TNAP-7 Image Discussion

One of the reasons for adding the AGC and SNR(db) signal lines at the bottom of Signal finder was to monitor data transponders. Notice the green line is broken instead of traveling all the way across? Sure "Eye Candy", but also "Information".

SF8008-4020_20260825150513.webp
 
Notice the Edision is set to the 16APSK side of the NOAA transponder. The AGC and SNR lines across the bottom right are wavy and have small breaks in them. Granted, it is nice to have a decent looking screen, but these added widgets are put there to be used instead of just adding eye candy. Normally on healthy FTA satellite transponders, small breaks in the AGC line as shown below would point to a possible hardware failure.

Edision-NOAA-4020_20260825153214.webp
 
I've hesitated upgrading to TNAP-7 and I'm still running TNAP-6 on my Edision MIO 4k. Sometimes when I unplug it, it won't boot up and it stops on the Edision screen. Usually I unplug it and try again. Today it took two attempts. I'm wondering if I have some lingering corrupted file(s). Should I do a factory reset and install TNAP-7 or will I mess something up? Is there an easier way than a factory reset to wipe all the files? I appreciate any hints or suggestions.

Carl
 
There are 4 slots in the receiver for images. Put TNAP 7 into a different slot and see how it does. No factory default needed. Feeds no longer exist for TNAP 6. Upgrade to TNAP 7. Need help, Ask....
 
It should be explained what is being shown at the bottom of the screen in nwwsviewer.

Metrics and Status Information:
NWWS text bulletins (ISI 2, PID 201) [RUNNING 0:41:09 quiet 0:00:34 last product 0:00:34 ago]
tuner: slot 0 LOST LOCK @121.1W snr 14.31 dB (0.00-15.05) lock 16%/2473s flaps 186 retunes 0
dvb0_0 pid 201 -> 224.1.1.1:1201 | frames 2487 (0.0/s) ip 2487 kept 934 (68 partial) | 91 bulletin(s)
capturing - dvb0_0 (adapter 0, if 0, pid 201, reused via ioctl) | DATA CONFIRMED on 224.1.1.1:1201


1. NWWS text bulletins (ISI 2, PID 201)
This identifies the stream and type of data the viewer is processing.
NWWS text bulletins = The viewer is currently looking for National Weather Service text products carried on the NOAAPORT/NWWS stream.
ISI 2 = Input Stream Identifier 2 is selected. On DVB-S2/Multistream transmissions, the ISI identifies the particular logical stream being received.
PID 201 = MPEG Transport Stream Packet Identifier 201 is the PID from which the DVB network/IP data is being extracted.
PID 201 is therefore the transport-stream path feeding this particular NWWS data session.

2. [RUNNING 0:41:09]
The viewer/capture process has been running for 41 minutes, 9 seconds.
This is the elapsed operating time for the current monitoring session, not necessarily the amount of time the tuner has been locked.

3. quiet 0:00:34
The monitored data path has been quiet for 34 seconds.
This counter increases while no new usable/relevant NWWS data activity is being received.
A short quiet period is normal because weather products are not necessarily transmitted continuously. A rapidly increasing quiet counter while the tuner is also reporting LOST LOCK can instead indicate that reception has stopped because the RF signal disappeared.

4. last product 0:00:34 ago
The most recent successfully recognized NWWS product was received 34 seconds ago.
This is a product-level timer. It tells you how long it has been since the viewer last completed or recognized a bulletin/product.
Note:
In this screenshot, quiet and last product are both 34 seconds, which means the most recent useful activity and the most recent completed product occurred at approximately the same time.

Tuner / RF reception metrics
5. tuner: slot 0
The Enigma2 tuner being monitored is Tuner slot 0.
On a receiver with multiple physical tuners, this tells you which tuner frontend is supplying the stream.

6. LOST LOCK
The tuner is not currently locked to the satellite transponder. This is an instantaneous state. It does not mean the entire session failed.
The same screenshot shows:
(a) thousands of received frames,
(b) confirmed IP data,
(c) 91 decoded bulletins,
So the tuner had lock earlier in the session and successfully received data before the lock was lost.

