I tried another extension cable,and also tried both USB 2 & USB 3 ports, with the same nil result, )
Everything is now shut down, but I will try direct plugging it in tomorrow morning - OTOH, I somehow doubt that that would make a difference).

 

[Llew]

I can't do anything with it at present, I'm redecorating the room where the power inserter normally is that feeds to the preamp. Never had any probs with power, or Zadig. Should get back to working with it soon.

 

[jeallen01]

Fired up the N2620g this morning with the dongle plugged directly into a USB2 port - no change in that dump1090 could find no RTLSDR device.

OTOH,
- when I opened Zadig, it could not see the "unindentified devices" or the FT232R USB UART which it could see last night
- when I looked at Device Manager, there was a change in that it could also not see the "unindentified devices" or the FT232R USB UART which it could see last night.

So, "something" appeared to happen when the dongle was unpowered and cooled down overnight which changed it from a device which was partially recognised by the PC to one where nothing was recognised.

Also, overnight, I got another email from Sarah at NooElec in reply to mine early yesterday that I had received but not installed the device - I will reply to that in a few minutes, and the response I hope to get on Monday should be "interesting"!!!!

 

[Archive-8]

I can't do anything with it at present, I'm redecorating the room where the power inserter normally is that feeds to the preamp. Never had any probs with power, or Zadig. Should get back to working with it soon.

Good luck with the decorating

Fired up the N2620g this morning with the dongle plugged directly into a USB2 port - no change in that dump1090 could find no RTLSDR device.

OTOH,
- when I opened Zadig, it could not see the "unindentified devices" or the FT232R USB UART which it could see last night
- when I looked at Device Manager, there was a change in that it could also not see the "unindentified devices" or the FT232R USB UART which it could see last night.

So, "something" appeared to happen when the dongle was unpowered and cooled down overnight which changed it from a device which was partially recognised by the PC to one where nothing was recognised.

Also, overnight, I got another email from Sarah at NooElec in reply to mine early yesterday that I had received but not installed the device - I will reply to that in a few minutes, and the response I hope to get on Monday should be "interesting"!!!!

Can you try the dongle on another pc? And see if you can rule out a dongle fault.

And is this the pc you had to reinstall?

 

[jeallen01]

Can you try the dongle on another pc? And see if you can rule out a dongle fault.

Following your suggestion, I've now tried on the X275 ->
- Zadig sees "something" but can't install a USB driver.
- Device Manager sees a USB "something" but has "stopped it" because it's not working.

And is this the pc you had to reinstall?

No, it was used on the N2620g nettop when it (and the previous one) went wrong, and the above test was on the X275 small desktop which is the one on which I had to completely reinstall everything (and which is currently running - very slowly! - ProgDVB to reinstall all the stuff for the twin DVB-S2 sat tuner card. Boy, does ProgDVB/this tuner card have a slow blindsearch!).

Therefore, IMHO, this dongle is also "stuffed".

 

[Archive-8]

Following your suggestion, I've now tried on the X275 ->
- Zadig sees "something" but can't install a USB driver.
- Device Manager sees a USB "something" but has "stopped it" because it's not working.


No, it was used on the N2620g nettop when it (and the previous one) went wrong, and the above test was on the X275 small desktop which is the one on which I had to completely reinstall everything (and which is currently running - very slowly! - ProgDVB to reinstall all the stuff for the twin DVB-S2 sat tuner card. Boy, does ProgDVB/this tuner card have a slow blindsearch!).

Therefore, IMHO, this dongle is also "stuffed".

It does sound like a dodgy dongle.

It would be nice to get to the bottom of what is causing the problem.

And why yours gets so hot, all 3 of mine get a little warm, but not hot!

I use 1 or 2 dongles in a windows 10 pc, and 1 in my pi.

My pre-amp and Raspberry Pi are powered from my virtual radar server pc via usb cables.

 

[jeallen01]

It does sound like a dodgy dongle.
It would be nice to get to the bottom of what is causing the problem. .

From what I can see/feel, the dongles are getting to surface temps of around 55C (and that was happening even before I added the line amp near the antenna, and the DC inserter to power it!) when not strapped to the SSA metal case, and around 30-35C when they are - to me that means that the internal temps are quite a bit higher, and probably some of those are very localised.
(FWIW, I did a lot of product safety testing before I retired, and that included extended temperature testing to ensure that products & critical components did not exceed safe limits for insulation and to be touched - so I kinda have a "calibrated finger" which knows when something gets near the safe-to-touch temps limits, such as around 60C for metal handles/cases!)
What is kinda puzzling is the way that the USB id/response changes after they have begun to "fail" and are then allowed to cool down - at which point they seem to fail completely. To me that means there is one or more of the following defects:
- one or more overheated electronic components failing, and it seems to be the RF ends that fail first, followed by the USB interfaces.
- soldered joints failing due to thermal stressing/cooling factors, which would be understandable if the joints (presumably most of the components are SMD) were not made correctly in the first place
- inadequate manufacturing process test & burn-in times to screen out dodgy units and identify the root causes of the problems

 

[jeallen01]

Update: As I said in an earlier post, the replacement unit failed in pretty short order - about 3hrs running, and then almost everything died!