7. @121.1W
The tuner is associated with the satellite orbital position 121.1° West.
This is the configured/identified orbital location for the NOAAPORT signal being monitored.

8. snr 14.31 dB
The displayed Signal-to-Noise Ratio is 14.31 dB.
SNR is the difference between the desired satellite signal level and the noise/interference level.
Higher SNR usually means more signal margin, while Lower SNR = less decoding margin.
The exact minimum SNR needed for reliable reception depends on modulation, FEC, symbol rate, receiver implementation, and signal conditions. For this particular transponder, SNR is fairly low. A decent 7.5 ft. or 8ft. dish should receive it. The SNR value should not be confused with raw signal strength or AGC.

9. (0.00-15.05)
This is the observed SNR range during the monitoring session:
Minimum observed: 0.00 dB, and Maximum observed: 15.05 dB
Note:
The 0.00 dB minimum can occur when the tuner loses lock or when the driver cannot return a meaningful SNR measurement during an unlocked state. It should therefore not automatically be interpreted as a valid locked-signal SNR reading. The useful figure here is that the tuner has reached approximately 15.05 dB at its best during this run.

10. lock 16%/2473s
This reports tuner lock availability over the monitored interval. Observed interval = 2473 seconds. Percent of that interval locked: 16%. 2473 Seconds is about 41 minutes, 13 seconds. At 16% lock availability, the receiver had usable tuner lock for roughly 396 seconds, or about 6 minutes 36 seconds during the measured interval. That is very poor lock continuity. It means the receiver spent about 84% of the monitored time unlocked. This is one of the most important diagnostics in the screenshot. How important it really is is currently unknown as our signal meter system is not designed for this.

11. flaps 186
A flap is a change in tuner lock state — for example, going from locked to unlocked or from unlocked back to locked. The counter shows 186 lock-state changes during this monitoring session. A high flap count means the signal/lock is unstable.

Important interpretation Note:
If each state transition is counted separately, one complete loss-and-recovery cycle can contribute two flaps.
Therefore, 186 flaps does not necessarily mean 186 separate outages.
It does show that the receiver has been repeatedly bouncing between lock and no-lock states.
Combined with only 16% lock availability, this indicates a very unstable RF reception condition. But it should be noted that we are using a FTA receiver, whose signal interface is designed for satellite tv and radio, Not for receiving packets.

12. retunes 0
The viewer has performed 0 automatic/manual retune recovery operations during this monitoring session.
This is useful for separating two different causes of interruptions such as lock instability originating from the tuner/signal. Or Lock changes caused by the application deliberately retuning. Because the value is zero here, the 186 flaps were not the result of viewer-triggered retune attempts.

DVB/IP capture metrics:
13. dvb0_0
dvb0_0 is the Linux DVB network interface being used by dvbnet. The name normally corresponds to DVB adapter 0, network interface number 0.
It is the virtual network interface through which MPE/IP data extracted from the satellite transport stream enters the Linux networking stack. (Note, there are two of these.)

14. pid 201
This confirms that dvb0_0 is listening to Transport Stream PID 201. That matches the PID shown in the viewer title/status line.
15. -> 224.1.1.1:1201 = The extracted traffic is being associated with the multicast endpoint, IP multicast address: 224.1.1.1

UDP port: 1201
This is the network destination being monitored for the NWWS data handled by this viewer.
The arrow means that PID 201 is ultimately yielding IP traffic for this multicast destination.

16. frames 2487
The capture layer has processed 2,487 DVB/MPE data frames during the session. A frame is a lower-level transport/network unit delivered from the DVB data path before the application reconstructs higher-level weather products. This counter shows that actual payload-bearing data has made it through the DVB capture path.