Since then I've had a number of exchanges of emails with Alex in NooElec Tech Support, and he's had another replacement sent to me, whilst I have sent him details of the line amp, dc inserter and filter/PSU and am still waiting for any explanations that he can come up with.

In the meantime, I've taken the line amp/dc inserter etc., out of the antenna line and am now "running in" this new unit with a long VRS soak-test (currently seeing about 20-25 planes, but, of course, without the range I got with the line amp) - that started about an hour ago, and will leave it running for the next days or so 0to see what happens . BTW, Alex said that it is normal for the SDR unit to get quite warm, and so the new one is not secured to the SSA case like the first two were.

You never know, this one might continue to work;)

 

[jeallen01]

My last post here was over 2 weeks ago, because, although I have sent NooElec various info, component specs and several gentle reminders, I have had SFA from them since that post (except for acknowledgements and "we will pass it on to the relevant department" messages).

Not the sort of responses I was beginning to feel were typical of the company!

 

[jeallen01]

My last post here was over 2 months ago as I have been very pre-occupied in doing lots of other things - but now FINALLY trying to get back to setting up the SDR & VRS etc - but, of course, I am running into all sorts of problems, such as:
- The N2620g on which the s/w was installed had a WIN 10 meltdown and I had to reinstall that from scratch, via the original W7 licence - but now it refuses to even start because, I think, the little front panel switch is duff!
- So now trying to get the stuff running on the N4620G (basically same as the N2620g but Icore 3 processor) running WIN 10 Pro.

It having been so long since I played with this stuff, I am virtually having to relearn absolutely everything, but I have now cross-loaded most files from another machine, run the Zadig driver installation file and installed VRS, Mlat Client (including editing the bat file) and Dump1090 .
Edited: Now Found Chris' instructions on dump1090 bat file editing and the file in use appears t be correct

Anyway, I'm now hoping to get a bit further along now the line.

BTW, before getting VRS fully working Dump1090 is now reporting some 20 or so planes in the vicinity - not really a big improvement considering that I finally "bit the bullet" and moved the co-linear from the end of the garden to the house eves just outside our "office" in order to dispense with the immediate need for a line amp (having apparently "killed" 2 Noolec devices with the sat line amp "solution earlier this year). I nearly killed myself in the process in fitting 18" T&K brackets just below the eves to hold that antenna and the discone (it meant standing on the pitched tiles on the top of the utility area end wall, about 10ft from the ground, to use an SDS drill to drill the anchor bolt holes whilst trying - successfully! - not to fall backwards to the ground or forwards through the polycarbonate roof of the utility area)
Edited: Now got the VRS server showing about 50-55 planes being tracked - but not yet got the splatter diagram working againn

 

[Archive-8]

My last post here was over 2 months ago as I have been very pre-occupied in doing lots of other things - but now FINALLY trying to get back to setting up the SDR & VRS etc - but, of course, I am running into all sorts of problems, such as:
- The N2620g on which the s/w was installed had a WIN 10 meltdown and I had to reinstall that from scratch, via the original W7 licence - but now it refuses to even start because, I think, the little front panel switch is duff!
- So now trying to get the stuff running on the N4620G (basically same as the N2620g but Icore 3 processor) running WIN 10 Pro.

It having been so long since I played with this stuff, I am virtually having to relearn absolutely everything, but I have now cross-loaded most files from another machine, run the Zadig driver installation file and installed VRS, Mlat Client (including editing the bat file) and Dump1090 .
Edited: Now Found Chris' instructions on dump1090 bat file editing and the file in use appears t be correct

Anyway, I'm now hoping to get a bit further along now the line.

BTW, before getting VRS fully working Dump1090 is now reporting some 20 or so planes in the vicinity - not really a big improvement considering that I finally "bit the bullet" and moved the co-linear from the end of the garden to the house eves just outside our "office" in order to dispense with the immediate need for a line amp (having apparently "killed" 2 Noolec devices with the sat line amp "solution earlier this year). I nearly killed myself in the process in fitting 18" T&K brackets just below the eves to hold that antenna and the discone (it meant standing on the pitched tiles on the top of the utility area end wall, about 10ft from the ground, to use an SDS drill to drill the anchor bolt holes whilst trying - successfully! - not to fall backwards to the ground or forwards through the polycarbonate roof of the utility area)
Edited: Now got the VRS server showing about 50-55 planes being tracked - but not yet got the splatter diagram working againn

the splat is done via the drop pin in virtual radar server.

 

[jeallen01]

Chris

Thanks but I can't remember how to use that approach - but what I have found is that the VRS "Initial Settings" "settings.html" are currently limited to just 2 - as below - as opposed to the long list that you showed in Post #96!