17. (0.0/s)
The current frame arrival rate is 0.0 frames per second. This is an instantaneous/recent rate, not the session average. Because the screenshot simultaneously says LOST LOCK, a current rate of 0.0/s is expected: no new frames are arriving while the tuner is unlocked. The total frame count remains at 2487 because those frames were received earlier.

18. ip 2487
The decoder has extracted 2,487 IP packets/datagrams from the received DVB data. In this screenshot the IP count equals the frame count of frames 2487 and ip 2487. That indicates that each counted DVB/MPE frame in this sample produced an IP packet recognized by the network decoder. This is good evidence that the DVB-to-IP conversion path is functioning.

19. kept 934
Of the extracted IP traffic, the viewer retained 934 packets/units for further NWWS processing. "Kept" means the traffic passed the viewer's relevance/filtering checks rather than being discarded as unrelated to the product stream being assembled. For this snapshot, 934 / 2487 ≈ 37.6% of the extracted IP packets were retained by the viewer for further processing. Not every packet on a multicast/data stream necessarily belongs to a complete NWWS text bulletin that the viewer wants to display.

20. (68 partial)
The viewer currently reports 68 partial/incomplete product assemblies. A partial item is data that has started to form a product or bulletin but is not yet complete enough to be delivered as a finished bulletin. Typical reasons include missing packets, reception interrupted by loss of tuner lock, joining a product after its beginning was already transmitted, packet loss during a lock flap, or a product still being assembled when the status was sampled. With the tuner showing only 16% lock and 186 flaps, incomplete products are not surprising. This metric is particularly useful when diagnosing whether an unstable RF path is causing fragmented or incomplete weather products.

21. 91 bulletin(s)
The viewer has successfully recognized/assembled 91 NWWS bulletins during the current monitoring session. This is the most important application-level success counter. It proves that the complete chain has worked at least intermittently.

Chain Diagram:
Satellite RF

DVB-S/S2 tuner

Transport Stream PID 201

dvbnet / dvb0_0

MPE/IP extraction

224.1.1.1:1201

NWWS parser

91 completed bulletins
So even though the tuner is currently reporting LOST LOCK, the data-reception and bulletin-decoding design itself is demonstrably working.

Capture-state line:
22. capturing - dvb0_0
The viewer's packet capture is currently enabled and attached to dvb0_0. This means the application has not stopped its network capture simply because tuner lock has temporarily disappeared. If the tuner regains lock and packets return to the interface, the application is already in position to receive them.

23. (adapter 0, if 0, pid 201, reused via ioctl)
This describes how the DVB network interface was obtained.

adapter 0
Linux DVB adapter number 0 is being used.

Typical device-tree relationship: /dev/dvb/adapter0/ ... if 0 DVB network interface number 0 is being used. That is why the interface is named dvb0_0.

pid 201
The interface is configured to extract PID 201.

reused via ioctl
The application found an existing DVB network interface already configured for the required PID and reused it instead of unnecessarily creating another interface. ioctl Refers to the Linux device-control calls used to query/control the DVB network device. This is normally desirable because it avoids creating duplicate dvbnet interfaces for the same PID.

24. DATA CONFIRMED on 224.1.1.1:1201
This is a strong positive verification message. It means the application has actually observed real data arriving for 224.1.1.1:1201. This is more meaningful than merely proving that the tuner exists, dvb0_0 exists, PID 201 was configured, or the interface is marked UP.

DATA CONFIRMED means payload traffic has actually made it through the DVB/IP chain to the expected multicast destination.

What this particular screenshot tells us is the NWWS viewer itself is working. Here is why:
frames 2487 DVB data frames were received.
ip 2487 those frames were successfully decoded into IP traffic.
DATA CONFIRMED on 224.1.1.1:1201 the expected multicast data was actually seen.
91 bulletin(s) real NWWS products were successfully decoded and displayed.

The main problem shown by the metrics is RF/tuner lock instability, not a completely broken DVB/IP pipeline.
The strongest evidence is:
LOST LOCK
lock 16%/2473s
flaps 186

Only about 16% of the monitoring period had tuner lock, and the lock state changed 186 times. That explains why
the current packet rate is 0.0/s, there are partial/incomplete products, and bulletin reception occurs intermittently rather than continuously.

Bottom-line diagnosis from this screenshot:
The NOAAPORT/NWWS receiving chain is proven to work, but the satellite lock is extremely unstable. To a degree though how unstable it really is not exactly known because our satellite signal system is not designed for this.

This plugin has been tested to work on Octagon sf8008 with Avalink tuner or demod, and Edision 4K receivers.

metrics_20260829201155.webp
 
Last edited:
The aiassistant plugin will make a debut in TNAP 7. This plugin was originally attempted in the latter part of 2025, and was then abandoned. It is one of the most difficult plugins to create as it is complex. There are free tiers of Gemini that will provide some AI use for free, and paid versions will also be accepted. The plugin is currently set for Gemini and Claude, but other models could be added.

One thing that may be of use here is writinig cron jobs that instruct the AI to start and do a task. The cron is a very powerful tool within itself when it comes to executing well written scripts. Pairing the cron capabilities that currently exist with a live AI could yield some interesting results that cost the user Nothing in money.

The aiassistant could also be a power troubleshooting tool that reads crash logs, explains what happened, then attempts repairs. It could also be useful for updating things automatically when the receiver is inactive, and possibly check for a satellite signal before a scheduled recording starts, and attempt to get a signal for the recording if none exists. There are lots of possibilities. What actually happens with this plugin will probably depend on user input or the lack thereof. It can always be put in the dumpster or retired.

Attached is a hand-off document which describes where the plugin is in development. It is not written for the general public per se, but it does have useful information and insights. The aiassistant is a very challenging plugin to produce, but the tooling and AI capabilities have improved since the end of 2025.

aiassistant_20260830155849.webp
 

Attachments

just clean flashed 8/27 version in osmio4k+, and tryong to make a backup, but "backup image" does not recongnize 16gb, 32gb Kingston usb !!!! only 64gb and/or 128gb are being recognize !?!?
They were all recognized in previouse versions !!!
 

Attachments

  • usb-backup.webp
    usb-backup.webp
    60.3 KB · Views: 2
To be clear, BOTH usb drives are recognized, but showing insufficient space.
Pick ONE usb drive and insert it into the receiver or remove one usb drive and leave the other. Post the output of these terminal commands:
lsusb
df -h
df -i
blkid
mount | grep /media
 
The aiassistant plugin will make a debut in TNAP 7. This plugin was originally attempted in the latter part of 2025, and was then abandoned. It is one of the most difficult plugins to create as it is complex. There are free tiers of Gemini that will provide some AI use for free, and paid versions will also be accepted. The plugin is currently set for Gemini and Claude, but other models could be added.

One thing that may be of use here is writinig cron jobs that instruct the AI to start and do a task. The cron is a very powerful tool within itself when it comes to executing well written scripts. Pairing the cron capabilities that currently exist with a live AI could yield some interesting results that cost the user Nothing in money.

The aiassistant could also be a power troubleshooting tool that reads crash logs, explains what happened, then attempts repairs. It could also be useful for updating things automatically when the receiver is inactive, and possibly check for a satellite signal before a scheduled recording starts, and attempt to get a signal for the recording if none exists. There are lots of possibilities. What actually happens with this plugin will probably depend on user input or the lack thereof. It can always be put in the dumpster or retired.

Attached is a hand-off document which describes where the plugin is in development. It is not written for the general public per se, but it does have useful information and insights. The aiassistant is a very challenging plugin to produce, but the tooling and AI capabilities have improved since the end of 2025.

View attachment 20197
Great idea.... when I try the plugin tomorrow, I hope AI won't get ticked off when it finds all the garbage on my receiver! 😁
 
The Edision DXer that likes to search for low symbol rates gets a treat in today's online update. First, The TNAP MOD driver got an update some days ago that allowed for more precise scanning and at the same time opened up symbol rates to around 300, possibly less with the right dish antenna setup. A professional PLL lnb and dish of at least 1.2 ku is suggested for the reception of small transponders in North America.

This effort consist of a modified driver, modified blindscan binary, and a tricked out blindscan. Only one driver has all the features, which is the TNAP MOD driver. When this driver is installed into the receiver, the blindscan plugin will allow a start symbol rate of 00 and then provide a couple more entries below that. All other drivers and all other receivers will have a start symbol rate of 1.

And it works! Shown below is the blindscan plugin set for low symbol rates plus the results of a blindscan. This is the only combination I know of in an enigma2 type receiver that allows the blindscan search for low symbol rate transponders. And then only One modded driver is capable.

Scan times may increase dramatically when searching for low symbol rate transponders.

Edision-low-symbol-rate-floor.webpEdision-low-symbol-rate-results.webp
 
Having the proper lnb and a generous size dish is almost a "must have" for low symbol rate reception. Interference, locking problems, and other issues can exist with low symbol rate transponders. Shown below is a 360 symbol rate radio channel playing on the Edision Mini4K. An Invacom DRO lnb and 1.2 meter dish are being used. While far from ideal, it does work to a degree.

WY-Sounds-360SR_20260831150052.webp
 
The setup process for the Aiassistant should not change. The models will be updated along with some other things. A corrected version will be available as time permits. A separate thread needs to be started for this plugin. It is a complex item that needs user input to make it better. There are some things that will be useful, but the design needs to be tweaked.

A question about the internal drive:
Code:
root@sf8008:~# aiassistant --debug How healthy is my hard drive? Does it need a trim?
  · thinking...
  · reference: storage
  · thinking...
  · $ lsblk -o NAME,SIZE,TYPE,FSTYPE,MOUNTPOINT,ROTA,MODEL
  · thinking...

  The assistant wants to run:
    cat /sys/block/sdd/queue/discard_granularity /sys/block/sdd/queue/discard_max_bytes /sys/block/sdd/queue/rotational; lsusb; df -h /media/hdd; tune2fs -l /dev/sdd1 | grep -E "Filesystem state|Mount count|Last check"
  Why: Checking drive properties and filesystem status
  Allow? [y/N] y
  · $ cat /sys/block/sdd/queue/discard_granularity /sys/block/sdd/queue/discard_max_bytes /sys/block/sdd/queue/rotational; lsusb; df -h /media/hdd; tune2fs -l /dev/sdd1 | grep -E "Filesystem state|Mount count|Last check"
  · thinking...

  The assistant wants to run:
    smartctl -a /dev/sdd; smartctl -d sat -a /dev/sdd
  Why: Checking if SMART is passed through the USB bridge
  Allow? [y/N] y
  · $ smartctl -a /dev/sdd; smartctl -d sat -a /dev/sdd
  · thinking...
  · $ dmesg | grep -iE "sdd|error|uas|reset|remount" | tail -n 30
  · thinking...

Your storage drive is a **1TB Kingston NVMe SSD (SNV2S1000G)** connected via a JMicron JMS583 USB bridge, mounted at `/media/hdd`.

### Health: Excellent
* **SMART Status:** Passed with **0% wear** (100% available spare remaining).
* **Errors:** 0 media/data integrity errors and 0 logged errors.
* **Temperature:** 38°C (normal operating range).
* **Usage:** 769 power-on hours, ~992 GB written total.
* **Filesystem:** Clean `ext4` filesystem with 892 GB (97%) free space and no I/O errors or resets in the kernel logs.

### TRIM: No
The drive does not need a trim, nor can it be trimmed. The USB bridge controller does not expose discard support to the kernel (`discard_max_bytes` is `0`), so `fstrim` is unsupported over this USB interface. Given the 0% wear level and substantial free space, lack of TRIM will not impact drive health or performance.

It needs some work, but it can be a useful item. And runs to a degree on a free tier. But the user does have to go through the process of key creation.
 
Back
Top